Thesis Title: Subtitle - UQ eSpace

Transcription

Thesis Title: Subtitle - UQ eSpace
Bioproduction and self-assembly of designer peptides
Nicholas Fletcher
Bachelor of Biotechnology (Honours)
A thesis submitted for the degree of Doctor of Philosophy at
The University of Queensland in 2014
Australian Institute for Bioengineering and Nanotechnology
Abstract
Self-assembly is a powerful method for producing controlled morphologies at the
nanometre scale. Peptides are biomolecules capable of self-assembly and have the
potential for use in a variety of applications such as emulsion and foam stabilisation,
wound healing and drug delivery. The investigation of peptide sequence-structure
relationships is a rapidly advancing field of study, which in many cases now allows
the targeted design of self-assembling peptides.
While self-assembling peptides show potential in several fields, their large-scale
adoption is limited by the high production expenses associated with conventional
solid-phase synthesis. A potentially cheaper and more renewable approach to
peptide production is bacterial expression. However, the bioproduction of peptides is
a non-trivial process, and generally involves the expression of peptides as part of a
fusion protein in which the target peptide is only a small portion of the expressed
product. An alternate approach involves peptide concatemers, in which the target
peptide makes up the majority of the expressed construct.
The work reported in this thesis focused on the design, characterisation and
bioproduction of self-assembling α-helical peptides. It was particularly focussed on
the development of amphiphillic α-helical peptides with applications as surfactants
and hydrogelators. The aim of this work was to further the field of de novo peptide
design, while also investigating an approach for peptide bioproduction. This work
aimed to combine the requirements for peptide bioproduction with end-use
functionality.
Chapter 2 details successful bioproduction of an anionic helical surfactant peptide
EDP-11, as part of a charge-paired heteroconcatemer with the cationic expression
partner RDP-4. The method utilised designed assembly of the constituent α-helical
peptides to generate a stably folded coiled-coil concatemer. The polypeptide
sequence was further optimized for molecular charge, hydropathy and predicted
protease resistance. This process allowed expression of a soluble concatemer that
accumulated to high levels (22% of total protein) in E. coli. The expressed
concatemer possessed extreme stability to heat and proteases, allowing isolation by
simple heat and pH precipitation, yielding concatemer at 130 mg per gram of dry cell
weight and >99% purity. Key parameters used in designing the heteroconcatemer
were then compared to those of all open reading frames of several reference
i
proteomes in an attempt to gain insight into the mechanistic basis for the high
stability of the designed miniprotein.
Following bioproduction using this concatemer approach, further processing was
required to produce monomeric peptides for use in surfactant applications. The
design of acid-cleavable aspartate-proline sites within the concatemer sequence
allowed for simple heat- and acid-mediated cleavage to give constituent peptides.
Chapter 3 details characterization of cleavage of the expressed concatemer by
coupled liquid chromatography-mass spectrometry and includes modelling of the
kinetic pathways involved. Chemical denaturation studies showed that cleavage
decreased the stability of the coiled coil from 38.9 to 32.8 kcal mol-1. Both intact and
cleaved concatemer possessed surfactant functionality, with each giving an
equilibrium interfacial pressure of 29 mN m-1 at the air-water interface. Both
concatemer cleavage and addition of guanidinium chloride to partially denature
coiled-coil structure resulted in enhanced rates of adsorption to the interface. The
cleaved products were also used to prepare heat-stable oil-in-water emulsions with
droplet sizes in the nanometre range.
Chapter 4 describes the design and characterisation of peptide AFD19, which was
designed as part of ongoing work on amphiphilic α-helical peptides. AFD19 selfassembled to form fibrils and hydrogels at weight fractions below 0.1%. Gelation
occurred in a pH-dependent manner, which was attributed to changes in molecular
charge. AFD19 gave free-flowing solutions at high molecular charge, gel formation
where the peptide charge was close to ±1, and precipitation when the charge
approached zero. Characterisation of AFD19 self-assembly indicated the peptide
assembled to give coiled-coil fibrils of approximately a hexamer in cross-section that
formed physical cross-links below a critical molecular charge. While AFD19
precipitated at close to neutral pH, redesign of the peptide sequence gave AFD36,
which underwent gelation at physiological pH and salt, increasing utility in the
biomaterial field. Small-angle X-ray scattering showed AFD36 to form fibrils of 3.83.9 nm diameter at pH 4.0-7.0. Coiled coils of both peptides possessed high
thermodynamic stability, with ΔG(H2O) values of 9.3-9.5 kcal mol-1 per monomer.
Chapter 5 details the investigation of α-helical peptide hydrogels as cell scaffolds
and therapeutic delivery vehicles. Rheological measurements of an AFD36 hydrogel
showed viscoelastic properties with an elastic modulus of 350 Pa. Further peptide
ii
design gave the sequence AFD49, and hydrogels formed by either this peptide or
AFD36 supported the growth of NIH/3T3 fibroblasts at levels similar to controls of
tissue culture polystyrene. AFD49 hydrogels also supported the proliferation of
encapsulated fibroblasts over several weeks, but were unable to support the growth
of induced pluripotent stem cells in the absence of coating with Matrigel. The ability
of AFD49 hydrogels to incorporate and release the hydrophobic drug all-trans
retinoic acid was also demonstrated.
iii
Declaration by author
This thesis is composed of my original work, and contains no material previously
published or written by another person except where due reference has been made
in the text. I have clearly stated the contribution by others to jointly-authored works
that I have included in my thesis.
I have clearly stated the contribution of others to my thesis as a whole, including
statistical assistance, survey design, data analysis, significant technical procedures,
professional editorial advice, and any other original research work used or reported
in my thesis. The content of my thesis is the result of work I have carried out since
the commencement of my research higher degree candidature and does not include
a substantial part of work that has been submitted to qualify for the award of any
other degree or diploma in any university or other tertiary institution. I have clearly
stated which parts of my thesis, if any, have been submitted to qualify for another
award.
I acknowledge that an electronic copy of my thesis must be lodged with the
University Library and, subject to the General Award Rules of The University of
Queensland, immediately made available for research and study in accordance with
the Copyright Act 1968.
I acknowledge that copyright of all material contained in my thesis resides with the
copyright holder(s) of that material. Where appropriate I have obtained copyright
permission from the copyright holder to reproduce material in this thesis.
iv
Publications during candidature
Journal articles
1. Fletcher, N. L.; Paquet, N.; Dickinson, E. L.; Dexter, A. F. Bioproduction of
Highly Charged Designer Peptide Surfactants via a Chemically Cleavable
Coiled-Coil Heteroconcatemer. Biotechnology and Bioengineering,2014,
DOI: 10.1002/bit.25446.
2. Fletcher, N. L.; Lockett, C. V.; Dexter, A. F. A pH-Responsive Coiled-Coil
Peptide Hydrogel. Soft Matter, 2011, 7, 10210-10218.
Conference abstracts
1. Fletcher, N. L.; Lockett, C. V.; Dexter, A. F. A pH-Responsive Coiled-Coil
Peptide Hydrogel. 9th Australian Peptide Conference, 2011, Poster
2. Fletcher, N. L.; Lockett, C. V.; Dexter, A. F. Supramolecular assembly of
amphiphillic peptides, RACI Student Conference, 2011, Presentation
3. Fletcher, N. L.; Dexter, A. Bioproduction of designer peptide surfactants, RACI
Student Conference, 2012, Presentation
4. Fletcher, N. L.; Dexter, A. Bioproduction of designer peptide surfactants,
BioNano Innovation Conference, 2012, Presentation
5. Fletcher, N. L.; Lockett, C. V.; Jack, K.; Dexter, A. F. Self-assembling
designer peptides for biomaterial applications, RACI Student Conference,
2013, Presentation
6. Fletcher, N. L.; Lockett, C. V.; Boehm, M.; Jack, K.; Dexter, A. F. Selfassembling designer peptides for biomaterial applications, NanoBio Australia
Conference, 2014, Poster
v
Publications included in this thesis
1. Portions of the work presented in Chapters 2 and 3 have been accepted for
publication;
Fletcher, N. L.; Paquet, N.; Dickinson, E. L.; Dexter, A. F. Bioproduction of Highly
Charged Designer Peptide Surfactants via a Chemically Cleavable Coiled-Coil
Heteroconcatemer. Biotechnology and Bioengineering, DOI: 10.1002/bit.25446
My contributions to this article were:
•
Optimisation of downstream processing of the concatemer;
•
Optimisation of the concatemer cleavage method;
•
Monitoring of the cleavage process and kinetic modelling of observed species;
•
Characterisation of properties of intact and cleaved concatemer by
thermodynamic modelling of chemical denaturation data and interfacial
tension measurements;
•
Preparation of the submitted manuscript in conjunction with Dexter.
Dexter was responsible for peptide and concatemer design as well as a supervisory
role that included experimental direction, assistance in kinetic model refinement and
aiding in manuscript preparation. Shake flask expression was done by Paquet while
Dickinson assisted in preliminary concatemer cleavage characterisation via liquid
chromatography-mass spectrometry.
Contributor
Fletcher, N. (Candidate)
Statement of contribution
Designed experiments (70%)
Wrote the paper (70%)
Paquet, N.
Conducted shake flask expression
Dickinson, E.
Assisted in preliminary concatemer
cleavage characterisation
Dexter, A
Designed experiments (30%)
Wrote the paper (30%)
vi
2. Portions of the work presented in Chapter 4 have been accepted for
publication:
Fletcher, N. L.; Lockett, C. V.; Dexter, A. F. A pH-responsive coiled-coil peptide
hydrogel. Soft Matter 2011, 7, 10210-10218.
My contributions to this article were:
•
Initial discovery and characterisation of gelling behaviour;
•
Preparation and imaging of electron microscopy samples;
•
Co-supervising and assisting the second author Lockett in her work.
Lockett conducted studies of AFD19 self-assembly via observation of changes in
circular dichroism spectra during dilution and thermal denaturation of the peptide.
Dexter was responsible for peptide design, dynamic light scattering measurements
and the majority of manuscript preparation.
Contributor
Fletcher, N. (Candidate)
Statement of contribution
Electron microscopy sample preparation
and imaging
Designed experiments (30%)
Wrote the paper (10%)
Lockett, C
Collected circular dichroism data under
my supervision
Dexter, A
Dynamic light scattering measurements
Designed experiments (70%)
Wrote the paper (90%)
vii
3. Portions of the work presented in Chapters 4 and 5 are being prepared for
publication:
Dexter, A. F.; Fletcher, N. F.; Boehm, M.; Filardo, F.; Jack, K. S.; A designed αhelical peptide gel for mammalian cell growth.
My contributions to this article were:
•
Small-angle X-ray scattering experiments and data interpretation;
•
Tissue culture experiments characterising 3T3 fibroblast growth on hydrogels;
•
Assisting in rheological data collection and data analysis;
•
Shared experimental design and manuscript preparation with Dexter.
Jack trained me in the use of the scattering instrument and assisted in interpretation
of X-ray scattering experiments. Boehm conducted and assisted in analysing
rheology experiments and Filardo gave guidance on tissue culture experimentation
as well as conducting initial tissue culture trials. Dexter was responsible for peptide
design, development of bicarbonate buffering system, collection of data for circular
dichroism studies of heat and chemical peptide denaturation and shared manuscript
preparation.
Contributor
Dexter, A
Statement of contribution
Designed experiments (50%)
Wrote the paper (60%)
Fletcher, N. (Candidate)
Designed experiments (50%)
Wrote the paper (40%)
Boehm, M
Conducted and assisted in analysing
rheology experiments
Filardo, F
Guidance on tissue culture
experimentation
Jack, K
Assisting in interpretation of X-ray
scattering results
viii
Contributions by others to the thesis
In addition to the contributions to publications listed above, the Author acknowledges
the following individuals who have contributed to this thesis:
Dr Annette Dexter for contributing to the conception and design of the project as well
as assisting in the analysis and interpretation of the research detailed in this thesis.
Biopython code for the analysis of polypeptide sequences as described in Chapter 2
was written by Dr Stefan Maetschke.
Statement of parts of the thesis submitted to qualify for the
award of another degree
None
ix
Acknowledgements
First and foremost I would like to thank my supervisors Dr Annette Dexter, Dr Kevin
Jack and Prof Andrew Whittaker. Thank you for all your guidance, support and
inspiration over the past few years, which has greatly enhanced by PhD experience.
I would also like to thank the University of Queensland and the Australian Institute for
Bioengineering and Nanotechnology for providing both the facilities to conduct this
body of research and financial support throughout my studies. I would also like to
extend a special thanks to Tony Miscamble for his continuous support and
assistance throughout both the good and the bad times, going above and beyond his
call of duty.
A big thank you to the entire Whittaker Group, past and present, who have been my
lab mates, friends and colleagues throughout the past three and a half years. To all
my friends who have been there during my PhD, helped me maintain my sanity and
assisted with scientific knowledge over the past few years; (in no particular order)
Nathan Boase, Mimi Chuang, Samuel Richardson, Thomas Bennet, Amanda
Pearce, Lewis Chambers, Lauren Butler, Oliver Squires, Anna Gemmell, Adrian
Fuchs, Simon Puttick, Johanna Coyle, Paul Luckman, Will Anderson, Anna
Cifuentes-Rius, Mirjana Dimitrijev, Anneke Dorgelo, Vinh Truong, Craig Bell, Kirsten
Lawrie, Emma West and Cathaye Robertson.
I am also indebted to Idriss Blakey and Kristofer Thurecht for their ongoing advice
and support, in matters both scientific and outside of the laboratory.
I would also like to thank my family for their unfailing support right from the start. I
would not be the person I am today without you.
Most of all, I want to thank my wife Jodie, who has been an endless source of
patience, friendship and love. She has been unceasingly understanding and
supportive throughout my PhD and without her I would not have been able to get to
this point, thank you.
x
Keywords
Peptide, alpha helix, concatemer, coiled coil, de novo design, hydrogel, selfassembly, fibril, tissue culture
Australian and New Zealand Standard Research
Classifications (ANZSRC)
ANZSRC code: 060113, Synthetic Biology, 30%
ANZSRC code: 030406, Proteins and Peptides, 40%
ANZSRC code: 030402, Biomolecular Modelling and Design, 30%
Fields of Research (FoR) Classification
FoR code: 0601, Biochemistry and Cell Biology, 30%
FoR code: 0304, Medicinal and Biomolecular Chemistry, 30%
FoR code: 0903, Biomedical Engineering, 40%
xi
Table of contents
Abstract ....................................................................................................................... i
Declaration by author .................................................................................................iv
Publications during candidature ................................................................................. v
Publications included in this thesis .............................................................................vi
Contributions by others to the thesis ..........................................................................ix
Statement of parts of the thesis submitted to qualify for the award of another degree
ix
Acknowledgements .................................................................................................... x
Keywords ...................................................................................................................xi
Australian and New Zealand Standard Research Classifications (ANZSRC).............xi
Fields of Research (FoR) Classification .....................................................................xi
Table of contents ....................................................................................................... xii
List of figures ........................................................................................................... xvii
List of tables ........................................................................................................... xxiv
Symbols and abbreviations ..................................................................................... xxv
Amino acid abbreviations ....................................................................................... xxix
1
Literature Review ................................................................................................ 1
1.1
Molecular self-assembly ................................................................................ 1
1.2
Peptide self-assembly ................................................................................... 3
1.2.1
Peptide amphiphiles ............................................................................... 3
1.2.2
β-sheet peptides ..................................................................................... 7
1.2.3
α-Helical peptides ................................................................................. 14
1.3
Applications of peptide self-assembly ......................................................... 23
1.3.1
Surfactants ........................................................................................... 23
1.3.2
Cell scaffolds ........................................................................................ 25
1.3.3
Therapeutic delivery ............................................................................. 33
1.4
Peptide production ...................................................................................... 36
1.4.1
Chemical synthesis ............................................................................... 36
1.4.2
Biological expression ............................................................................ 37
1.5
Research aims and thesis overview ............................................................ 44
1.6
References .................................................................................................. 48
xii
2
Bioproduction of Highly Charged Designer Peptide Surfactants: Expression and
Chromatography-Free Purification of a Coiled-Coil Heteroconcatemer ................... 68
2.1
Introduction ................................................................................................. 68
2.2
Experimental ............................................................................................... 69
2.2.1
Materials ............................................................................................... 69
2.2.2
Peptide charge predictions ................................................................... 70
2.2.3
Cloning and expression of concatemers ............................................... 70
2.2.4
Downstream processing ....................................................................... 70
2.2.5
SDS-PAGE ........................................................................................... 71
2.2.6
Mass spectrometry ............................................................................... 71
2.2.7
Preparation of uninduced E. coli extract ............................................... 71
2.2.8
Electronic circular dichroism ................................................................. 72
2.2.9
Bioinformatic analysis ........................................................................... 73
2.3
3
Results and Discussion ............................................................................... 73
2.3.1
De novo sequence design .................................................................... 73
2.3.2
Bacterial expression ............................................................................. 79
2.3.3
Purification of Het2-6 heteroconcatemer .............................................. 84
2.3.4
Protein structure ................................................................................... 87
2.3.5
Thermodynamic stability ....................................................................... 88
2.3.6
Bioinformatics ....................................................................................... 93
2.4
Conclusions................................................................................................. 97
2.5
References .................................................................................................. 99
Bioproduction of Highly Charged Designer Peptide Surfactants: Chemical
Cleavage of a Parent Heteroconcatemer and Analysis of Kinetic Pathways .......... 103
3.1
Introduction ............................................................................................... 103
3.2
Experimental ............................................................................................. 103
3.2.1
Materials ............................................................................................. 103
3.2.2
Heteroconcatemer cleavage and kinetic modelling ............................ 104
3.2.3
Peptide charge predictions ................................................................. 105
3.2.4
Electronic circular dichroism ............................................................... 105
3.2.5
SDS-PAGE ......................................................................................... 105
3.2.6
Interfacial tension ............................................................................... 105
3.2.7
Emulsion preparation and characterisation......................................... 106
xiii
3.3
3.3.1
Acid hydrolysis of Het2-6 .................................................................... 106
3.3.2
Product characterisation ..................................................................... 115
3.3.3
Secondary structure ........................................................................... 116
3.3.4
Thermodynamic stability ..................................................................... 117
3.3.5
Gel electrophoresis ............................................................................. 120
3.3.6
Surfactant function .............................................................................. 122
3.4
Conclusions............................................................................................... 124
3.5
Future directions ....................................................................................... 125
3.5.1
Design parameters ............................................................................. 125
3.5.2
Bioproduction...................................................................................... 126
3.5.3
Het2-6 redesign .................................................................................. 127
3.6
4
Results and Discussion ............................................................................. 106
References ................................................................................................ 130
Characterisation of a pH-responsive coiled-coil peptide hydrogel ................... 132
4.1
Introduction ............................................................................................... 132
4.2
Experimental ............................................................................................. 133
4.2.1
Materials ............................................................................................. 133
4.2.2
Electronic circular dichroism ............................................................... 134
4.2.3
Peptide charge predictions ................................................................. 134
4.2.4
AFD19 gel formation ........................................................................... 134
4.2.5
Electron microscopy ........................................................................... 134
4.2.6
Plate reader assays ............................................................................ 135
4.2.7
Small-angle X-ray scattering............................................................... 136
4.2.8
AFD36 gel melting .............................................................................. 136
4.3
Results and Discussion ............................................................................. 137
4.3.1
Peptide Design ................................................................................... 137
4.3.2
Peptide characterisation ..................................................................... 138
4.3.3
AFD19 secondary structure ................................................................ 138
4.3.4
Characterisation of coiled-coil assembly............................................. 141
4.3.5
AFD19 gelling as a function of charge ................................................ 145
4.3.6
Fibril formation .................................................................................... 146
4.3.7
Thermal stability ................................................................................. 148
4.3.8
Mechanism of gelation ........................................................................ 149
xiv
4.3.9
5
Peptide re-design for physiological gelling.......................................... 154
4.3.10
Self-assembly and gel formation ..................................................... 155
4.3.11
Investigation of fibril structure .......................................................... 158
4.3.12
Bicarbonate controlled pH adjustment for gelation .......................... 161
4.3.13
Thermal denaturation of AFD36 ...................................................... 163
4.3.14
Guanidinium denaturation of AFD36 ............................................... 165
4.4
Conclusions............................................................................................... 170
4.5
References ................................................................................................ 171
α-Helical peptide hydrogels for mammalian cell growth and drug delivery ...... 177
5.1
Introduction ............................................................................................... 177
5.2
Experimental ............................................................................................. 178
5.2.1
Materials ............................................................................................. 178
5.2.2
Gel rheology ....................................................................................... 178
5.2.3
Ultraviolet-visible absorption spectroscopy ......................................... 179
5.2.4
Electronic circular dichroism ............................................................... 179
5.2.5
Peptide charge predictions ................................................................. 179
5.2.6
AFD36 treatment for cell culture ......................................................... 179
5.2.7
AFD49 treatment for cell culture ......................................................... 179
5.2.8
Protease digestion of peptide hydrogel .............................................. 180
5.2.9
Fibroblast cell culture .......................................................................... 180
5.2.10
Cytotoxicity of free peptide assay .................................................... 181
5.2.11
Induced pluripotent stem cell culture ............................................... 181
5.2.12
Recombinant vitronectin fragment ................................................... 182
5.2.13
Microscopy ...................................................................................... 182
5.2.14
All-trans retinoic acid analysis ......................................................... 182
5.3
Results and Discussion ............................................................................. 183
5.3.1
Mechanical characterisation of an AFD36 hydrogel ........................... 183
5.3.2
Chemical contaminants ...................................................................... 188
5.3.3
Cell Growth on AFD36 ........................................................................ 190
5.3.4
Long term growth on AFD36............................................................... 192
5.3.5
Trypsin digestion ................................................................................ 195
5.3.6
Peptide design .................................................................................... 197
5.3.7
AFD49 characterisation ...................................................................... 198
xv
5.3.8
Cytoxicity of peptide released from hydrogels .................................... 202
5.3.9
Effect of length of equilibration time for AFD49 gels ........................... 202
5.3.10
Presence of serum in pre-soaking media ........................................ 204
5.3.11
Fibroblast attachment without serum............................................... 206
5.3.12
Fibroblast attachment in the presence of free RGD ........................ 209
5.3.13
Effects of hydrogel pre-equilibration on fibroblast attachment and
proliferation ..................................................................................................... 210
5.3.14
Culture of encapsulated fibroblasts ................................................. 211
5.3.15
Induced pluripotent stem cell growth on AFD49 gels ...................... 218
5.3.16
iPSC growth on Matrigel coated hydrogels ..................................... 223
5.3.17
Drug loading into peptide hydrogels ................................................ 225
5.3.18
ATRA release from peptide hydrogels............................................. 228
5.4
Conclusions............................................................................................... 229
5.5
Future directions ....................................................................................... 231
5.5.1
Characterisation of fibril forming peptides........................................... 231
5.1.1
Hydrogel forming peptides as cell scaffold materials .......................... 231
5.5.2
Hydrogel forming peptides for therapeutic delivery............................. 233
5.5.3
Bioproduction of hydrogel forming peptides........................................ 233
5.6
References ................................................................................................ 236
xvi
List of figures
Figure 1-1. Classes of self-assembling peptides. A) Peptide amphiphiles, B) β-sheet
forming peptides, C) α-helix forming peptides. Dashed bonds indicate peptide
backbone hydrogen bonding pattern. Purple shading highlights secondary structures
of β-sheet and α-helix. The dashed box in A and C highlight hydrophobic regions of
the peptides. In B the hydrophobic region extends out of the image towards the
viewer. Atoms are coloured according to element; carbon (grey), oxygen (red),
nitrogen (blue) and hydrogen (white).......................................................................... 2
Figure 1-2. Example chemical structures of A) an alkyl-tail PA and B) a nanofibre
forming alkyl-tail PA.................................................................................................... 6
Figure 1-3. Schematic representation of amphipathic β-sheet and bilayer assembly.
A) β-strand, B) β-sheet in which hydrogen bonding occurs parallel with the plane of
the page (as in Figure 1-1B), C) β-sheet bilayer. Hydrophobic and hydrophilic
residue side chains are represented as red and blue spheres respectively. For
simplicity the polypeptide backbone is represented as a line. .................................... 8
Figure 1-4. Representations of periodicity of α-helices in A) isolated and B) coiledcoil arrangements, and C) schematic illustrating interactions between helices in a
parallel dimeric coiled coil. ....................................................................................... 16
Figure 1-5. Schematic showing three design approaches for coiled-coil fibril
formation: A) forced offsetting of helices by placement of hydrogen bonding residues
within the hydrophobic core and asymmetric charged residue distribution,27, 28, 31, 165,
166, 168-173
B) end-to-end stacking of coiled coils with offset helices,174 C) forced
offsetting of helices by non-continuous hydrophobic faces of helices.29, 175 Hydrogen
bonds within the hydrophobic core are shown as lock-and-key arrangements;
horizontal lines within each peptide mark the ends of heptads. ............................... 20
Figure 2-1. Heteroconcatemer design. A) Helical wheel diagram showing amino acid
positioning in an 18-mer peptide designed to associate with a hydrophobic interface,
where boxed positions indicate the hydrophobic face of the peptide, B) Helical wheel
diagrams for EDP-11 and RDP-4 amphipathic core sequences, C) Design of a
tetrameric heteroconcatemer comprising EDP-11 (white) and RDP-4 (grey) helices
connected by unstructured linkers, D) Heptad-based projection of core helices
corresponding to C), based on EDP-11 (top left and bottom right) and RDP-4 (top
right and bottom left). ............................................................................................... 78
xvii
Figure 2-2. Expression testing of DNA constructs coding for concatemers Het2-5
and Het2-6. M is molecular weight marker. T is total protein and S is soluble protein
at each time point. Red box highlights over-expressed protein. ............................... 81
Figure 2-3. Expression testing of DNA construct coding for concatemer Het2-7. M is
molecular weight marker. T is total protein and S is soluble protein at each time
point. ........................................................................................................................ 82
Figure 2-4. Overexpression of Het2-6 in E. coli. SDS-PAGE gel showing Het2-6
expression in shake flasks. M is molecular weight marker. T is total protein and S is
soluble protein at each time point. ............................................................................ 83
Figure 2-5. SDS-PAGE analysis of Het2-6 fractions across purification steps. Lanes
show: 1), 10) molecular weight marker, 2) total lysate, 3) clarified lysate, 4) heattreated lysate, 5) pH 5.0 supernatant, 6) pH 5.0 pellet, 7) MgSO4-treated
supernatant, 8) second pH 5.0 supernatant, 9) second pH 5.0 pellet. ..................... 84
Figure 2-6. ECD spectra of Het2-6 (100 µM) at pH 3.0 (solid line) and pH 8.0
(dashed). .................................................................................................................. 88
Figure 2-7. ECD analysis of intact Het2-6 and E. coli soluble protein chemical
denaturation at pH 8.0. Het2-6 samples (●) were monitored at 90 °C and 222 nm,
while E. coli KC6 soluble protein (○) was monitored at 20 °C and 230 nm. Fits show
a three-state unfolding transition for Het2-6 and non-linear regression for E. coli
protein. ..................................................................................................................... 89
Figure 2-8. Schematic of proposed unfolding pathway of Het2-6. N fully folded Het26, I, unfolding intermediate with high helical content, U, fully unfolded Het2-6.
Cylinders indicate helical structure. Constituent peptides coloured as white (EDP-11)
or grey (RDP-4). ....................................................................................................... 92
Figure 2-9. Calculated bioinformatic parameters for selected organisms and random
polypeptide sequences. A) pI, B) GRAVY and C) instability index values for ORFs of
Escherichia coli, Thermus thermophilus, Pyrococcus furiosus and Colwellia
psychrerythraea. Solid line shows calculated parameters for random protein
sequences. Dashed vertical lines show the corresponding values for Het2-6. ......... 94
Figure 3-1. Liquid chromatography elution profiles showing products of Het2-6
cleavage after 0, 6 and 24 h at 70 °C. Peak positions for authentic standard peptides
are indicated. .......................................................................................................... 107
xviii
Figure 3-2. Sequential acid hydrolysis pathways A) C-terminal to aspartate at a DP
site and B) N-terminal to aspartate, leading to loss of the C-terminal aspartate
residue. .................................................................................................................. 111
Figure 3-3. Proposed minimal cleavage pathway for Het2-6. M represents the Nterminal MD dipeptide, E is EDP-11, L is linker and R is RDP-4, while E- represents
EP-11 and R- represents RP-4. Line thicknesses are approximately proportional to
the magnitude of first-order rate constants derived from modelling. ....................... 113
Figure 3-4. Time course of Het2-6 cleavage. Separate panels show species present
at A) high, B) intermediate and C) low concentrations. Symbols indicate species
concentrations determined by LCMS, while solid lines indicate data from the model
fit. (M)-E-L-R-L-E represents a combination of the species M-E-L-R-L-E and E-L-RL-E due to the inability to separately quantify these species (as discussed in text).
............................................................................................................................... 114
Figure 3-5. Effect of aspartate loss on molecular charge. Predicted molecular charge
for 2:2 complexes of EDP-11/RDP-4 (solid line) and EP-11/RP-4 (dashed) .......... 115
Figure 3-6. Electronic circular dichroism spectra of cleaved Het2-6 at pH 3 (solid)
and pH 8 (dashed).................................................................................................. 117
Figure 3-7. G.HCl-dependent denaturation of intact and cleaved Het2-6.
Denaturation of intact (●) and cleaved Het2-6 (○) was carried out at 90 °C and pH 8.
The fits are to a three-state unfolding transition for intact Het2-6 (reproduced from
Chapter 2) and disassembly of a tetrameric coiled coil for cleaved Het2-6. ........... 119
Figure 3-8. Gel electrophoresis of Het2-6 and cleavage products. A) Coomassie
blue R-250 stain; B) silver stained gel after partial destaining and C) lanthanum
chloride stain. Lanes are: 1, MW standard; 2, intact Het2-6; 3, cleaved Het2-6; 4,
EDP-11 and 5, RDP-4. ........................................................................................... 121
Figure 3-9. Interfacial activity of Het2-6 (red line), cleaved Het2-6 (black line), BSA
(green line) and AM1 (blue line). ............................................................................ 123
Figure 4-1. Helical wheel representations of; A) canonical 18 position helical wheel
representation of a facially amphiphilic α-helix associated with a hydrophobic
interface (boxed); B) the peptide AFD19 originally designed as a peptide surfactant
with expanded hydrophobic face (boxed). .............................................................. 137
Figure 4-2. ECD spectra of self-buffered AFD19 (1.75 mM) at 20 °C in gelled (solid
line, pH 5.9) and ungelled (dashed, pH 3.4) states. ............................................... 139
xix
Figure 4-3. Heptad-based helical wheel projection of peptide AFD19. Positions a, d,
e and g are occupied by non-polar residues. Positions b, c and f are occupied by
hydrophilic residues with self-similar heptad repeats. ............................................ 140
Figure 4-4. Electronic circular dichroism spectroscopy study of the self-assembly of
AFD19 into α-helical structures as a function of peptide concentration at pH values of
2.0 (○), 3.0 (▲), 4.0 (□) and 5.0 (▼). Lines are a fitted two-state dissociation model
described in text. .................................................................................................... 141
Figure 4-5. Linear least-squares fits of AFD19 assembly using Equation 4-5 at pH
values of 2.0 (○), 3.0 (▲), 4.0 (□) and 5.0 (▼). ...................................................... 143
Figure 4-6. Predicted per peptide molecular charge of AFD19 and observed phase
changes in self-buffered AFD19 (1.2 mM; 0.3% (w/v)) solutions as a function of pH.
............................................................................................................................... 146
Figure 4-7. A-E) Negative staining TEM images of 0.2 mg mL-1 AFD19 at pH 3.0,
4.0, 5.0, 6.0 and 7.0 respectively (scale bars 100 nm); F) Cryo-TEM image of 1.0 mg
mL-1 AFD19 gel at pH 6 (scale bar 50 nm). ............................................................ 147
Figure 4-8. Secondary structure of AFD19 as assessed at 222 nm using ECD as a
function of temperature during heating (solid line) and cooling (dashed) at different
pH values. The concentration of peptide was 200 µM and the buffer ionic strength
was maintained at 0.01. Solution pH as indicated by colour. ................................. 149
Figure 4-9. Proposed scheme for AFD19 self-assembly into fibrils that form
physically cross-linked hydrogels. Stage a involves the assembly of monomeric
peptides into staggered coiled-coil fibrils. Stage b then involves the formation of a
three-dimensional network of fibrils where association is controlled by peptide charge
as a function of pH. Horizontal lines within each peptide mark the ends of heptads.
............................................................................................................................... 150
Figure 4-10. Comparison of heptad based helical wheel projections of peptides
AFD19 and AFD36 with the serine-lysine substitution highlighted in red. .............. 155
Figure 4-11. Predicted charge per peptide of AFD19 (dotted) and AFD36 (solid line)
as a function of pH. ................................................................................................ 156
Figure 4-12. ECD spectra of self-buffered AFD36 (1.6 mM) at 20 °C in gelled (solid
line, pH 7.0) and non-gelled (dashed, pH 3.1) states. ............................................ 157
Figure 4-13. Small-angle X-ray scattering by AFD36 as a function of pH. Data were
collected for 0.5% (w/v) peptide at pH 4.0 (○), 5.0 (□), 6.0 (◇) and 7.0 (△). Solid
xx
lines show fits to a flexible-cylinder model. Data sets above pH 4.0 are displaced
vertically progressively by a factor of ten for clarity. ............................................... 159
Figure 4-14. pH changes over time for AFD36 gels prepared on titration with
different concentrations of bicarbonate. Solid line, 20.0 mM NaHCO3; dotted line,
22.5 mM NaHCO3; dashed line, 25 mM NaHCO3 (Final concentrations). Inset: local
gelling and pH inhomogeneity on addition of NaOH to acidic solution of AFD36
containing phenol red. The peptide concentration in each case was 4.2 mM. ....... 161
Figure 4-15. Thermal stability of secondary structure for gelled AFD36. Thick line,
heating curve; thin line, cooling curve. Inset: Determination of gel melting
temperature via a ball drop test with photos taken at 30 °C (upper left) and 80 °C
(lower right). Arrow indicates melting temperature. The peptide concentration was
4.2 mM in the gel. ................................................................................................... 164
Figure 4-16. Guanidinium chloride denaturation of AFD19 (○) and AFD36 (▲) at 20
°C. Lines represent fits to the data obtained as described in text........................... 166
Figure 5-1. Rheological characterisation of a 1.4 mM (0.35 % (w/v)) AFD36 hydrogel
A) over time during gelation (x-axis is time from measurement initialization), and B)
as a function of frequency (two samples). In both plots G′; triangles, G″; squares. 185
Figure 5-2. Absorbance spectra of AFD36 solutions at steps during purification; asreceived peptide (solid line), peptide following urea and charcoal treatment (dotted),
dialysed product (dashed). Spectra normalised to 1 mM peptide. All spectra collected
in 10 mm path length cuvette. ................................................................................ 189
Figure 5-3. Growth of NIH/3T3 fibroblasts on AFD36 gels and TCPS controls.
Representative images of cells cultured on A) TCPS control surface or B) AFD36
hydrogel at 3 days (scale bars 100 µm). C) Cell growth measured via AlamarBlue
fluorescence assay on TCPS (black bars) and AFD36 hydrogels (grey bars) at 1-4
days ±SD (n=3).
*p
values ≤ 0.05, relative to the control, as determined by Student’s T-test. Peptide gels
were 1.4 mM AFD36 final.
191
Figure 5-4. Representative images of NIH/3T3 fibroblasts grown on AFD36
hydrogels and TCPS controls. Days cultured as indicated (scale bars 100 µm).
Peptide gels were 1.4 mM AFD36 final. ................................................................. 193
Figure 5-5. Live-dead staining of NIH/3T3 fibroblasts grown for 21 days on A) TCPS
or B-D) AFD36 hydrogels (scale bars 100 µm). ..................................................... 194
xxi
Figure 5-6. Digestion of a 0.5 mM (0.12% (w/v)) AFD36 gel by trypsin. Helical
secondary structure monitored via ECD at 222 nm. ............................................... 196
Figure 5-7. Comparison of heptad based helical wheel projections of peptides
AFD36 and AFD49. Sequence changes are highlighted in red. ............................. 197
Figure 5-8. ECD spectra of self-buffered AFD49 (1.6 mM) at 20 °C in gelled (solid
line, pH 7.0) and non-gelled (dashed, pH 2.8) states. ............................................ 199
Figure 5-9. Absorbance spectra of AFD49 stock solutions before (solid line) and
after dialysis using Method 1 (dotted) and Method 2 (dashed). Spectra normalized to
1 mM peptide and 10 mm path length. ................................................................... 200
Figure 5-10. Growth of NIH/3T3 fibroblasts on AFD49 gels and TCPS controls. A)
Cell growth measured via AlamarBlue fluorescence assay on TCPS and AFD49
hydrogels at 3 days ±SD (n=3). Representative images of cells cultured on B) TCPS;
or AFD49 hydrogels prepared from peptide dialysed using either C) Method 1 or D)
Method 2 (discussed in text) (scale bars 100 µm). Peptide gels were 3.75 mM
AFD49 final. ........................................................................................................... 201
Figure 5-11. NIH/3T3 fibroblast growth on TCPS and AFD49 gels equilibrated in cell
culture media for varying times as indicated, measured via AlamarBlue fluorescence
assay at 3 days ±SD (n=3). Peptide gels were 3.75 mM AFD49 final. * p values ≤
0.05, as determined by Student’s T-test. ................................................................ 203
Figure 5-12. Growth of NIH/3T3 fibroblasts on AFD49 gels and TCPS controls using
different equilibration conditions. Representative images of cells cultured for 3 days
on A) AFD49 hydrogels or B) TCPS control surfaces pre-equilibrated with media as
indicated (scale bars 100 µm). C) Cell growth measured via AlamarBlue
fluorescence assay on AFD49 hydrogels (black bars) and TCPS (grey bars) at 3
days ±SD (n=3). Peptide gels were 3.75 mM AFD49 final. * p values ≤ 0.05, relative
to samples of same substrate material, as determined by Student’s T-test. .......... 205
Figure 5-13. The effects of serum on spreading of NIH/3T3 fibroblasts on TCPS or
AFD49 surfaces. Representative images of cells after 24 hours when they were
added in A) serum-containing media, or B) serum-free media to surfaces preequilibrated as indicated. (scale bars 200 µm). Peptide gels were 1 mM AFD49 final.
............................................................................................................................... 208
Figure 5-14. Effect of free cyclic-RGD on NIH/3T3 fibroblast adhesion to TCPS and
AFD49. Fibroblasts incubated on TCPS in A) culture media, B) 100 µM cyclic-RGD
xxii
in culture media, or on AFD49 in C) culture media, D) 100 µM cyclic-RGD in culture
media. (scale bars 100 µm). Peptide gels were 2 mM AFD49 final. ....................... 210
Figure 5-15. Effects of hydrogel encapsulation on NIH/3T3 fibroblast growth.
NIH/3T3 fibroblasts cultured A) on TCPS, B) on TCPS with AFD49 hydrogel cast on
top, C) on AFD49 hydrogel, D) on AFD49 hydrogel with second gel cast on top.
(scale bars 100 µm) Peptide concentration was 2.1 mM in the base gels and 2 mM in
the top layered gels. ............................................................................................... 213
Figure 5-16. Confocal microscopy of live-dead staining of NIH/3T3 fibroblasts in
layered AFD49 hydrogels at day 7. Images A) top-down, B) and C) side-on. (green,
live; red, dead). Peptide concentration was 2.1 mM in the base gels and 2 mM in the
top layered gels. ..................................................................................................... 215
Figure 5-17. Representative images of NIH/3T3 fibroblast growth encapsulated in
AFD49 hydrogels at longer times. (scale bars 100 µm). Peptide concentration was
2.1 mM in the base gels and 2 mM in the top layered gels. ................................... 216
Figure 5-18. Confocal microscopy of live-dead staining of NIH/3T3 fibroblasts in
layered AFD49 hydrogels at 42 days. Images A) top-down, B) and C) side-on.
(green, live; red, dead). Peptide concentration was 2.1 mM in the base gels and 2
mM in the top layered gels. .................................................................................... 217
Figure 5-19. Growth of iPSCs on AFD49 hydrogels and Matrigel coated TCPS
controls. A) Matrigel coated TCPS, B) AFD49 hydrogels, C) AFD49 hydrogels cast
with 10 µg mL-1 vitronectin fragment, D) AFD49 hydrogels cast with 20 µg mL-1
vitronectin fragment, E) AFD49 hydrogels cast with 50 µg mL-1 vitronectin fragment.
Each fluorescence image (green, EGFP fluorescence) is of the same location as the
phase contrast image above. (scale bars 100 µm). Hydrogels were 2 mM AFD49
final......................................................................................................................... 222
Figure 5-20. Growth of iPSCs on Matrigel coated surfaces. A) and B) TCPS, C) and
D) AFD49 hydrogels. Each fluorescence image (B and D; Green, EGFP
fluorescence) is of the same location as the phase contrast image above (A and C).
(scale bars 100 µm). Hydrogels were 2 mM AFD49 final. ...................................... 224
Figure 5-21. Chemical structure of A) ATRA and B) 13-cis retinoic acid. .............. 226
Figure 5-22. ECD of ATRA in AFD49 solution on exposure to room light. ATRA initial
concentration was 290 µM, AFD49 concentration was 7.5 mM. Spectra collected at
times as indicated by colour legend. ...................................................................... 227
xxiii
List of tables
Table 1-1. Cleavage sites and resulting peptide terminus modifications of common
enzymatic and chemical cleavage methods ............................................................. 40
Table 1-2. Research aims of each chapter within this thesis ................................... 45
Table 2-1. Table of computed properties of designed concatemers ........................ 79
Table 2-2. Quantification of Het2-6 purification ........................................................ 86
Table 2-3. Fit parameters for Het2-6 denaturation with G.HCl ................................. 91
Table 3-1. Species observed by LCMS and their potential and assigned identities.
............................................................................................................................... 109
Table 3-2. Potential redesign of EDP-11 and RDP-4 ............................................. 128
Table 4-1. Parameters used to fit ECD data for AFD19 self-assembly under nongelling conditions .................................................................................................... 144
Table 4-2. Diameters of AFD36 fibrils at different pH values obtained by SAXS fitted
to a flexible cylinder model ..................................................................................... 158
Table 4-3. Fitting parameters for modelling G.HCl denaturation of AFD19 and
AFD36 .................................................................................................................... 168
Table 5-1. Table of previously reported self-assembling peptide hydrogels
referenced in this chapter ....................................................................................... 186
xxiv
Symbols and abbreviations
1-cyano-4-dimethylaminopyridinium tetrafluoroborate
CDAP
2-(N-morpholino)ethanesulfonic acid
MES
2-nitro-5-thiocyanobenxoic acid
NTCB
4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid
HEPES
4',6-diamidino-2-phenylindole
DAPI
Accessible surface area
ASA
Acetonitrile
ACN
All-trans retinoic acid
ATRA
Bovine serum albumin
BSA
Change in Gibbs free energy
ΔG
Change in Gibbs free energy for transition x
ΔGx
Coiled coil of oligomerization state n
n-mer
Concatemer linker sequence
L
Cyanogen bromide
CNBr
Denaturant concentration
[D]
Dimethyl sulfoxide
DMSO
Dipeptide instability weight value
DIWV
Dipeptide of residues x and y
xiyi+1
Dulbeccos modified Eagle medium
DMEM
EDP-11 peptide
E
EP-11
E-
Electronic circular dichroism
ECD
Enhanced green fluorescent protein
EGFP
Equilibrium constant for disassembly
Kd
Extracellular matrix
ECM
xxv
Fluorescein diacetate
FDA
Fluorescein isothiocyanate
FITC
Formic acid
FA
Fraction folded
Ff
Free monomer
mon
Glutathione S-transferase
GST
Grand average hydropathy
GRAVY
Guanidine hydrochloride
G.HCl
Human embryonic stem cell
hESC
Ideal gas constant
R
Induced pluripotent stem cell
iPSC
Instability index
II
Interfacial tension
IFT
Intermediate state
I
Isoelectic point
pI
Isopropyl-β-D-thiogalactoside
IPTG
Ketosteroid isomerase
KSI
Liquid chromatography-mass spectrometry
LCMS
Lithium dodecyl sulfate
LDS
Loss moduli
G″
Luria broth
LB
m value
m
Maltose-binding protein
MBP
Mean residue ellipticity
[θ]
Mean residue ellipticity at 222 nm
[θ]222
Mean residue ellipticity of assembled state x
[θ]x
xxvi
Measured ellipticity
θobs
Methionine-aspartate dipeptide
M
Momentum transfer vector
q
Native state
N
Number of monomer units
n
Number of peptide bonds
nb
Number of residues of anionic residue j
Nj
Number of residues of cationic residue i
Ni
Open reading frame
ORF
Pathlength
l
Peptide amphiphile
PA
Phosphate buffered saline
PBS
pKa of anionic residue j
pKaj
pKa of cationic residue i
pKai
Propidium iodide
PI
Protein concentration
c
RDP-4 peptide
R
RP-4 peptide
R-
Residue i
i
Reversed-phase high-performance liquid chromatography
RP-HPLC
Rho-associated kinase
ROCK
Sequence length
LP
Size exclusion chromatography
SEC
Small ubiquitin-related modifier
SUMO
Small-angle X-ray scattering
SAXS
Somatomedin B
SMB
xxvii
Storage moduli
G′
Temperature
T
Temperature of 50% denaturation
Tm
Tetramethylrhodamine
TRITC
Tissue culture polystyrene
TCPS
Tobacco etch virus
TEV
Total concentration of peptide
[P]tot
Transmission electron microscopy
TEM
Trifluoroacetic acid
TFA
Tris(hydroxymethyl)aminomethane
Tris
Unfolded state
U
xxviii
Amino acid abbreviations
Alanine
A
Arginine
R
Asparagine
N
Aspartic acid
D
Cysteine
C
Glutamine
Q
Glutamic acid
E
Glycine
G
Histidine
H
Isoleucine
I
Leucine
L
Lysine
K
Methionine
M
Phenylalanine
F
Proline
P
Serine
S
Threonine
T
Tryptophan
W
Tyrosine
Y
Valine
V
xxix
1
Literature Review
1.1 Molecular self-assembly
Self-assembly is the spontaneous organisation of components into larger assemblies
based only on local interactions. In the case of molecular assembly this process is
governed by non-covalent forces that can include electrostatic, hydrophobic and van der
Waals interactions, metal-ligand and hydrogen bonds, as well as aromatic π-stacking.1-3
Although these interactions are individually weak, the sum of multiple interactions can
generate highly stable and defined assemblies at the nanometre scale. Molecular selfassembly is common in both natural and synthetic systems, ranging from the assembly of
lipid bilayers4 and actin protein filaments5 to the designed assembly of supramolecular
polymers6-8 and virus-like particles.9, 10
In recent years the targeted design of self-assembling molecules has become a
burgeoning field of study. This has rapidly enhanced understanding of the factors
governing the assembly of monomeric building blocks to give structurally well-defined
architectures.11, 12 Understanding of these factors has assisted in the development of selfassembling materials with potential in many applications such as drug delivery,13 tissue
engineering14 and targeted diagnostic imaging.15 Advantages of self-assembling materials
relative to covalently bonded equivalents include the responsive nature of the selfassembly process, as well as the ability to tune the properties of the assembled material
for individual applications by modification of the assembling monomers, and potential for
multiple functionalities to be incorporated through the combination of varied monomers.11
Self-assembling biomolecules, such as nucleic acids, carbohydrates, polypeptides and
lipids, are of particular interest in many applications due to their higher biocompatibility and
biodegradability in comparison to synthetic materials.16, 17
1
Figure 1-1. Classes of self-assembling peptides. A) Peptide amphiphiles, B) β-sheet forming
peptides, C) α-helix forming peptides. Dashed bonds indicate peptide backbone hydrogen
bonding pattern. Purple shading highlights secondary structures of β-sheet and α-helix. The
dashed box in A and C highlight hydrophobic regions of the peptides. In B the hydrophobic
region extends out of the image towards the viewer. Atoms are coloured according to element;
carbon (grey), oxygen (red), nitrogen (blue) and hydrogen (white).
Designed peptides are particularly attractive components for the self-assembly of
supramolecular structures, with potential in applications such as emulsion and foam
stabilisation,18, 19 wound healing20, 21 and drug delivery.22, 23 These informational polymers
may be produced with specific sequences, generally using synthetic methods based on
solid-phase synthesis.24 Molecular complexity may be engineered into these molecules by
producing sequences of amino acids of varied properties. The ease of access to
oligopeptides of defined sequence has enabled investigation of sequence-structure
relationships.25 While the precise link between molecular architecture and assembled
supramolecular structure is still an area of ongoing study, in many cases it is now possible
to use established design principles to produce peptides that self-assemble to form
specific structures. This self-assembly is often highly dependent on the peptide
environment and may be responsive to external stimuli such as changes in solvent,26
solution pH,26, 27 temperature,28 salt concentration29, 30 or chelating metals.18, 31 Designed
peptides also possess the advantage of being amenable to bioproduction, making them a
potentially renewable material for large-scale applications.32, 33
2
Self-assembling peptides may be classified by the types of structures they form. Three of
the most common classes of self-assembling peptides are peptide amphiphiles, β-sheet
forming peptides and α-helix forming peptides (Figure 1-1). A key driving force for selfassembly in aqueous environments is the hydrophobic interaction. This drives hydrophobic
moieties to segregate so as to minimise contact with the bulk aqueous phase.34 In peptide
self-assembly, this driving force for association is generally opposed by the solvation of
polar moieties and, where they are charged, electrostatic interactions.35 The selfassembling peptides discussed in this review are relatively short amphiphilic sequences
(Figure 1-1). Design of these amphiphilic peptides allows for forces favouring and
disfavouring association to be controlled, and in turn direct peptide self-assembly. Section
1.2 of this review gives an overview of the three self-assembling peptide classes and
discusses key examples of each. Many self-assembling peptides have potential for use
across a variety of applications. Section 1.3 focuses on the use of these self-assembling
peptides as surfactants, cell scaffolds and therapeutic delivery vehicles. One of the
barriers to the adoption of self-assembling peptides in many applications is the cost of
production. Section 1.4 discusses production methodologies for designed self-assembling
peptides, with a focus on bioproduction techniques.
1.2 Peptide self-assembly
1.2.1 Peptide amphiphiles
Peptide amphiphiles (PAs) are a structurally simple class of self-assembling peptides that
are designed to possess a lipid-like structure comprising a hydrophilic head and
hydrophobic tail. The hydrophilic head is composed of one or more charged amino acids,
while the tail can be either a string of hydrophobic amino acids or an alkyl chain.36 PAs
assemble similarly to lipid molecules, with hydrophobic tails associating to minimize
exposure to water, and have well-defined critical aggregation concentrations.37 Selfassembly of PAs has been reported to produce well-ordered nanostructures such as
bilayers,38, 39 micelles,40 vesicles,37 nanotubes,37 nanorods39 and nanofibres.41
In PAs composed only of amino acids, the hydrophobic tail generally comprises 3-9
hydrophobic residues such as alanine, valine or leucine.36 The hydrophilic head group is
composed of one or more charged amino acids with lysine or aspartate being the most
common. The hydrophilic head group is generally at the C-terminus of these peptides and
3
the terminal carboxylate may be either uncapped or capped, depending on the desire for
an anionic charge. The hydrophobic tail is generally at the N-terminus, with the terminal
amino acid capped to prevent interruption of hydrophobic interactions. However, the
orientation of the hydrophobic and hydrophilic moieties can be swapped, in some cases
without any significant effect on nanostructure assembly.42 PA self-assembly into defined
nanostructures is pH dependent due to changes in the protonation state of charge-carrying
residues. In highly charged states, electrostatic repulsion prevents self-assembly.37 As this
charge is decreased, PA solubility is lowered and the electrostatic repulsion between
peptides is lessened, allowing self-assembly. Further decreases in charge then leads to
amorphous aggregation.39, 42
The assembly of PAs occurs similarly to that of lipid molecules, where hydrophobic tails
associate away from water, leaving the charged groups facing the bulk aqueous solution.
However, unlike alkyl chain surfactants, the hydrophobic amino acids may form
intermolecular hydrogen bonds as well as hydrophobic interactions. This leads to
assembled PAs adopting a higher degree of organisation than would be present in lipid
assemblies.37, 39
PA self-assembly has been well characterised in solution. Studies by the Zhang group and
others have shown PAs of six to twelve amino acids, including one to two anionic or
cationic residues with the remainder being hydrophobic residues, to assemble into ordered
nanotubes and vesicles at neutral pH.37, 42-45 A preference for formation of vesicles over
nanotubes was seen with increasing PA hydrophilicity,37 however in changing PA
hydropathicity, the peptide residue composition is modified. In doing so, other factors such
as the size of the hydrophilic head group likely also play a role in structure determination.
The assembly of these varied PAs illustrates that peptides with a similar distribution of
hydrophobic and hydrophilic amino acids may self-assemble into comparable structures,
even when their primary sequence varies. Further studies have shown longer PAs of ~ 30
amino acids, with a higher proportion of charged residues, to form core-shell micelles at
neutral pH.40 These micelles show potential in gene delivery applications, as is discussed
in Section 1.3.3.
In addition to characterisation of PAs in solution, studies of their assembly at interfaces
have been reported. For example, the peptides Ac-V6K1-2-CONH2 were reported to form
predominantly bilayer structures at the silica-water interface.38, 46, 47 These adsorbed
4
cationic peptides also bound DNA, demonstrating their potential for gene delivery
applications.46 Xu et al.39 studied the assembly of the peptide series (Ac-AnK-NH2, n = 3, 6,
9) at solid surfaces and showed that increasing the length of the hydrophobic tail gave
changes in assembled nanostructures from stacked bilayers to nanotubes and nanorods.
These transitions between structures were attributed to changes in the packing of the
hydrophobic tails and interactions of the charged head groups. Some reports such as this
have correlated observed transitions in assembled PA nanostructures with molecular
architecture using adaptations of the packing parameter48 commonly used to describe
conventional surfactant assembly.39, 45 However, these studies are so far limited, and
require further experimental work to allow prediction of assembled structures.
Rather than a series of hydrophobic residues, PAs may incorporate alkyl chains of typically
12-16 carbon atoms in length at the N-terminus (Figure 1-2A).41, 49-52 Alkyl-tail PAs
assemble in a similar manner to the PAs previously discussed, with the hydrophobic
moiety segregating away from water and driving assembly. Assembly of alkyl-tail PAs has
been reported to produce structures such as monolayers,50 spherical micelles49 and
cylindrical micelles (nanofibres).41, 52
Of the alkyl-tail PAs, those that form nanofibres have been particularly well studied.41, 52-58
A key structural component of these PAs is the presence of a β-sheet forming region
adjacent to the alkyl-chain tail (Figure 1-2B). Intermolecular hydrogen bonding between
these first few residues occurs primarily down the longitudinal axis of the nanofibre, and
stabilises the fibre arrangement relative to a spherical micelle structure.54 The interactions
of this region also affect the ultimate mechanical properties of assembled structures, with
less twisted β-sheet structures generating stiffer materials.53 Reversible cross-linking may
be incorporated into alkyl-tail PAs by including cysteine residues adjacent to the alkyl
chain. The formation of intermolecular disulfides stabilises nanofibres to disassembly.41
5
Figure 1-2. Example chemical structures of A) an alkyl-tail PA and B) a nanofibre forming alkyltail PA.
The incorporation of one or more charged residues in the hydrophilic head provides both
solubility and electrostatic interactions that influence structure formation.41, 52 Nanofibreforming PAs with either cationic or anionic charges can be triggered to self-assemble by
lowering the electrostatic repulsion between charged PAs. This can be achieved by
adjusting pH to partially neutralise charge, charge-screening through the addition of
electrolyte solutions such as cell culture media, or in the case of anionic PAs, the addition
of divalent metal ions.41, 55, 56 Furthermore, mixed systems of oppositely charged PAs
assemble under conditions where both species possess charge.57 In some cases,
assembled nanofibres are able to form physical cross-links, resulting in hydrogel
formation.55, 57, 58 Hydrogels formed by these peptides have potential use in several
biomedical applications as is discussed in Section 1.3.
In PAs, residues further away from the core are less involved in nanostructure formation.
Alkyl-tail PA self-assembly tolerates a range of amino acids to be incorporated within the
hydrophilic head, and allows greater sequence variation to be incorporated into alkyl-tail
6
PA design compared to amino acid only PAs. This has been used for the incorporation of
functional motifs for specific applications.41, 54 For example, PAs have been designed as
mineralisation templates and cellular scaffolds by incorporation of crystal nucleation and
cellular recognition motifs respectively.41, 52 In some cases, incorporation of functional
motifs partially destabilises PA self-assembly, likely by altering the length of the
assembling peptide and disrupting peptide-peptide interactions.59 However, by including a
proportion of non-functionalised PA, self-assembly can be restored. The functionalization
of alkyl-tail PAs for specific applications has been well studied by both the Stupp group
and others, and is discussed further in Section 1.3.
Peptide amphiphiles are a structurally simple class of self-assembling peptide. The
potential for modification of their structure is relatively limited due to requirements for selfassembly. However, functionality is able to be integrated into many of these designs by
incorporation of functional motifs within the hydrophilic head. Several such PAs,
predominantly alkyl-tail PAs, have found utility in a range of applications as is discussed in
Section 1.3. Peptides designed to assemble to form secondary structures such as βsheets or α-helices are more complex in design and are discussed in the following
sections.
1.2.2 β-sheet peptides
The β-sheet structure was discerned from studies of proteins such as silk fibroin and βkeratin, which contain a high proportion of this secondary structure.60-62 Structural
assignment based on X-ray diffraction data and theoretical calculations allowed the
conformation of polypeptides in β-sheets to be proposed. β-sheets are comprised of
multiple polypeptide β-strands positioned in either parallel or anti-parallel configurations,
with hydrogen bonding between backbone amide groups of neighbouring β-strands (Figure
1-1B). In β-strands, polypeptides adopt an almost fully extended conformation. In this
arrangement the side chains of sequential amino acids will be oriented alternating above
and below the plane of the β-sheet.62 These early studies also indicated lamination of βsheet structures to occur, with interactions between laminated sheets dictated by the
properties of residues at each face.62
In early attempts to produce peptides that adopted stable laminated β-sheet structures,
oligopeptides with repeating hydrophilic-hydrophobic sequences were investigated.
7
Studies of oligopeptides such as poly(VK),63 poly(EA),64 poly(KF)65 and poly(YK)66 showed
these sequences to form β-sheet structures under conditions where the charge on the
peptide was lowered by adjusting pH or screened by the addition of salt to reduce
repulsive electrostatic interactions. This was attributed to the formation of β-sheet bilayers,
in which hydrophobic interactions between non-polar residues stabilise association, and
hydrophilic amino acids are displayed on the exterior (Figure 1-3).63
Figure 1-3. Schematic representation of amphipathic β-sheet and bilayer assembly. A) β-strand, B)
β-sheet in which hydrogen bonding occurs parallel with the plane of the page (as in Figure 1-1B),
C) β-sheet bilayer. Hydrophobic and hydrophilic residue side chains are represented as red and blue
spheres respectively. For simplicity the polypeptide backbone is represented as a line.
Amino acids have distinct conformational preferences that influence their propensity to
adopt secondary structures such as α-helices and β-sheets.67 Notably, the cyclic structure
of proline makes this residue incompatible with the formation of either of these secondary
structures. The propensity of amino acids for β-sheet formation has been studied by
statistical analysis of residue composition of proteins with known structure68-70 and hostguest studies,71 resulting in thermodynamic scales for β-sheet propensity. While there is
some variation across the different studies, residues such as tyrosine, threonine,
isoleucine and valine are seen to have high β-sheet propensity, while glycine, aspartate
and alanine do not.68-71 The design of β-sheet forming peptides often targets the
incorporation of high β-sheet propensity residues as well as hydrophobic-hydrophilic
residue patterning consistent with the desired assembled structure.
The advent of solid-phase peptide synthesis enabled investigation of short designed
peptides of defined sequence. Short peptides composed of alternating hydrophobichydrophilic residues showed a high tendency for β-sheet formation, similar to earlier
oligopeptides, and allowed better study of factors influencing self-assembly. For example
8
the peptide length is a key determinant of self-assembly.72-75 Although some variation was
seen with amino acid composition and solution conditions, a minimum peptide length of
approximately 12 residues is observed for β-sheet formation. Typical designed β-sheet
forming peptides are between 12-20 residues in length.76
A complete understanding of the factors influencing the assembly of β-sheet forming
peptides is yet to be achieved. However, the influence of assembly-stabilising and destabilising interactions has been the focus of significant research. Several studies have
investigated the balance between stabilising hydrophobic interactions and hydrogen
bonds, and destabilising electrostatic interactions.77-81 For example, studies have shown
that increasing the hydrophobicity of the non-polar residues in alternating hydrophobichydrophilic β-sheet forming peptides favoured self-assembly due to increased stabilising
hydrophobic interactions.77, 78 Additionally, Dong et al.79 developed a series of “frustrated”
amphiphilic peptides, in which the lengths of a central β-sheet forming domain and flanking
cationic destabilising domains were varied. At neutral pH the peptide Ac-K2(QL)6K2CONH2 assumed a β-sheet secondary structure and formed nanofibres consistent with βsheet bilayers. Addition of salts lowered electrostatic repulsion and increased fibre lengths,
while assembly in phosphate-buffered saline gave self-supporting hydrogels. Decreasing
the length of the central domain lowered the number of stabilising interactions, while
increasing the length of the flanking domains increased destabilising electrostatic
interactions, such that in either case β-sheet and fibre formation was largely inhibited.
These reports exemplify the fine balance between hydrogen bonding and hydrophobic
effects that promote self-assembly and electrostatic repulsive effects that oppose selfassembly.
Many β-sheet peptides such as Ac-K2(QL)6K2-CONH2 assemble to form fibrils due to the
repeating nature of their intermolecular hydrogen bonding network. Hydrogel formation
occurs above a critical concentration for some of these peptides under appropriate solution
conditions due to the formation of physical cross-links between fibrils.26, 82-88 Several
research groups have developed series of β-sheet and hydrogel forming peptides based
on successive modifications of β-sheet forming peptide designs and are now discussed.
The series of fibril and hydrogel forming peptides reported by the Zhang group are some of
the most studied in the literature. The first of these peptides, EAK16 (Ac-((AEAEAKAK)2)CONH2)) was designed based on a repeating AEAEAKAK sequence of the Z-DNA binding
9
protein Zuotin.89 EAK16 self-assembly gave a β-sheet structure at low concentrations that
was extremely stable to thermal and chemical denaturation, as well as insensitive to
changes in pH.30 Addition of NaCl to peptide solutions produced a macroscopic
membrane composed of interwoven fibres with diameters corresponding to approximately
the length of one peptide, consistent with individual peptides being oriented
perpendicularly to the fibre axis.30, 75 The substitution of arginine for lysine and aspartate
for glutamate in EAK16 gave the peptide RAD16-II (Ac-((RARADADA)2)-CONH2), which
displayed self-assembly similar to EAK16. This outcome highlighted the importance of
charged residue distribution rather than specific residue composition in these peptides.85
Shortening of the repeating unit of RAD16-II gave the peptide RAD16-I (Ac-((RADA)4)CONH2), which maintains the alternating hydrophilic-hydrophobic amino acid sequence
and assembles similarly to EAK16 and RAD16-II peptides.84 For these three peptides,
assembly was proposed to proceed through formation of anti-parallel β-sheet bilayers due
to intermolecular electrostatic repulsive interactions in the parallel arrangement.90 All three
peptides readily self-assemble to form hydrogels when introduced into a solution of
monovalent salt, physiological buffer or cell culture media, giving these peptides potential
utility in several biomedical applications, as is discussed in Section 1.3.84, 85 As with many
physical hydrogels, these peptide hydrogels are self-healing following mechanical
disruption due to the non-covalent nature of fibre formation and fibre cross-links.90
Peptide self-assembly can also be linked to conformational changes in the monomer
structure. An example of this is the series of β-hairpin forming MAX peptides such as
MAX1 (H2N-VKVKVKVKVDPPTKVKVKVKV-CONH2) developed by Schneider and
colleagues.87 The design of these 20-residue peptides was based on two arms that
possess a high tendency for β-sheet formation due to their alternating hydrophobichydrophilic residue composition, similar to the EAK and RAD designs discussed
previously. These arms flank a central tetrapeptide (VDPPT) that is designed to adopt a
type II′ turn structure to enable β-hairpin formation. The ability of the two arms to form
intramolecular hydrogen bonds is dependent on reducing the electrostatic repulsion
between lysine residues by adjusting pH or increasing ionic strength.87, 91 β-hairpin folding
occurs simultaneously with the formation of single monomer wide bilayer fibrils via lateral
β-sheet formation and hydrophobic facial association. Several variants of MAX1 have been
reported in which effects of residue substitution in the hydrophobic and hydrophilic faces
have been studied.92-94 For example, the replacement of one lysine with a glutamate
10
residue gave the peptide MAX8, which showed self-assembly that was faster and occurred
at lower pH values than that of MAX1 due to decreased levels of repulsive positive
charge.92, 94 Both MAX1 and MAX8 peptides self-assemble to form hydrogels on addition
of cell culture media,87, 95, 96 and have potential in several biomedical applications, as is
discussed in Section 1.3.
Polypeptide sequences with glutamine repeats possess a high tendency for β-sheet
formation. Aggregates formed by these polyQ tracts have been linked to
neurodegenerative diseases such as Huntington’s disease and Kennedy disease.97, 98 This
predisposition for β-sheet formation is attributed to the ability of glutamine to form interchain hydrogen bonds via both backbone amides and their polar side chains.98, 99 The
design of peptides P11-2 (originally named DN1; Ac-QQRFQWQFEQQ-CONH2)26, 100 by
the Aggeli group and Q11 (Ac-QQKFQFQFEQQ-CONH2)83 by the Collier group used the
β-sheet forming tendency of this amino acid to direct peptide assembly. The alternating
hydrophobic-hydrophilic residues in the centre of the peptides were included to promote βsheet assembly, while the inclusion of cationic and anionic residues near opposing peptide
termini were used to dictate an antiparallel β-sheet alignment due to inter-molecular
charge-charge interactions. The positioning of the three hydrophobic residues towards the
centre of the peptides provided a hydrophobic driving force for assembly as well as
intermolecular recognition via π-π interactions.
Peptide Q11 assembled to form a β-sheet structure in solution at low concentrations, and
electron microscopy revealed fibrils that were a single peptide in diameter.83 Increased
ionic strength enhanced β-sheet assembly, with aqueous solutions of Q11 undergoing
rapid hydrogel formation in monovalent and divalent salts or phosphate buffered saline.
Fibrils and hydrogels of Q11 have shown utility in several biomedical applications as is
discussed in Section 1.3.
Peptide P11-2 also formed β-sheet structures in solution at low concentrations, with
thermostable gels formed at acidic pH.26, 82, 100 A series of P11 class peptides were
subsequently produced using similar design principles, but incorporated alternate charge
carrying residues to give changes in the pH dependence of assembly.82, 101 For these
peptides, assembly followed a nucleated pathway; above a critical concentration peptide
monomers assembled into β-sheet tapes with further increases in concentration leading to
11
higher-order structures of ribbons (pairs of tapes stacked back to back via hydrophobic
interactions), fibrils (bundles of stacked ribbons) and fibres (entwined fibrils).102
The effect of peptide charge on β-sheet self-assembly has been investigated by the Aggeli
group. Characterisation of a series of P11 gelling peptides capable of carrying either one or
multiple charges allowed the proposal that in solutions of low ionic strength, a charge of ±1
per peptide was required to decrease electrostatic repulsion and allow fibril formation and
physical cross-linking interactions, while also maintaining peptide solubility.82 This
provided pH responsive self-assembly as the charge carrying groups on self-assembling
peptides were titrated. Increased ionic strength decreased electrostatic interactions via
charge screening and altered the pH of gelation. This is exemplified by peptide P11-4 (AcQQRFEWEFEQQ-CONH2), which transitions from a fluid to a gel phase at pH 6.0 upon
the addition of salt.82 More recent experimentation by the Aggeli group has generated a
much larger number of P11 peptides,101, 103 some of which are capable of gelation in cell
culture media. One key observation was that in cell culture media, which possesses a high
ionic strength, a charge per peptide of ±2 was required to maintain solubility. This is due to
the screening of molecular charge and stronger hydrophobic driving force for assembly.
These experiments also showed P11-4 to be capable of gel formation at physiological pH in
cell culture media, making this peptide amenable to use in tissue culture models as is
discussed in Section 1.3. Hydrogels formed by peptide P11-4 have also been demonstrated
to promote remineralisation and hydroxyapatite nucleation. This has shown utility in the
treatment of dental caries,104 and formulations involving P11-4 have progressed to clinical
safety trials with promising results.105
As well as assembly in solution, β-sheet peptides may be designed to assemble at
interfaces. Rapaport et al.106 reported ordered monolayers of designed β-sheet peptides 7
to 17 amino acids in length at the air-water interface as confirmed by grazing-incidence Xray diffraction. These peptides were of the form H2N-PE(FE)nP-COOH (PFE-n), where n =
2-7. This series of peptides form anti-parallel β-sheets where the phenylalanine residues
adsorb to the interface while the glutamate residues face the bulk aqueous solution. Lower
solution pH leads to protonation of acidic side-chains, which decreases peptide aqueous
solubility and increases the potential for intermolecular hydrogen bonding of the carboxyl
groups. Both of these factors result in a more stable association at the interface.
Monolayers of these peptides were both compressible and elastic, with recovery of original
12
structure following compression attributed to a reversible buckling of the peptide
backbone.107, 108
Further experimentation by the same group has generated several designs that direct the
assembly of other molecules. The peptide PFD-5 assembles to give both monolayers and
bilayer fibrils, of which the fibrils are able to form hydrogels at neutral pH.88 This peptide
was also shown to template calcium phosphate mineralisation, making it potentially
applicable in bone tissue regeneration. Vinod et al.109 recently reported the peptide PFK-5,
which incorporated the cationic lysine as the charged hydrophilic residue. This peptide
assembled at air-water interfaces and was able to template gold nanofibre film formation.
Finally, by selective placement of residues on the hydrophilic face of the β-sheet peptides,
catalytic arrangements of residues were formed.110 This enabled the direction of small
molecule adsorption at peptide monolayers with potential in detection applications.
Larger β-sheet forming polypeptides may also be designed using similar principles to
those discussed above. A 30-residue peptide BS30 was designed as a triple-stranded βsheet peptide that formed a monolayer at the air-water interface.111 In this design, 8residue repeating sequences, following the hydrophobic-hydrophilic residue pattern
(above), were linked via proline-alanine dipeptides. These dipeptides were designed to
form type II β-turns, which resulted in an anti-parallel arrangement of adjacent β-strands.
The triple-stranded peptide assembled into extended hydrogen bonded ribbons, with
several packing configurations possible between triple-stranded units.
The design of monomeric β-sheet based structures has been hindered by the tendency of
this secondary structure for aggregate and fibril formation. However, there has been some
progress in the design of small β-sheet units.112-118 For example, the 20-residue peptide
Betanova was designed to adopt an antiparallel three-stranded β-sheet motif using tworesidue β-turn linkers.113 This peptide formed a discrete, well-folded conformation despite
the absence of any covalent crosslinking. Several designs reported by multiple groups at
almost the same time utilised similar design principles and also gave discrete 3-stranded
β-sheet motifs.114-116 Ongoing work in this area has produced β-structured mini-proteins
based on both naturally occurring motifs stabilised by disulfide bonds117 and modification
of amyloid forming β-hairpin sequences.118 Investigation of such discrete β-sheet
structures may increase understanding of complex interactions in native and designed
proteins, as well as enabling de novo design of structures with increased complexity.
13
Peptides designed to adopt β-sheet conformations have been well studied, and as
discussed above, several research groups have developed series of peptides based on
particular designs. The rationale behind the design of these peptides generally focuses on
a combination of residue patterning and the selection of amino acids that have high βsheet propensity. Studies of β-sheet forming peptides have predominantly focused on
those capable of fibril formation. This is likely due in part to the tendency of β-sheet
peptides to form these extended structures owing to the repeating nature of their
intermolecular hydrogen bonding network. β-Sheet forming peptides show utility in many
applications, as is further discussed in Section 1.3. An alternate secondary structure
commonly targeted in peptide design is the α-helix, as is now discussed.
1.2.3 α-Helical peptides
The α-helical secondary structure was first proposed based on X-ray scattering studies of
α-keratin, myosin and other fibrous proteins,119-121 as well as synthetic polypeptides122
found to be rich in helical structure. In an α-helical conformation, the polypeptide backbone
is folded into a right-handed helix where each amino acid corresponds to an approximately
100° turn in the helix (Figure 1-1C). This results in 3.6 residues per helix turn and a
periodicity of 5 turns or 18 residues (Figure 1-4A). In this arrangement the backbone
amide groups are aligned axially and hydrogen bonds form between amides spaced four
residues apart (i, i + 4).
Further studies of helices in proteins suggested that many of these helices are in coiledcoil arrangements.123, 124 In a coiled coil, two or more helices associate by wrapping
around one another in a left-handed coil in either a parallel or anti-parallel arrangement.
This arrangement results in a smaller 3.5 residues per helix turn, and a periodicity of 7
residues, referred to as a heptad, in which positions are labelled a, b, c, d, e, f, and g
(Figure 1-4B). Stabilisation of helix association was proposed to be due to hydrophobic
interactions between the helices.124 A repeating periodicity of hydrophobic residues
(conventionally at positions a and d) produces a hydrophobic face on each helix
compatible with coiled-coil formation.
Similar to studies of β-sheet forming peptides, the advent of solid-phase synthesis allowed
studies of short defined peptide sequences, and investigation of factors influencing the
folding of peptides into α-helical structures. Isolated α-helices tend to be
14
thermodynamically unstable in solution.125 However, it was found that α-helical structure in
peptides could be stabilised by intramolecular constraints such as covalent bonds between
i and either i + 4 or i + 7 residues126, 127 or metal ion coordination between residues at i and
i + 4 positions.128, 129 In addition to these intramolecular constraints, helices may also be
stabilised by association with interfaces. If hydrophobic residues are placed at alternating i
+ 3 and i + 4 positions (residues 1, 4, 8, 11, 15 and 18 in Figure 1-4A) and hydrophilic
residues at the remaining positions, a facially amphiphilic helix is generated that can
associate laterally at air/oil-water interfaces.130 Over short distances of 1-3 heptads, the
positioning of hydrophobic residues in helices for coiled-coil formation are also compatible
with the hydrophobic residue positioning for facially amphiphilic helices capable of
association with interfaces. One well studied facially amphiphilic peptide is Lac21 (AcMKQLADS LMQLARQ VSRLESA-CONH2), which was derived from the coiled-coil forming
domain of the Lac repressor protein131 and assembles at air-water interfaces.132 Several
modifications to this peptide have been made to investigate their effect on assembly at
interfaces and are discussed in Section 1.3.1.
Similar to amino acid β-sheet propensity, statistical analysis of proteins of known
structure,68-70 and host-guest studies using both protein and peptide sequences67, 133, 134
led to the development of thermodynamic scales for the helix-forming propensity of each
amino acid. While some variation is present across these studies, residues such as
alanine, arginine, lysine and leucine are considered to have high α-helix propensity, while
glycine, cysteine and aspartate have low propensity for such structures. The design of αhelix and coiled-coil forming peptides generally targets a combination of sequences with
hydrophobic-hydrophilic residue patterning compatible with the desired structure, as well
as selection of amino acids of high α-helix propensity.
The study of peptides that assemble to form discrete coiled coils has been the focus of a
large body of research. This work initially used peptides with sequences taken from
leucine zipper motifs of native proteins such as GCN4,135, 136 and the paired Fos and
Jun,137, 138 which were known to adopt coiled-coil structures. This work has led to an
understanding of many factors affecting coiled-coil assembly such as peptide length,
hydrophobic residue identity and the placement of charged residues, as is discussed
below. Following on from this work has been the design of de novo peptides that adopt αhelix and coiled-coil structures as is also discussed.
15
Figure 1-4. Representations of periodicity of α-helices in A) isolated and B) coiled-coil
arrangements, and C) schematic illustrating interactions between helices in a parallel dimeric
coiled coil.
16
As found for β-sheet structures (Section 1.2.2), coiled-coil stability is strongly dependent
on the length of the constituent α-helical peptides.131, 139, 140 For example, Fairman et al.131
showed that increasing the length of the Lac21 peptide to the four-heptad Lac28 or the
five-heptad Lac35 increased coiled-coil stability by approximately 4 kcal mol-1 monomer-1
per added heptad. This is attributed to increased numbers of stabilising interactions as well
as lowering the proportion of amino acids in less stabilised positions at helix termini that
take part in coiled-coil end-fraying.139, 141 Designed α-helix and coiled coil forming peptides
are generally of three to six heptads, or 21 to 42 residues in length.142
The identity of hydrophobic residues at heptad a and d positions also influences coiled-coil
assembly due to packing of residues within the hydrophobic core. Hodges and colleagues
investigated the effect of residue substitutions at positions a and d of the central heptad of
a five-heptad dimeric coiled coil.143, 144 These studies showed a correlation between
residue hydrophobicity and the resulting coiled-coil stability. This highlights the importance
of hydrophobic interactions in stabilising coiled-coil assembly. While some variation was
seen between positions, these studies showed the highest coiled-coil stability for peptides
in which residues such as leucine, isoleucine or methionine were substituted at a and d
positions. The packing of residues in the hydrophobic core also influences the
oligomerization state of the coiled coil. By varying combinations of isoleucine, leucine or
valine at a and d positions of the dimeric leucine zipper peptide GCN4-p1, Harbury et al.145
showed transitions between dimeric, trimeric and tetrameric coiled coils. The general
relationships between hydrophobic residue identity, coiled-coil stability and oligomerization
state are likely to be complex, and vary between individual peptides. However, these
results suggest that the packing of residues within the hydrophobic core of peptide coiled
coils is an important factor in coiled-coil assembly.
In both native proteins and designed coiled coils, positions e and g that flank the
hydrophobic residues are often occupied by ionisable amino acid residues. In an
assembled coiled coil, residues at positions e and g of constituent peptides will be brought
into close proximity and interactions between them can dictate oligomerization selectivity
(Figure 1-4C).142 This is exemplified in homo-dimerization, where placement of oppositely
charged residues at e and g will destabilise the anti-parallel arrangement of helices, and
favour parallel dimerization due to electrostatic interactions.146, 147 Similarly, the selective
placement of charged residues at e and g positions can provide specificity in hetero17
oligomerization interactions.137, 139, 140, 148, 149 The electrostatic interactions between helices
can also confer pH-responsive coiled-coil assembly. Repulsion of like-charged residues at
e and g positions of two helices will destabilise coiled-coil assembly, therefore allowing
assembly only at pH values where charge on these residues is decreased.149, 150
The hydrophobic core may also be expanded from positions a and d to include flanking
positions e and g. This expansion is associated with an increase in the oligomerization
state of coiled coils due to the larger hydrophobic face for association. For example, a
change from a dimeric to a tetrameric coiled coil was reported on replacement of the three
charged residues at g positions of the GCN4-pR peptide with hydrophobic alanine or
valine.151 Further expansion of the hydrophobic core to include the e position gave the
peptide GCN4-pAA, which assembled to form a heptameric coiled coil.152 More recently,
Zaccai et al.153 utilised the relationship between expansion of the hydrophobic core and
higher oligomerization state to design the hexamer forming peptide CC-Hex by placing
hydrophobic residues at a, d and e positions. In coiled-coil forming peptides, the selective
placement of charged or hydrophobic residues at positions e and g allows a degree of
control over assembly to be engineered into the design.
Designed coiled coils have been extensively utilised as polypeptide scaffolds for the
production of catalytically active protein “maquettes”.154, 155 These structures generally
involve the assembly of tetrameric coiled coils 4-5 heptads in length that are designed
utilising similar principles to those already discussed, where hydrophobic interactions drive
helix association.156, 157 Control over helix orientation is generally provided by covalent
linkage of helix termini through disulfide bonds154, 158 or short polypeptide linkers.159-161
Functionality is incorporated by including residues within the core of the coiled coil that
bind redox-active metal ions or cofactors.154, 158, 160-163 Membrane association and
integration may also be designed into these maquettes. This is achieved by increasing the
hydrophobic residue content on the exterior of the coiled coils, and allows investigation of
trans-membrane processes in simple designed systems.164 While natural protein
engineering is hindered by a high degree of structural complexity, the simplicity of
maquette design facilitates easier and more rapid investigations of structure-function
relationships in molecular binding and catalysis.
In addition to the discrete peptide coiled coils discussed so far, several examples of coiledcoil fibrils have been reported. In these systems, coiled-coil forming peptides associate
18
with end-to-end overlaps to give longitudinal assembly (Figure 1-5). In some systems the
formation of physical cross-links between these coiled-coil fibrils can produce hydrogels,
similar to those of PA and β-sheet designs discussed previously. The majority of published
work on fibril forming α-helical peptides has been completed by the Woolfson group,28, 165172
however these peptides predominantly form insoluble fibrils rather than hydrogels. The
Woolfson group’s fibril forming designs are based on assembly of paired 28-residue
peptides such as SAF-p1 (H2N-K IAALKQK IASLKQE IDALEYE NDALEQ-COOH) and
SAF-p2 (H2N-K IRALKAK NAHLKQE IAALEQE IAALEQ-COOH). These peptides were
designed with an asymmetric distribution of charged residues at e and g positions along
the helix axis to promote staggered assembly. An asparagine residue was also included in
an a position of each peptide, which is only able to form structure-stabilising intermolecular
hydrogen bonds in an offset arrangement of helices (Figure 1-5A).165 SAF-p1/p2 peptides
and several subsequent generations of designed sequences self-assembled at neutral pH
to give insoluble rigid fibres of micrometres in length and ≥45 nm in width. These fibres
were highly ordered and resulted from the lateral association of peptide coiled-coil
filaments that were each expected to be ~ 2 nm in diameter.165, 166 The presence of salts
inhibited peptide assembly, demonstrating the importance of electrostatic interactions in
the design.165 Due to the destabilising placement of polar residues in the coiled-coil core,
peptide fibrils were thermally unstable and required low temperatures for assembly.165, 167
Assembled morphology was controlled by the addition of modified peptides designed to
introduce kinks168 and branches169 into the fibrils. More recent work by the same group has
given the peptides hSAFAAA-p1 and hSAFAAA-p2.28 These peptides follow the same basic
design as the SAF sequences, except that b, c and f positions of both peptides are
populated with alanine residues to reduce electrostatic interactions and promote
hydrophobic interactions between peptide fibrils. Assembly of these peptides gave thinner,
more flexible fibrils that were thermally stable and produce hydrogels at physiological pH
and salt. These hydrogel forming peptides have been investigated for biomedical
applications as is discussed in Section 1.3.2.
19
Figure 1-5. Schematic showing three design approaches for coiled-coil fibril formation: A) forced
offsetting of helices by placement of hydrogen bonding residues within the hydrophobic core and
asymmetric charged residue distribution,27, 28, 31, 165, 166, 168-173 B) end-to-end stacking of coiled coils
with offset helices,174 C) forced offsetting of helices by non-continuous hydrophobic faces of
helices.29, 175 Hydrogen bonds within the hydrophobic core are shown as lock-and-key
arrangements; horizontal lines within each peptide mark the ends of heptads.
20
A similar approach to specify helix offsetting and fibril formation was used in the peptide
designs of the Conticello group.27, 31, 173 The peptides YZ1 (Ac-E IAQLEKE IQALEKE
NAQLEKK IQALRYK IAQLREK NQALRE-CONH2) and TZ1H (Ac-E IAQHEKE IQAIEKK
IAQHEYK IQAIEEK IAQHKEK IQAIK-CONH2) were designed to form dimeric and trimeric
coiled-coil fibrils respectively. Both peptide designs utilised asymmetrical charge
distribution along the helix axis as well as multiple polar residues within the hydrophobic
core to specify helix offsetting. In the case of TZ1H, assembly was designed to be pH
dependent, with protonation of histidine residues at d positions causing assembly inhibition
at low pH.27 Further experiments showed these internal histidine residues to coordinate
silver ions that stabilised coiled-coil assembly.31 Assembly of both YZ1 and TZ1H gave
insoluble fibrils of ≥20 nm, showing a high degree of lateral association similar to the SAF
peptide designs.
Another system showing offsetting of helices was reported by Liu et al..152 The peptide
GCN4-pAA was observed to form a heptameric coiled coil with a single-residue shift in
registry between adjacent helices, resulting in an offset of a full heptad between helices 1
and 7. Helix offsetting was hypothesised to be driven by preferential packing of
hydrophobic residues at positions a, d, e and g as well as a ring of hydrogen-bonding
residues within the coiled-coil core. Xu et al.174 recently reported redesign of this peptide to
give 7HSAP1 (H2N-KLAQA VEKLARA VEKLAYA NEKLARA VEKLAQA VE-COOH). Selfassembly of 7HSAP1 resulted in staggered coiled coils that were stacked via end-to-end
overlap of helices (Figure 1-5B) to give thermostable coiled-coil fibrils that did not laterally
associate in solution. At lower concentrations, fibrils formed from 7HSAP1 were seen to be
unstable and showed breaks and partial dissociation. This is likely to be due to the small
amount of overlap afforded between coiled coils in this arrangement.
In another approach to helix offsetting, the Fairman group has reported the development of
peptides that use non-continuous hydrophobic faces to dictate end-to-end overlap (Figure
1-5C).29 This design utilised two copies of the first two heptads of the GCN4 sequence
separated by a two alanine residue linker to give the peptide CpA (Ac-CKQLEDK IEELLSK
AA CKQLEDK IEELLSK-CONH2). Incorporation of the alanine spacer results in an
approximately 200° rotational offset between the two-heptad modules, making these
peptides only able to assemble into a staggered conformation rather than flush-ended
coils. Addition of salts to enhance hydrophobic interactions was required for assembly of
21
these coiled coils. This was due to the short two-heptad overlap providing only a weak
driving force for association. Assembly in NaCl solutions gave nanofilaments of only a few
nm in diameter with lengths of several hundred nm, while addition of the divalent anion
sulfate gave laterally associated filaments. Further experimentation showed that
substitution of isoleucine at the first a positions of the CpA peptide increased the
hydrophobic driving force for association and promoted coiled-coil assembly at lower NaCl
concentrations.175
Several coiled-coil fibril forming peptides have also been reported in which no attempt was
made to specify helix offsetting in the original design. These designed peptides all have a
high degree of self-similarity in the constituent heptad repeats. For example, peptide α3
(H2N-(LETLAKA)3-COOH) reported by Kojima et al.176 and Takei et al.177 comprises three
exact heptad repeats, peptide αFPP (H2N-QLAREL(QQLAREL)4-COOH) and derivatives
reported by Potekhin et al.178 and Melnik et al.179 comprise four exact and one incomplete
heptad repeat, while Peptide 1 (H2N-EIKQLES EISKLEQ EIQSLEK-COOH) and
derivatives reported by Dong et al.180 possess three heptads with a slightly lower degree of
self-similarity. These peptides all form thin fibrils of <10 nm in diameter, with self-assembly
of peptides αFPP and Peptide 1 producing hydrogels at acidic pH. In studies of αFPP and
Peptide 1 derivatives, replacement of charged residues with uncharged residues in solvent
exposed positions resulted in laterally aggregated fibrils, likely due to reduced
intermolecular repulsive forces.179, 180 No detailed study has been made of the mechanism
of fibril formation in these peptides. However, one consequence of the self-similar heptads
is that helices shifted out of register by increments of a heptad will have similar inter-helical
interactions as those in an in-register arrangement. This may lower the energetic barrier to
end-to-end overlaps, possibly promoting fibril formation.178
α-Helix and coiled-coil forming peptides have been well studied, and many of the factors
affecting their assembly, such as residue patterning and α-helix propensity, have been
investigated. In many cases, this now allows for the de novo design of α-helix and coiledcoil forming peptides. This is demonstrated by the growing number of reported surfactant
peptides, protein maquettes and fibril forming peptides, as discussed above. Compared to
PA and β-sheet forming peptides, α-helix forming peptides that assemble to give fibrils and
hydrogels are less well developed. However, as described in this section, this is an area of
study attracting recent attention. α-Helix forming peptides, as well as PAs and β-sheet
22
forming designs, have demonstrated utility in several applications as surfactants, cell
scaffold materials and therapeutic delivery vehicles, as is now discussed.
1.3 Applications of peptide self-assembly
1.3.1 Surfactants
There has been growing interest in biosurfactants as alternatives to conventional
surfactants that are derived from petrochemicals. This interest is due to biosurfactants
lower toxicity, higher biodegradability and potential for renewable production.181, 182
Peptides in particular are of interest due to their stimuli responsiveness and designable
assembly.183 Due to the amphiphilic structure of the peptide designs discussed in Section
1.2, many of these peptides associate at protein surfaces and air/oil-water interfaces.
Peptide self-assembly at these surfaces and interfaces shows utility in applications such
as protein, foam and emulsion stabilisation, which are now discussed.
Amino-acid only PAs have been investigated for their ability to solubilise and stabilise
membrane proteins in solution. PAs such as Ac-A6K-CONH2 stabilised the photosynthetic
complex photosystem I for several weeks, thereby increasing its potential utility in devices
such as photodetectors and photovoltaic cells.184-186 Similar PAs were used to stabilise E.
coli glycerol-3-phosphate dehydrogenase187 and G-protein coupled receptor bovine
rhodopsin188 without causing structural damage or loss of activity. PAs increased protein
stability over several days relative to many conventional detergents such as the commonly
used octyl glucoside.187 The increased stability of membrane proteins solubilised by PAs
may potentially be due to the more membrane-like environment produced by the assembly
of these lipid-like peptides compared to conventional detergents. This increased protein
stability may accelerate study of membrane protein structure. More recent work has used
similar PAs to solubilise and stabilise membrane proteins such as olfactory receptors189
and several G-protein coupled receptors190 during cell-free production, enhancing the
yields of these otherwise difficult to produce proteins.
Lac21 is a facially amphiphilic α-helical peptide (Section 1.2.3) that associates at air/oilwater interfaces and lowers interfacial tension.132 The surfactant properties of a series of
reversibly cross-linkable α-helical peptides based on the sequence of Lac21 has been well
studied, predominantly by the Middelberg group.18, 19, 191, 192 Peptide AM1 was designed
based on the sequence of Lac21, where residues 9 and 20 were replaced with histidine
23
residues.18, 191 This positioning leaves the histidine residues facing the aqueous phase, but
oriented towards neighbouring helices at the interface. At neutral pH in the presence of
metal ions such as Zn(II), these histidine residues form intermolecular cross-links, resulting
in a mechanically strong interfacial film that stabilises emulsions and foams.18, 191 The
addition of two more histidine residues at positions 13 and 17 produced the peptide AFD4.
This resulted in histidine residue spacings of i + 3 and i + 4, which are compatible with
both intermolecular and intramolecular metal ion cross-linking.19 Under identical crosslinking conditions AFD4 gave stronger interfacial films and improved emulsion and foam
stability compared to AM1.19, 192 Film formation by both AM1 and AFD4 was reversible
through the addition of chelating agents or lowering pH to inhibit intramolecular crosslinking. Loss of film cross-linking resulted in rapid emulsion coalescence or foam
collapse.18, 19, 191, 192 Neutron reflectometry showed AM1 concentration at the interface is
not significantly altered by film formation or loss. This indicated that the switching of
emulsion and foam stabilisation is due to loss of the mechanical properties of the
interfacial film rather than surfactant desorption.18, 19, 192
The functionality of α-helical surfactant peptides possessing a high carboxylate content
has also been investigated. The peptide Lac21E was designed based on the Lac21
sequence,193 with the substitution of glutamate at positions 2, 3, 9, 10, 16 and 17 that are
oriented towards bulk aqueous solution. This peptide adsorbed to the air-water interface
and lowered interfacial tension as a function of molecular charge.194 At low pH, where
glutamate residues are partially neutralised, Lac21E readily adsorbed to the interface,
lowering interfacial tension and forming an interfacial film. Conversely, at high pH, where
glutamate residues are deprotonated, peptide desorbed from the interface and film
formation was lost.
The rate of adsorption of surfactant peptides at interfaces is dependent on the assembled
state of the peptide in bulk solution. As discussed in Section 1.2.3, the peptide Lac28
assembles into a tetrameric coiled coil more readily than the shorter Lac21. Middelberg et
al.132 reported that at moderate to high surface coverage of the octane-water interface,
Lac28 showed slower adsorption kinetics than Lac21 due to the requirement for Lac28
tetramer dissociation. Similarly, DAMP4, which was designed as a tetrameric miniprotein
of DAMP1 repeats, displayed much slower adsorption to the air-water interface than the
24
monomeric DAMP1 surfactant peptide. This was attributed to the requirement for
rearrangement of the larger and more structured DAMP4.195
There have been few studies of PA or β-sheet peptide assembly at air/oil-water interfaces,
with most reports instead focusing on their assembly in solution (Sections 1.2.1 and 1.2.2).
The few reports on the assembly of these peptides at interfaces are generally focused on
the formation of monolayers rather than their surfactant functionality.46, 50, 106 One report by
Dexter196 investigated the surfactant properties of a series of β-sheet peptides based on
the designs of Rapaport et al.106 with the sequence Ac-PX(FX)3P-CONH2, where X was an
ionic residue. By varying the charged residue composition and solution pH, the assembly
of these peptides at molecular-charge values of ±4 to 0 was assessed. At low molecular
charge, peptides adsorbed readily at the interface and formed interfacial films, presumably
by assembly of an ordered β-sheet. Lower peptide adsorption but better emulsification was
seen at higher molecular charge. This suggests these β-sheet peptides stabilise emulsions
primarily by conferring a high droplet surface charge rather than the film-formation
mechanism seen for the AM1 and AFD4 α-helical peptides.
The responsive self-assembly at interfaces of both the α-helical and β-sheet forming
peptides discussed here allow surfactant functionality to be rapidly “switched”. This can be
achieved by adjusting solution pH or the addition of a chelating agent, depending on the
peptide. This provides stimuli-responsive control over emulsion or foam stability, which is
of growing interest for many applications ranging from oil recovery and demulsification to
cosmetics and textile treatment.197, 198
As discussed here, designed self-assembling peptides show potential utility in several
applications such as protein, emulsion and foam stabilisation. One of the benefits of these
peptides over alternate biosurfactants is their designable and responsive assembly. These
peptides also possess the potential for renewable production using bacterial expression,
as discussed in Section 1.4. As well as assembly at interfaces, many self-assembling
peptides are capable of fibril and hydrogel formation (Section 1.2). These hydrogel forming
peptides have shown utility in several biomedical applications and are now discussed.
1.3.2 Cell scaffolds
Matrices that support cell growth are of interest for applications such as wound healing,199
tissue engineering200, 201 and modelling disease states in vitro.202 Commonly used cell
25
growth matrices are naturally sourced and include decellularised extracellular matrices,203,
204
collagen gels205, 206 or secreted protein mixtures such as Matrigel.207, 208 However, these
naturally-sourced matrices have several drawbacks such as the high potential for
immunogenic rejection, batch-to-batch variability and a lack of the ability to modulate their
characteristics.202, 209, 210 The use of defined synthetic matrices such as polymer- or
peptide-based hydrogels offers the potential to overcome many of these issues.206, 209, 210
Peptide-based materials are particularly appealing due to their biocompatibility,84, 85 stimuli
responsiveness211 and the ability to rationally design and control their self-assembly.28, 86
Hydrogels formed by PA and β-sheet peptides in particular have been well characterised
as cell scaffolds and are now discussed in detail.
Several peptide hydrogels based on β-sheet structures have been demonstrated to
support cellular attachment and growth in the absence of any further functionalization.84, 85,
95, 212
For example, hydrogels of the peptides EAK16 and RAD16-II supported the
attachment of a variety of primary cells and cell lines including fibroblasts and
keratinocytes.85 In subsequent experiments, hydrogels of RAD16-II and RAD16-I
supported extensive neurite outgrowth and active synapse formation by several neuronal
cell types.84 Additionally, hydrogels of the peptides PFD-5 and MAX1 were reported to
support the attachment and proliferation of osteosarcoma and fibroblast cells
respectively.95, 212 Furthermore, MAX peptides are also antibacterial, with efficacy against
both gram positive and negative bacteria, potentially increasing the utility of these peptides
as wound dressings.213, 214
To date, the most widely studied self-assembling peptides for cell culture matrices are the
RAD16 peptides developed by Zhang and colleagues. RAD16-I peptide hydrogels have
now been commercialised and are sold under the trade name Puramatrix by Corning. This
has expanded the ease of access to this defined substrate for ongoing research.
In a cells native environment within the body it is surrounded by the extracellular matrix
(ECM). The ECM is a complex material composed primarily of a mixture of proteins and
polysaccharides that provides structural support as well as cell signalling and adhesion
domains.215, 216 Common components of the ECM include fibronectin, laminin, collagen
and vitronectin.216 These macromolecules include specific integrin recognition motifs
involved in cellular attachment. While some peptide hydrogels are able to support cellular
attachment and proliferation with no further functionalization (above), others such as those
26
formed by PAs55 and peptide Q11217 show poor cellular attachment and proliferation. One
common strategy used to improve cellular interactions with scaffold materials such as
peptide hydrogels is to functionalise these materials with cell-recognition motifs such as
the fibronectin-derived RGD and RGDS or the laminin-derived YIGSR and IKVAV
sequences.21, 55, 59, 217-225
The functionalization of alkyl-tail PA nanofibres by incorporating RGDS and IKVAV
sequences in the hydrophilic head has been extensively explored.21, 55, 59, 218-221 For
example, PAs functionalised with a variety of RGDS ligands supported the attachment and
proliferation of fibroblasts, with the more mobile branched RGDS motifs showing enhanced
cell adhesion and migration.219 As discussed in Section 1.2.1, the functionalization of PAs
partially destabilises assembly, however mixtures of PAs functionalised with RGDS motifs
and non-functionalised PAs gave recovery of material properties and supported the
proliferation of bone marrow derived stem cells.221
Hydrogels of functionalised Q11 peptide have also been investigated as defined
substrates for cell culture. Incorporation of Q11 peptides functionalised at the N-terminus
with RGD or laminin-derived IKVAV and YIGSR motifs did not significantly affect the
assembly of the peptide at all ratios investigated. While poor cellular attachment and
proliferation was seen on hydrogels of Q11 alone, hydrogels that incorporated
functionalised peptide supported the attachment and proliferation of fibroblasts and human
umbilical vein endothelial cells.217, 223-225
The effect of functionalising the C-terminus of RAD16-I with a variety of biologically active
motifs has been assessed for several cell lines.222, 226-228 While the functionalization of
RAD16-I with short motifs did not interfere with self-assembly, longer sequences of more
than 6 residues significantly decreased assembly and required the incorporation of nonfunctionalised peptide to recover material properties.226 The functionalization of RAD16-I
with several physiologically relevant motifs enhanced the attachment and proliferation of
murine neural stem cells227 and precursor osteoblasts.222 Additionally, primary human
fibroblasts proliferated and were stimulated to produce their own ECM when cultured on
hydrogels of RAD16-I functionalised with RGD and laminin adhesion motifs.228 In each of
these examples, cells seeded on hydrogels of functionalized RAD16-I infiltrated the
peptide scaffold, demonstrating the potential of these gels in wound healing applications
where cells from surrounding tissues may migrate into the peptide matrix.222, 227, 228
27
Modification of peptide hydrogels can also be targeted towards controlling material
degradation rates. For example a series of MAX peptides were designed to include
metalloproteinase cleavable sites to allow cellular degradation of the gels over time, which
enhanced cell proliferation and migration into the gels.229 Variation of the cleavage
susceptible site gave altered rates of digestion, allowing materials to be designed with
specific rates of degradation. Notably, incorporation of cleavable sites also decreased
peptide self-assembly, and it is likely that the folded or partially unfolded state of the
peptide will also influence proteolytic degradation.
There has been significant work done towards the functionalization of peptides to enhance
cellular interactions. However, it is still unclear as to why some peptide systems such as
PAs and Q11 peptides appear to require this functionalization for cellular attachment and
proliferation, while others do not. It is expected that materials will become non-specifically
functionalised with adsorbed proteins when they are exposed to biological systems such
as serum containing media used in tissue culture experiments.201 These adsorbed proteins
may facilitate cellular attachment in lieu of any specific functionalization. It is therefore
possible that protein adsorption, and consequently cellular attachment, occurs less readily
on the materials that require specific functionalization. However further experimentation
will be required to determine the exact reason for poor cellular attachment to these
surfaces.
The hydrogels of modified PA, RAD16-I, Q11 and MAX peptides discussed here highlight
the potential for design of these self-assembling peptides to give enhanced cellular
interactions with these defined cell scaffold materials. Appropriate biologically active
sequences are able to be selected to enhance interactions with specific cell types (above).
In some cases the incorporation of biologically relevant motifs can interfere with peptide
self-assembly. However, it is possible to overcome this by incorporating a proportion of
non-functionalised assembling peptide.219, 221, 222, 226, 228 The optimal proportion of nonfunctionalised peptide is a factor that is expected to be sequence dependent and will likely
need to be optimised empirically for each self-assembly peptide and functional motif.
The cell microenvironment can play an important role in determining cellular proliferation,
morphology and response to therapeutic drugs.230-234 The culture of cells in 2D or 3D
environments is a strong influence in determining cell responses and proliferation rates
due to factors such as drug penetration, nutrient diffusion and cell-cell contacts. The
28
culture of cells within 3D matrices rather than on 2D substrates is therefore an important
step in providing in vitro models that better mimic the conditions experienced by cells
within their native environment, as well as characterising cellular responses to scaffold
materials in 3D prior to in vivo work.202
The culture of cells encapsulated within self-assembling peptide hydrogels has been a
target of ongoing research. For example, RAD16-II peptide gels stimulated angiogenesis
of encapsulated human micro-vascular endothelial cells,235 while studies of RAD16-I
hydrogels showed them to support the growth and differentiation of encapsulated neuronal
stem cells236, 237 and hepatocyte progenitor cells.238 Hydrogels of alkyl-tail PAs
functionalised with RGDS or IKVAV recognition sequences supported the proliferation of
encapsulated bone marrow derived stem cells221 or mouse neural progenitor cells55
respectively. In separate studies, hydrogels of MAX8 supported the growth of
encapsulated primary bovine chondrocytes.239 Furthermore, redesign of MAX8 to give
peptides with lower positive charge produced similar hydrogel assembly, but enhanced
growth of encapsulated cells, indicating that a less charged matrix may benefit cellular
proliferation.239 Collier and colleagues have also reported hydrogels of the peptide Q11 to
support the encapsulation and growth of fibroblasts, as well as murine stem cells when
functionalised with RGD motifs.240 Hydrogels of several P11 type peptides, including P11-4,
were reported to support the growth of encapsulated fibroblasts.241, 242 Additionally, due to
the benign methods used to trigger crosslinking in these peptide systems, gel formation is
generally able to occur in the presence of cells. This allows the cells to be homogeneously
distributed throughout the gel matrix, producing a more uniform material and cellular
environment.55, 86, 235-240 These results demonstrate hydrogels formed by designed selfassembling peptides to be promising materials for the culture of several cell types with
potential as both in vitro model systems and in regenerative medicine applications.
However, further clinical studies will be required to determine how these materials may
translate to model systems and human treatments.
It has been shown that cell growth and function are impacted by the mechanical properties
of the underlying substrate.243-246 Peptide hydrogels are generally relatively weak
compared to chemically cross-linked materials and have elastic moduli in the range of
pascals to kilopascals,28, 87, 91, 247 which may limit their utility to soft tissue applications.248,
249
The network density and resulting hydrogel mechanical properties can be controlled by
29
the concentration of self-assembling peptide.236, 250, 251 This is exemplified by peptide
MAX1, which at concentrations of 0.7, 1.4 and 2.1 wt% produced hydrogels with elastic
moduli of approximately 0.25, 1.6 and 4 kPa respectively.250 These moduli are comparable
to soft tissues of the body such as the lymph node and brain, and approach the moduli of
liver and breast tissue at higher concentrations.248 Solution conditions such as pH, ionic
strength and temperature may also be used to control hydrogel mechanical properties.87, 91
However, these factors are generally pre-determined by requirements for cell growth and
are therefore less useful for specifying mechanical properties of cell scaffold materials.
The type of cross-links between fibrils can influence the resulting mechanical properties.
For example, hydrogels formed by hSAF peptides designed to form fibril-fibril hydrogenbonds melted on warming to physiological temperature, while those stabilised by
hydrophobic interactions were strengthened.28 Additionally, the formation of more specific
interactions between fibrils can increase mechanical strength. For example, the elastic
modulus of hydrogels formed by the peptide PLE-5 at 1.8% (w/v) was as high as 60-90 kPa
when Ca2+ was included at the time of gelation, likely due to the formation of coordinated
ion crosslinks.212 Covalent cross-linking of peptide hydrogels by chemical217, 252or
enzymatic253 means can also be used to increase mechanical strength. Jung et al.217
showed cross-linking of 4.6 % (w/v) Q11 peptide hydrogels using native chemical
ligation254 to increase elastic moduli approximately five-fold. The mechanical properties of
peptide hydrogels may therefore be selected for individual applications by adjusting
peptide concentrations or by forming more stable cross-links between assembled fibrils.
The physical nature of the cross-links in non-covalently cross-linked peptide hydrogels
also means they are generally shear thinning and self-healing, meaning these materials
are injectable, enabling simpler delivery in biomedical applications.86, 87, 90
The success of peptide hydrogels in in vitro studies has led to interest in their use for
tissue engineering applications. However, few of these systems are yet to be
characterised in vivo. For example, while in vitro characterisation of MAX1 and MAX8
showed these peptides to be non-cytotoxic and not elicit macrophage activation,96 to date
no in vivo testing of these hydrogels has been reported. In comparison, the EAK16 and
RAD16 peptides developed by Zhang and colleagues have been much better
characterised in vivo. Preliminary studies showed that injection of EAK16 and RAD16-II
peptide hydrogels into rabbits and rats gave no toxic effects or immunological response,
30
indicating high biocompatibility.84, 85 This led to further studies of these hydrogels in several
in vivo models.
Studies of RAD16-II and RAD16-I hydrogels in burn and diabetic wound models showed
hydrogels to enhance angiogenesis and wound closure, demonstrating the potential of
these hydrogels in wound healing applications.255, 256 Characterisation of RAD16-I
hydrogels in neural injury and spinal cord damage models showed improved tissue and
axon regeneration while also reducing scar formation.20, 247, 257 In one study of damage to
the optic tract of hamsters, axon growth through hydrogels at the wound site gave
functional recovery in the majority of treated animals.20 Furthermore, encapsulation and
delivery of neuronal stem cells inside RAD16-I hydrogels enhanced regeneration of
damaged rat brain tissue.247 These studies demonstrate the potential of these peptide
hydrogels to both enhance tissue regrowth and deliver stem cells in neural wound healing
applications, however further clinical studies will be required to determine how this efficacy
may translate to humans. Alkyl-tail PAs have also been characterised by the Stupp group
for their utility in neural wound healing models. Hydrogels of these peptides showed
promising results in a murine spinal cord damage model21 and also delivered encapsulated
neural progenitor cells to these injury sites to enhance tissue regeneration.218 The utility of
peptide hydrogels is increased in many of these tissue engineering models by the ability to
shear thin preformed hydrogels or trigger gel formation in situ. This facilitates the injection
of the peptide material in a liquid state allowing the material to fill gaps and enhance
subsequent integration with host tissue.21, 86, 257
Despite the success of RAD16 hydrogels in in vivo trials, the progression of this material to
clinical trials has been limited, with only one reported study investigating this materials
efficacy as a haemostatic agent.258 One potential reason for this slow progression may be
that there remain concerns about the applicability of β-sheet based materials for clinical
use. One reason for this is that β-Sheet fibrils possess structural and potentially
pathological similarities with misfolded protein aggregates implicated in several
neurodegenerative and amyloid diseases such as Alzheimer’s disease, Parkinson’s
disease and prion diseases including spongiform encephelopathies.259-261 Progression of
several of these diseases appears to involve the induction of native protein unfolding by a
β-sheet template, and there exists the potential for materials formed by β-sheet peptides to
initiate such misfolding. Indeed, experiments by Westermark et al.262 showed induction of
31
AA-amyloid deposits in mice during inflammation following subcutaneous injection of
amyloid-like fibril forming peptides, including RAD16-I. While these experiments
demonstrate the potential for amyloid nucleation by these peptides, how these results may
translate to human risks is currently unknown. Furthermore, the long lead time for many of
these neurodegenerative diseases will make such health risks difficult to rule out.263, 264
In addition to their implicated role in disease pathogenesis, recent studies have also
shown several examples in which the formation of amyloid structures is part of normal
biological functions for many organisms, ranging from micro-organisms to humans.
Examples of functions carried out by amyloid structures include biofilm formation,
regulation of catabolic pathways and sequestration of toxic intermediates.265-267 Given the
diverse roles played by amyloid structures, exactly how amyloid structures formed or
nucleated by peptides may induce disease or influence native biological processes is likely
a complex question which will require further study to gain a better understanding. While
there still remains questions regarding the clinical application of amyloid-like fibril-forming
peptides, one alternate, and possibly safer, route to self-assembling peptide hydrogels is
to avoid these peptides and instead utilise α-helical peptides that form coiled-coil fibrils.
A limited number of hydrogels formed from the assembly of α-helical peptides have been
reported to date (Section 1.2.3). Of these, few have been assessed for their ability to
support cell growth, largely due to hydrogel assembly not occurring under physiological
conditions. The hSAFAAA peptides designed by the Woolfson group form hydrogels under
conditions of physiological pH and salt.28 However, these hydrogels were unstable in cell
culture media. A substitution of one alanine residue for a tryptophan residue was required
to increase hydrophobic interactions to produce gels stable enough for cell culture
experiments. These hydrogels supported the growth of rat adrenal pheochromocytoma
cells at levels similar to that of Matrigel controls. While this demonstrated the potential for
α-helical peptide hydrogels to support mammalian cell growth, this tumour-derived cell line
is likely to be less demanding in terms of growth conditions than non-tumour-derived
alternatives.268 Investigation of the growth of cells with a more native phenotype will be
required to better determine the utility of these materials for many applications. More
recently, a cyclic coiled-coil peptide Cycl_one was reported that utilises a similar approach
to helix offsetting as the SAF peptides. Assembly of this peptide gave mesoscopic fibrillar
networks that did not gel but supported the growth of human dermal fibroblasts.
32
Proliferation on these surfaces was further enhanced by the incorporation of the lamininderived YIGSR recognition motif.269
Hydrogels of self-assembled peptides have demonstrated utility as cellular scaffolds for
both in vitro and in vivo use. As discussed in this section, several alternate peptide
designs, which are predominantly PA and β-sheet based, have been investigated with
promising results. However, there is no standard approach to testing these materials and a
variety of cell types, indicators of cytocompatibility and measures of proliferation have
been used that complicate comparisons between systems. The general outcomes from
this body of work suggest that these relatively weak matrices are best suited to soft tissue
applications, with several reports showing promising results for neural tissue engineering
in particular. In addition to their use as cell scaffold materials, self-assembling peptides
have shown utility in other biomedical applications such as therapeutic delivery, as is now
discussed.
1.3.3 Therapeutic delivery
There is a large body of ongoing research towards the development of more effective,
targeted and controlled therapeutic delivery systems.270-272 Self-assembling peptides are
promising materials for therapeutic delivery vehicles. They possess advantages due to
their biocompatibility,84, 85, 96 stimuli responsiveness31, 40 and the possibility to rationally
design their assembly.23, 28
Their amphiphilic structure and tendency to form micelles and liposomes make PAs
attractive targets for therapeutic delivery systems.23, 40, 273-275 An example of this are PAs
with an alanine tail and the cationic H5K15 head group that form micelles capable of binding
DNA for gene delivery applications and show similar efficacy to polyethylenimine.40 In this
design the lysine residues are utilised to bind DNA, while histidine residues are included to
provide pH-responsiveness upon endosomal uptake, causing DNA release. Incorporation
of the more hydrophobic residue phenylalanine in the PA tail gave peptide FA32, which
assembles into micelles that are also capable of loading the chemotherapeutic
doxorubicin.23 Micelles of FA32 co-delivered both encapsulated doxorubicin and bound
DNA in in vitro studies. This demonstrates the potential for these PAs to be used as a dual
therapeutic delivery system.23, 40 However, to date little work has been done towards
33
functionalization of such micelles for targeted delivery, which would enhance their
specificity and efficacy for in vivo applications.270
Peptide-based scaffolds are well suited for the delivery of therapeutics and can be
designed to give controlled release of proteins and small molecules. The ability to
assemble peptide scaffolds in situ also enables co-injection of bioactive molecules and
self-assembling peptides to achieve localised delivery of therapeutics. Hydrogels formed
by functionalised alkyl-tail PAs have been developed for growth factor and small molecule
delivery. For example, heparin-binding PAs were produced by incorporation of the binding
sequence (LRKKLGKA) in the hydrophilic head.276 Heparin is a natural biopolymer known
to interact with several growth factors, and when bound to PA nanofibres may present and
release these molecules to promote bioactivity. Hydrogels formed by heparin-binding PAs
have been characterised in both in vivo and in vitro studies on angiogenesis,276, 277 islet
transplantation278, 279 and cardiovascular disease280 with positive results. In another
approach for therapeutic delivery from PA hydrogels, attachment of the anti-inflammatory
drugs nabumetone or dexamethasone to PA nanofibres through a degradable hydrazone
linkage gave slow controlled release under physiological conditions281, 282 and showed
promise in in vivo testing.282
Nanofibres formed by an alkyl-tail PA have also been used to physically encapsulate the
chemotherapeutic doxorubicin for delivery.283 In this case, nanofibre disassembly and
therapeutic release was triggered via enzymatic modification of the PA by an extracellular
cancer biomarker. Doxorubicin was also reported to associate with the β-sheet peptide
PFD-5 designed by the Rapaport group.284 Doxorubicin enhanced folding of PFD-5 in
solution and was incorporated into peptide hydrogels. Sustained release of the
chemotherapeutic occurred over several days with demonstrated cytotoxicity to an
osteosarcoma cell line. Macroscopic disintegration of the hydrogels was observed in some
cases, possibly due to the Ca2+ used to trigger gelation leaching from the gel. This gel
instability is a barrier to the application of these gels in vivo, where reproducible material
stability is required.
Hydrogels of the fibril-forming RAD16 and MAX peptides have also been investigated for
therapeutic delivery.22, 251, 285-289 For example, it was shown that the release of physically
encapsulated small molecules from RAD16-I hydrogels was dependent both on peptide
concentration and the strength of electrostatic interactions between model compounds and
34
the peptide scaffold.285 Similar gels also showed controlled release of physically
entrapped proteins such as lysozyme, trypsin inhibitor, bovine serum albumin and
immunoglobulin G without causing denaturation of loaded biomolecules.286 The rate of
biomolecule release was dependent on both protein size and peptide scaffold
concentration. Furthermore, Gelain et al.287 showed that the addition of C-terminal charged
sequences to RAD16-I peptides altered electrostatic interactions with loaded molecules.
Design of such peptide-therapeutic interactions was utilised to give controlled release of
active cytokines from hydrogels over several days. More recently, Koutsopoulos et al.288
reported the long-term release of human antibodies from layered peptide scaffolds of
RAD16-I and the related peptide Ac-(KLDL)3-CONH2.
MAX peptide hydrogels showed controlled release of positively charged and neutral
biomolecules such as lysozyme and immunoglobulin G, with the rate of release governed
by steric interactions with the network.251, 289 Conversely, negatively charged biomolecules
interacted strongly with the positively charged peptide matrix and were released at much
lower levels.289 Altunbas et al.22 have also reported the controlled release of the antitumorigenic molecule curcumin from MAX8 peptide hydrogels. Released therapeutic was
demonstrated to maintain bioactivity in in vitro assays. These examples demonstrate that
the control of interactions between loaded molecules and peptide hydrogels may be used
to regulate release rates. This has the potential for future applications where therapeutic
release may be modulated depending on specific clinical requirements.
Nanofibres formed by the β-sheet peptide Q11 have been investigated as vaccine delivery
vehicles.290-294 In these systems, covalent linkage of epitopes or whole proteins to a short
linker sequence at the termini of the peptide allowed display of these molecules at high
density on peptide fibres. Importantly, the attachment of these molecules did not inhibit
Q11 peptide assembly. When investigated in vivo the peptide fibres were seen to act as an
adjuvant, generating strong immune responses to displayed vaccine molecules such as
malarial epitopes290 that in some cases lasted for over one year, while no immune
response was seen to the peptide itself.291
The use of α-helical peptides in therapeutic delivery applications has been less well
studied than PA or β-sheet peptides. However, their designable assembly as well as the
potential for hydrophobic drug encapsulation295-297 make α-helical peptides attractive
targets for therapeutic delivery. For geometric reasons, higher-order coiled coils contain an
35
internal tubular cavity, as observed in heptamer,152, 174 hexamer,153 pentamer298 and even
tetramer structures.145, 298 The binding of hydrophobic biologically active molecules alltrans retinoic acid, vitamin A and vitamin D3 has been shown within such a hydrophobic
channel of the COMPcc pentameric coiled coil.295, 296 This demonstrates the potential for
encapsulation and delivery of these molecules utilising coiled-coil forming peptides.
Coiled-coil forming peptides have also been investigated as epitope presenting molecules
for vaccines. For example the Burkhard group has reported peptides that consist of two
linked coiled-coil domains that assembled into peptide nanoparticles.299-301 Similar to the
Q11 fibres discussed previously, the functionalization of these peptides with epitopes
elicited specific immunogenic response to the repetitively displayed antigens. This
demonstrates the potential of this system for use in vaccine production.
Self-assembling peptides show potential in applications for therapeutic delivery both as
small micelle delivery vehicles and scaffold materials for controlled release. These scaffold
materials also have demonstrated applications as epitope presenting materials for
vaccines. Similar to the trends observed in cell-scaffold materials, the majority of work in
this space has utilised either PAs or β-sheet forming peptides, with few α-helix forming
peptide systems thus far reported. However interest in α-helical based systems is
expanding.
While this section has outlined potential applications of self-assembling peptides ranging
from emulsion and foam stabilisation to tissue engineering and therapeutic delivery, there
still remain limitations in translating these materials to commercial products. One key
limitation to the adoption of these materials in many applications is the cost of peptide
production. Current production methods are centred on high-cost solid-phase synthesis
methods. However the biological expression of designer peptides may offer a cheaper and
more renewable alternative, as is discussed in Section 1.4.
1.4 Peptide production
1.4.1 Chemical synthesis
The majority of peptides reported to date for practical applications have been prepared by
solid-phase chemical synthesis. This technique was originally introduced by Merrifield and
involves the stepwise addition of amino acids to a surface bound oligopeptide.302 The
current state of development of these techniques has been reviewed extensively24, 303 and
36
only a brief summary is included here as a complete review of this field falls outside the
focus of this project. In a typical solid-phase synthesis process, each amino acid is
temporarily protected at the N-terminal amino group. In each step the N-terminal protecting
group of the bound oligopeptide is removed before the next sequential amino acid is
introduced for addition. Amino acid side chains are also protected during synthesis in order
to avoid undesired side reactions and are deprotected following recovery of the peptide
from the resin. The surface attachment of the growing peptide chain allows removal of
non-reacted derivatives and coupling reagents via simple washing procedures. Solidphase peptide synthesis is generally straightforward and may be automated to enable the
synthesis of desired peptide sequences by non-specialist chemists.
While this technique offers versatility in the range of sequences able to be synthesised, it
does possess several key drawbacks. For example, there still exist technical difficulties in
synthesising long oligopeptides (>20-30 amino acids) at large scale.24 Additionally, largescale production of peptides is expensive and has the potential for significant
environmental impact due to the generation of toxic by-products and large volumes of
chemical waste.24, 304, 305 While high-value products such as pharmaceuticals and
biomedical devices can be feasibly produced by solid-phase synthesis, adoption of
peptides for large-scale material applications is only viable if they can be produced at low
cost. Biological expression of designed peptides offers a production methodology that may
be able to achieve renewable and economically viable production at large scale.306
1.4.2 Biological expression
While bioproduction is an appealing method for peptide production, the expression of short
peptides is a non-trivial process. Peptides that are expressed in isolation are generally
unstructured and therefore susceptible to proteolysis in the intracellular environment.32, 307
One common approach to producing a more stable construct for expression is to couple
peptides to a larger carrier protein. This approach facilitates expression of the peptide as
part of a folded fusion protein and minimises proteolysis.32 Several detailed reviews exist
of available fusion protein partners of which only a few are highlighted here.308, 309
Commonly used fusion partners for soluble product expression include glutathione stransferase (GST, 26 kDa), maltose-binding protein (MBP, 42kDa), thioredoxin A (12 kDa)
and small ubiquitin-related modifier (SUMO, 12 kDa). These highly expressed proteins
enhance production, protect the fused peptide from proteolysis and increase product
37
solubility. Peptides may also be attached to a carrier protein such as ketosteroid
isomerase (KSI, 13 kDa), which forms insoluble inclusion bodies that protect products from
intracellular degradation and can simplify recovery after expression. One strategy to
enhance recovery following expression is to incorporate short sequences such as
polyhistidines, antibody epitopes or streptavidin binding tags as part of the fusion protein
construct.308 These tags enhance product recovery by enabling the use of affinity
chromatography to recover the fusion construct. The fusion partners GST and MBP also
possess the added advantage that these facilitate the use of affinity chromatography
directly, without the need for additional affinity tags.308
Expression of peptides as fusion proteins is an easily adaptable method that enables
biological production of short peptides. However, due to the relatively small size of most
peptides compared to the carrier proteins, the target sequence only comprises a small
portion of the expressed product (ca. 3-15% by weight for a typical 2 kDa peptide). As the
total amount of heterologous protein that can be produced by a cell is finite, this low weight
fraction becomes a limiting factor in production yields. One effective approach used to
increase the proportion of target peptide in the expressed product is the incorporation of
multiple copies of the target sequence.310-313 However, the number of incorporated repeats
cannot be increased indefinitely, and the higher repeat number constructs tend to display
lower expression.310-313 The optimal repeat numbers for expression must therefore be
determined empirically for each expression system. The exact reason for decreased
expression of higher repeat number constructs is not clear. However, it may relate to the
instability of high repeat number expression plasmids and long RNA transcripts in the
cell.312, 314
To further enhance potential bioproduction yields, peptides may be expressed as tandem
repeats or “concatemers” in the absence of a fusion partner.313 In this approach the target
peptide comprises almost the entire expressed product. Similar to multiple copies in fusion
proteins, this approach also requires empirical optimisation of repeat numbers.313-315
Furthermore, expression of such designed constructs will generally succeed only if the
resulting concatemer can form a stably folded structure. Targeted design approaches such
as those outlined in Section 1.2 may be utilised to give peptide and concatemer
sequences that will fold into stable structures for expression.316-318 Both theoretical
considerations and a lack of examples in the literature suggest tandem expression of
38
highly cationic or anionic peptides to be unfavourable. Expression of these highly charged
sequences is constrained by charge destabilisation preventing stable folding within the
cell, and potential host toxicity in the case of cationic sequences.315, 319 At the same time,
highly charged peptides are valuable targets as functional surfactants and antimicrobial
peptides. In some cases the pairing of highly-charged peptides with oppositely-charged
expression partners to form a less charged complex has been used to overcome these
effects.315, 319
Several factors influencing expression in the host organism can also be accounted for in
the design of expression constructs and their coding sequences.320, 321 Variation in residue
composition to avoid overloading amino acid biosynthesis pathways can enhance the
potential for protein overexpression.322, 323 Optimisation of the nucleic acid sequence
coding for the expression construct may be used to avoid undesired restriction, promoter
or ribosomal binding sites,320 while balancing of relative nucleotide content can also be
used to enhance oligonucleotide construct stability.320 Heterologous protein expression
may also be enhanced by matching codon usage frequency to the host expression
organism, as rare codon usage may decrease translation rates.324 Avoiding sequence
repeats can also reduce undesired nucleic acid structure that may otherwise lower
expression.320 Several algorithms exist for optimisation of nucleic acid sequences that
greatly simplify this process when designing expression construct coding sequences.320, 324
In both fusion protein and concatemer expression, cleavage of the expressed construct is
required to give functional monomeric peptides. This can be accomplished by either
enzymatic or chemical cleavage methods. Table 1-1 lists common cleavage approaches
and their associated cleavage sites. Both enzymatic and chemical cleavage methods
require design of the expression construct sequence to both include targeted cleavage
sites, and avoid undesired cleavage within the target peptide. The functionality of the
peptide target is generally highly dependent on amino acid sequence (Sections 1.2 and
1.3). The selection of cleavage methods that minimise the required sequence
modifications are therefore necessary to preserve functionality of the expressed peptide.
39
Table 1-1. Cleavage sites and resulting peptide terminus modifications of common enzymatic
and chemical cleavage methods
Cleavage agent
Type
Cleavage site
Terminus modifications
Enterokinasea
Enzymatic DDDDK|X
-
Factor Xab
Enzymatic I(E/D)GR|X
-
(X≠R or P)
SUMO proteasec
Enzymatic (SUMO protein)|X
-
(X≠P)
Tobacco etch virus
Enzymatic ENLYFQ|(S/G)
-
Thrombine
Enzymatic LVPR|GS
-
3C proteasef
Enzymatic ETLFQ|GP
-
Hydroxylamineg
Chemical
(TEV) proteased
N|G
C-terminal
hydroxamate
CNBrh
CDAP/NTCBi
Chemical
Chemical
M|X
C-terminal homoserine
(X≠S or T)
lactone
X|C
Iminothiazolidine
modified cysteine
Acid and heatj
Chemical
D|P
-
Where | represents cleavage location and X represents any amino acid except as noted. a Hosfield
and Lu.328 b Nagai and Thogersen.329 c Malakhov et al..330 d Dougherty et al.331 and Carrington
and Dougherty.332 e Chang.333 f Cordingley et al..334
h
g
Bornstein and Balian335 and Smith et al..336
Kuliopulos and Walsh310 and Riley et al..311 i Tang et al.337 and Wu and Watson.338 j Li et al.339
and Zhang et al.340
40
Enzymatic methods are generally higher cost and require the incorporation of longer
recognition sequences than chemical cleavage alternatives (Table 1-1). Furthermore,
proteases such as enterokinase and factor Xa can give non-specific cleavage, thereby
lowering peptide recovery.325, 326 However, enzymatic cleavage generates free amino acid
termini and requires relatively benign cleavage conditions, thereby limiting amino acid
modifications during processing. Chemical cleavage is generally a simpler and cheaper
option; however, cleavage agents such as hydroxylamine, cyanogen bromide (CNBr), 1cyano-4-dimethylaminopyridinium tetrafluoroborate (CDAP) and 2-nitro-5thiocyanobenxoic acid (NTCB) produce non-native amino acid termini (Table 1-1).
Furthermore, side-reactions and the harsher conditions used in chemical cleavage
methods can result in amino acid modifications such as deamidation and oxidation.327
Environmental concerns with toxic waste production can also limit the use of reagents
such as CNBr in large-scale production. The scale of production, requirements for product
peptide sequence and sensitivity to chemical modifications will therefore influence
cleavage method selection during expression construct design.
Purification of cleaved peptides may be achieved by several methods depending on
factors such as cost, scale and purity required. The majority of published methods for the
purification of expressed peptides utilise reversed phase, affinity or ion-exchange
chromatography, either in single or sequential steps.274, 306, 311, 312, 317, 341-343 This is due to
the high level of purity attained with these methods as well as technique simplicity and
adaptability for varied peptides. While these methods are effective at the laboratory scale
for early peptide characterisation, chromatography-based approaches add significant cost
to overall peptide production and are impractical for large-scale processes.344, 345 An
alternative approach is to utilise chromatography-free methodologies, which use selective
precipitation steps induced by changes in solution pH, salt concentration or
temperature.346-348 These methods utilise knowledge of the target peptide physicochemical
properties to achieve separation from other cellular material. These techniques offer the
potential for low-cost purification of expressed peptide and are easily scalable. However,
the level of purity achieved in these methods may be lower than that achieved with
conventional chromatography.345-348 While this may be acceptable for industrial
applications, the production of therapeutic peptides necessitates the higher purity currently
only achievable with chromatography.
41
The biological production of only a few self-assembling peptides has been reported to
date. The majority of these have utilised a fusion protein approach and involved recovery
via chromatography based methods as described above. For example, peptide
amphiphiles PA2 and PA7 were expressed as SUMO fusion partners. Cleavage via SUMO
protease gave monomeric peptides that self-assembled to form vesicles.274 Reed et al.341
reported expression of a tetrameric RAD16-I construct as a cellulose-binding domain
fusion protein. A glutamic acid residue was incorporated at the C-terminus of each peptide
to enable endoproteinase GluC mediated cleavage that produced monomeric peptide that
assembled similarly to synthetic RAD16-I.
The expression of P11 peptides has been reported in several instances. Hartmann et al.312
first reported the expression of tandem repeats of P11-2 separated by cysteine residues as
KSI fusion proteins that were cleaved by CDAP. In this case the single peptide repeat
gave the highest P11-2 recovery. However, the self-assembly of the expressed peptide was
not reported. In another approach used by the Middelberg group, P11-2 was expressed as
a thioredoxin fusion protein that was cleaved by TEV protease. Chromatography-free
processes gave recovery of expressed peptide that self-assembled similarly to synthetic
P11-2.347 P11-4 has also been expressed by the Aggeli group as either tandem- or singlerepeat KSI fusion proteins with methionine residues inserted for cleavage with CNBr.311, 348
Recovery of the single-repeat fusion protein was achieved using chromatography-free
methods.348 Of the recovered peptides, only P11-4 produced by the single-repeat fusion
protein assembled similarly to synthetically produced P11-4. This was due to the C-terminal
homoserine lactone formed during cleavage that altered the net peptide charge and
modified the assembly of peptides produced as tandem repeats.311
Self-assembling α-helical peptides have also been biologically produced. For example,
studies of the coiled-coil fibril forming peptide α3 used peptide expressed as an adenylate
kinase fusion protein that was cleaved by CNBr.342 The biological production of the αhelical peptide surfactants Lac21 and AM1 has also been investigated. Lac21 was
expressed as a KSI fusion protein that had terminal cysteine residues incorporated for
cleavage with CDAP.306 Surfactant functionality of the product peptide was lower than that
of synthetically produced peptide due to terminus modifications resulting from the chemical
cleavage methods used. Kaar et al.343 reported expression of the surfactant peptide AM1
as an MBP fusion protein with a glycine residue incorporated at the N-terminus (GAM1) to
42
enable cleavage by TEV protease. Recovered peptide showed similar interfacial properties
to synthetically produced GAM1. However, a comparison to AM1 peptide was not
reported. DAMP4 was then designed as a tetrameric coiled-coil concatemer of the AM1
sequence separated by acid-cleavable linkers which was recovered using a
chromatographic approach.317 More recent work by the same group has produced a
chromatography-free process for recovery of “industrial grade” DAMP4.346 However,
deamidation of asparagine residues occurred during cleavage of the expressed protein
that affected the interfacial properties of the recovered peptide.349
Several of these biologically produced peptides display altered self-assembly compared to
synthetically produced equivalents.306, 311, 349 These examples illustrate the potential for
sequence modifications to affect peptide properties. Given the limited range of cleavage
methods available, it seems likely that compatibility of peptide sequences with these
methods will be a limiting factor in determining sequences that are amenable to
bioproduction. Future design of bioproduction processes may be able to overcome some
of these issues by redesign of target sequences to be more compatible with expression
and cleavage, or investigation of alternate cleavage methods.
The examples outlined here represent a variety of approaches to peptide bioproduction
and demonstrate the ability to design expression systems for self-assembling peptides.
However as mentioned above, peptide recovered from bioproduction processes is
potentially not equivalent to synthetically produced material due to modifications
introduced during expression and cleavage. While this is currently a limitation to the field,
the bioproduction of compatible sequences potentially enables the renewable production
of these peptides at large scale and lower cost. Chromatography-free methods in
particular may simplify the scale-up of bioproduction methods and increase the potential
for large-scale production.346-348 The approaches detailed here focus on the adaptation of
existing self-assembling peptides to biological production methodologies. An alternate
strategy is to incorporate features for biological production during the de novo design of
self-assembling peptides. In this way it may be possible to reconcile end-use properties
with expressibility and provide a framework for the design of application-targeted
expressible peptides.
43
1.5 Research aims and thesis overview
The work reported in this thesis focused on the design, characterisation and bioproduction
of self-assembling α-helical peptides. It was particularly focused on peptides that act as
switchable surfactants or self-assemble to form fibrils and hydrogels. These peptides show
potential utility in several applications such as emulsion and foam stabilisation, tissue
engineering and drug delivery, as discussed in Section 1.3.
As discussed in Section 1.4, one of the barriers to the large-scale adoption of these selfassembling peptides for materials applications, such as foam and emulsion stabilisation, is
the high cost of solid-phase synthesis. Bioproduction methodologies may offer a low-cost
route for production of such peptides. The recombinant expression of designer peptides is
a currently emerging field of research, making this an appealing research focus. As
described in Section 1.4.2, one particularly promising approach for biological production of
peptides is as designed protein concatemers. By combining design considerations for
stable concatemer folding and expression with peptide end-use properties, this work
aimed to produce an overall design framework for application targeted designer peptides
which are amenable to bioproduction. This aimed to both expand the availability of
designer peptides for large-scale applications such as emulsion and foam stabilisation,
while also increasing the understanding of factors effecting concatemer folding and design.
This is then summarized as research hypothesis 1; that by utilising peptide design
methods that consider factors for both expression and end-use properties, a low-cost
method for producing peptides for large-scale applications may be achieved.
As discussed in Section 1.2, peptides that form α-helices and discrete coiled coils have
been the subject of extensive studies. This now provides a solid background for future
design of α-helix and coiled-coil forming peptides. However, in comparison to the
established fields of fibril forming PAs and β-sheet forming peptides, the study of α-helical
based fibrils and hydrogels is less well developed. At present, the majority of reported αhelix based fibrils form insoluble materials (Section 1.2.3). This has limited the
investigation of these materials in biomedical applications (Section 1.3). The investigation
of peptides that form α-helical based fibrils and hydrogels with utility as cellular scaffolds
and therapeutic delivery vehicles therefore represents an opportunity to expand the
44
knowledge base of the disciple and formed a secondary aim of the work reported in this
thesis.
This is then summarized as research hypothesis 2; that by characterising the selfassembly and resulting physical properties of α-helical peptide hydrogels, as well as
successive redesign of peptide sequences, a better understanding of these materials may
Table 1-2. Research aims of each chapter within this thesis
Chapter
Research Aims
1) To design and assess expression of de novo peptide
concatemers intended as bioproduction constructs for
surfactant peptides.
2
2) To develop a low-cost recovery and purification method
for expressed concatemers, consistent with the aim to
develop a production method for peptides for largescale applications
3) To cleave the expressed peptide concatemer from
Chapter 2 to release constituent surfactant peptides.
3
4) To characterise the pathways involved in heat and
acid-mediated cleavage of the concatemer.
5) To assess the thermodynamic and interfacial properties
of the cleaved products.
6) To characterise the self-assembly of a novel hydrogel
forming α-helical peptide.
4
7) To devise a model for the fibril and hydrogel formation
of the peptide based on biophysical characterisation of
self-assembly.
8) To utilise the proposed model of peptide self-assembly
to design related peptides with desired properties.
9) To characterise the mechanical properties of hydrogels
produced by peptides designed in Chapter 4.
5
10) To investigate the peptide hydrogels as both 2D and
3D cell scaffold materials.
11) To assess the ability of hydrogel forming peptides
designed in Chapters 4 and 5 to bind and release
therapeutics.
45
be achieved which will allow for design of α-helical hydrogels with utility as cellular
scaffolds and therapeutic delivery vehicles.
The research focus of this project was therefore twofold. It initially aimed to design peptide
sequences amenable to biological production while also possessing functionality. The goal
of this work was to increase the availability of designer peptides by lowering production
costs. The second focus of the project was to design and characterise novel selfassembling peptides to expand the understanding of this field of biophysical chemistry.
This project ultimately encompassed each step of the pathway from biological production
through to the use of self-assembling peptides in novel applications.
Table 1-2 shows the research aims of each chapter within this thesis and the outcomes of
each chapter are described below.
Chapter 2 describes the successful bioproduction of an anionic helical surfactant peptide
EDP-11 as part of a charge-paired heteroconcatemer with the cationic partner RDP-4. The
design of the constituent peptides and overall bioproduction process considered both the
end-use properties of individual peptides and the ability of the concatemer to generate a
stably folded coiled-coil structure for expression. The polypeptide sequence was further
optimised for net charge, hydropathy and predicted protease resistance. The designed
heteroconcatemer expressed at high levels in E. coli and was extremely stable to heat and
proteases, allowing isolation of the concatemer by simple heat and pH precipitation steps.
Following expression using this concatemer approach, cleavage was required to produce
monomeric α-helical peptides. The design of acid-cleavable aspartate-proline sites within
the concatemer sequence allowed for simple heat and acid cleavage to give constituent
peptides, as detailed in Chapter 3. Quantification of intermediate species over time
allowed modelling of the kinetic pathways involved, which showed previously unobserved
preferences in cleavage sites. Both intact and cleaved heteroconcatemer were
demonstrated to possess surfactant functionality, with the rate of adsorption to the airwater interface controlled by the assembled state in solution.
In Chapter 4 the peptide AFD19 is reported as part of ongoing work on the design of αhelical surfactant peptides. This peptide unexpectedly self-assembled to form coiled-coil
fibrils and underwent gelation as a function of solution pH. This chapter describes
characterisation of the assembly of AFD19 and proposal of a model for self-assembly.
46
Based on this model, AFD19 was redesigned to give the peptide AFD36, which forms
hydrogels at physiological pH and salt, giving potential utility in the biomaterial field. Both
AFD19 and AFD36 possess high thermodynamic stability and results from transmission
electron microscopy and small-angle X-ray scattering are consistent with the peptides selfassembling to form unbundled higher-order coiled-coil fibrils.
Chapter 5 describes the investigation of α-helical peptide hydrogels as potential cellular
scaffolds and therapeutic delivery vehicles. Rheological characterisation of AFD36
hydrogels showed weak viscoelastic properties potentially suitable for soft tissue
engineering. Redesign of AFD36 gave peptide AFD49, and both supported the growth of
NIH/3T3 fibroblasts at levels similar to those of TCPS controls. AFD49 hydrogels
supported the proliferation of encapsulated NIH/3T3 fibroblasts over several weeks.
Culture of induced pluripotent stem cells on AFD49 hydrogels indicated non-functionalized
AFD49 hydrogels to be unsuitable as substrates for induced pluripotent stem cells. AFD49
hydrogels were also investigated as therapeutic delivery vehicles for the hydrophobic drug
all-trans-retinoic acid.
47
1.6 References
1.
Whitesides, G. M.; Grzybowski, B. Self-Assembly at All Scales. Science 2002, 295,
2418-2421.
2.
Whitesides, G. M.; Mathias, J. P.; Seto, C. T. Molecular Self-Assembly and
Nanochemistry - a Chemical Strategy for the Synthesis of Nanostructures. Science 1991,
254, 1312-1319.
3.
Lehn, J. M. Toward Self-Organization and Complex Matter. Science 2002, 295,
2400-2403.
4.
Simons, K.; Sampaio, J. L. Membrane Organization and Lipid Rafts. Cold Spring
Harb. Perspect. Biol. 2011, 3.
5.
Holmes, K. C.; Popp, D.; Gebhard, W.; Kabsch, W. Atomic Model of the Actin
Filament. Nature 1990, 347, 44-49.
6.
Brunsveld, L.; Folmer, B. J. B.; Meijer, E. W.; Sijbesma, R. P. Supramolecular
Polymers. Chem. Rev. 2001, 101, 4071-4097.
7.
Sijbesma, R. P.; Beijer, F. H.; Brunsveld, L.; Folmer, B. J. B.; Hirschberg, J.; Lange,
R. F. M.; Lowe, J. K. L.; Meijer, E. W. Reversible Polymers Formed from SelfComplementary Monomers Using Quadruple Hydrogen Bonding. Science 1997, 278,
1601-1604.
8.
Folmer, B. J. B.; Sijbesma, R. P.; Versteegen, R. M.; van der Rijt, J. A. J.; Meijer, E.
W. Supramolecular Polymer Materials: Chain Extension of Telechelic Polymers Using a
Reactive Hydrogen-Bonding Synthon. Adv. Mater. 2000, 12, 874-878.
9.
Hamley, I. W. Self-Assembly of Amphiphilic Peptides. Soft Matter 2011, 7, 41224138.
10.
Zeltins, A. Construction and Characterization of Virus-Like Particles: A Review. Mol.
Biotechnol. 2013, 53, 92-107.
11.
Petkau-Milroy, K.; Brunsveld, L. Supramolecular Chemical Biology; Bioactive
Synthetic Self-Assemblies. Org. Biomol. Chem. 2013, 11, 219-232.
12.
Thiruvengadathan, R.; Korampally, V.; Ghosh, A.; Chanda, N.; Gangopadhyay, K.;
Gangopadhyay, S. Nanomaterial Processing Using Self-Assembly-Bottom-up Chemical
and Biological Approaches. Reports on Progress in Physics 2013, 76.
13.
Verma, G.; Hassan, P. A. Self Assembled Materials: Design Strategies and Drug
Delivery Perspectives. Phys. Chem. Chem. Phys. 2013, 15, 17016-17028.
14.
Matson, J. B.; Stupp, S. I. Self-Assembling Peptide Scaffolds for Regenerative
Medicine. Chem. Commun. 2012, 48, 26-33.
15.
Louie, A. Y. Multimodality Imaging Probes: Design and Challenges. Chem. Rev.
2010, 110, 3146-3195.
16.
Van Vlierberghe, S.; Dubruel, P.; Schacht, E. Biopolymer-Based Hydrogels as
Scaffolds for Tissue Engineering Applications: A Review. Biomacromolecules 2011, 12,
1387-408.
17.
Busseron, E.; Ruff, Y.; Moulin, E.; Giuseppone, N. Supramolecular Self-Assemblies
as Functional Nanomaterials. Nanoscale 2013, 5, 7098-7140.
18.
Dexter, A. F.; Malcolm, A. S.; Middelberg, A. P. J. Reversible Active Switching of
the Mechanical Properties of a Peptide Film at a Fluid-Fluid Interface. Nat. Mater. 2006, 5,
502-506.
19.
Dexter, A. F.; Middelberg, A. P. J. Switchable Peptide Surfactants with Designed
Metal Binding Capacity. J. Phys. Chem. C 2007, 111, 10484-10492.
20.
Ellis-Behnke, R. G.; Liang, Y. X.; You, S. W.; Tay, D. K. C.; Zhang, S. G.; So, K. F.;
Schneider, G. E. Nano Neuro Knitting: Peptide Nanofiber Scaffold for Brain Repair and
48
Axon Regeneration with Functional Return of Vision. Proc. Natl. Acad. Sci. U. S. A. 2006,
103, 5054-5059.
21.
Tysseling-Mattiace, V. M.; Sahni, V.; Niece, K. L.; Birch, D.; Czeisler, C.; Fehlings,
M. G.; Stupp, S. I.; Kessler, J. A. Self-Assembling Nanofibers Inhibit Glial Scar Formation
and Promote Axon Elongation after Spinal Cord Injury. J. Neurosci. 2008, 28, 3814-3823.
22.
Altunbas, A.; Lee, S. J.; Rajasekaran, S. A.; Schneider, J. P.; Pochan, D. J.
Encapsulation of Curcumin in Self-Assembling Peptide Hydrogels as Injectable Drug
Delivery Vehicles. Biomaterials 2011, 32, 5906-5914.
23.
Wiradharma, N.; Tong, Y. W.; Yang, Y. Y. Self-Assembled Oligopeptide
Nanostructures for Co-Delivery of Drug and Gene with Synergistic Therapeutic Effect.
Biomaterials 2009, 30, 3100-3109.
24.
Zompra, A. A.; Galanis, A. S.; Werbitzky, O.; Albericio, F. Manufacturing Peptides
as Active Pharmaceutical Ingredients. Future Med. Chem. 2009, 1, 361-377.
25.
Ryadnov, M. G., Prescriptive Peptide Design. In Amino Acids, Peptides and
Proteins, Vol 37, Royal Soc Chemistry: Cambridge, 2012; Vol. 37, pp 190-237.
26.
Aggeli, A.; Bell, M.; Boden, N.; Keen, J. N.; Knowles, P. F.; McLeish, T. C. B.;
Pitkeathly, M.; Radford, S. E. Responsive Gels Formed by the Spontaneous SelfAssembly of Peptides into Polymeric Beta-Sheet Tapes. Nature 1997, 386, 259-262.
27.
Zimenkov, Y.; Dublin, S. N.; Ni, R.; Tu, R. S.; Breedveld, V.; Apkarian, R. P.;
Conticello, V. P. Rational Design of a Reversible Ph-Responsive Switch for Peptide SelfAssembly. J. Am. Chem. Soc. 2006, 128, 6770-6771.
28.
Banwell, E. F.; Abelardo, E. S.; Adams, D. J.; Birchall, M. A.; Corrigan, A.; Donald,
A. M.; Kirkland, M.; Serpell, L. C.; Butler, M. F.; Woolfson, D. N. Rational Design and
Application of Responsive Alpha-Helical Peptide Hydrogels. Nat. Mater. 2009, 8, 596-600.
29.
Wagner, D. E.; Phillips, C. L.; Ali, W. M.; Nybakken, G. E.; Crawford, E. D.; Schwab,
A. D.; Smith, W. F.; Fairman, R. Toward the Development of Peptide Nanofilaments and
Nanoropes as Smart Materials. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 12656-12661.
30.
Zhang, S. G.; Lockshin, C.; Cook, R.; Rich, A. Unusually Stable Beta-Sheet
Formation in an Ionic Self-Complementary Oligopeptide. Biopolymers 1994, 34, 663-672.
31.
Dublin, S. N.; Conticello, V. P. Design of a Selective Metal Ion Switch for SelfAssembly of Peptide-Based Fibrils. J. Am. Chem. Soc. 2008, 130, 49-+.
32.
Itakura, K.; Hirose, T.; Crea, R.; Riggs, A. D.; Heyneker, H. L.; Bolivar, F.; Boyer, H.
W. Expression in Escherichia-Coli of a Chemically Synthesized Gene for Hormone
Somatostatin. Science 1977, 198, 1056-1063.
33.
Kyle, S.; Aggeli, A.; Ingham, E.; McPherson, M. J. Production of Self-Assembling
Biomaterials for Tissue Engineering. Trends Biotechnol. 2009, 27, 423-433.
34.
Meyer, E. E.; Rosenberg, K. J.; Israelachvili, J. Recent Progress in Understanding
Hydrophobic Interactions. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 15739-15746.
35.
Israelachvili, J.; Wennerstrom, H. Role of Hydration and Water Structure in
Biological and Colloidal Interactions. Nature 1996, 379, 219-225.
36.
Zhao, X. B.; Pan, F.; Xu, H.; Yaseen, M.; Shan, H. H.; Hauser, C. A. E.; Zhang, S.
G.; Lu, J. R. Molecular Self-Assembly and Applications of Designer Peptide Amphiphiles.
Chem. Soc. Rev. 2010, 39, 3480-3498.
37.
Meng, Q. B.; Kou, Y. Y.; Ma, X.; Liang, Y. J.; Guo, L.; Ni, C. H.; Liu, K. L. Tunable
Self-Assembled Peptide Amphiphile Nanostructures. Langmuir 2012, 28, 5017-5022.
38.
Han, S. Y.; Xu, W. W.; Cao, M. W.; Wang, J. Q.; Xia, D. H.; Xu, H.; Zhao, X. B.; Lu,
J. R. Interfacial Adsorption of Cationic Peptide Amphiphiles: A Combined Study of in Situ
Spectroscopic Ellipsometry and Liquid Afm. Soft Matter 2012, 8, 645-652.
49
39.
Xu, H.; Wang, J.; Han, S. Y.; Wang, J. Q.; Yu, D. Y.; Zhang, H. Y.; Xia, D. H.; Zhao,
X. B.; Waigh, T. A.; Lu, J. R. Hydrophobic-Region-Induced Transitions in Self-Assembled
Peptide Nanostructures. Langmuir 2009, 25, 4115-4123.
40.
Wiradharma, N.; Khan, M.; Tong, Y. W.; Wang, S.; Yang, Y. Y. Self-Assembled
Cationic Peptide Nanoparticles Capable of Inducing Efficient Gene Expression in Vitro.
Adv. Funct. Mater. 2008, 18, 943-951.
41.
Hartgerink, J. D.; Beniash, E.; Stupp, S. I. Peptide-Amphiphile Nanofibers: A
Versatile Scaffold for the Preparation of Self-Assembling Materials. Proc. Natl. Acad. Sci.
U. S. A. 2002, 99, 5133-5138.
42.
von Maltzahn, G.; Vauthey, S.; Santoso, S.; Zhang, S. U. Positively Charged
Surfactant-Like Peptides Self-Assemble into Nanostructures. Langmuir 2003, 19, 43324337.
43.
Santoso, S.; Hwang, W.; Hartman, H.; Zhang, S. Self-Assembly of Surfactant-Like
Peptides with Variable Glycine Tails to Form Nanotubes and Nanovesicles. Nano Lett.
2002, 2, 687-691.
44.
Nagai, A.; Nagai, Y.; Qu, H. J.; Zhang, S. G. Dynamic Behaviors of Lipid-Like SelfAssembling Peptide a(6)D and a(6)K Nanotubes. J. Nanosci. Nanotechnol. 2007, 7, 22462252.
45.
Vauthey, S.; Santoso, S.; Gong, H. Y.; Watson, N.; Zhang, S. G. Molecular SelfAssembly of Surfactant-Like Peptides to Form Nanotubes and Nanovesicles. Proc. Natl.
Acad. Sci. U. S. A. 2002, 99, 5355-5360.
46.
Pan, F.; Zhao, X.; Perumal, S.; Waigh, T. A.; Lu, J. R.; Webster, J. R. P. Interfacial
Dynamic Adsorption and Structure of Molecular Layers of Peptide Surfactants. Langmuir
2009, 26, 5690-5696.
47.
Zhao, X. B.; Pan, F.; Perumal, S.; Xu, H.; Lu, J. R.; Webster, J. R. P. Interfacial
Assembly of Cationic Peptide Surfactants. Soft Matter 2009, 5, 1630-1638.
48.
Nagarajan, R. Molecular Packing Parameter and Surfactant Self-Assembly: The
Neglected Role of the Surfactant Tail. Langmuir 2002, 18, 31-38.
49.
Makovitzki, A.; Baram, J.; Shai, Y. Antimicrobial Lipopolypeptides Composed of
Palmitoyl Di- and Tricationic Peptides: In Vitro and in Vivo Activities, Self-Assembly to
Nanostructures, and a Plausible Mode of Action. Biochemistry 2008, 47, 10630-10636.
50.
Berndt, P.; Fields, G. B.; Tirrell, M. Synthetic Lipidation of Peptides and AminoAcids - Monolayer Structure and Properties. J. Am. Chem. Soc. 1995, 117, 9515-9522.
51.
Yu, Y. C.; Berndt, P.; Tirrell, M.; Fields, G. B. Self-Assembling Amphiphiles for
Construction of Protein Molecular Architecture. J. Am. Chem. Soc. 1996, 118, 1251512520.
52.
Hartgerink, J. D.; Beniash, E.; Stupp, S. I. Self-Assembly and Mineralization of
Peptide-Amphiphile Nanofibers. Science 2001, 294, 1684-1688.
53.
Pashuck, E. T.; Cui, H. G.; Stupp, S. I. Tuning Supramolecular Rigidity of Peptide
Fibers through Molecular Structure. J. Am. Chem. Soc. 2010, 132, 6041-6046.
54.
Paramonov, S. E.; Jun, H. W.; Hartgerink, J. D. Self-Assembly of PeptideAmphiphile Nanofibers: The Roles of Hydrogen Bonding and Amphiphilic Packing. J. Am.
Chem. Soc. 2006, 128, 7291-7298.
55.
Silva, G. A.; Czeisler, C.; Niece, K. L.; Beniash, E.; Harrington, D. A.; Kessler, J. A.;
Stupp, S. I. Selective Differentiation of Neural Progenitor Cells by High-Epitope Density
Nanofibers. Science 2004, 303, 1352-1355.
56.
Yuwono, V. M.; Hartgerink, J. D. Peptide Amphiphile Nanofibers Template and
Catalyze Silica Nanotube Formation. Langmuir 2007, 23, 5033-5038.
50
57.
Niece, K. L.; Hartgerink, J. D.; Donners, J.; Stupp, S. I. Self-Assembly Combining
Two Bioactive Peptide-Amphiphile Molecules into Nanofibers by Electrostatic Attraction. J.
Am. Chem. Soc. 2003, 125, 7146-7147.
58.
Greenfield, M. A.; Hoffman, J. R.; de la Cruz, M. O.; Stupp, S. I. Tunable Mechanics
of Peptide Nanofiber Gels. Langmuir 2010, 26, 3641-3647.
59.
Jun, H. W.; Paramonov, S. E.; Dong, H.; Forraz, N.; McGuckin, C.; Hartgerink, J. D.
Tuning the Mechanical and Bioresponsive Properties of Peptide-Amphiphile Nanofiber
Networks. J. Biomater. Sci., Polym. Ed. 2008, 19, 665-676.
60.
Pauling, L.; Corey, R. B. Configurations of Polypeptide Chains with Favored
Orientations around Single Bonds - 2 New Pleated Sheets. Proc. Natl. Acad. Sci. U. S. A.
1951, 37, 729-740.
61.
Pauling, L.; Corey, R. B. 2 Rippled-Sheet Configurations of Polypeptide Chains, and
a Note About the Pleated Sheets. Proc. Natl. Acad. Sci. U. S. A. 1953, 39, 253-256.
62.
Marsh, R. E.; Corey, R. B.; Pauling, L. An Investigation of the Structure of Silk
Fibroin. Biochim. Biophys. Acta 1955, 16, 1-34.
63.
Brack, A.; Orgel, L. E. Beta-Structures of Alternating Polypeptides and Their
Possible Prebiotic Significance. Nature 1975, 256, 383-387.
64.
Rippon, W. B.; Chen, H. H.; Walton, A. G. Spectroscopic Characterization of
Poly(Glu-Ala). J. Mol. Biol. 1973, 75, 369-375.
65.
Seipke, G.; Arfmann, H. A.; Wagner, K. G. Synthesis and Properties of Alternating
Poly(Lys-Phe) and Comparison with Random Copolymer Poly(Lys51, Phe49).
Biopolymers 1974, 13, 1621-1633.
66.
Stpierre, S.; Ingwall, R. T.; Verlander, M. S.; Goodman, M. Conformational Studies
of Sequential Polypeptides Containing Lysine and Tyrosine. Biopolymers 1978, 17, 18371848.
67.
O'Neil, K. T.; Degrado, W. F. A Thermodynamic Scale for the Helix-Forming
Tendencies of the Commonly Occurring Amino Acids. Science 1990, 250, 646-651.
68.
Kallberg, Y.; Gustafsson, M.; Persson, B.; Thyberg, J.; Johansson, J. Prediction of
Amyloid Fibril-Forming Proteins. J. Biol. Chem. 2001, 276, 12945-12950.
69.
Chou, P. Y.; Fasman, G. D. Conformational Parameters for Amino-Acids in Helical,
Beta-Sheet, and Random Coil Regions Calculated from Proteins. Biochemistry 1974, 13,
211-222.
70.
Chou, P. Y.; Fasman, G. D. Empirical Predictions of Protein Conformation. Annu.
Rev. Biochem. 1978, 47, 251-276.
71.
Smith, C. K.; Withka, J. M.; Regan, L. A Thermodynamic Scale for the Beta-Sheet
Forming Tendencies of the Amino-Acids. Biochemistry 1994, 33, 5510-5517.
72.
Caplan, M. R.; Schwartzfarb, E. M.; Zhang, S. G.; Kamm, R. D.; Lauffenburger, D.
A. Control of Self-Assembling Oligopeptide Matrix Formation through Systematic Variation
of Amino Acid Sequence. Biomaterials 2002, 23, 219-227.
73.
Osterman, D.; Mora, R.; Kezdy, F. J.; Kaiser, E. T.; Meredith, S. C. A Synthetic
Amphiphilic Beta-Strand Tridecapeptide - a Model for Apolipoprotein-B. J. Am. Chem. Soc.
1984, 106, 6845-6847.
74.
Osterman, D. G.; Kaiser, E. T. Design and Characterization of Peptides with
Amphiphilic Beta-Strand Structures. J. Cell. Biochem. 1985, 29, 57-72.
75.
Zhang, S. G.; Holmes, T.; Lockshin, C.; Rich, A. Spontaneous Assembly of a SelfComplementary Oligopeptide to Form a Stable Macroscopic Membrane. Proc. Natl. Acad.
Sci. U. S. A. 1993, 90, 3334-3338.
76.
Bowerman, C. J.; Nilsson, B. L. Review Self-Assembly of Amphipathic Beta-Sheet
Peptides: Insights and Applications. Biopolymers 2012, 98, 169-184.
51
77.
Bowerman, C. J.; Liyanage, W.; Federation, A. J.; Nilsson, B. L. Tuning Beta-Sheet
Peptide Self-Assembly and Hydrogelation Behavior by Modification of Sequence
Hydrophobicity and Aromaticity. Biomacromolecules 2011, 12, 2735-2745.
78.
Caplan, M. R.; Moore, P. N.; Zhang, S. G.; Kamm, R. D.; Lauffenburger, D. A. SelfAssembly of a Beta-Sheet Protein Governed by Relief of Electrostatic Repulsion Relative
to Van Der Waals Attraction. Biomacromolecules 2000, 1, 627-631.
79.
Dong, H.; Paramonov, S. E.; Aulisa, L.; Bakota, E. L.; Hartgerink, J. D. SelfAssembly of Multidomain Peptides: Balancing Molecular Frustration Controls
Conformation and Nanostructure. J. Am. Chem. Soc. 2007, 129, 12468-12472.
80.
Aulisa, L.; Dong, H.; Hartgerink, J. D. Self-Assembly of Multidomain Peptides:
Sequence Variation Allows Control over Cross-Linking and Viscoelasticity.
Biomacromolecules 2009, 10, 2694-2698.
81.
Wang, K.; Keasling, J. D.; Muller, S. J. Effects of the Sequence and Size of NonPolar Residues on Self-Assembly of Amphiphilic Peptides. Int. J. Biol. Macromol. 2005, 36,
232-240.
82.
Aggeli, A.; Bell, M.; Carrick, L. M.; Fishwick, C. W. G.; Harding, R.; Mawer, P. J.;
Radford, S. E.; Strong, A. E.; Boden, N. Ph as a Trigger of Peptide Beta-Sheet SelfAssembly and Reversible Switching between Nematic and Isotropic Phases. J. Am. Chem.
Soc. 2003, 125, 9619-9628.
83.
Collier, J. H.; Messersmith, P. B. Enzymatic Modification of Self-Assembled Peptide
Structures with Tissue Transglutaminase. Bioconjug. Chem. 2003, 14, 748-755.
84.
Holmes, T. C.; de Lacalle, S.; Su, X.; Liu, G. S.; Rich, A.; Zhang, S. G. Extensive
Neurite Outgrowth and Active Synapse Formation on Self-Assembling Peptide Scaffolds.
Proc. Natl. Acad. Sci. U. S. A. 2000, 97, 6728-6733.
85.
Zhang, S. G.; Holmes, T. C.; Dipersio, C. M.; Hynes, R. O.; Su, X.; Rich, A. SelfComplementary Oligopeptide Matrices Support Mammalian-Cell Attachment. Biomaterials
1995, 16, 1385-1393.
86.
Haines-Butterick, L.; Rajagopal, K.; Branco, M.; Salick, D.; Rughani, R.; Pilarz, M.;
Lamm, M. S.; Pochan, D. J.; Schneider, J. P. Controlling Hydrogelation Kinetics by Peptide
Design for Three-Dimensional Encapsulation and Injectable Delivery of Cells. Proc. Natl.
Acad. Sci. U. S. A. 2007, 104, 7791-7796.
87.
Schneider, J. P.; Pochan, D. J.; Ozbas, B.; Rajagopal, K.; Pakstis, L.; Kretsinger, J.
Responsive Hydrogels from the Intramolecular Folding and Self-Assembly of a Designed
Peptide. J. Am. Chem. Soc. 2002, 124, 15030-15037.
88.
Segman-Magidovich, S.; Grisaru, H.; Gitli, T.; Levi-Kalisman, Y.; Rapaport, H.
Matrices of Acidic Beta-Sheet Peptides as Templates for Calcium Phosphate
Mineralization. Adv. Mater. 2008, 20, 2156-2161.
89.
Zhang, S. G.; Lockshin, C.; Herbert, A.; Winter, E.; Rich, A. Zuotin, a Putative ZDNA Binding-Protein in Saccharomyces-Cerevisiae. EMBO J. 1992, 11, 3787-3796.
90.
Yokoi, H.; Kinoshita, T.; Zhang, S. G. Dynamic Reassembly of Peptide Rada16
Nanofiber Scaffold. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 8414-8419.
91.
Ozbas, B.; Kretsinger, J.; Rajagopal, K.; Schneider, J. P.; Pochan, D. J. SaltTriggered Peptide Folding and Consequent Self-Assembly into Hydrogels with Tunable
Modulus. Macromolecules 2004, 37, 7331-7337.
92.
Rajagopal, K.; Lamm, M. S.; Haines-Butterick, L. A.; Pochan, D. J.; Schneider, J. P.
Tuning the Ph Responsiveness of Beta-Hairpin Peptide Folding, Self-Assembly, and
Hydrogel Material Formation. Biomacromolecules 2009, 10, 2619-2625.
52
93.
Pochan, D. J.; Schneider, J. P.; Kretsinger, J.; Ozbas, B.; Rajagopal, K.; Haines, L.
Thermally Reversible Hydrogels Via Intramolecular Folding and Consequent SelfAssembly of a De Novo Designed Peptide. J. Am. Chem. Soc. 2003, 125, 11802-11803.
94.
Larsen, T. H.; Branco, M. C.; Rajagopal, K.; Schneider, J. P.; Furst, E. M.
Sequence-Dependent Gelation Kinetics of Beta-Hairpin Peptide Hydrogels.
Macromolecules 2009, 42, 8443-8450.
95.
Kretsinger, J. K.; Haines, L. A.; Ozbas, B.; Pochan, D. J.; Schneider, J. P.
Cytocompatibility of Self-Assembled Ss-Hairpin Peptide Hydrogel Surfaces. Biomaterials
2005, 26, 5177-5186.
96.
Haines-Butterick, L. A.; Salick, D. A.; Pochan, D. J.; Schneider, J. P. In Vitro
Assessment of the Pro-Inflammatory Potential of Beta-Hairpin Peptide Hydrogels.
Biomaterials 2008, 29, 4164-4169.
97.
Ross, C. A.; McInnis, M. G.; Margolis, R. L.; Li, S. H. Genes with Triplet Repeats Candidate Mediators of Neuropsychiatric Disorders. Trends Neurosci. 1993, 16, 254-260.
98.
Perutz, M. F.; Johnson, T.; Suzuki, M.; Finch, J. T. Glutamine Repeats as Polar
Zippers - Their Possible Role in Inherited Neurodegenerative Diseases. Proc. Natl. Acad.
Sci. U. S. A. 1994, 91, 5355-5358.
99.
Stott, K.; Blackburn, J. M.; Butler, P. J. G.; Perutz, M. Incorporation of Glutamine
Repeats Makes Protein Oligomerize - Implications for Neurodegenerative Diseases. Proc.
Natl. Acad. Sci. U. S. A. 1995, 92, 6509-6513.
100. Aggeli, A.; Bell, M.; Boden, N.; Keen, J. N.; McLeish, T. C. B.; Nyrkova, I.; Radford,
S. E.; Semenov, A. Engineering of Peptide Beta-Sheet Nanotapes. J. Mater. Chem. 1997,
7, 1135-1145.
101. Carrick, L. M.; Aggeli, A.; Boden, N.; Fisher, J.; Ingham, E.; Waigh, T. A. Effect of
Ionic Strength on the Self-Assembly, Morphology and Gelation of Ph Responsive BetaSheet Tape-Forming Peptides. Tetrahedron 2007, 63, 7457-7467.
102. Aggeli, A.; Nyrkova, I. A.; Bell, M.; Harding, R.; Carrick, L.; McLeish, T. C. B.;
Semenov, A. N.; Boden, N. Hierarchical Self-Assembly of Chiral Rod-Like Molecules as a
Model for Peptide Beta-Sheet Tapes, Ribbons, Fibrils, and Fibers. Proc. Natl. Acad. Sci.
U. S. A. 2001, 98, 11857-11862.
103. Boden, N.; Agelli, A.; Ingham, E.; Kirkham, J. Beta Sheet Tapes Ribbons in Tissue
Engineering. US 07700721 B2, 04/20/2010.
104. Kirkham, J.; Firth, A.; Vernals, D.; Boden, N.; Robinson, C.; Shore, R. C.; Brookes,
S. J.; Aggeli, A. Self-Assembling Peptide Scaffolds Promote Enamel Remineralization. J.
Dent. Res. 2007, 86, 426-430.
105. Brunton, P. A.; Davies, R. P. W.; Burke, J. L.; Smith, A.; Aggeli, A.; Brookes, S. J.;
Kirkham, J. Treatment of Early Caries Lesions Using Biomimetic Self-Assembling Peptides
- a Clinical Safety Trial. Br. Dent. J. 2013, 215, E6.
106. Rapaport, H.; Kjaer, K.; Jensen, T. R.; Leiserowitz, L.; Tirrell, D. A. TwoDimensional Order in Beta-Sheet Peptide Monolayers. J. Am. Chem. Soc. 2000, 122,
12523-12529.
107. Isenberg, H.; Kjaer, K.; Rapaport, H. Elasticity of Crystalline Beta-Sheet
Monolayers. J. Am. Chem. Soc. 2006, 128, 12468-12472.
108. Vaiser, V.; Rapaport, H. Compressibility and Elasticity of Amphiphilic and Acidic ΒSheet Peptides at the Air−Water Interface. The Journal of Physical Chemistry B 2010,
115, 50-56.
109. Vinod, T. P.; Zarzhitsky, S.; Morag, A.; Zeiri, L.; Levi-Kalisman, Y.; Rapaport, H.;
Jelinek, R. Transparent, Conductive, and Sers-Active Au Nanofiber Films Assembled on
an Amphiphilic Peptide Template. Nanoscale 2013, 5, 10487-10493.
53
110. Yaakobi, K.; Liebes-Peer, Y.; Kushmaro, A.; Rapaport, H. Designed Amphiphilic ΒSheet Peptides as Templates for Paraoxon Adsorption and Detection. Langmuir 2013, 29,
6840-6848.
111. Rapaport, H.; Möller, G.; Knobler, C. M.; Jensen, T. R.; Kjaer, K.; Leiserowitz, L.;
Tirrell, D. A. Assembly of Triple-Stranded Β-Sheet Peptides at Interfaces. J. Am. Chem.
Soc. 2002, 124, 9342-9343.
112. Quinn, T. P.; Tweedy, N. B.; Williams, R. W.; Richardson, J. S.; Richardson, D. C.
Betadoublet - De-Novo Design, Synthesis, and Characterization of a Beta-Sandwich
Protein. Proc. Natl. Acad. Sci. U. S. A. 1994, 91, 8747-8751.
113. Kortemme, T.; Ramirez-Alvarado, M.; Serrano, L. Design of a 20-Amino Acid,
Three-Stranded Beta-Sheet Protein. Science 1998, 281, 253-256.
114. Schenck, H. L.; Gellman, S. H. Use of a Designed Triple-Stranded Antiparallel BetaSheet to Probe Beta-Sheet Cooperativity in Aqueous Solution. J. Am. Chem. Soc. 1998,
120, 4869-4870.
115. De Alba, E.; Santoro, J.; Rico, M.; Jimenez, M. A. De Novo Design of a Monomeric
Three-Stranded Antiparallel Beta-Sheet. Protein Sci. 1999, 8, 854-865.
116. Sharman, G. J.; Searle, M. S. Cooperative Interaction between the Three Strands of
a Designed Antiparallel Beta-Sheet. J. Am. Chem. Soc. 1998, 120, 5291-5300.
117. Ottesen, J. J.; Imperiali, B. Design of a Discretely Folded Mini-Protein Motif with
Predominantly Beta-Structure. Nat. Struct. Biol. 2001, 8, 535-539.
118. Wang, W. X.; Hecht, M. H. Rationally Designed Mutations Convert De Novo
Amyloid-Like Fibrils into Monomeric Beta-Sheet Proteins. Proc. Natl. Acad. Sci. U. S. A.
2002, 99, 2760-2765.
119. Pauling, L.; Corey, R. B. 2 Hydrogen-Bonded Spiral Configurations of the
Polypeptide Chain. J. Am. Chem. Soc. 1950, 72, 5349-5349.
120. Pauling, L.; Corey, R. B.; Branson, H. R. The Structure of Proteins - 2 HydrogenBonded Helical Configurations of the Polypeptide Chain. Proc. Natl. Acad. Sci. U. S. A.
1951, 37, 205-211.
121. Perutz, M. F. New X-Ray Evidence on the Configuration of Polypeptide Chains.
Nature 1951, 167, 1053-1054.
122. Cochran, W.; Crick, F. H. C. Evidence for the Pauling-Corey Alpha-Helix in
Synthetic Polypeptides. Nature 1952, 169, 234-235.
123. Crick, F. H. C. Is Alpha-Keratin a Coiled Coil. Nature 1952, 170, 882-883.
124. Crick, F. H. C. The Packing of Alpha-Helices - Simple Coiled-Coils. Acta
Crystallogr. 1953, 6, 689-697.
125. Scholtz, J. M.; Baldwin, R. L. The Mechanism of Alpha-Helix Formation by
Peptides. Annu. Rev. Biophys. Biomol. Struct. 1992, 21, 95-118.
126. Jackson, D. Y.; King, D. S.; Chmielewski, J.; Singh, S.; Schultz, P. G. GeneralApproach to the Synthesis of Short Alpha-Helical Peptides. J. Am. Chem. Soc. 1991, 113,
9391-9392.
127. Condon, S. M.; Morize, I.; Darnbrough, S.; Burns, C. J.; Miller, B. E.; Uhl, J.; Burke,
K.; Jariwala, N.; Locke, K.; Krolikowski, P. H.; Kumar, N. V.; Labaudiniere, R. F. The
Bioactive Conformation of Human Parathyroid Hormone. Structural Evidence for the
Extended Helix Postulate. J. Am. Chem. Soc. 2000, 122, 3007-3014.
128. Ghadiri, M. R.; Choi, C. Secondary Structure Nucleation in Peptides - TransitionMetal Ion Stabilized Alpha-Helices. J. Am. Chem. Soc. 1990, 112, 1630-1632.
129. Ghadiri, M. R.; Fernholz, A. K. Peptide Architecture - Design of Stable Alpha-Helical
Metallopeptides Via a Novel Exchange-Inert Ruiii Complex. J. Am. Chem. Soc. 1990, 112,
9633-9635.
54
130. Boncheva, M.; Vogel, H. Formation of Stable Polypeptide Monolayers at Interfaces:
Controlling Molecular Conformation and Orientation. Biophys. J. 1997, 73, 1056-1072.
131. Fairman, R.; Chao, H. G.; Mueller, L.; Lavoie, T. B.; Shen, L. Y.; Novotny, J.;
Matsueda, G. R. Characterization of a New 4-Chain Coiled-Coil: Influence of Chain Length
on Stability. Protein Sci. 1995, 4, 1457-1469.
132. Middelberg, A. P. J.; Radke, C. J.; Blanch, H. W. Peptide Interfacial Adsorption Is
Kinetically Limited by the Thermodynamic Stability of Self Association. Proc. Natl. Acad.
Sci. U. S. A. 2000, 97, 5054-5059.
133. Blaber, M.; Zhang, X. J.; Matthews, B. W. Structural Basis of Amino-Acid AlphaHelix Propensity. Science 1993, 260, 1637-1640.
134. Pace, C. N.; Scholtz, J. M. A Helix Propensity Scale Based on Experimental Studies
of Peptides and Proteins. Biophys. J. 1998, 75, 422-427.
135. Oshea, E. K.; Klemm, J. D.; Kim, P. S.; Alber, T. X-Ray Structure of the Gcn4
Leucine Zipper, a 2-Stranded, Parallel Coiled Coil. Science 1991, 254, 539-544.
136. Oshea, E. K.; Rutkowski, R.; Kim, P. S. Evidence That the Leucine Zipper Is a
Coiled Coil. Science 1989, 243, 538-542.
137. Oshea, E. K.; Rutkowski, R.; Kim, P. S. Mechanism of Specificity in the Fos-Jun
Oncoprotein Heterodimer. Cell 1992, 68, 699-708.
138. Oshea, E. K.; Rutkowski, R.; Stafford, W. F.; Kim, P. S. Preferential Heterodimer
Formation by Isolated Leucine Zippers from Fos and Jun. Science 1989, 245, 646-648.
139. Litowski, J. R.; Hodges, R. S. Designing Heterodimeric Two-Stranded Alpha-Helical
Coiled-Coils: The Effect of Chain Length on Protein Folding, Stability and Specificity. J.
Pept. Res. 2001, 58, 477-492.
140. De Crescenzo, G.; Litowski, J. R.; Hodges, R. S.; O'Connor-McCourt, M. D. RealTime Monitoring of the Interactions of Two-Stranded De Novo Designed Coiled-Coils:
Effect of Chain Length on the Kinetic and Thermodynamic Constants of Binding.
Biochemistry 2003, 42, 1754-1763.
141. Zhou, N. E.; Kay, C. M.; Hodges, R. S. Synthetic Model Proteins - Positional Effects
of Interchain Hydrophobic Interactions on Stability of 2-Stranded Alpha-Helical CoiledCoils. J. Biol. Chem. 1992, 267, 2664-2670.
142. Apostolovic, B.; Danial, M.; Klok, H. A. Coiled Coils: Attractive Protein Folding
Motifs for the Fabrication of Self-Assembled, Responsive and Bioactive Materials. Chem.
Soc. Rev. 2010, 39, 3541-3575.
143. Wagschal, K.; Tripet, B.; Lavigne, P.; Mant, C.; Hodges, R. S. The Role of Position
a in Determining the Stability and Oligomerization State of Alpha-Helical Coiled Coils: 20
Amino Acid Stability Coefficients in the Hydrophobic Core of Proteins. Protein Sci. 1999, 8,
2312-2329.
144. Tripet, B.; Wagschal, K.; Lavigne, P.; Mant, C. T.; Hodges, R. S. Effects of SideChain Characteristics on Stability and Oligomerization State of a De Novo-Designed Model
Coiled-Coil: 20 Amino Acid Substitutions in Position "D". J. Mol. Biol. 2000, 300, 377-402.
145. Harbury, P. B.; Zhang, T.; Kim, P. S.; Alber, T. A Switch between 2-Stranded, 3Stranded and 4-Stranded Coiled Coils in Gcn4 Leucine-Zipper Mutants. Science 1993,
262, 1401-1407.
146. Lau, S. Y. M.; Taneja, A. K.; Hodges, R. S. Synthesis of a Model Protein of Defined
Secondary and Quaternary Structure - Effect of Chain-Length on the Stabilization and
Formation of 2-Stranded Alpha-Helical Coiled-Coils. J. Biol. Chem. 1984, 259, 3253-3261.
147. Su, J. Y.; Hodges, R. S.; Kay, C. M. Effect of Chain-Length on the Formation and
Stability of Synthetic Alpha-Helical Coiled Coils. Biochemistry 1994, 33, 15501-15510.
55
148. Moll, J. R.; Ruvinov, S. B.; Pastan, I.; Vinson, C. Designed Heterodimerizing
Leucine Zippers with a Range of Pis and Stabilities up to 10(-15) M. Protein Sci. 2001, 10,
649-655.
149. Zhou, N. E.; Kay, C. M.; Hodges, R. S. The Net Energetic Contribution of
Interhelical Electrostatic Attractions to Coiled-Coil Stability. Protein Eng. 1994, 7, 13651372.
150. Zhou, N. E.; Kay, C. M.; Hodges, R. S. The Role of Interhelical Ionic Interactions in
Controlling Protein-Folding and Stability - De-Novo Designed Synthetic 2-Stranded AlphaHelical Coiled-Coils. J. Mol. Biol. 1994, 237, 500-512.
151. Deng, Y. Q.; Liu, J.; Zheng, Q.; Eliezer, D.; Kallenbach, N. R.; Lu, M. Antiparallel
Four-Stranded Coiled Coil Specified by a 3-3-1 Hydrophobic Heptad Repeat. Structure
2006, 14, 247-255.
152. Liu, J.; Zheng, Q.; Deng, Y. Q.; Cheng, C. S.; Kallenbach, N. R.; Lu, M. A SevenHelix Coiled Coil. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 15457-15462.
153. Zaccai, N. R.; Chi, B.; Thomson, A. R.; Boyle, A. L.; Bartlett, G. J.; Bruning, M.;
Linden, N.; Sessions, R. B.; Booth, P. J.; Brady, R. L.; Woolfson, D. N. A De Novo Peptide
Hexamer with a Mutable Channel. Nat. Chem. Biol. 2011, 7, 935-941.
154. Robertson, D. E.; Farid, R. S.; Moser, C. C.; Urbauer, J. L.; Mulholland, S. E.;
Pidikiti, R.; Lear, J. D.; Wand, A. J.; Degrado, W. F.; Dutton, P. L. Design and Synthesis of
Multi-Heme Proteins. Nature 1994, 368, 425-431.
155. Lichtenstein, B. R.; Farid, T. A.; Kodali, G.; Solomon, L. A.; Anderson, J. L. R.;
Sheehan, M. M.; Ennist, N. M.; Fry, B. A.; Chobot, S. E.; Bialas, C.; Mancini, J. A.;
Armstrong, C. T.; Zhao, Z. Y.; Esipova, T. V.; Snell, D.; Vinogradov, S. A.; Discher, B. M.;
Moser, C. C.; Dutton, P. L. Engineering Oxidoreductases: Maquette Proteins Designed
from Scratch. Biochem. Soc. Trans. 2012, 40, 561-566.
156. Fry, H. C.; Lehmann, A.; Saven, J. G.; DeGrado, W. F.; Therien, M. J.
Computational Design and Elaboration of a De Novo Heterotetrameric Alpha-Helical
Protein That Selectively Binds an Emissive Abiological (Porphinato)Zinc Chromophore. J.
Am. Chem. Soc. 2010, 132, 3997-4005.
157. Summa, C. M.; Rosenblatt, M. M.; Hong, J. K.; Lear, J. D.; DeGrado, W. F.
Computational De Novo Design, and Characterization of an a(2)B(2) Diiron Protein. J. Mol.
Biol. 2002, 321, 923-938.
158. Koder, R. L.; Anderson, J. L. R.; Solomon, L. A.; Reddy, K. S.; Moser, C. C.; Dutton,
P. L. Design and Engineering of an O-2 Transport Protein. Nature 2009, 458, 305-U64.
159. Regan, L.; Degrado, W. F. Characterization of a Helical Protein Designed from 1st
Principles. Science 1988, 241, 976-978.
160. Handel, T. M.; Williams, S. A.; Degrado, W. F. Metal-Ion Dependent Modulation of
the Dynamics of a Designed Protein. Science 1993, 261, 879-885.
161. Faiella, M.; Andreozzi, C.; de Rosales, R. T. M.; Pavone, V.; Maglio, O.; Nastri, F.;
DeGrado, W. F.; Lombardi, A. An Artificial Di-Iron Oxo-Protein with Phenol Oxidase
Activity. Nat. Chem. Biol. 2009, 5, 882-884.
162. McAllister, K. A.; Zou, H. L.; Cochran, F. V.; Bender, G. M.; Senes, A.; Fry, H. C.;
Nanda, V.; Keenan, P. A.; Lear, J. D.; Saven, J. G.; Therien, M. J.; Blasie, J. K.; DeGrado,
W. F. Using Alpha-Helical Coiled-Coils to Design Nanostructured Metalloporphyrin Arrays.
J. Am. Chem. Soc. 2008, 130, 11921-11927.
163. Cochran, F. V.; Wu, S. P.; Wang, W.; Nanda, V.; Saven, J. G.; Therien, M. J.;
DeGrado, W. F. Computational De Novo Design and Characterization of a Four-Helix
Bundle Protein That Selectively Binds a Nonbiological Cofactor. J. Am. Chem. Soc. 2005,
127, 1346-1347.
56
164. Korendovych, I. V.; Senes, A.; Kim, Y. H.; Lear, J. D.; Fry, H. C.; Therien, M. J.;
Blasie, J. K.; Walker, F. A.; DeGrado, W. F. De Novo Design and Molecular Assembly of a
Transmembrane Diporphyrin-Binding Protein Complex. J. Am. Chem. Soc. 2010, 132,
15516-15518.
165. Pandya, M. J.; Spooner, G. M.; Sunde, M.; Thorpe, J. R.; Rodger, A.; Woolfson, D.
N. Sticky-End Assembly of a Designed Peptide Fiber Provides Insight into Protein
Fibrillogenesis. Biochemistry 2000, 39, 8728-8734.
166. Papapostolou, D.; Smith, A. M.; Atkins, E. D. T.; Oliver, S. J.; Ryadnov, M. G.;
Serpell, L. C.; Woolfson, D. N. Engineering Nanoscale Order into a Designed Protein
Fiber. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 10853-10858.
167. Gribbon, C.; Channon, K. J.; Zhang, W. J.; Banwell, E. F.; Bromley, E. H. C.;
Chaudhuri, J. B.; Oreffo, R. O. C.; Woolfson, D. N. Magicwand: A Single, Designed
Peptide That Assembles to Stable, Ordered Alpha-Helical Fibers. Biochemistry 2008, 47,
10365-10371.
168. Ryadnov, M. G.; Woolfson, D. N. Engineering the Morphology of a Selfassembling
Protein Fibre. Nat. Mater. 2003, 2, 329-332.
169. Ryadnov, M. G.; Woolfson, D. N. Introducing Branches into a Self-Assembling
Peptide Fiber. Angew. Chem., Int. Ed. 2003, 42, 3021-3023.
170. Ryadnov, M. G.; Ceyhan, B.; Niemeyer, C. M.; Woolfson, D. N. "Belt and Braces": A
Peptide-Based Linker System of De Novo Design. J. Am. Chem. Soc. 2003, 125, 93889394.
171. Papapostolou, D.; Bromley, E. H. C.; Bano, C.; Woolfson, D. N. Electrostatic
Control of Thickness and Stiffness in a Designed Protein Fiber. J. Am. Chem. Soc. 2008,
130, 5124-5130.
172. Smith, A. M.; Banwell, E. F.; Edwards, W. R.; Pandya, M. J.; Woolfson, D. N.
Engineering Increased Stability into Self-Assembled Protein Fibers. Adv. Funct. Mater.
2006, 16, 1022-1030.
173. Zimenkov, Y.; Conticello, V. P.; Guo, L.; Thiyagarajan, P. Rational Design of a
Nanoscale Helical Scaffold Derived from Self-Assembly of a Dimeric Coiled Coil Motif.
Tetrahedron 2004, 60, 7237-7246.
174. Xu, C. F.; Liu, R.; Mehta, A. K.; Guerrero-Ferreira, R. C.; Wright, E. R.; DuninHorkawicz, S.; Morris, K.; Serpell, L. C.; Zuo, X. B.; Wall, J. S.; Conticello, V. P. Rational
Design of Helical Nanotubes from Self-Assembly of Coiled-Coil Lock Washers. J. Am.
Chem. Soc. 2013, 135, 15565-15578.
175. Tsang, B. P.; Bretscher, H. S.; Kokona, B.; Manning, R. S.; Fairman, R.
Thermodynamic Analysis of Self-Assembly in Coiled-Coil Biomaterials. Biochemistry 2011,
50, 8548-8558.
176. Kojima, S.; Kuriki, Y.; Yoshida, T.; Yazaki, K.; Miura, K. Fibril Formation by an
Amphipathic Alpha-Helix-Forming Polypeptide Produced by Gene Engineering. Proc. Jpn.
Acad. Ser. B Phys. Biol. Sci. 1997, 73, 7-11.
177. Takei, T.; Hasegawa, K.; Imada, K.; Namba, K.; Tsumoto, K.; Kuriki, Y.; Yoshino,
M.; Yazaki, K.; Kojima, S.; Takei, T.; Ueda, T.; Miura, K. Effects of Chain Length of an
Amphipathic Polypeptide Carrying the Repeated Amino Acid Sequence (Letlaka)(N) on
Alpha-Helix and Fibrous Assembly Formation. Biochemistry 2013, 52, 2810-2820.
178. Potekhin, S. A.; Melnik, T. N.; Popov, V.; Lanina, N. F.; Vazina, A. A.; Rigler, P.;
Verdini, A. S.; Corradin, G.; Kajava, A. V. De Novo Design of Fibrils Made of Short AlphaHelical Coiled Coil Peptides. Chem. Biol. 2001, 8, 1025-1032.
57
179. Melnik, T. N.; Villard, V.; Vasiliev, V.; Corradin, G.; Kajava, A. V.; Potekhin, S. A.
Shift of Fibril-Forming Ability of the Designed Alpha-Helical Coiled-Coil Peptides into the
Physiological Ph Region. Protein Eng. 2003, 16, 1125-1130.
180. Dong, H.; Paramonov, S. E.; Hartgerink, J. D. Self-Assembly of Alpha-Helical
Coiled Coil Nanofibers. J. Am. Chem. Soc. 2008, 130, 13691-13695.
181. Banat, I. M.; Franzetti, A.; Gandolfi, I.; Bestetti, G.; Martinotti, M. G.; Fracchia, L.;
Smyth, T. J.; Marchant, R. Microbial Biosurfactants Production, Applications and Future
Potential. Appl. Microbiol. Biotechnol. 2010, 87, 427-444.
182. Cameotra, S. S.; Makkar, R. S. Recent Applications of Biosurfactants as Biological
and Immunological Molecules. Curr. Opin. Microbiol. 2004, 7, 262-266.
183. Dexter, A. F.; Middelberg, A. P. J. Peptides as Functional Surfactants. Ind. Eng.
Chem. Res. 2008, 47, 6391-6398.
184. Das, R.; Kiley, P. J.; Segal, M.; Norville, J.; Yu, A. A.; Wang, L. Y.; Trammell, S. A.;
Reddick, L. E.; Kumar, R.; Stellacci, F.; Lebedev, N.; Schnur, J.; Bruce, B. D.; Zhang, S.
G.; Baldo, M. Integration of Photosynthetic Protein Molecular Complexes in Solid-State
Electronic Devices. Nano Lett. 2004, 4, 1079-1083.
185. Kiley, P.; Zhao, X. J.; Vaughn, M.; Baldo, M. A.; Bruce, B. D.; Zhang, S. G. SelfAssembling Peptide Detergents Stabilize Isolated Photosystem I on a Dry Surface for an
Extended Time. PLoS Biol. 2005, 3, 1180-1186.
186. Matsumoto, K.; Vaughn, M.; Bruce, B. D.; Koutsopoulos, S.; Zhang, S. G. Designer
Peptide Surfactants Stabilize Functional Photosystem-I Membrane Complex in Aqueous
Solution for Extended Time. J. Phys. Chem. B 2009, 113, 75-83.
187. Yeh, J. I.; Du, S.; Tortajada, A.; Paulo, J.; Zhang, S. Peptergents: Peptide
Detergents That Improve Stability and Functionality of a Membrane Protein, Glycerol-3Phosphate Dehydrogenase†. Biochemistry 2005, 44, 16912-16919.
188. Zhao, X.; Nagai, Y.; Reeves, P. J.; Kiley, P.; Khorana, H. G.; Zhang, S. Designer
Short Peptide Surfactants Stabilize G Protein-Coupled Receptor Bovine Rhodopsin. Proc.
Natl. Acad. Sci. U. S. A. 2006, 103, 17707-17712.
189. Corin, K.; Baaske, P.; Ravel, D. B.; Song, J. Y.; Brown, E.; Wang, X. Q.; Wienken,
C. J.; Jerabek-Willemsen, M.; Duhr, S.; Luo, Y.; Braun, D.; Zhang, S. G. Designer LipidLike Peptides: A Class of Detergents for Studying Functional Olfactory Receptors Using
Commercial Cell-Free Systems. PLoS One 2011, 6, e25067.
190. Wang, X. Q.; Corin, K.; Baaske, P.; Wienken, C. J.; Jerabek-Willemsen, M.; Duhr,
S.; Braun, D.; Zhang, S. G. Peptide Surfactants for Cell-Free Production of Functional G
Protein-Coupled Receptors. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 9049-9054.
191. Malcolm, A. S.; Dexter, A. F.; Middelberg, A. P. J. Foaming Properties of a Peptide
Designed to Form Stimuli-Responsive Interfacial Films. Soft Matter 2006, 2, 1057-1066.
192. Malcolm, A. S.; Dexter, A. F.; Katakdhond, J. A.; Karakashev, S. I.; Nguyen, A. V.;
Middelberg, A. P. J. Tuneable Control of Interfacial Rheology and Emulsion Coalescence.
ChemPhysChem 2009, 10, 778-781.
193. Fairman, R.; Chao, H. G.; Lavoie, T. B.; Villafranca, J. J.; Matsueda, G. R.;
Novotny, J. Design of Heterotetrameric Coiled Coils: Evidence for Increased Stabilization
by Glu(-)-Lys(+) Ion Pair Interactions. Biochemistry 1996, 35, 2824-2829.
194. Middelberg, A. P. J.; He, L.; Dexter, A. F.; Shen, H. H.; Holt, S. A.; Thomas, R. K.
The Interfacial Structure and Young's Modulus of Peptide Films Having Switchable
Mechanical Properties. J. Royal Soc. Interface 2008, 5, 47-54.
195. Dwyer, M. D.; He, L. Z.; James, M.; Nelson, A.; Middelberg, A. P. J. Insights into the
Role of Protein Molecule Size and Structure on Interfacial Properties Using Designed
Sequences. J. Royal Soc. Interface 2013, 10, 20120987.
58
196. Dexter, A. F. Interfacial and Emulsifying Properties of Designed Beta-Strand
Peptides. Langmuir 2010, 26, 17997-18007.
197. Brown, P.; Butts, C. P.; Eastoe, J. Stimuli-Responsive Surfactants. Soft Matter
2013, 9, 2365-2374.
198. Rosslee, C.; Abbott, N. L. Active Control of Interfacial Properties. Curr. Opin. Colloid
Interface Sci. 2000, 5, 81-87.
199. Clark, R. A. F.; Ghosh, K.; Tonnesen, M. G. Tissue Engineering for Cutaneous
Wounds. J. Investig. Dermatol. 2007, 127, 1018-1029.
200. Cui, H. G.; Webber, M. J.; Stupp, S. I. Self-Assembly of Peptide Amphiphiles: From
Molecules to Nanostructures to Biomaterials. Biopolymers 2010, 94, 1-18.
201. Cao, Y.; Croll, T. I.; Lees, J. G.; Tuch, B. E.; Cooper-White, J. J. Scaffolds, Stem
Cells, and Tissue Engineering: A Potent Combination! Aust. J. Chem. 2005, 58, 691-703.
202. Pampaloni, F.; Reynaud, E. G.; Stelzer, E. H. K. The Third Dimension Bridges the
Gap between Cell Culture and Live Tissue. Nat. Rev. Mol. Cell Biol. 2007, 8, 839-845.
203. Choi, Y. C.; Choi, J. S.; Kim, B. S.; Kim, J. D.; Yoon, H. I.; Cho, Y. W. Decellularized
Extracellular Matrix Derived from Porcine Adipose Tissue as a Xenogeneic Biomaterial for
Tissue Engineering. Tissue Eng., Part C 2012, 18, 866-876.
204. Hoshiba, T.; Lu, H. X.; Kawazoe, N.; Chen, G. P. Decellularized Matrices for Tissue
Engineering. Expert Opin. Biol. Ther. 2010, 10, 1717-1728.
205. Patino, M. G.; Neiders, M. E.; Andreana, S.; Noble, B.; Cohen, R. E. Collagen as an
Implantable Material in Medicine and Dentistry. J. Oral Implantol. 2002, 28.
206. Hoffman, A. S. Hydrogels for Biomedical Applications. Adv. Drug Delivery. Rev.
2012, 64, 18-23.
207. Gorelik, J. V.; Paramonov, B. A.; Blinova, M. I.; Diakonov, I. A.; Kukhareva, L. V.;
Pinaev, G. P. Matrigel Increases the Rate of Split Wound Healing and Promotes
Keratinocyte 'Take' in Deep Wounds in Rats. Cytotechnology 2000, 32, 79-86.
208. Uemura, M.; Refaat, M. M.; Shinoyama, M.; Hayashi, H.; Hashimoto, N.; Takahashi,
J. Matrigel Supports Survival and Neuronal Differentiation of Grafted Embryonic Stem CellDerived Neural Precursor Cells. J. Neurosci. Res. 2010, 88, 542-551.
209. Rosso, F.; Marino, G.; Giordano, A.; Barbarisi, M.; Parmeggiani, D.; Barbarisi, A.
Smart Materials as Scaffolds for Tissue Engineering. J. Cell. Physiol. 2005, 203, 465-470.
210. Villa-Diaz, L. G.; Ross, A. M.; Lahann, J.; Krebsbach, P. H. Concise Review: The
Evolution of Human Pluripotent Stem Cell Culture: From Feeder Cells to Synthetic
Coatings. Stem Cells 2013, 31, 1-7.
211. Chockalingam, K.; Blenner, M.; Banta, S. Design and Application of StimulusResponsive Peptide Systems. Protein Eng. Des. Sel. 2007, 20, 155-161.
212. Rapaport, H.; Grisaru, H.; Silberstein, T. Hydrogel Scaffolds of Amphiphilic and
Acidic Beta-Sheet Peptides. Adv. Funct. Mater. 2008, 18, 2889-2896.
213. Salick, D. A.; Kretsinger, J. K.; Pochan, D. J.; Schneider, J. P. Inherent Antibacterial
Activity of a Peptide-Based Beta-Hairpin Hydrogel. J. Am. Chem. Soc. 2007, 129, 1479314799.
214. Gupta, K.; Jang, H.; Harlen, K.; Puri, A.; Nussinov, R.; Schneider, J. P.; Blumenthal,
R. Mechanism of Membrane Permeation Induced by Synthetic Beta-Hairpin Peptides.
Biophys. J. 2013, 105, 2093-2103.
215. Chan, B. P.; Leong, K. W. Scaffolding in Tissue Engineering: General Approaches
and Tissue-Specific Considerations. Eur. Spine J. 2008, 17, S467-S479.
216. Frantz, C.; Stewart, K. M.; Weaver, V. M. The Extracellular Matrix at a Glance. J.
Cell Sci. 2010, 123, 4195-4200.
59
217. Jung, J. P.; Jones, J. L.; Cronier, S. A.; Collier, J. H. Modulating the Mechanical
Properties of Self-Assembled Peptide Hydrogels Via Native Chemical Ligation.
Biomaterials 2008, 29, 2143-2151.
218. Berns, E. J.; Sur, S.; Pan, L. L.; Goldberger, J. E.; Suresh, S.; Zhang, S. M.;
Kessler, J. A.; Stupp, S. I. Aligned Neurite Outgrowth and Directed Cell Migration in SelfAssembled Monodomain Gels. Biomaterials 2014, 35, 185-195.
219. Storrie, H.; Guler, M. O.; Abu-Amara, S. N.; Volberg, T.; Rao, M.; Geiger, B.; Stupp,
S. I. Supramolecular Crafting of Cell Adhesion. Biomaterials 2007, 28, 4608-4618.
220. Beniash, E.; Hartgerink, J. D.; Storrie, H.; Stendahl, J. C.; Stupp, S. I. SelfAssembling Peptide Amphiphile Nanofiber Matrices for Cell Entrapment. Acta Biomater.
2005, 1, 387-397.
221. Webber, M. J.; Tongers, J.; Renault, M. A.; Roncalli, J. G.; Losordo, D. W.; Stupp,
S. I. Development of Bioactive Peptide Amphiphiles for Therapeutic Cell Delivery. Acta
Biomater. 2010, 6, 3-11.
222. Horii, A.; Wang, X. M.; Gelain, F.; Zhang, S. G. Biological Designer Self-Assembling
Peptide Nanofiber Scaffolds Significantly Enhance Osteoblast Proliferation, Differentiation
and 3-D Migration. PLoS One 2007, 2, e190.
223. Jung, J. P.; Nagaraj, A. K.; Fox, E. K.; Rudra, J. S.; Devgun, J. M.; Collier, J. H. CoAssembling Peptides as Defined Matrices for Endothelial Cells. Biomaterials 2009, 30,
2400-2410.
224. Jung, J. P.; Moyano, J. V.; Collier, J. H. Multifactorial Optimization of Endothelial
Cell Growth Using Modular Synthetic Extracellular Matrices. Integr. Biol. 2011, 3, 185-196.
225. Gasiorowski, J. Z.; Collier, J. H. Directed Intermixing in Multicomponent SelfAssembling Biomaterials. Biomacromolecules 2011, 12, 3549-3558.
226. Genove, E.; Shen, C.; Zhang, S. G.; Semino, C. E. The Effect of Functionalized
Self-Assembling Peptide Scaffolds on Human Aortic Endothelial Cell Function.
Biomaterials 2005, 26, 3341-3351.
227. Gelain, F.; Bottai, D.; Vescovi, A.; Zhang, S. G. Designer Self-Assembling Peptide
Nanofiber Scaffolds for Adult Mouse Neural Stem Cell 3-Dimensional Cultures. PLoS One
2006, 1, e119.
228. Kumada, Y.; Zhang, S. G. Significant Type I and Type Iii Collagen Production from
Human Periodontal Ligament Fibroblasts in 3d Peptide Scaffolds without Extra Growth
Factors. PLoS One 2010, 5, e10305.
229. Giano, M. C.; Pochan, D. J.; Schneider, J. P. Controlled Biodegradation of SelfAssembling Beta-Hairpin Peptide Hydrogels by Proteolysis with Matrix Metalloproteinase13. Biomaterials 2011, 32, 6471-6477.
230. Mao, Y.; Schwarzbauer, J. E. Stimulatory Effects of a Three-Dimensional
Microenvironment on Cell-Mediated Fibronectin Fibrillogenesis. J. Cell Sci. 2005, 118,
4427-36.
231. Serebriiskii, I.; Castelló-Cros, R.; Lamb, A.; Golemis, E. A.; Cukierman, E.
Fibroblast-Derived 3d Matrix Differentially Regulates the Growth and DrugResponsiveness of Human Cancer Cells. Matrix Biol. 2008, 27, 573-585.
232. Millerot-Serrurot, E.; Guilbert, M.; Fourre, N.; Witkowski, W.; Said, G.; Van Gulick,
L.; Terryn, C.; Zahm, J. M.; Garnotel, R.; Jeannesson, P. 3d Collagen Type I Matrix
Inhibits the Antimigratory Effect of Doxorubicin. Cancer Cell International 2010, 10.
233. Meads, M. B.; Gatenby, R. A.; Dalton, W. S. Environment-Mediated Drug
Resistance: A Major Contributor to Minimal Residual Disease. Nat. Rev. Cancer 2009, 9,
665-74.
60
234. Ratner, B. D.; Bryant, S. J. Biomaterials: Where We Have Been and Where We Are
Going. Annu. Rev. Biomed. Eng. 2004, 6, 41-75.
235. Narmoneva, D. A.; Oni, O.; Sieminski, A. L.; Zhang, S. G.; Gertler, J. P.; Kamm, R.
D.; Lee, R. T. Self-Assembling Short Oligopeptides and the Promotion of Angiogenesis.
Biomaterials 2005, 26, 4837-4846.
236. Cunha, C.; Panseri, S.; Villa, O.; Silva, D.; Gelain, F. 3d Culture of Adult Mouse
Neural Stem Cells within Functionalized Self-Assembling Peptide Scaffolds. Int. J.
Nanomedicine 2011, 6, 943-955.
237. Koutsopoulos, S.; Zhang, S. G. Long-Term Three-Dimensional Neural Tissue
Cultures in Functionalized Self-Assembling Peptide Hydrogels, Matrigel and Collagen I.
Acta Biomater. 2013, 9, 5162-5169.
238. Semino, C. E.; Merok, J. R.; Crane, G. G.; Panagiotakos, G.; Zhang, S. G.
Functional Differentiation of Hepatocyte-Like Spheroid Structures from Putative Liver
Progenitor Cells in Three-Dimensional Peptide Scaffolds. Differentiation 2003, 71, 262270.
239. Sinthuvanich, C.; Haines-Butterick, L. A.; Nagy, K. J.; Schneider, J. P. Iterative
Design of Peptide-Based Hydrogels and the Effect of Network Electrostatics on Primary
Chondrocyte Behavior. Biomaterials 2012, 33, 7478-7488.
240. Tian, Y. F.; Devgun, J. M.; Collier, J. H. Fibrillized Peptide Microgels for Cell
Encapsulation and 3d Cell Culture. Soft Matter 2011, 7, 6005-6011.
241. Maude, S.; Miles, D. E.; Felton, S. H.; Ingram, J.; Carrick, L. M.; Wilcox, R. K.;
Ingham, E.; Aggeli, A. De Novo Designed Positively Charged Tape-Forming Peptides:
Self-Assembly and Gelation in Physiological Solutions and Their Evaluation as 3d Matrices
for Cell Growth. Soft Matter 2011, 7, 8085-8099.
242. Kyle, S.; Felton, S. H.; McPherson, M. J.; Aggeli, A.; Ingham, E. Rational Molecular
Design of Complementary Self-Assembling Peptide Hydrogels. Adv. Healthcare Mater.
2012, 1, 640-645.
243. Peyton, S. R.; Ghajar, C. M.; Khatiwala, C. B.; Putnam, A. J. The Emergence of
Ecm Mechanics and Cytoskeletal Tension as Important Regulators of Cell Function. Cell
Biochem. Biophys. 2007, 47, 300-320.
244. Sieminski, A. L.; Was, A. S.; Kim, G.; Gong, H.; Kamm, R. D. The Stiffness of
Three-Dimensional Ionic Self-Assembling Peptide Gels Affects the Extent of Capillary-Like
Network Formation. Cell Biochem. Biophys. 2007, 49, 73-83.
245. Stevenson, M. D.; Piristine, H.; Hogrebe, N. J.; Nocera, T. M.; Boehm, M. W.; Reen,
R. K.; Koelling, K. W.; Agarwal, G.; Sarang-Sieminski, A. L.; Gooch, K. J. A SelfAssembling Peptide Matrix Used to Control Stiffness and Binding Site Density Supports
the Formation of Microvascular Networks in Three Dimensions. Acta Biomater. 2013, 9,
7651-7661.
246. Rowlands, A. S.; George, P. A.; Cooper-White, J. J. Directing Osteogenic and
Myogenic Differentiation of Mscs: Interplay of Stiffness and Adhesive Ligand Presentation.
Am. J. Physiol. Cell Physiol. 2008, 295, C1037-C1044.
247. Cheng, T.-Y.; Chen, M.-H.; Chang, W.-H.; Huang, M.-Y.; Wang, T.-W. Neural Stem
Cells Encapsulated in a Functionalized Self-Assembling Peptide Hydrogel for Brain Tissue
Engineering. Biomaterials 2013, 34, 2005-2016.
248. Collier, J. H.; Rudra, J. S.; Gasiorowski, J. Z.; Jung, J. P. Multi-Component
Extracellular Matrices Based on Peptide Self-Assembly. Chem. Soc. Rev. 2010, 39, 34133424.
61
249. Kloxin, A. M.; Kloxin, C. J.; Bowman, C. N.; Anseth, K. S. Mechanical Properties of
Cellularly Responsive Hydrogels and Their Experimental Determination. Adv. Mater. 2010,
22, 3484-3494.
250. Ozbas, B.; Rajagopal, K.; Schneider, J. P.; Pochan, D. J. Semiflexible Chain
Networks Formed Via Self-Assembly of Beta-Hairpin Molecules. Phys. Rev. Lett. 2004, 93,
268106
251. Branco, M. C.; Pochan, D. J.; Wagner, N. J.; Schneider, J. P. Macromolecular
Diffusion and Release from Self-Assembled Beta-Hairpin Peptide Hydrogels. Biomaterials
2009, 30, 1339-1347.
252. Hsu, L.; Cvetanovich, G. L.; Stupp, S. I. Peptide Amphiphile Nanofibers with
Conjugated Polydiacetylene Backbones in Their Core. J. Am. Chem. Soc. 2008, 130,
3892-3899.
253. Bakota, E. L.; Aulisa, L.; Galler, K. M.; Hartgerink, J. D. Enzymatic Cross-Linking of
a Nanofibrous Peptide Hydrogel. Biomacromolecules 2011, 12, 82-87.
254. Dawson, P. E.; Muir, T. W.; Clarklewis, I.; Kent, S. B. H. Synthesis of Proteins by
Native Chemical Ligation. Science 1994, 266, 776-779.
255. Meng, H.; Chen, L. Y.; Ye, Z. Y.; Wang, S. T.; Zhao, X. J. The Effect of a SelfAssembling Peptide Nanofiber Scaffold (Peptide) When Used as a Wound Dressing for the
Treatment of Deep Second Degree Burns in Rats. J. Biomed. Mater. Res., Part B 2009,
89B, 379-391.
256. Cho, H.; Balaji, S.; Sheikh, A. Q.; Hurley, J. R.; Tian, Y. F.; Collier, J. H.;
Crombleholme, T. M.; Narmoneva, D. A. Regulation of Endothelial Cell Activation and
Angiogenesis by Injectable Peptide Nanofibers. Acta Biomater. 2012, 8, 154-164.
257. Guo, J. S.; Leung, K. K. G.; Su, H. X.; Yuan, Q. J.; Wang, L.; Chu, T. H.; Zhang, W.
M.; Pu, J. K. S.; Ng, G. K. P.; Wong, W. M.; Dai, X.; Wu, W. T. Self-Assembling Peptide
Nanofiber Scaffold Promotes the Reconstruction of Acutely Injured Brain. NanomedNanotechnol 2009, 5, 345-351.
258. Masuhara, H.; Fujii, T.; Watanabe, Y.; Koyama, N.; Tokuhiro, K. Novel Infectious
Agent-Free Hemostatic Material (Tdm-621) in Cardiovascular Surgery. Ann. Thorac.
Cardiovasc. Surg. 2012, 18, 444-451.
259. Cherny, I.; Gazit, E. Amyloids: Not Only Pathological Agents but Also Ordered
Nanomaterials. Angewandte Chemie-International Edition 2008, 47, 4062-4069.
260. Chiti, F.; Dobson, C. M. Protein Misfolding, Functional Amyloid, and Human
Disease. Annu. Rev. Biochem. 2006, 75, 333-366.
261. Cohen, F. E.; Prusiner, S. B. Pathologic Conformations of Prion Proteins. Annu.
Rev. Biochem. 1998, 67, 793-819.
262. Westermark, P.; Lundmark, K.; Westermark, G. T. Fibrils from Designed NonAmyloid-Related Synthetic Peptides Induce Aa-Amyloidosis During Inflammation in an
Animal Model. PLoS One 2009, 4, e6041.
263. Aguzzi, A.; Sigurdson, C.; Heikenwaelder, M., Molecular Mechanisms of Prion
Pathogenesis. In Annual Review of Pathology-Mechanisms of Disease, Annual Reviews:
Palo Alto, 2008; Vol. 3, pp 11-40.
264. Roses, A. D.; Saunders, A. M. Perspective on a Pathogenesis and Treatment of
Alzheimer’s Disease. Alzheimers Dement. 2006, 2, 59-70.
265. Fowler, D. M.; Koulov, A. V.; Balch, W. E.; Kelly, J. W. Functional Amyloid - from
Bacteria to Humans. Trends Biochem. Sci. 2007, 32, 217-224.
266. Chapman, M. R.; Robinson, L. S.; Pinkner, J. S.; Roth, R.; Heuser, J.; Hammar, M.;
Normark, S.; Hultgren, S. J. Role of Escherichia Coli Curli Operons in Directing Amyloid
Fiber Formation. Science 2002, 295, 851-855.
62
267. Maji, S. K.; Perrin, M. H.; Sawaya, M. R.; Jessberger, S.; Vadodaria, K.; Rissman,
R. A.; Singru, P. S.; Nilsson, K. P. R.; Simon, R.; Schubert, D.; Eisenberg, D.; Rivier, J.;
Sawchenko, P.; Vale, W.; Riek, R. Functional Amyloids as Natural Storage of Peptide
Hormones in Pituitary Secretory Granules. Science 2009, 325, 328-332.
268. Grau, C. M.; Greene, L. A. Use of Pc12 Cells and Rat Superior Cervical Ganglion
Sympathetic Neurons as Models for Neuroprotective Assays Relevant to Parkinson's
Disease. Methods Mol. Biol. 2012, 846, 201-211.
269. Bella, A.; Ray, S.; Shaw, M.; Ryadnov, M. G. Arbitrary Self-Assembly of Peptide
Extracellular Microscopic Matrices. Angew. Chem., Int. Ed. 2012, 51, 428-431.
270. Torchilin, V. P. Recent Advances with Liposomes as Pharmaceutical Carriers. Nat.
Rev. Drug Discovery 2005, 4, 145-160.
271. Kumari, A.; Yadav, S. K.; Yadav, S. C. Biodegradable Polymeric Nanoparticles
Based Drug Delivery Systems. Colloids Surf., B 2010, 75, 1-18.
272. Pack, D. W.; Hoffman, A. S.; Pun, S.; Stayton, P. S. Design and Development of
Polymers for Gene Delivery. Nat. Rev. Drug Discovery 2005, 4, 581-593.
273. Seow, W. Y.; Yang, Y. Y. A Class of Cationic Triblock Amphiphilic Oligopeptides as
Efficient Gene-Delivery Vectors. Adv. Mater. 2009, 21, 86-90.
274. van Hell, A. J.; Costa, C.; Flesch, F. M.; Sutter, M.; Jiskoot, W.; Crommelin, D. J. A.;
Hennink, W. E.; Mastrobattista, E. Self-Assembly of Recombinant Amphiphilic
Oligopeptides into Vesicles. Biomacromolecules 2007, 8, 2753-2761.
275. Chung, E. J.; Cheng, Y.; Morshed, R.; Nord, K.; Han, Y.; Wegscheid, M. L.;
Auffinger, B.; Wainwright, D. A.; Lesniak, M. S.; Tirrell, M. V. Fibrin-Binding, Peptide
Amphiphile Micelles for Targeting Glioblastoma. Biomaterials 2014, 35, 1249-1256.
276. Rajangam, K.; Behanna, H. A.; Hui, M. J.; Han, X. Q.; Hulvat, J. F.; Lomasney, J.
W.; Stupp, S. I. Heparin Binding Nanostructures to Promote Growth of Blood Vessels.
Nano Lett. 2006, 6, 2086-2090.
277. Chow, L. W.; Bitton, R.; Webber, M. J.; Carvajal, D.; Shull, K. R.; Sharma, A. K.;
Stupp, S. I. A Bioactive Self-Assembled Membrane to Promote Angiogenesis. Biomaterials
2011, 32, 1574-1582.
278. Stendahl, J. C.; Wang, L. J.; Chow, L. W.; Kaufman, D. B.; Stupp, S. I. Growth
Factor Delivery from Self-Assembling Nanofibers to Facilitate Islet Transplantation.
Transplantation 2008, 86, 478-481.
279. Chow, L. W.; Wang, L. J.; Kaufman, D. B.; Stupp, S. I. Self-Assembling
Nanostructures to Deliver Angiogenic Factors to Pancreatic Islets. Biomaterials 2010, 31,
6154-6161.
280. Webber, M. J.; Han, X. Q.; Murthy, S. N. P.; Rajangam, K.; Stupp, S. I.; Lomasney,
J. W. Capturing the Stem Cell Paracrine Effect Using Heparin-Presenting Nanofibres to
Treat Cardiovascular Diseases. J. Tissue Eng. Regen. Med. 2010, 4, 600-610.
281. Matson, J. B.; Stupp, S. I. Drug Release from Hydrazone-Containing Peptide
Amphiphiles. Chem. Commun. 2011, 47, 7962-7964.
282. Webber, M. J.; Matson, J. B.; Tamboli, V. K.; Stupp, S. I. Controlled Release of
Dexamethasone from Peptide Nanofiber Gels to Modulate Inflammatory Response.
Biomaterials 2012, 33, 6823-6832.
283. Webber, M. J.; Newcomb, C. J.; Bitton, R.; Stupp, S. I. Switching of Self-Assembly
in a Peptide Nanostructure with a Specific Enzyme. Soft Matter 2011, 7, 9665-9672.
284. Zarzhitsky, S.; Rapaport, H. The Interactions between Doxorubicin and Amphiphilic
and Acidic Β-Sheet Peptides Towards Drug Delivery Hydrogels. J. Colloid Interface Sci.
2011, 360, 525-531.
63
285. Nagai, Y.; Unsworth, L. D.; Koutsopoulos, S.; Zhang, S. G. Slow Release of
Molecules in Self-Assembling Peptide Nanofiber Scaffold. J. Control. Release 2006, 115,
18-25.
286. Koutsopoulos, S.; Unsworth, L. D.; Nagaia, Y.; Zhang, S. G. Controlled Release of
Functional Proteins through Designer Self-Assembling Peptide Nanofiber Hydrogel
Scaffold. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 4623-4628.
287. Gelain, F.; Unsworth, L. D.; Zhang, S. G. Slow and Sustained Release of Active
Cytokines from Self-Assembling Peptide Scaffolds. J. Control. Release 2010, 145, 231239.
288. Koutsopoulos, S.; Zhang, S. G. Two-Layered Injectable Self-Assembling Peptide
Scaffold Hydrogels for Long-Term Sustained Release of Human Antibodies. J. Control.
Release 2012, 160, 451-458.
289. Branco, M. C.; Pochan, D. J.; Wagner, N. J.; Schneider, J. P. The Effect of Protein
Structure on Their Controlled Release from an Injectable Peptide Hydrogel. Biomaterials
2010, 31, 9527-9534.
290. Rudra, J. S.; Mishra, S.; Chong, A. S.; Mitchell, R. A.; Nardin, E. H.; Nussenzweig,
V.; Collier, J. H. Self-Assembled Peptide Nanofibers Raising Durable Antibody Responses
against a Malaria Epitope. Biomaterials 2012, 33, 6476-6484.
291. Rudra, J. S.; Sun, T.; Bird, K. C.; Daniels, M. D.; Gasiorowski, J. Z.; Chong, A. S.;
Collier, J. H. Modulating Adaptive Immune Responses to Peptide Self-Assemblies. ACS
Nano 2012, 6, 1557-1564.
292. Chen, J. J.; Pompano, R. R.; Santiago, F. W.; Maillat, L.; Sciammas, R.; Sun, T.;
Han, H. F.; Topham, D. J.; Chong, A. S.; Collier, J. H. The Use of Self-Adjuvanting
Nanofiber Vaccines to Elicit High-Affinity B Cell Responses to Peptide Antigens without
Inflammation. Biomaterials 2013, 34, 8776-8785.
293. Rudra, J. S.; Tian, Y. F.; Jung, J. P.; Collier, J. H. A Self-Assembling Peptide Acting
as an Immune Adjuvant. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 622-627.
294. Hudalla, G. A.; Modica, J. A.; Tian, Y. F.; Rudra, J. S.; Chong, A. S.; Sun, T.;
Mrksich, M.; Collier, J. H. A Self-Adjuvanting Supramolecular Vaccine Carrying a Folded
Protein Antigen. Adv. Healthcare Mater. 2013, 2, 1114-1119.
295. Guo, Y.; Bozic, D.; Malashkevich, V. N.; Kammerer, R. A.; Schulthess, T.; Enger, J.
All-Trans Retinol, Vitamin D and Other Hydrophobic Compounds Bind in the Axial Pore of
the Five-Stranded Coiled-Coil Domain of Cartilage Oligomeric Matrix Protein. EMBO J.
1998, 17, 5265-5272.
296. Ozbek, S.; Engel, J.; Stetefeld, J. Storage Function of Cartilage Oligomeric Matrix
Protein: The Crystal Structure of the Coiled-Coil Domain in Complex with Vitamin D-3.
EMBO J. 2002, 21, 5960-5968.
297. McFarlane, A. A.; Orriss, G. L.; Stetefeld, J. The Use of Coiled-Coil Proteins in Drug
Delivery Systems. Eur. J. Pharmacol. 2009, 625, 101-107.
298. Liu, J.; Zheng, Q.; Deng, Y. Q.; Kallenbach, N. R.; Lu, M. Conformational Transition
between Four and Five-Stranded Phenylalanine Zippers Determined by a Local Packing
Interaction. J. Mol. Biol. 2006, 361, 168-179.
299. Schroeder, U.; Graff, A.; Buchmeier, S.; Rigler, P.; Silvan, U.; Tropel, D.; Jockusch,
B. M.; Aebi, U.; Burkhard, P.; Schoenenberger, C. A. Peptide Nanoparticles Serve as a
Powerful Platform for the Immunogenic Display of Poorly Antigenic Actin Determinants. J.
Mol. Biol. 2009, 386, 1368-1381.
300. Pimentel, T.; Yan, Z.; Jeffers, S. A.; Holmes, K. V.; Hodges, R. S.; Burkhard, P.
Peptide Nanoparticles as Novel Immunogens: Design and Analysis of a Prototypic Severe
Acute Respiratory Syndrome Vaccine. Chem. Biol. Drug Des. 2009, 73, 53-61.
64
301. Raman, S.; Machaidze, G.; Lustig, A.; Aebi, U.; Burkhard, P. Structure-Based
Design of Peptides That Self-Assemble into Regular Polyhedral Nanoparticles. NanomedNanotechnol 2006, 2, 95-102.
302. Merrifield, R. B. Solid Phase Peptide Synthesis .1. Synthesis of a Tetrapeptide. J.
Am. Chem. Soc. 1963, 85, 2149-2154.
303. Guzman, F.; Barberis, S.; Illanes, A. Peptide Synthesis: Chemical or Enzymatic.
Electron. J. Biotechn. 2007, 10, 279-314.
304. Marx, V. Watching Peptide Drugs Grow Up. Chem. Eng. News 2005, 83, 17-24.
305. Corradin, G.; Kajava, A. V.; Verdini, A. Long Synthetic Peptides for the Production
of Vaccines and Drugs: A Technological Platform Coming of Age. Sci. Transl. Med. 2010,
2, 50rv3.
306. Morreale, G.; Lee, E. G.; Jones, D. B.; Middelberg, A. P. J. Bioprocess-Centered
Molecular Design (Bmd) for the Efficient Production of an Interfacially Active Peptide.
Biotechnol. Bioeng. 2004, 87, 912-923.
307. Hannig, G.; Makrides, S. C. Strategies for Optimizing Heterologous Protein
Expression in Escherichia Coli. Trends Biotechnol. 1998, 16, 54-60.
308. Young, C. L.; Britton, Z. T.; Robinson, A. S. Recombinant Protein Expression and
Purification: A Comprehensive Review of Affinity Tags and Microbial Applications.
Biotechnol. J. 2012, 7, 620-634.
309. Terpe, K. Overview of Tag Protein Fusions: From Molecular and Biochemical
Fundamentals to Commercial Systems. Appl. Microbiol. Biotechnol. 2003, 60, 523-533.
310. Kuliopulos, A.; Walsh, C. T. Production, Purification, and Cleavage of Tandem
Repeats of Recombinant Peptides. J. Am. Chem. Soc. 1994, 116, 4599-4607.
311. Riley, J. M.; Aggeli, A.; Koopmans, R. J.; McPherson, M. J. Bioproduction and
Characterization of a Ph Responsive Self-Assembling Peptide. Biotechnol. Bioeng. 2009,
103, 241-251.
312. Hartmann, B. M.; Kaar, W.; Falconer, R. J.; Zeng, B.; Middelberg, A. P. J.
Expression and Purification of a Nanostructure-Forming Peptide. J. Biotechnol. 2008, 135,
85-91.
313. Shen, S. H. Multiple Joined Genes Prevent Product Degradation in EscherichiaColi. Proc. Natl. Acad. Sci. U. S. A. 1984, 81, 4627-4631.
314. Lee, J. H.; Kim, M. S.; Cho, J. H.; Kim, S. C. Enhanced Expression of Tandem
Multimers of the Antimicrobial Peptide Buforin Ii in Escherichia Coli by the Dead-Box
Protein and Trxb Mutant. Appl. Microbiol. Biotechnol. 2002, 58, 790-796.
315. Kim, H. K.; Chun, D. S.; Kim, J. S.; Yun, C. H.; Lee, J. H.; Hong, S. K.; Kang, D. K.
Expression of the Cationic Antimicrobial Peptide Lactoferricin Fused with the Anionic
Peptide in Escherichia Coli. Appl. Microbiol. Biotechnol. 2006, 72, 330-338.
316. Nikkhah, M.; Jawad-Alami, Z.; Demydchuk, M.; Ribbons, D.; Paoli, M. Engineering
of Beta-Propeller Protein Scaffolds by Multiple Gene Duplication and Fusion of an
Idealized Wd Repeat. Biomol. Eng. 2006, 23, 185-194.
317. Middelberg, A. P. J.; Dimitrijev-Dwyer, M. A Designed Biosurfactant Protein for
Switchable Foam Control. ChemPhysChem 2011, 12, 1426-1429.
318. Hecht, M. H.; Richardson, J. S.; Richardson, D. C.; Ogden, R. C. Denovo Design,
Expression, and Characterization of Felix - a 4-Helix Bundle Protein of Native-Like
Sequence. Science 1990, 249, 884-891.
319. Lee, J. H.; Minn, I.; Park, C. B.; Kim, S. C. Acidic Peptide-Mediated Expression of
the Antimicrobial Peptide Buforin Ii as Tandem Repeats in Escherichia Coli. Protein Expr.
Purif. 1998, 12, 53-60.
65
320. Raab, D.; Graf, M.; Notka, F.; Schodl, T.; Wagner, R. The Geneoptimizer Algorithm:
Using a Sliding Window Approach to Cope with the Vast Sequence Space in
Multiparameter DNA Sequence Optimization. Syst. Synth. Biol. 2010, 4, 215-25.
321. Welch, M.; Villalobos, A.; Gustafsson, C.; Minshull, J. You're One in a Googol:
Optimizing Genes for Protein Expression. J. Royal Soc. Interface 2009, 6, S467-S476.
322. Ramirez, D. M.; Bentley, W. E. Enhancement of Recombinant Protein-Synthesis
and Stability Via Coordinated Amino-Acid Addition. Biotechnol. Bioeng. 1993, 41, 557-565.
323. Ruiz, J.; Fernandez-Castane, A.; de Mas, C.; Gonzalez, G.; Lopez-Santin, J. From
Laboratory to Pilot Plant E. Coli Fed-Batch Cultures: Optimizing the Cellular Environment
for Protein Maximization. J. Ind. Microbiol. Biotechnol. 2013, 40, 335-343.
324. Sharp, P. M.; Li, W. H. The Codon Adaptation Index - a Measure of Directional
Synonymous Codon Usage Bias, and Its Potential Applications. Nucleic Acids Res. 1987,
15, 1281-1295.
325. Liew, O. W.; Chong, J. P. C.; Yandle, T. G.; Brennan, S. O. Preparation of
Recombinant Thioredoxin Fused N-Terminal Procnp: Analysis of Enterokinase Cleavage
Products Reveals New Enterokinase Cleavage Sites. Protein Expr. Purif. 2005, 41, 332340.
326. Jenny, R. J.; Mann, K. G.; Lundblad, R. L. A Critical Review of the Methods for
Cleavage of Fusion Proteins with Thrombin and Factor Xa. Protein Expr. Purif. 2003, 31,
1-11.
327. Hwang, P. M.; Pan, J. S.; Sykes, B. D. Targeted Expression, Purification, and
Cleavage of Fusion Proteins from Inclusion Bodies in Escherichia Coli. FEBS Lett. 2014,
588, 247-252.
328. Hosfield, T.; Lu, Q. Influence of the Amino Acid Residue Downstream of
(Asp)(4)Lys on Enterokinase Cleavage of a Fusion Protein. Anal. Biochem. 1999, 269, 1016.
329. Nagai, K.; Thogersen, H. C. Generation of Beta-Globin by Sequence-Specific
Proteolysis of a Hybrid Protein Produced in Escherichia-Coli. Nature 1984, 309, 810-812.
330. Malakhov, M. P.; Mattern, M. R.; Malakhova, O. A.; Drinker, M.; Weeks, S. D.; Butt,
T. R. Sumo Fusions and Sumo-Specific Protease for Efficient Expression and Purification
of Proteins. J. Struct. Funct. Genomics 2004, 5, 75-86.
331. Dougherty, W. G.; Carrington, J. C.; Cary, S. M.; Parks, T. D. Biochemical and
Mutational Analysis of a Plant-Virus Polyprotein Cleavage Site. EMBO J. 1988, 7, 12811287.
332. Carrington, J. C.; Dougherty, W. G. A Viral Cleavage Site Cassette - Identification of
Amino-Acid Sequences Required for Tobacco Etch Virus Polyprotein Processing. Proc.
Natl. Acad. Sci. U. S. A. 1988, 85, 3391-3395.
333. Chang, J. Y. Thrombin Specificity - Requirement for Apolar Amino-Acids Adjacent
to the Thrombin Cleavage Site of Polypeptide Substrate. Eur. J. Biochem. 1985, 151, 217224.
334. Cordingley, M. G.; Callahan, P. L.; Sardana, V. V.; Garsky, V. M.; Colonno, R. J.
Substrate Requirements of Human Rhinovirus 3c Protease for Peptide Cleavage Invitro. J.
Biol. Chem. 1990, 265, 9062-9065.
335. Bornstein, P.; Balian, G., [14] Cleavage at Asngly Bonds with Hydroxylamine. In
Methods Enzymol., Hirs, C. H. W.; Serge, N. T., Eds. Academic Press: 1977; Vol. Volume
47, pp 132-145.
336. Smith, B. J., Chemical Cleavage of Proteins at Asparaginyl-Glycyl Peptide Bonds.
In Protein Protocols Handbook, Third Edition, Walker, J. M., Ed. Humana Press Inc:
Totowa, 2009; pp 899-903.
66
337. Tang, H. Y.; Speicher, D. W. Identification of Alternative Products and Optimization
of 2-Nitro-5-Thiocyanatobenzoic Acid Cyanylation and Cleavage at Cysteine Residues.
Anal. Biochem. 2004, 334, 48-61.
338. Wu, J.; Watson, J. T. Optimization of the Cleavage Reaction for Cyanylated
Cysteinyl Proteins for Efficient and Simplified Mass Mapping. Anal. Biochem. 1998, 258,
268-276.
339. Li, A. Q.; Sowder, R. C.; Henderson, L. E.; Moore, S. P.; Garfinkel, D. J.; Fisher, R.
J. Chemical Cleavage at Aspartyl Residues for Protein Identification. Anal. Chem. 2001,
73, 5395-5402.
340. Zhang, S. F.; Basile, F. Site-Specific Pyrolysis-Induced Cleavage at Aspartic Acid
Residue in Peptides and Proteins. J. Proteome Res. 2007, 6, 1700-1704.
341. Reed, D. C.; Barnard, G. C.; Anderson, E. B.; Klein, L. T.; Gerngross, T. U.
Production and Purification of Self-Assembling Peptides in Ralstonia Eutropha. Protein
Expr. Purif. 2006, 46, 179-188.
342. Kojima, S.; Kuriki, Y.; Sato, Y.; Arisaka, F.; Kumagai, I.; Takahashi, S.; Miura, K.
Synthesis of Alpha-Helix-Forming Peptides by Gene Engineering Methods and Their
Characterization by Circular Dichroism Spectra Measurements. Biochimica Et Biophysica
Acta-Protein Structure and Molecular Enzymology 1996, 1294, 129-137.
343. Kaar, W.; Hartmann, B. M.; Fan, Y.; Zeng, B.; Lua, L. H. L.; Dexter, A. F.; Falconer,
R. J.; Middelberg, A. P. J. Microbial Bio-Production of a Recombinant Stimuli-Responsive
Biosurfactant. Biotechnol. Bioeng. 2009, 102, 176-187.
344. Azevedo, A. M.; Rosa, P. A. J.; Ferreira, I. F.; Aires-Barros, M. R. ChromatographyFree Recovery of Biopharmaceuticals through Aqueous Two-Phase Processing. Trends
Biotechnol. 2009, 27, 240-247.
345. Przybycien, T. M.; Pujar, N. S.; Steele, L. M. Alternative Bioseparation Operations:
Life Beyond Packed-Bed Chromatography. Curr. Opin. Biotechnol. 2004, 15, 469-478.
346. Dwyer, M. D.; Brech, M.; Yu, L.; Middelberg, A. P. J. Intensified Expression and
Purification of a Recombinant Biosurfactant Protein. Chem. Eng. Sci. 2014, 105, 12-21.
347. Hartmann, B. M.; Kaar, W.; Yoo, I. K.; Lua, L. H. L.; Falconer, R. J.; Middelberg, A.
P. J. The Chromatography-Free Release, Isolation and Purification of Recombinant
Peptide for Fibril Self-Assembly. Biotechnol. Bioeng. 2009, 104, 973-985.
348. Kyle, S.; Aggeli, A.; Ingham, E.; McPherson, M. J. Recombinant Self-Assembling
Peptides as Biomaterials for Tissue Engineering. Biomaterials 2010, 31, 9395-9405.
349. Dimitrijev-Dwyer, M.; He, L.; James, M.; Nelson, A.; Wang, L.; Middelberg, A. P. J.
The Effects of Acid Hydrolysis on Protein Biosurfactant Molecular, Interfacial, and Foam
Properties: Ph Responsive Protein Hydrolysates. Soft Matter 2012, 8, 5131-5139.
67
2 Bioproduction of Highly Charged Designer Peptide
Surfactants: Expression and Chromatography-Free
Purification of a Coiled-Coil Heteroconcatemer
2.1 Introduction
Designed self-assembling peptides show potential utility in several fields as described in
Chapter 1. One particularly promising area of research involves the use of designed selfassembling peptides as low-toxicity and highly biodegradable responsive surfactants.1-3
Work within the candidate’s research group has shown that amphipathic helical peptides
with a high content of carboxylate-containing amino acid residues confer emulsion stability
in high salt. This is due to the stabilising role of strong headgroup hydration and is
particularly true for peptides with carboxylate sidechains designed to align close the
interface (unpublished data). Peptides of this kind are desirable for use in applications
such as intravenous delivery of hydrophobic drugs as well as formulation of industrial
drilling and cutting fluids, where exposure to salt may destabilise emulsions formed by
conventional surfactants.4
Adoption of peptides for material applications is only viable if they can be produced at low
cost. As described in Chapter 1, the biological expression of designed peptides,
particularly as concatemers, offers an attractive alternative to expensive chemical
synthesis methods. Bioproduction methods should be able to be optimised to achieve
renewable and economically viable processes at large scale.5 However, the strong
negative charge present on peptides with a high carboxylate content is unfavourable for
expression as tandem repeats, due to electrostatic repulsion preventing stable folding.
One way to overcome this destabilisation is to reduce net charge by expressing the
charged peptide with an oppositely charged partner.6, 7
This chapter describes a bioproduction process for a highly anionic surfactant peptide.
This process was designed to use low-cost approaches, and therefore avoided the use of
fusion partners or enzymatic cleavage and targeted minimal reagent requirements. The
process was also designed to be easily scalable to provide commercially relevant
quantities of peptide. To this end the use of chromatographic purification was avoided and
selective precipitation steps were used in downstream processing.
68
Maquette-type artificial proteins were designed in which the highly anionic surfactant
peptide EDP-11 is paired with a cationic expression partner RDP-4 to form a coiled-coil
heteroconcatemer for expression. The polypeptide sequences were optimised for net
charge, hydropathy and predicted protease resistance (via the Guruprasad instability
index).8 Constructs with differing ratios (2:1 2:2 and 3:2) of anionic to cationic partner were
tested to investigate the effects of net charge and hydrophobic core packing on
concatemer expression.
The 2:2 construct, designated Het2-6, expressed as a soluble concatemer that
accumulated to high levels in E. coli. The concatemer showed extreme stability to heat and
proteolysis, allowing chromatography-free isolation by simple heat and pH precipitation
steps. Chemical denaturation studies showed the concatemer to have extremely high
thermodynamic stability. Finally, the bioinformatic parameters of the designed
heteroconcatemer were then compared to all open reading frames of four reference
microbes. This showed that this design lies in the extreme hydrophobic range for soluble
proteins while also possessing an unusually low instability index.
2.2 Experimental
2.2.1 Materials
Reagents and chemicals were of analytical grade unless otherwise indicated. Water was
obtained from an Elga Purelab Classic system (Veolia, Saint Maurice, France) and had a
resistivity of >18.2 MΩ.cm. Guanidine hydrochloride (G.HCl) (≥99%),
tris(hydroxymethyl)aminomethane (Tris) (≥99%) and 2-(N-morpholino)ethanesulfonic acid
(MES) hydrate (Biotech, performance certified) were purchased from Sigma-Aldrich.
Magnesium sulfate (≥98%) and citric acid (≥99%) were from Ajax Finechem. Protein
concentrations were determined by quantitative amino acid analysis (Australian Proteome
Analysis Facility, Sydney).
69
2.2.2 Peptide charge predictions
The net charge on peptide sequences was calculated as previously described9 using
equation 2-1:
𝑍 = � 𝑁𝑖
𝑖
10𝑝𝐾𝑎𝑖
10𝑝𝐻
−
�
𝑁
𝑗
10𝑝𝐻 + 10𝑝𝐾𝑎𝑖
10𝑝𝐻 + 10𝑝𝐾𝑗
(2-1)
𝑗
where Ni is the number of residues of a given cationic residue and pKai is the
corresponding pKa value, while Nj and pKaj are the values for the anionic residues. The
charge contributions from the N and C-termini were calculated similarly. Ionizing group pKa
values were: arginine, 12.5; lysine, 10.8; N-terminus, 7.8; histidine, 6.1; glutamate, 4.5;
aspartate, 4.1; C-terminus, 3.6. Ionizable sites were assumed to be non-interacting.
2.2.3 Cloning and expression of concatemers
DNA coding sequences were codon-optimised and synthesised by GENEART
(Regensburg, Germany). Genes were cloned into pET-48b(+) (Novagen) using NdeI and
BamHI restriction sites. Constructs were verified by DNA sequencing. For expression
testing of Het2-5, Het2-6 and Het2-7, DNA constructs were transformed into E. coli
BL21(DE3) and grown at 26 °C in 50 mL Luria broth (LB) containing 15 µg mL-1
kanamycin. Following induction with 1 mM isopropyl-β-D-thiogalactoside (IPTG), samples
were taken at 2 and 4 h and overnight. To test Het2-6 expression at larger scale, a single
colony was used to inoculate 50 mL starter culture in LB containing 15 μg mL-1 kanamycin.
This culture was then used to inoculate 2 × 500 mL shake flask cultures, which were
grown at 32 °C in the same medium. The cultures were grown to OD600 = 0.6 and induced
with 1 mM IPTG. Samples were taken at 0, 2, 5 and 8 h and analysed via SDS-PAGE. For
larger-scale shake flask expression, a single colony was used to inoculate 100 mL starter
culture in 2 × LB with 30 μg mL-1 kanomycin. This culture was used to inoculate 750 mL
shake flask cultures in the same medium. The cultures were grown to OD660 = 1.0 at 37 °C
and induced with 1 mM IPTG. The cultures were harvested at 4 h.
2.2.4 Downstream processing
Cell pellets were resuspended in 5 mL of H2O per gram of wet cell weight and lysed by 3
passes through an Emulsiflex-C5 valve homogeniser (Avestin, Ottawa, Canada) at 19,000
psi. The lysate was centrifuged at 20,000 g for 10 minutes at 20 °C. The clarified lysate
was heated to 90 °C for 10 minutes then centrifuged again at 20,000 g for 10 minutes at
70
20 °C to remove heat-precipitated cellular proteins. The resulting supernatant was
adjusted to pH 5.0 and centrifuged at 20,000 g for 10 minutes at 20 °C to harvest the
desired protein. The pellet was taken up in water and the pH was adjusted to 9.0.
Magnesium sulfate was added to 1 M and the solution was heated for 1 hour at 90 °C then
centrifuged at 20,000 g for 10 minutes at 20 °C. The pellet containing nucleic acid and
lipopolysaccharide was discarded and the supernatant was again adjusted to pH 5.0 and
centrifuged at 20,000 g for 10 minutes at 20 °C to recover Het2-6. The pellet was taken up
in water and the pH adjusted to 2.0. The final protein concentration was adjusted to ~10
mg mL-1 based on electronic circular dichroism spectra.
2.2.5 SDS-PAGE
NuPAGE 4-12% Bis-Tris gels, lithium dodecyl sulfate (LDS) sample buffer, reducing agent,
antioxidant and SeeBlue Plus2 Prestained Standard (Invitrogen, Grand Island, NY) were
used following standard protocols for analysis of purification steps. Analysis of gel images
used Image Lab software (Bio-Rad Laboratories, Hercules, CA, USA).
2.2.6 Mass spectrometry
Purified protein samples prepared at 2.5 mg mL-1 in water were diluted 3:20 with 5%
acetonitrile (ACN) 0.1% (v/v) trifluoroacetic acid (TFA) and injected into an Agilent
1100/1200 capillary LC system. Samples were desalted on an Agilent 300SB-C8 trap 0.3
× 5 mm (5 min) then eluted using a gradient of 0-60% buffer B at 10 µL min-1. Buffer A was
0.1% formic acid (v/v) in 5% ACN (v/v) 95% water (v/v) and Buffer B was 0.1% formic acid
in ACN. Eluted protein was analysed on-line on a QSTAR pulsar i over m/z 6002000. Data were reconstructed using Bayesian Protein Reconstruct Tool in BioTools
(Bruker).
2.2.7 Preparation of uninduced E. coli extract
E. coli KC6 cells were grown in 1 L of defined medium10 in 2 L shake flasks at 37 °C, and
harvested at OD600 5–6 (6,000 g, 30 min). The pellet was washed twice in cold S30 buffer
(10 mM Tris acetate, 14 mM magnesium acetate, 60 mM potassium acetate, pH 8.2), then
resuspended in 1 mL S30 buffer per gram of wet cells. The cells were lysed by a single
pass through an Emulsiflex-C5 valve homogeniser at 17,500 psi. The lysate was cleared
by centrifugation (30,000 g, 4 °C, 1 h). The extract (estimated protein concentration 60 mg
mL-1)11 was aliquotted, flash frozen, and stored at -80 °C until use. On thawing, protease
71
inhibitor cocktail (Sigma P8465) was added to give final concentrations of 5.75 mM 4-(2aminoethyl)benzenesulfonyl fluoride hydrochloride, 25 mM ethylenediaminetetraacetic
acid, 0.5 mM bestatin, 0.075 mM pepstatin A and 0.075 mM E-64. The cellular extract was
stable to proteolysis over 19 h as assessed by SDS-PAGE and electronic circular
dichroism.
2.2.8 Electronic circular dichroism
Electronic circular dichroism spectra were recorded on a Jasco J-815 spectropolarimeter
in 0.1-1 mm quartz cuvettes. The temperature was regulated using a Jasco CDF-426S/15
controller attached to a Zalman Reserator XT cooling system. All measurements used a
data integration time of 4 s and a band width of 1 nm. Spectral scans were recorded at a
scan speed of 50 nm min-1 and data pitch of 0.1 nm over 190-260 nm and were averaged
over 3 scans. Single wavelength measurements were recorded over 60 s with a 5 s data
pitch.
Spectra of purified Het2-6 were recorded at 20 °C using 100 µM protein in 10 mM Na+
citrate pH 3.0 or 10 mM Tris.HCl pH 8.0. To study chemical denaturation, 0.4 mg mL-1
Het2-6 solutions in 1.0-7.5 G.HCl were prepared by mixing separate protein stocks
prepared in 10 mM Tris.HCl pH 8.0 containing either 0 or 8.0 M G.HCl. Samples were
allowed to equilibrate for 10 min at 20, 50 or 90 °C prior to single wavelength readings at
222 nm. Reference samples of uninduced E. coli extract in 1.0-6.0 M G.HCl were prepared
by mixing solutions of cellular extract (est. 1.2 mg mL-1 protein) prepared in 10 mM MES
pH 7.0 containing either 0 or 7.0 M G.HCl. E. coli total protein samples were allowed to
equilibrate for 10 min at 20 °C prior to single wavelength readings at 230 nm.
For monitoring the effect of SDS-PAGE sample preparation on Het2-6 secondary
structure, samples were prepared at ca. 1.25 mg mL-1 in loading buffer and heated to 80
°C for 10 minutes. Spectra were recorded at 20 °C after heating and compared to spectra
of Het2-6 prepared in the absence of loading buffer and without heating.
Mean residue ellipticity ([θ]) for Het2-6 samples was calculated using:
[θ] =
100 ∙ 𝜃𝑜𝑏𝑠
𝑛𝑏 ∙ 𝑙 ∙ 𝑐
(2-2)
where θobs is the measured ellipticity in degrees, nb is the number of peptide bonds, l is the
pathlength in cm and c is the protein concentration in M. For thermodynamic analysis, data
72
were subjected to least-squares fitting in Igor Pro 6.2.2.2 software (WaveMetrics, Oregon,
USA).
2.2.9 Bioinformatic analysis
Whole organism open reading frame (ORF) sequence data for Escherichia coli (taxon
identifier 469008), Thermus thermophilus (taxon identifier 300852) Pyrococcus furiosus
(taxon identifier 186497) and Colwellia psychrerythraea (taxon identifier 167879) were
taken from the Uniprot database.12 For reference data on random peptide compositions,
random sequences were generated using scripts written in Python (Python Software
Foundation) to yield the same distribution of polypeptide length and amino acid
composition, and the same total number of sequences as found in all ORFs of Escherichia
coli. Calculations of pI,13 instability index8 and grand average hydropathy (GRAVY)14 for
polypeptide sequences also employed Python scripts.
2.3 Results and Discussion
2.3.1 De novo sequence design
The goal of the work presented in this chapter was to generate a bioproduction process for
a surfactant peptide with a high level of negative charge. The design approach involved
pairing this anionic peptide with a cationic partner in a heteroconcatemer, where helical
peptides were connected with unstructured linkers to generate a stable coiled-coil
structure for expression. Acid-sensitive sites were incorporated in the designed
heteroconcatemer to allow cleavage to monomeric peptides after downstream processing.
It was expected that the target anionic peptide could be recovered from the expression
partner by selective pH precipitation following cleavage.
2.3.1.1 Anionic peptide
The surfactant peptides reported in this chapter were designed using similar principles to
those used to design the surfactant α-helical peptides discussed in Chapter 1.1, 15 The
peptide sequences were built around a canonical α-helix structure, with 3.6 residues per
turn resulting in an 18 position repeat (Figure 2-1A). Hydrophobic residues were placed at
positions 1, 4, 8, 11, 15 and 18 to give a hydrophobic face (Figure 2-1A boxed). The
remaining positions were populated with hydrophilic residues to give a facially amphiphilic
helix compatible with association at interfaces. As discussed in Chapter 1, the
73
arrangement of hydrophobic and hydrophilic residues for such a facially amphiphilic
structure is compatible with the heptad repeats of coiled-coil structures over lengths of 1-3
heptads.16 This compatibility allows peptides designed to function at a planar interface to
also be incorporated into a folded structure to enhance stability during bioproduction.
The anionic peptide EDP-11 was designed with an emphasis on carboxylate content and
helical structure. Work within the candidate’s research group has shown that incorporation
of multiple carboxylate-containing residues in an α-helical surfactant peptide yields
emulsifiers with high salt resistance. This is particularly true if helix-favouring glutamate
residues are placed in positions aligned close to the interface, corresponding to positions
3, 5, 7, 12, 14 and 16 of the 18-mer (unpublished data). The remaining positions can be
filled with hydrophilic and hydrophobic residues consistent with the design of an
amphipathic helix to give an 18-residue α-helical core sequence.
In selecting specific amino acid residues for the 18-mer core sequence, emphasis was
placed on residues with high helix propensity and chemical stability, to obtain robust
products with a well-defined secondary structure.17, 18 Accordingly, proline and glycine
were avoided due to their helix-breaking character, while cysteine was not chosen due to
redox instability.19 Asparagine and glutamine were omitted due to sidechain amide lability
under the acidic conditions required for concatemer cleavage.19, 20 The hydrophobic
residues phenylalanine, tyrosine and tryptophan were not chosen on the expectation that
their large size might interfere with packing of the hydrophobic core of the coiled coil. In
addition, the sidechains of tyrosine, methionine and tryptophan are susceptible to
oxidation, which could affect yields, processing conditions or the shelf life of products
containing the peptides.
These considerations left the amino acid residues valine, leucine or isoleucine to populate
the hydrophobic positions of the core amphipathic helix. Of these, leucine was emphasised
due to its high helix propensity, while a smaller proportion of isoleucine and valine were
included to avoid overloading cellular pathways for leucine production. For non-glutamate
hydrophilic positions, the residues remaining to select from were polar serine and
threonine, cationic lysine and histidine, and the small aliphatic residue alanine. Based on
considerations of charge, solubility and helix propensity, the remaining sites were
populated with alanine and serine.
74
Expression of insoluble inclusion bodies is useful in some bioexpression systems, allowing
ready separation of desired products from cellular proteins. However, inclusion bodies are
generally solubilised using high concentrations of denaturing agents, which adds to the
cost of downstream processing and was therefore undesirable. Sequences of high
solubility were therefore targeted in this design process. Polypeptide solubility was
predicted based on sequence hydrophobicity by calculating the grand average of
hydropathy (GRAVY).14 Positive values of GRAVY indicate more hydrophobic sequences,
and ongoing studies indicate that GRAVY values greater than 0.6 are associated with
insolubility, even for charged peptides (unpublished data). Sequence composition was
adjusted to obtain hydropathy values of less than 0.6 while maintaining the hydrophobic
face required for coiled-coil formation and surfactant functionality.14
In addition to optimisation of parameters for predicted solubility, designed sequences were
optimised using the Guruprasad instability index8 with the aim of improving protease
resistance. The instability index was developed from a statistical analysis of 44 proteins
found to have either long (>16 h) or short (<5 h) in vivo half-lives.8 Analysis of the
dipeptide composition of these protein sequences allowed the assignment of numerical
values to apparently stabilising or destabilising amino acid pairs. For example, the residue
pair methionine-histidine was found to be destabilising (pair value +58.28), while the
glutamate-tryptophan pair was found to be stabilising (pair value -14.03). The instability
index (II) for a given sequence can be calculated by averaging the pair values for each
dipeptide combination within a sequence and applying a scaling factor:
𝐿−1
10
𝐼𝐼 = � � � 𝐷𝐼𝑊𝑉(𝑥𝑖 𝑦𝑖+1 )
𝐿𝑃
(2-3)
𝑖=1
where LP is the sequence length, xiyi+1 is a dipeptide and DIWV is an associated dipeptide
instability weight value.8 Reference instability index values were also derived that were
proposed to predict polypeptide stability in vivo. A calculated instability index of greater
than 40 is taken to predict polypeptide instability. The physicochemical basis for the
instability index is not well understood, and few further reports have studied the correlation
between this parameter and observed protein in vivo stability.21
To the candidate’s knowledge the index has not been tested as an element of de novo
peptide design. However, it has been adopted as a simple predictor for in vivo stability of
polypeptide sequences,22-24 and has been incorporated into several polypeptide property
75
prediction packages.25, 26 In the design of peptide sequences reported in this thesis, the
instability index values of test sequences were minimised by manually replacing
“destabilising” dipeptide sequences with other residue pairs having similar
physicochemical properties. For example, an initial anionic peptide design included the
dipeptides glutamate-isoleucine and isoleucine-glutamate within the tripeptide glutamateisoleucine-glutamate. Both dipeptides are considered to be destabilising, with DIWVs of
+20.26 and +44.94, respectively. An isoleucine to leucine replacement gave the dipeptides
glutamate-leucine and leucine-glutamate, each with DIWV +1.0, meaning that this
sequence element is considered to be neither significantly stabilising nor destabilising for
protease resistance.
An important element in this overall design was a strategy for concatemer cleavage to
allow recovery of the product peptides. For cost reasons as well as avoiding the
incorporation of long recognition sequences, it was desirable to avoid enzymatic cleavage
in downstream processing.27, 28 Of the available chemical cleavage methods (Chapter 1;
Section 1.4), acid cleavage at aspartate-proline sites20, 29 was chosen as it is inexpensive,
simple and employs relatively non-toxic reagents. At acidic pH and moderate
temperatures, cleavage between aspartic acid and proline is favoured due to the basicity
of the proline secondary amine group, while non-aspartate bonds in the polypeptide
backbone are essentially fully stable.29 Aspartate-proline dipeptides were incorporated at
the termini of peptide sequences to facilitate peptide recovery. Also included in peptide
design was a glycine at the N-terminus of each helix, adjacent to the cleavage dipeptide,
to provide increased local flexibility that may favour cleavage.30
Residue patterning for an amphiphilic helix as outlined above, followed by stepwise
optimisation of the instability index, gave an 18-residue core sequence that was extended
to 22 residues to obtain greater helix stability.31 Addition of end residues consistent with an
aspartate-proline chemical cleavage strategy yielded peptide EDP-11 (PG IAELEAE
LSAVEAE LEAILAE LD, MW 2596; Figure 2-1B). This sequence had a predicted
molecular charge of -8.1 at pH 7, instability index of +13.9 and GRAVY value of 0.488,
indicating a highly negatively charged peptide that should be stable and soluble in vivo.
2.3.1.2 Cationic peptide
The design of a complementary charged peptide for incorporation into a heteroconcatemer
with EDP-11 used cationic arginine as the main charged residue. The choice of arginine
76
was driven by the highly solubilising properties of this residue14 and its higher oxidative
stability relative to lysine. Using a patterning and optimisation process similar to that
described above, the peptide sequence RDP-4 (PG IRALARA IRALARA VRALIRA VRD,
MW 2826) was obtained (Figure 2-1B). RDP-4 was predicted to have a charge of +5.9 at
pH 7, an instability index of 6.4 and GRAVY value of 0.412, indicating a highly positively
charged peptide that is expected to be stable and soluble in vivo.
2.3.1.3 Linker and heteroconcatemer
The constituent peptides EDP-11 and RDP-4 were both expected to be highly charged
under physiological conditions. For concatemer sequences, which constituted a
combination of the highly charged peptide sequences, calculation of sequence pI13 and
predicted molecular charge at pH 7 were used to predict isoelectric points and levels of
charge under physiological conditions. By adjusting charged residue selection, isoelectric
points close to pH 7 were avoided, while a net molecular charge of at least ± 1 per
concatemer helix was targeted to provide solubility.
Charge-compensated concatemers for expression were designed with alternating EDP-11
(E) and RDP-4 (R) sequences connected by a flexible, charge-neutral peptide linker (L),
PGRGMD to allow folding of the constituent helices into an anti-parallel coiled coil (Figure
2-1C). Linker ends were constrained by the same criteria for mild acid cleavage at aspartic
acid as outlined above. A further glycine was included for linker flexibility30 and length,
while arginine was included to help balance the overall protein charge. The hydrophilic
residues of the linker sequence were also expected to increase the solubility of the
concatemer.
A methionine-aspartate (M) dipeptide was included at the start of the polypeptide
sequence for initiation of expression, while also allowing for subsequent removal of
methionine by acid cleavage. As with the concatemer examples discussed in Chapter 1,
the number of incorporated peptide repeats was expected to affect construct expression
and require optimisation.7, 32 Three different heteroconcatemers were designed using
differing peptide ratios aimed at maximising the yield of the desired product EDP-11.
These were trimeric Het2-5 (M-E-L-R-L-E), tetrameric Het2-6 (M-E-L-R-L-E-L-R) and
pentameric Het2-7 (M-E-L-R-L-E-L-R-L-E).
77
As an illustration of the expected structural fold, Figure 2-1D shows EDP-11 and RDP-4 in
heptad-based helical wheel projections for a tetrameric coiled coil, Het2-6. It can be seen
that charged glutamate and arginine residues lie in largely water-exposed positions, with
hydrophobic residues buried in the core. The e, g and c positions of the RDP-4 peptide are
predominantly populated by the non-polar residue alanine. While this residue does not
possess a strong hydrophobic character, this expanded non-polar face may be expected
to promote coiled-coil assembly.
Figure 2-1. Heteroconcatemer design. A) Helical wheel diagram showing amino acid positioning in
an 18-mer peptide designed to associate with a hydrophobic interface, where boxed positions
indicate the hydrophobic face of the peptide, B) Helical wheel diagrams for EDP-11 and RDP-4
amphipathic core sequences, C) Design of a tetrameric heteroconcatemer comprising EDP-11
(white) and RDP-4 (grey) helices connected by unstructured linkers, D) Heptad-based projection of
core helices corresponding to C), based on EDP-11 (top left and bottom right) and RDP-4 (top right
and bottom left).
78
Table 2-1. Table of computed properties of designed concatemers
Concatemer
Het2-5
Het2-6
Het2-7
4.3
4.9
4.5
Chargeb
-11.1
-5.1
-13.1
Instability
Indexc
9.5
8.2
8.7
GRAVYe
0.183
0.153
0.146
pIa
a
Following Bjellqvist et al..13 b Charge per concatemer at pH 7, as described in experimental
section. c Following Guruprasad et al..8 d Following Kyte & Doolittle.14
Values for pI, predicted molecular charge at pH 7, instability index and GRAVY were
computed for each heteroconcatemer design (Table 2-1). Values of instability index and
GRAVY were similar across all three concatemers as expected based on the properties of
the constituent peptides. However, the molecular charge at neutral pH varied due to
differing contributions from anionic and cationic helices, with Het2-6 showing the lowest
charge while still maintaining a net charge per helix of at least one unit, which was
believed necessary to maintain solubility in the folded protein. The values of pI, predicted
molecular charge, instability index and GRAVY for all three heteroconcatemers predicted
sequences that would be soluble under physiological conditions and be stable in vivo.
2.3.2 Bacterial expression
Expression of the three designed heteroconcatemers, Het2-5, Het 2-6 and Het2-7, was
tested under IPTG induction in E. coli. Samples were taken pre-induction as well as after 4
hours, 8 hours and overnight induction (Figure 2-2 and Figure 2-3). Constructs coding for
trimeric Het2-5 and pentameric Het2-7, intended for excess EDP-11 production, showed
no over-expression and were not investigated further. In contrast, strong expression of a
protein at ca. 10 kDa was seen for the DNA construct coding for the charge-balanced
tetramer Het2-6.
The failed expression of Het2-5 and Het2-7 was surprising, given the successful overexpression seen from a DNA construct coding for Het2-6 that was designed using the
same process as the failed concatemers. In the absence of further characterisation it is not
79
possible to conclude the reason for this failed expression; however several proposals may
be made. Failure of expression of Het2-5 and Het2-7 may be a result of poor packing of
residues within the hydrophobic core in the trimeric and pentameric coiled coils compared
to the tetrameric Het2-6 (as discussed in Chapter 1). Alternately, Het2-5 and Het2-7 both
contain an excess of the anionic peptide EDP-11, which results in a higher level of net
charge under physiological conditions compared to Het2-6. This increased level of charge
may destabilise the heteroconcatemer folding due to electrostatic interactions.
Destabilisation by either poor hydrophobic core packing or excess charge would result in a
partially unfolded heteroconcatemer structure that may be more susceptible to degradation
in the cell and result in no observable over-expression. This is a point which will be
returned to in Section 2.3.6, where further analysis is incorporated.
Following the successful demonstration of protein over-expression from the construct
coding for heteroconcatemer Het2-6, expression was then assessed at larger scale. Figure
2-4 shows samples taken at 0, 2, 5 and 8 hours following IPTG induction of E. coli
transformed with a plasmid coding for Het2-6 and grown in a shake flask. Strong
overexpression of a protein having a molecular weight close to 10 kDa was seen, which at
8 h constitutes 22% of total cellular protein. Based on this observation, expression was
further scaled up in shake flasks and the overexpressed protein was purified for further
characterisation as discussed in Section 2.3.3.
80
Figure 2-2. Expression testing of DNA constructs coding for concatemers Het2-5 and Het2-6. M is
molecular weight marker. T is total protein and S is soluble protein at each time point. Red box
highlights over-expressed protein.
81
Figure 2-3. Expression testing of DNA construct coding for concatemer Het2-7. M is molecular
weight marker. T is total protein and S is soluble protein at each time point.
82
Figure 2-4. Overexpression of Het2-6 in E. coli. SDS-PAGE gel showing Het2-6 expression in
shake flasks. M is molecular weight marker. T is total protein and S is soluble protein at each
time point.
83
2.3.3 Purification of Het2-6 heteroconcatemer
It was desirable for the surfactant peptide bioproduction process to be low cost and easily
scalable to provide commercially relevant quantities of peptide for material applications.
The recovery process therefore avoided the use of buffers or protease inhibitors to
minimise associated costs, while also avoiding chromatography steps to increase
scalability. On workup of freeze-thawed cell pellets from shake flask expression, cellular
protein bands in SDS-PAGE (Figure 2-5, Lanes 2 and 3) were less intense than in
samples taken during culturing (Figure 2-4).This indicated a high degree of autolysis of
cellular proteins, likely by proteases released from cellular compartments during freezing
and thawing of the cells. In contrast to cellular proteins, the over-expressed protein (at ca.
7 kDa) appears resistant to degradation, comprising 60% of residual protein after
Figure 2-5. SDS-PAGE analysis of Het2-6 fractions across purification steps. Lanes show:
1), 10) molecular weight marker, 2) total lysate, 3) clarified lysate, 4) heat-treated lysate, 5)
pH 5.0 supernatant, 6) pH 5.0 pellet, 7) MgSO4-treated supernatant, 8) second pH 5.0
supernatant, 9) second pH 5.0 pellet.
84
homogenisation. This proteolysis provides an initial purification step to reduce levels of
native proteins in the lysate.
Initial attempts to titrate the lysate directly to acidic pH for concatemer cleavage showed
poor solubility of the over-expressed protein at low pH. This was attributed to association
and co-precipitation of the over-expressed protein with other cellular components,
indicating further purification was required. During an attempt to selectively precipitate
cellular contaminants, it was found that when the clarified lysate was heated to 90 °C, the
remaining cellular protein aggregated and could be removed by centrifugation. In contrast,
the over-expressed protein remained in the supernatant (Lane 4), indicating a high degree
of thermal stability. This heat induced precipitation provided a simple purification step to
further reduce levels of native proteins in the lysate.
The heat-treated lysate was then adjusted to pH 5.0 in an attempt to selectively precipitate
the over-expressed protein at the expected isoelectric point of the heteroconcatemer Het26 (Table 2-1). This led to precipitation of the over-expressed protein (Lane 5) along with
removal of some nucleic acid (Table 2-2; see below). The over-expressed protein could be
redissolved at pH 9.0 (Lane 6) and had an estimated purity of >99% by Coomassie Blue
staining.
Mass spectrometry of the purified protein gave a molecular weight of 12879.5, matching
the expected value for Het2-6 (12877.9). The lower apparent molecular weight of the
protein on SDS-PAGE (7-10 kDa) appears to be an artefact of incomplete denaturation
prior to electrophoresis. Electronic circular dichroism (ECD) showed that heating Het2-6 in
loading buffer, as done for SDS-PAGE sample preparation, gives only a 6% loss of helical
structure measured at 222 nm. Retention of a compact folded structure would provide a
decreased hydrodynamic cross-section for separation through the polyacrylamide network.
The anomalous electrophoretic behaviour of Het2-6 is similar to that previously seen for
other highly stable proteins.33
85
Table 2-2. Quantification of Het2-6 purification
Het2-6
(mg)a
Yieldb
Nucleic
acid (mg)c
Clarified lysate
274
-
246
Heat-treated lysate
182
100%
209
0
-
48
Redissolved Het2-6 at pH 9.0
166
91%
153
MgSO4-treated supernatant
147
81%
146
Second supernatant at pH 5.0
7
-
138
Redissolved protein at pH 2.0
133
73%
6
Sample
Supernatant at pH 5.0
All values given per gram of dry cell weight.
a
Het2-6 yield estimated from ellipticity at 222 nm
b
Yield relative to initial heat precipitation step
c
Nucleic acid concentration estimated by absorption at 260 nm
While the initial purification steps yielded Het2-6 free of other cellular proteins, large
amounts of nucleic acid were still present (Table 2-2). This represented an almost equal
concentration on a weight basis and decreased Het2-6 solubility under the acidic
conditions required for cleavage. It appeared the nucleic acids were binding to the arginine
tracts of the heteroconcatemer, leading to co-precipitation at pH 5.0, and preventing
separation by pH precipitation. Most of the remaining nucleic acid was able to be removed
by heating Het2-6 with 1 M MgSO4 at 90 °C, followed by a second protein precipitation
step at pH 5.0 (Table 2-2). This effective separation of Het2-6 from nucleic acid is likely
due to complexation of magnesium ions by nucleic acid, thereby preventing these sites
from being available for association with the heteroconcatemer.34 The final preparation
contained nucleic acid at less than 5% weight ratio to protein and was soluble at 50 mg
mL-1 protein at pH 2.0. Some loss of Het2-6 was observed during MgSO4 treatment, but
the overall recovery of protein was still high.
86
As Het2-6 does not possess aromatic residues suitable for UV spectroscopic analysis,
ECD was used to monitor yields during purification. The ellipticity at 222 nm for a pure
protein is directly proportional to concentration, with a proportionality constant (mean
residue ellipticity, [θ]222) that can be determined by quantitative amino acid analysis.
Measurements of raw ellipticity in the initial lysate include contributions from other proteins
with helical secondary structure, but after heat treatment Het2-6 is essentially the only
protein present. Consistent estimates of protein concentration were obtained using either
ellipticity at 222 nm, Coomassie staining of SDS-PAGE gels or quantitative amino acid
analysis.
Quantification of purified Het2-6 by amino acid analysis (Table 2-2) showed that
approximately 73% of the expressed protein was recovered from the initial lysate. 133 mg
of Het2-6 was obtained per gram of dry cell weight, equivalent to 150 mg per litre of culture
in shake flasks. This is favourable as compared to charge-pairing constructs reported in
the literature that were isolated using chromatography methods and yielded 60-107 mg of
peptide per litre of culture.6, 7 Notably, this volumetric yield of Het2-6 was obtained using
low cell-density culture in shake flasks. It is anticipated that much higher yields could be
obtained in high cell-density fermenters, as would be utilized in large-scale production.35, 36
This series of workup steps provides a simple, inexpensive and effective way to recover
Het2-6 at high purity following expression. The lack of any buffers or protease inhibitors in
this process, as well as the chromatography-free approach that was developed, lowers the
costs associated with purification and increases this method’s viability for large-scale and
commercial applications.
2.3.4 Protein structure
Het2-6 folds into a strongly helical structure as judged by ECD. Figure 2-6 shows ECD
spectra for purified Het2-6 at pH 3.0 and 8.0. The double minimum at 208 and 222 nm and
maximum at 192 nm are characteristic of an α-helix. The value of [θ]222 is -29,600 deg cm2
dmol-1 at pH 3.0 and 20 °C, indicating Het2-6 to be predominantly helical under these
conditions,37 in agreement with the overall design. At pH 8.0 the value of [θ]222 is slightly
smaller at -26,600 deg cm2 dmol-1, suggesting a slight unfolding of the helical structure on
ionization of the glutamate residues, potentially due to increased electrostatic repulsion.
87
Figure 2-6. ECD spectra of Het2-6 (100 µM) at pH 3.0 (solid line) and pH 8.0 (dashed).
2.3.5 Thermodynamic stability
Handling of Het2-6 during purification showed the protein to be highly thermostable. It was
able to retain structure and solubility on heating to 90 °C, under conditions where other
cellular proteins denature and aggregate. Chemical denaturation was employed to
characterise the thermodynamic stability of Het2-6 in more detail, with comparison to a
reference extract comprising total E. coli soluble protein. Figure 2-7 shows the effects of
guanidinium chloride (G.HCl) on the secondary structure of Het2-6 and E. coli total protein.
The secondary structure of E. coli cellular proteins is already partly lost at 1 M G.HCl at 20
°C, with denaturation complete by 6 M G.HCl. In contrast, Het2-6 does not unfold fully at
either 20 or 50 °C in the presence of 8 M G.HCl; it was therefore necessary to carry out
chemical denaturation studies of Het2-6 at 90 °C, where heat plays an additional role in
protein unfolding. In the absence of G.HCl, the value of [θ]222 at 90 °C for Het2-6 is -20,600
deg cm2 dmol-1, approximately 23% lower than when measured at 20 °C. While this
88
Figure 2-7. ECD analysis of intact Het2-6 and E. coli soluble protein chemical denaturation at pH
8.0. Het2-6 samples (●) were monitored at 90 °C and 222 nm, while E. coli KC6 soluble protein (○)
was monitored at 20 °C and 230 nm. Fits show a three-state unfolding transition for Het2-6 and
non-linear regression for E. coli protein.
suggests a degree of unfolding, addition of 1-4 M G.HCl does not further affect the helix
content, but higher concentrations of G.HCl induce denaturation in a two-phase process.
The G.HCl concentration required to achieve 50% unfolding was 7.0 M for Het2-6 and 1.7
M for E. coli cellular extract. This highlights the extreme thermodynamic stability of Het2-6
compared to native proteins in bacterial cells. This is accentuated by the fact that in the
case of Het2-6 the 50% unfolding value is determined from denaturation at 90 °C.
Quantitative determination of the thermodynamic stability of a protein by chemical
denaturation commonly assumes a two-state model in which the stability of the folded
state can be calculated by plotting the fraction of protein in a folded state (Ff) as a function
of denaturant concentration.31, 38 Quantitative determination of the thermodynamic stability
of proteins in E. coli cellular extracts from chemical denaturation data was not possible.
89
Analysis was complicated by the ECD signal of these samples resulting from a mixture of
secondary structures present in all proteins in the extracts. In the case of Het2-6, fitting of
the denaturation curve to a two-state model was attempted, however a good fit was not
attained. Chemical denaturation data were therefore analysed using a three-state
unfolding model adapted from Morjana et al.,39 where unfolding is taken to proceed
through an observable intermediate (I) between the native (N) and fully unfolded (U)
states:
𝐾𝐼→𝑈
𝐾𝑁→𝐼
𝑁 ⇌ 𝐼 ⇌ 𝑈
(2-4)
The mean residue ellipticity measured by circular dichroism ([θ]obs) at a given denaturant
concentration ([D]) was fit using the equation:
[𝜃]𝑜𝑏𝑠 =
[𝜃]𝑁 + [𝜃]𝐼 ∙ 𝑒
1+𝑒
−(Δ𝐺𝑁→𝐼 −𝑚𝑁→𝐼 [D])
RT
−(Δ𝐺𝑁→𝐼 −𝑚𝑁→𝐼 [D])
RT
+ [𝜃]𝑈 ∙ 𝑒
+𝑒
−(Δ𝐺𝑁→𝐼 −𝑚𝑁→𝐼 [D])
RT
−(Δ𝐺𝑁→𝐼 −𝑚𝑁→𝐼 [D])
RT
∙𝑒
∙𝑒
−(Δ𝐺𝐼→𝑈 −𝑚𝐼→𝑈 [D])
RT
−(Δ𝐺𝐼→𝑈 −𝑚𝐼→𝑈 [D])
RT
(2-5)
where [θ]N represents the mean residue ellipticity of the native state, [θ]I the mean residue
ellipticity of the intermediate state and [θ]U that of the unfolded state. ΔGN→I and ΔGI→U are
the free energies of the N→I and I→U transitions extrapolated to [D] = 0. The m value is
an empirical constant corresponding to the slope of a plot of ΔG versus [D], which for a
two-step process is divided into mN→I and mI→U, the m values for the N→I and I→U
transitions respectively. Data were plotted as Ff, calculated as:
𝐹𝑓 =
[𝜃]𝑜𝑏𝑠 − [𝜃]𝑈
[𝜃]𝑁 − [𝜃]𝑈
(2-6)
Table 2-3 shows parameters obtained by fitting Het2-6 denaturation to the three-state
model. The summed ΔGN→U of 38.9 kcal mol-1 or 9.7 kcal mol-1 helix-1 is high for a tetramer
of 23-24 residue helices. Maquette tetramers based on 27-residue α-helices had a much
lower ΔG of 6.4 kcal mol-1 helix-1,40 while Fairman et al.31 reported the Lac peptide
tetramer coiled coils in which the stability per helix ranged from 3.9 kcal mol-1 (21 residues)
to 7.6 kcal mol-1 (28 residues) to 12.9 kcal mol-1 (35 residues). Dimer coiled coils are
typically less stable, with a reported free energy of unfolding per helix of 6.0 kcal mol-1 for
35-residue helices and 4.1 kcal mol-1 for 28-residue helices.41 Of note is the fact that these
reference stability values were determined at room temperature, which again shows the
thermodynamic stability of Het2-6 at 90 °C to be very high compared to related coiled-coil
structures.
90
Table 2-3. Fit parameters for Het2-6 denaturation with G.HCl
ΔGN→I(H2O)a
ΔGI→U(H2O)a
mN→Ib
mI→Ub
[θ]Nc
7.7
31.2
1.27
4.38
-20.6 -14.0
a
kcal mol-1 per coiled coil
b
kcal mol-1 M-1
c
× 10-3 deg cm2 dmol-1 at 222 nm
[θ]Ic
[θ]Uc
-3.6
The m value has been employed by other workers to estimate changes in the accessible
surface area (ASA) of proteins on unfolding.38, 42 These estimates are based on
correlations between the surface area buried on folding (from crystallographic data) and
measured m values for a range of proteins. The m value determined for Het2-6 is large
(5.65 kcal mol-1 M-1 summed across two transitions) compared to similar sized proteins
and peptide coiled coils.31, 42 Applying literature correlations to the large m value calculated
for Het2-6 gives a value for the change in ASA that is not physically meaningful and would
exceed the total calculated surface area of Het2-6.43 Correlations across different proteins
are relatively scattered, particularly for denaturation using G.HCl rather than urea, where
ionic strength plays an additional role in unfolding.42, 44 High m values, comparable to
those obtained here, have been reported for mutants of staphylococcal nuclease (149
residues)45 and T4 lysozyme (163 residues),46 although these m values are not
accompanied by the same high thermodynamic stability found for Het2-6. In addition, m
values used in ASA correlations are determined at room temperature, and extrapolation to
90 °C relies on assumptions about the effects of temperature on protein-denaturant
binding that may not be justifiable.47 The absolute values of change in ASA calculated from
m values determined in the fitting of Het2-6 denaturation are therefore likely notmeaningful.
91
Figure 2-8. Schematic of proposed unfolding pathway of Het2-6. N fully folded Het2-6, I,
unfolding intermediate with high helical content, U, fully unfolded Het2-6. Cylinders indicate
helical structure. Constituent peptides coloured as white (EDP-11) or grey (RDP-4).
At the same time, the relative values of m for the two unfolding steps may be used to infer
information about the changes involved in each step. The N→I transition involves a
smaller value of m (22% of the total across two steps), suggesting that the hydrophobic
face of each helix may still be largely excluded from the aqueous environment in the
intermediate state. This could occur by opening of the tetramer to dimeric coiled coils still
having a largely helical structure (Figure 2-8), as also supported by the smaller change in
ellipticity in the first stage of unfolding. The I→U transition would then involve complete
exposure of hydrophobic residues to solvent. As with the m values, the free energy change
involved in the N→I transition represents a small fraction of the total free energy of
unfolding. This is consistent with the proposal that the structural changes in the first
unfolding step are relatively small. The larger change in free energy in the I→U transition
is consistent with coiled-coil stability being largely dependent on hydrophobic interactions
between helices, which would be disrupted upon unfolding of the intermediate state.
92
2.3.6 Bioinformatics
As discussed above, key parameters used in the design of the sequence of Het2-6 were
pI, GRAVY and the Guruprasad instability index. Prediction of isoelectric point values were
utilised to avoid insolubility at physiological pH, GRAVY values of less than 0.6 were
targeted to increase solubility at all pH values, and instability index values were minimised
in an attempt to increase in vivo stability. Given the unusually high proteolytic and
thermodynamic stability of Het2-6, it was of interest to understand the extent to which the
physicochemical parameters for the heteroconcatemer differed from typical bacterial
proteins.
The pI, GRAVY and instability index values for all open reading frames (ORFs) of four
reference organisms were compared to designed sequences, as well as a series of
randomly-generated polypeptide sequences, to obtain a better understanding of the
mechanistic basis for the high stability of Het2-6. Escherichia coli was chosen as a
reference organism as it is the expression host and one of the most commonly used
expression systems in industrial processes.48 The remaining three reference organisms
were extremophiles selected to examine the effect of specialisation on key protein
properties. Thermus thermophilus and Pyrococcus furiosus were selected as
(hyper)thermophiles, while Colwellia psychrerythraea is a psychrophile.49-51 Similar profiles
for pI, GRAVY and instability index were found across all four model organisms (Figure
2-9). This indicates that the organisms native thermal environment does not strongly select
for these parameters. The same parameters were also calculated for a series of random
sequences with the same average amino acid composition as E. Coli, to examine the
extent to which proteins within the proteome fall into different composition classes.
The physicochemical properties of designed sequences deviate from those of the bacterial
proteome in several regards. For the complete proteomes, calculated pI values fall in a
bimodal distribution on either side of intracellular pH with modal values of pH 5.6 and 9.4
(Figure 2-9A). A similar distribution is found for random sequences having the same
average amino acid composition as the E. coli proteome (modal values of pH 5.5 and 9.3).
This distribution appears to be a simple consequence of the pKa values of amino acid
sidechains, which lie well above or below physiological pH, with the exception of histidine.
93
Virtually no bacterial open reading frames have predicted pI values below 4.0 or above
12.0. The pI values of RDP-4 (12.4) and EDP-11 (3.3) lie at the extremes of the
distribution seen for bacterial proteomes. This reflects the design process that included a
high proportion of cationic residues in RDP-4 and anionic residues in EDP-11. The pI of
Het2-6 is 4.90, which lies at the lower end of the distribution seen for bacterial proteomes,
but to a lesser degree than the non-expressed constructs Het2-5 (pI 4.3) or Het2-7 (pI
4.5). The low pI values of the designed concatemers reflect the design of sequences with
Figure 2-9. Calculated bioinformatic parameters for selected organisms and random polypeptide
sequences. A) pI, B) GRAVY and C) instability index values for ORFs of Escherichia coli,
Thermus thermophilus, Pyrococcus furiosus and Colwellia psychrerythraea. Solid line shows
calculated parameters for random protein sequences. Dashed vertical lines show the
corresponding values for Het2-6.
94
pI values that are not near physiological pH, as well as the high carboxylate residue
content of these sequences due to the incorporation of anionic surfactant peptide EDP-11.
GRAVY values for the bacterial proteomes show a bimodal distribution (with modal values
of -0.19 and 0.80 in E. coli), while random polypeptide sequences have a monomodal
distribution with a modal value of -0.03 (Figure 2-9B). The bimodal distribution of GRAVY
values in the bacterial proteomes represents the division between membrane-bound and
soluble proteins that show dissimilar levels of hydrophobic amino acid incorporation. A
value of 0.45 has previously been associated with the threshold between integral
membrane and soluble proteins,52 and is consistent with the distribution seen here. This
distinction is not reflected in random amino acid sequences, which have amino acid
compositions that are an average of those in membrane-bound and soluble proteins. The
GRAVY value of 0.153 for Het2-6 lies toward the hydrophobic end of the range for nonintegral membrane proteins. This is a result of the design process, which employed highly
amphipathic α-helices to drive self-assembly, while still targeting net solubility of the
polypeptide structure by inclusion of hydrophilic residues in the linker sequences. The nonexpressed constructs Het2-5 and Het2-7 have GRAVY values of 0.183 and 0.146
respectively, which are comparable to the value for Het2-6 due to the similarity in design.
The GRAVY values for EDP-11 and RDP-4, at 0.488 and 0.412 respectively, lie closer to
the values for membrane-bound cellular proteins. This is due to the high content of
hydrophobic amino acids used to provide an amphiphilic character and the lack of
hydrophilic linker sequences that are present in the designed concatemers.
For the Guruprasad instability index, the ORFs of E. coli show a monomodal distribution,
with a mean of 37.8 ± 10.8, similar to the other three reference species, while random
polypeptide sequences give values of 36.9 ± 8.8 (Figure 2-9C). These values all lie close
to the threshold value of 40 proposed to provide a distinction between in vivo stability and
instability. In the absence of further information it is difficult to conclude if this reflects
selection by the organisms for proteins that are neither perpetually stable due to
accumulation of defective proteins, nor completely unstable due to high metabolic costs.
Alternatively, the similarity between random sequences and proteome distributions may
indicate that protein instability as measured by the instability index is not selected for or
against in these organisms. The instability index values for the designed sequences Het26 (8.2), EDP-11 (13.9) and RDP-4 (6.4) all lie more than two standard deviations below the
95
mean of the bacterial proteomes. This would appear to predict high in vivo stability, as was
indeed observed in practice for Het2-6. However, Het2-5 and Het2-7 also had similarly low
instability index values (Het2-5, 9.5; Het2-7, 8.7) yet showed no expression, potentially
due to poor concatemer folding and degradation in the cell.
There have been several different algorithms developed to predict the stability of folded
proteins.53 One such tool is FoldIndex, which predicts the level of disorder in a polypeptide
based on the average hydrophobicity of its amino acids and the net charge,54 effectively
comparing stabilisation due to hydrophobic interactions with the destabilising effects of
high charge. The index was developed by statistical analysis of 275 natively folded
proteins and 91 proteins known to be natively unfolded.55 It was found that the two classes
of protein could be distinguished based on the net charge at pH 7.0 (|<R>|) and the mean
hydrophobicity14 (<H>), rescaled to a range of 0-1.
𝐹𝑜𝑙𝑑𝐼𝑛𝑑𝑒𝑥 = 2.785 < 𝐻 > − |< 𝑅 >| − 1.151
(2-7)
For this index, positive values represent sequences likely to be folded, while negative
values represent sequences likely to be disordered. Due to a high hydrophobic content,
the index predicts Het2-6 to be folded, with a value of 0.248. The index also predicts
folding for the constituent peptides EDP-11 and RDP-4, with values of 0.072 and 0.138
respectively. However, trimeric Het2-5 (value 0.176) and pentameric Het2-7 (value 0.189)
are also predicted to be stable, though expression failed in these two cases.
It is likely that the observed stability of Het2-6 can be accounted for by a combination of
different factors. The calculated values for pI and GRAVY lie within the normal range for
bacterial ORFs, which predicts good solubility as observed in practice. However, the
instability index lies at the lowest extreme observed in bacterial ORFs, consistent with a
polypeptide that is not readily digested by endogenous proteases, as also observed in
practice. At the same time, Het2-6 has been designed to fold into a well-ordered coiled-coil
structure, a design that has been validated by circular dichroism spectroscopy. In addition,
FoldIndex, which has a well-characterised physicochemical basis, predicts
heteroconcatemer stability based on the high hydrophobic content of the polypeptide.
From this it can be inferred that minimisation of the Guruprasad instability index, along with
stabilisation by a well-ordered hydrophobic core, may contribute to the unusually high
proteolytic and thermodynamic stability of Het2-6. At the same time, the lack of expression
of the similarly designed Het2-5 and Het2-7 suggest that optimisation of these design
96
parameters alone is not necessarily sufficient to give highly expressed stable
concatemers.
It was initially postulated (Section 2.2.3) that high net charge and consequent poor folding
may be responsible for the lack of expression of Het2-5 and Het2-7. However the
prediction of stable folding by calculation of the FoldIndex suggests that that excess net
molecular charge is likely not the key determining factor in lack of expression of Het2-5
and Het2-7. Poor packing of residues within the hydrophobic core in the trimeric (Het2-5)
and pentameric (Het2-7) coiled coils compared to the tetrameric Het2-6 (as discussed in
Chapter 1) is one potential factor which may play a role in destabilising coiled-coil folding.
This would likely result in a less stably folded construct within the cell which may be more
susceptible to proteolytic degradation and consequently show no over-expression. Further
investigation will be required to determine the precise reason for the lack of expression of
these concatemers; however factors such as specific hydrophobic packing within the core
likely influence concatemer stability and expression.
2.4 Conclusions
This chapter reports the successful bioproduction of the designed heteroconcatemer Het26, which comprises the strongly anionic surfactant peptide EDP-11 and cationic expression
partner RDP-4. This indicates that charge pairing can offer a successful strategy for the
expression of designer anionic peptides by providing stably folded structures for
downstream processing.
The design of polypeptide sequences combined consideration of both surfactant
functionality and the formation of a stably folded structure for expression. In the design of
the concatemer Het2-6, the polypeptide sequence was optimised using the parameters of
pI, GRAVY and instability index to enhance function and expression. Het2-6 accumulated
to high levels in E. coli and displayed extremely high thermodynamic stability and protease
resistance. By making use of these usual properties, a chromatography-free downstream
process was developed that yielded Het2-6 at >99% purity in the absence of protease
inhibitors or buffers. The simplicity, minimal reagent requirements and lack of
chromatographic methods make this process low-cost and readily scalable.
The bioinformatic parameters of pI, GRAVY and instability index of Het2-6 were also
compared with those of all ORFs of four reference microbes, as well as random
97
polypeptide sequences. This showed that Het2-6 lies in the high hydrophobic range for
soluble proteins, while also possessing an unusually low instability index. This allowed
inference that minimisation of sequence instability index values as well as stabilisation by
the high hydrophobic content of the coiled-coil structure may contribute to the unusually
high stability of Het2-6. Future design processes similar to that outlined here may make
use of these calculated bioinformatic parameters of designed sequences and bacterial
proteomes as reference points. By targeting desired values corresponding to in vivo
stability, hydrophobicity and solubility, along with an understanding of factors stabilising
particular secondary structures, it may possible to generate de novo designs with
properties suitable for expression. At the same time, the lack of expression of the
concatemers Het2-5 and Het2-7 suggest that optimisation of these design parameters
alone is not necessarily sufficient to give highly expressed and stable concatemers, and
that future studies will be required to further refine this design process.
While this chapter details the design and expression of the heteroconcatemer Het2-6,
further processing is required to recover the product peptides. Chapter 3 focuses on the
heat- and acid-mediated cleavage of the expressed concatemer to give monomeric
surfactant peptides, and also reports characterisation of their interfacial properties.
98
2.5 References
1.
Dexter, A. F.; Malcolm, A. S.; Middelberg, A. P. J. Reversible Active Switching of
the Mechanical Properties of a Peptide Film at a Fluid-Fluid Interface. Nat. Mater. 2006, 5,
502-506.
2.
Malcolm, A. S.; Dexter, A. F.; Katakdhond, J. A.; Karakashev, S. I.; Nguyen, A. V.;
Middelberg, A. P. J. Tuneable Control of Interfacial Rheology and Emulsion Coalescence.
ChemPhysChem 2009, 10, 778-781.
3.
Dexter, A. F. Interfacial and Emulsifying Properties of Designed Beta-Strand
Peptides. Langmuir 2010, 26, 17997-18007.
4.
Bos, M. A.; van Vliet, T. Interfacial Rheological Properties of Adsorbed Protein
Layers and Surfactants: A Review. Adv. Colloid Interface Sci. 2001, 91, 437-471.
5.
Morreale, G.; Lee, E. G.; Jones, D. B.; Middelberg, A. P. J. Bioprocess-Centered
Molecular Design (Bmd) for the Efficient Production of an Interfacially Active Peptide.
Biotechnol. Bioeng. 2004, 87, 912-923.
6.
Lee, J. H.; Minn, I.; Park, C. B.; Kim, S. C. Acidic Peptide-Mediated Expression of
the Antimicrobial Peptide Buforin Ii as Tandem Repeats in Escherichia Coli. Protein Expr.
Purif. 1998, 12, 53-60.
7.
Kim, H. K.; Chun, D. S.; Kim, J. S.; Yun, C. H.; Lee, J. H.; Hong, S. K.; Kang, D. K.
Expression of the Cationic Antimicrobial Peptide Lactoferricin Fused with the Anionic
Peptide in Escherichia Coli. Appl. Microbiol. Biotechnol. 2006, 72, 330-338.
8.
Guruprasad, K.; Reddy, B. V. B.; Pandit, M. W. Correlation between Stability of a
Protein and Its Dipeptide Composition - a Novel-Approach for Predicting Invivo Stability of
a Protein from Its Primary Sequence. Protein Eng. 1990, 4, 155-161.
9.
Fletcher, N. L.; Lockett, C. V.; Dexter, A. F. A Ph-Responsive Coiled-Coil Peptide
Hydrogel. Soft Matter 2011, 7, 10210-10218.
10.
Zawada, J.; Richter, B.; Huang, E.; Lodes, E.; Shah, A.; Swartz, J. R., High-Density,
Defined Media Culture for the Production of Escherichia Coli Cell Extracts. In Fermentation
Biotechnology, American Chemical Society: 2003; Vol. 862, pp 142-156.
11.
Goerke, A. R.; Swartz, J. R. Development of Cell-Free Protein Synthesis Platforms
for Disulfide Bonded Proteins. Biotechnol. Bioeng. 2008, 99, 351-367.
12.
Consortium, T. U. Update on Activities at the Universal Protein Resource (Uniprot)
in 2013. Nucleic Acids Res. 2013, 41, D43-D47.
13.
Bjellqvist, B.; Hughes, G. J.; Pasquali, C.; Paquet, N.; Ravier, F.; Sanchez, J. C.;
Frutiger, S.; Hochstrasser, D. The Focusing Positions of Polypeptides in Immobilized Ph
Gradients Can Be Predicted from Their Amino-Acid-Sequences. Electrophoresis 1993, 14,
1023-1031.
14.
Kyte, J.; Doolittle, R. F. A Simple Method for Displaying the Hydropathic Character
of a Protein. J. Mol. Biol. 1982, 157, 105-132.
15.
Dexter, A. F.; Middelberg, A. P. J. Switchable Peptide Surfactants with Designed
Metal Binding Capacity. J. Phys. Chem. C 2007, 111, 10484-10492.
16.
Apostolovic, B.; Danial, M.; Klok, H. A. Coiled Coils: Attractive Protein Folding
Motifs for the Fabrication of Self-Assembled, Responsive and Bioactive Materials. Chem.
Soc. Rev. 2010, 39, 3541-3575.
17.
O'Neil, K. T.; Degrado, W. F. A Thermodynamic Scale for the Helix-Forming
Tendencies of the Commonly Occurring Amino Acids. Science 1990, 250, 646-651.
18.
Blaber, M.; Zhang, X. J.; Matthews, B. W. Structural Basis of Amino-Acid AlphaHelix Propensity. Science 1993, 260, 1637-1640.
99
19.
Reubsaet, J. L. E.; Beijnen, J. H.; Bult, A.; van Maanen, R. J.; Marchal, J. A. D.;
Underberg, W. J. M. Analytical Techniques Used to Study the Degradation of Proteins and
Peptides: Chemical Instability. J. Pharm. Biomed. Anal. 1998, 17, 955-978.
20.
Dimitrijev-Dwyer, M.; He, L.; James, M.; Nelson, A.; Wang, L.; Middelberg, A. P. J.
The Effects of Acid Hydrolysis on Protein Biosurfactant Molecular, Interfacial, and Foam
Properties: Ph Responsive Protein Hydrolysates. Soft Matter 2012, 8, 5131-5139.
21.
Jennissen, H. P. Ubiquitin and the Enigma of Intracellular Protein-Degradation. Eur.
J. Biochem. 1995, 231, 1-30.
22.
Lindgren, M.; Gallet, X.; Soomets, U.; Hallbrink, M.; Brakenhielm, E.; Pooga, M.;
Brasseur, R.; Langel, U. Translocation Properties of Novel Cell Penetrating Transportan
and Penetratin Analogues. Bioconjug. Chem. 2000, 11, 619-626.
23.
Lubec, G.; Afjehi-Sadat, L.; Yang, J. W.; John, J. P. P. Searching for Hypothetical
Proteins: Theory and Practice Based Upon Original Data and Literature. Prog. Neurobiol.
2005, 77, 90-127.
24.
McMahon, M. T.; Gilad, A. A.; DeLiso, M. A.; Berman, S. D. C.; Bulte, J. W. M.; van
Zijl, P. C. M. New "Multicolor" Polypeptide Diamagnetic Chemical Exchange Saturation
Transfer (Diacest) Contrast Agents for Mri. Magn. Reson. Med. 2008, 60, 803-812.
25.
Thomas, S.; Karnik, S.; Barai, R. S.; Jayaraman, V. K.; Idicula-Thomas, S. Camp: A
Useful Resource for Research on Antimicrobial Peptides. Nucleic Acids Res. 2010, 38,
D774-D780.
26.
Artimo, P.; Jonnalagedda, M.; Arnold, K.; Baratin, D.; Csardi, G.; de Castro, E.;
Duvaud, S.; Flegel, V.; Fortier, A.; Gasteiger, E.; Grosdidier, A.; Hernandez, C.; Ioannidis,
V.; Kuznetsov, D.; Liechti, R.; Moretti, S.; Mostaguir, K.; Redaschi, N.; Rossier, G.;
Xenarios, I.; Stockinger, H. Expasy: Sib Bioinformatics Resource Portal. Nucleic Acids
Res. 2012, 40, W597-W603.
27.
Jenny, R. J.; Mann, K. G.; Lundblad, R. L. A Critical Review of the Methods for
Cleavage of Fusion Proteins with Thrombin and Factor Xa. Protein Expr. Purif. 2003, 31,
1-11.
28.
Young, C. L.; Britton, Z. T.; Robinson, A. S. Recombinant Protein Expression and
Purification: A Comprehensive Review of Affinity Tags and Microbial Applications.
Biotechnol. J. 2012, 7, 620-634.
29.
Inglis, A. S. Cleavage at Aspartic-Acid. Methods Enzymol. 1983, 91, 324-332.
30.
Huang, F.; Nau, W. M. A Conformational Flexibility Scale for Amino Acids in
Peptides. Angew. Chem., Int. Ed. 2003, 42, 2269-2272.
31.
Fairman, R.; Chao, H. G.; Mueller, L.; Lavoie, T. B.; Shen, L. Y.; Novotny, J.;
Matsueda, G. R. Characterization of a New Four-Chain Coiled-Coil: Influence of Chain
Length on Stability. Protein Sci. 1995, 4, 1457-1469.
32.
Lee, J. H.; Kim, M. S.; Cho, J. H.; Kim, S. C. Enhanced Expression of Tandem
Multimers of the Antimicrobial Peptide Buforin Ii in Escherichia Coli by the Dead-Box
Protein and Trxb Mutant. Appl. Microbiol. Biotechnol. 2002, 58, 790-796.
33.
Shiell, B. J.; Tachedjian, M.; Bruce, K.; Beddome, G.; Farn, J. L.; Hoyne, P. A.;
Michalski, W. P. Expression, Purification and Characterization of Recombinant
Phospholipase B from Moraxella Bovis with Anomalous Electrophoretic Behavior. Protein
Expr. Purif. 2007, 55, 262-272.
34.
Lyons, J. W.; Kotin, L. Effect of Magnesium Ion on Secondary Structure of
Deoxyribonucleic Acid. J. Am. Chem. Soc. 1965, 87, 1781-1785.
35.
Lee, S. Y. High Cell-Density Culture of Escherichia Coli. Trends Biotechnol. 1996,
14, 98-105.
36.
Shiloach, J.; Fass, R. Growing E-Coli to High Cell Density - a Historical Perspective
on Method Development. Biotechnol. Adv. 2005, 23, 345-357.
100
37.
Scholtz, J. M.; Qian, H.; York, E. J.; Stewart, J. M.; Baldwin, R. L. Parameters of
Helix-Coil Transition Theory for Alanine-Based Peptides of Varying Chain Lengths in
Water. Biopolymers 1991, 31, 1463-1470.
38.
Scholtz, J. M.; Grimsley, G. R.; Pace, C. N., Solvent Denaturation of Proteins and
Interpretations of the M Value. In Methods in Enzymology: Biothermodynamics, Pt B,
Johnson, M. L.; Holt, J. M.; Ackers, G. K., Eds. Elsevier Academic Press Inc: San Diego,
2009; Vol. 466, pp 549-565.
39.
Morjana, N. A.; McKeone, B. J.; Gilbert, H. F. Guanidine-Hydrochloride Stabilization
of a Partially Unfolded Intermediate During the Reversible Denaturation of Protein
Disulfide Isomerase. Proc. Natl. Acad. Sci. U. S. A. 1993, 90, 2107-2111.
40.
Gibney, B. R.; Johansson, J. S.; Rabanal, F.; Skalicky, J. J.; Wand, A. J.; Dutton, P.
L. Global Topology & Stability and Local Structure & Dynamics in a Synthetic Spin-Labeled
Four-Helix Bundle Protein. Biochemistry 1997, 36, 2798-2806.
41.
Litowski, J. R.; Hodges, R. S. Designing Heterodimeric Two-Stranded Alpha-Helical
Coiled-Coils: The Effect of Chain Length on Protein Folding, Stability and Specificity. J.
Pept. Res. 2001, 58, 477-492.
42.
Myers, J. K.; Pace, C. N.; Scholtz, J. M. Denaturant M-Values and Heat Capacity
Changes - Relation to Changes in Accessible Surface Areas of Protein Unfolding. Protein
Sci. 1995, 4, 2138-2148.
43.
Rose, G. D.; Geselowitz, A. R.; Lesser, G. J.; Lee, R. H.; Zehfus, M. H.
Hydrophobicity of Amino-Acid Residues in Globular-Proteins. Science 1985, 229, 834-838.
44.
Monera, O. D.; Kay, C. M.; Hodges, R. S. Protein Denaturation with GuanidineHydrochloride or Urea Provides a Different Estimate of Stability Depending on the
Contributions of Electrostatic Interactions. Protein Sci. 1994, 3, 1984-1991.
45.
Shortle, D.; Meeker, A. K. Mutant Forms of Staphylococcal Nuclease with Altered
Patterns of Guanidine Hydrochloride and Urea Denaturation. Proteins: Struct., Funct.,
Genet. 1986, 1, 81-89.
46.
Zhang, T.; Bertelsen, E.; Benvegnu, D.; Alber, T. Circular Permutation of T4Lysozyme. Biochemistry 1993, 32, 12311-12318.
47.
Zweifel, M. E.; Barrick, D. Relationships between the Temperature Dependence of
Solvent Denaturation and the Denaturant Dependence of Protein Stability Curves.
Biophys. Chem. 2002, 101, 221-237.
48.
Jana, S.; Deb, J. K. Strategies for Efficient Production of Heterologous Proteins in
Escherichia Coli. Appl. Microbiol. Biotechnol. 2005, 67, 289-298.
49.
Oshima, T.; Imahori, K. Description of Thermus-Thermophilus (Yoshida and
Oshima) Comb. Nov., a Nonsporulating Thermophilic Bacterium from a Japanese Thermal
Spa. Int. J. Syst. Bacteriol. 1974, 24, 102-112.
50.
Fiala, G.; Stetter, K. O. Pyrococcus-Furiosus Sp. Nov. Represents a Novel Genus
of Marine Heterotrophic Archaebacteria Growing Optimally at 100°C. Arch. Microbiol.
1986, 145, 56-61.
51.
Bowman, J. P.; Gosink, J. J.; McCammon, S. A.; Lewis, T. E.; Nichols, D. S.;
Nichols, P. D.; Skerratt, J. H.; Staley, J. T.; McMeekin, T. A. Colwellia Demingiae Sp. Nov.,
Colwellia Hornerae Sp. Nov., Colwellia Rossensis Sp. Nov. And Colwellia Psychrotropica
Sp. Nov.: Psychrophilic Antarctic Species with the Ability to Synthesize Docosahexaenoic
Acid (22 : 6 Omega 3). Int. J. Syst. Bacteriol. 1998, 48, 1171-1180.
52.
Lobry, J. R. Influence of Genomic G+C Content on Average Amino-Acid
Composition of Proteins from 59 Bacterial Species. Gene 1997, 205, 309-316.
53.
Deng, X.; Eickholt, J.; Cheng, J. L. A Comprehensive Overview of Computational
Protein Disorder Prediction Methods. Mol. BioSyst. 2012, 8, 114-121.
101
54.
Prilusky, J.; Felder, C. E.; Zeev-Ben-Mordehai, T.; Rydberg, E. H.; Man, O.;
Beckmann, J. S.; Silman, I.; Sussman, J. L. Foldindex((C)): A Simple Tool to Predict
Whether a Given Protein Sequence Is Intrinsically Unfolded. Bioinformatics 2005, 21,
3435-3438.
55.
Uversky, V. N.; Gillespie, J. R.; Fink, A. L. Why Are "Natively Unfolded" Proteins
Unstructured under Physiologic Conditions? Proteins: Struct., Funct., Genet. 2000, 41,
415-427.
102
3 Bioproduction of Highly Charged Designer Peptide
Surfactants: Chemical Cleavage of a Parent Heteroconcatemer
and Analysis of Kinetic Pathways
3.1 Introduction
Chapter 2 details the expression of the highly anionic surfactant peptide EDP-11 as part of
a charge-paired heteroconcatemer with the cationic partner RDP-4. The tetrameric coiledcoil concatemer comprises alternating copies of each peptide connected by short linker
sequences. Within the concatemer, each peptide is flanked by acid-sensitive cleavage
sites to allow cleavage to monomeric peptides. This chapter details the acid and heat
mediated cleavage of the expressed heteroconcatemer to give designed surfactant
peptides. Liquid chromatography-mass spectrometry was used to identify and quantify
products and intermediates during cleavage. Observation of intermediate species allowed
the proposal of cleavage pathways as well as kinetic modelling to provide information
about the relative rates of cleavage at each of the chemically similar cleavage sites. This
showed a previously unobserved preference for cleavage at the C-termini of peptide
helices. Chemical denaturation of the cleaved products showed the removal of linkers
between helices decreased the coiled-coil stability, but the tetramer retained a remarkably
high stability compared to designed coiled coils. Surface activity studies at the air-water
interface showed both intact and cleaved Het2-6 to lower interfacial tension to a similar
extent, with slightly more rapid adsorption for the cleaved peptides. Partial denaturation of
either intact or cleaved Het2-6 by guanidinium chloride increased the rate of adsorption to
levels similar to that of unstructured surfactant peptide AM1. The cleaved peptides were
also used to prepare heat-stable emulsions with droplet sizes in the nanometre range.
3.2 Experimental
3.2.1 Materials
Reagents and chemicals were of analytical grade unless otherwise indicated. Water was
from an Elga Purelab Classic system (Veolia, Saint Maurice, France). Guanidine
hydrochloride (G.HCl) (≥99%) and tris(hydroxymethyl)aminomethane (Tris) (≥99%) were
purchased from Sigma-Aldrich. Citric acid (≥99%) was from Ajax Finechem. Peptides
EDP-11 (trifluoroacetate salt, H2N-PG IAELEAE LSAVEAE LEAILAE LD-COOH, FW
2596), RDP-4 (trifluoroacetate salt, H2N-PG IRALARA IRALARA VRALIRA VRD-COOH,
103
FW 2826) and AM1 (trifluoroacetate salt, Ac-MKQLADS LHQLARQ VSRLEHA-CONH2,
FW 2473) were custom synthesised and reversed-phase high-performance liquid
chromatography (RP-HPLC) purified by GenScript (Piscataway, NJ). Peptides EP-11
(trifluoroacetate salt, H2N-PG IAELEAE LSAVEAE LEAILAE L-COOH, FW 2481) and RP4 (trifluoroacetate salt, H2N-PG IRALARA IRALARA VRALIRA VR-COOH, FW 2711) were
custom synthesised and RP-HPLC purified by Peptide 2.0 (Chantilly, VA). The final purity
was >95% in each case. Peptide concentrations were determined by quantitative amino
acid analysis (Australian Proteome Analysis Facility, Sydney). Bovine serum albumin
(premium grade) was from Ausgenex (Molendinar, Australia). Group II base oil (Jurong
150) was from Mobil Oil Australia. Tetrameric heteroconcatemer Het2-6 was expressed in
E. coli and recovered using chromatography-free methods as described in Chapter 2.
3.2.2 Heteroconcatemer cleavage and kinetic modelling
Cleavage was carried out by heating Het2-6 (10 mg mL-1, pH 2) to 70 °C for 1-24 h.
Samples were frozen at -80 °C until analysis. Liquid chromatography-mass spectrometry
(LCMS) used a Waters 2695 separations module with a Waters 2489 UV detector (Waters
Corporation, Milford, MA). Samples at 1 mg mL-1 were applied to a Jupiter 5 µm C4 300 Å
column (Phenomenex) at a flow rate of 0.2 mL min-1, using a linear gradient of 10-100%
(v/v) buffer B over 45 min with absorbance of the eluate monitored at 214 nm. Buffer A
was 0.01% (v/v) trifluoroacetic acid (TFA) and buffer B was 0.01% (v/v) TFA in ACN.
Eluate was injected directly into a Quattro micro API tandem quadrupole system in positive
ion mode. Peptide species were quantified by integration of UV peak areas using synthetic
RDP-4 as a standard. In two cases where multiple species co-eluted (R and R-, E- and LR-L-E), quantification was by integration of specific ion counts using synthetic peptide
equivalents as standards.
Pathway modelling was based on step-wise cleavage of 18 polypeptide species observed
during Het2-6 cleavage, using an initial basis set of 25 possible reactions. Individual steps
were assumed to occur with first-order kinetics at a fixed pH and temperature. Numerical
integration of concentrations was carried out with a step size of 0.1 h. Least-squares fitting
was carried out using Microsoft Excel Solver. To assist in fitting of the overall kinetic model
to minor species, differential weighting was applied for species present at low (M-E-L-R-LE-L, M-E-L-R-L, L-R-L-E, L-E-L, M-E-; 100% weighting), intermediate (M-E-L-R-L-E, E-LR-L-E, L-R-L-E-L-R, M-E-L-R, L-E-L-R, E-L-R, M-E; 50% weighting) or high concentrations
(Het2-6, L-R, R, R-, L-E, E, E-; 10% weighting).
104
3.2.3 Peptide charge predictions
The net charge on the peptide was calculated as described in Chapter 2.
3.2.4 Electronic circular dichroism
ECD measurements were collected and mean residue ellipticity calculated as described in
Chapter 2.
ECD spectra for cleaved Het2-6 (1.3 mg mL-1) were recorded at 20 °C in 10 mM Na+
citrate pH 3 or 10 mM Tris.HCl pH 8. For chemical denaturation studies, ellipticity at 222
nm was recorded for cleaved Het2-6 (0.4 mg mL-1) in 1.0-7.5 M guanidinium chloride
(G.HCl) in 10 mM Tris.HCl pH 8, made by mixing separate stocks prepared in 0 or 8 M
G.HCl. Samples were allowed to equilibrated for 10 minutes at the selected temperature
prior to single wavelength measurements.
For thermodynamic analysis, data were subjected to least-squares fitting using Microsoft
Excel Solver.
3.2.5 SDS-PAGE
NuPAGE 4-12% Bis-Tris gels, LDS sample buffer, reducing agent, antioxidant and
SeeBlue Plus2 Prestained Standards were used as per manufacturer’s instructions
(Invitrogen, Grand Island, NY). For Coomassie blue and lanthanum chloride staining,
samples were loaded at ca. 2 µg per lane. Following SDS-PAGE, peptides were detected
by first soaking gels for 10 min in 50 mM LaCl3 and visualising under white light epiillumination. Gels were then stained using 0.1% (w/v) Coomassie blue R-250 (Bio-Rad) in
40% (v/v) methanol, 10% (v/v) acetic acid for 30 min followed by destaining in 40% (v/v)
methanol, 10% (v/v) acetic acid. Silver staining samples were loaded at ca. 0.5 µg per lane
and stained following SDS-PAGE using a Bio-Rad Silver Stain Plus kit (Bio-Rad
Laboratories, Hercules, CA). Partial destaining of silver stained gels was achieved by
soaking gels in 2 g L-1 Na2S2O3, 2 g L-1 K3[Fe(CN)6]. Analysis of gel images used Image
Lab software (Bio-Rad Laboratories, Hercules, CA).
3.2.6 Interfacial tension
Interfacial tension (IFT) data were collected on a Dataphysics OCA20 contact angle
system (Dataphysics, Filderstadt, Germany). Drops (ca. 25 µL) of cleaved or intact Het2-6,
bovine serum albumin or AM1 (1 mg mL-1 in 10 mM Tris.HCl pH 8) were formed in air
using a straight dosing needle in an 8 mL quartz cuvette. Measurements of buffer alone
105
gave IFT 70.2 ± 0.3 mN m-1, while 4 M G.HCl in buffer gave a slightly higher value of 71.4
± 0.4 mN m-1, as expected for a chaotropic salt.1 IFT measurements were collected at 5
sec intervals over 10 min, averaged over 3 measurements. Equilibrium surface pressure
values were determined from the y-intercept of a plot of surface pressure vs. (time)-1/2.2
3.2.7 Emulsion preparation and characterisation
Cleaved Het2-6 (2.5 mg mL-1) in 10 mM triethanolamine.HCl pH 9 was pre-mixed with 5%
(v/v) Jurong 150 base oil for 1 min at 24, 000 rpm using an IKA T25 digital Ultra-Turrax
mixer fitted with a 24 mm dispersing tool. The resulting coarse emulsion was processed
through an Avestin Emulsiflex C5 homogeniser for 3 passes at 16, 000 psi. Droplet sizing
was carried out using dynamic light scattering (Zetasizer Nano ZS, Malvern Instruments,
Worcestershire). The sample was heated for 3 h at 90 °C in a heating block and size
measurements were repeated after this time.
3.3 Results and Discussion
3.3.1 Acid hydrolysis of Het2-6
Chapter 2 described the design, expression and purification of heteroconcatemer Het2-6,
a tetramer comprising alternating copies of peptides EDP-11 and RDP-4 connected by
short unstructured linkers. Within Het2-6, each peptide is flanked by aspartate-proline
dipeptides susceptible to cleavage at acid pH with mild heating.3, 4 The heteroconcatemer
sequence can be represented as M-E-L-R-L-E-L-R, where M is the initial MD dipeptide, E
is EDP-11, R is RDP-4 and L is the linker. Complete cleavage of all seven DP sites in
Het2-6 will produce two copies each of EDP-11 and RDP-4, one MD dipeptide and three
copies of the linker. During the design of Het2-6, an emphasis was placed on amino acid
residues that display a high degree of chemical stability to prevent modification of amino
acid sidechains under the cleavage conditions (Chapter 2).
Several different reaction conditions for the cleavage of aspartate-proline bonds have
previously been reported. Common to these reaction conditions are pH values below the
pKa of the aspartate sidechain and temperatures of 60 °C or higher.3-6 Cleavage of Het2-6
was carried out at 70 °C and pH 2, where the solubility of the concatemer exceeds 50 mg
mL-1. At lower cleavage temperatures, similar product compositions were observed but
with a slower overall rate of reaction (not shown).
106
Figure 3-1. Liquid chromatography elution profiles showing products of Het2-6 cleavage after
0, 6 and 24 h at 70 °C. Peak positions for authentic standard peptides are indicated.
107
Samples of Het2-6 cleavage were collected over 24 hours. Cleavage samples were
analysed by liquid chromatography-mass spectrometry (LCMS; Figure 3-1). Quantification
of eluting species was by comparison to synthetically produced peptide standards.
Identification of species used comparison of observed mass-to-charge ratios to those
expected from all potential products of cleavage at aspartate-proline sites within Het2-6
(Table 3-1). Of the observed species, only M-E-L-R-L-E and E-L-R-L-E were unable to be
separately quantified. These two species co-eluted and the low concentrations involved
and lack of synthetically produced equivalents of these longer polypeptides for comparison
prevented individual quantification.
A total of 18 different oligopeptide species were identified and quantified including intact
Het2-6, monomeric peptides and partly-cleaved intermediates. Importantly, several
species that would be expected to arise if cleavage occurred equally at all DP positions
were either not present, or present in only trace amounts. This greatly simplified the
proposal of a cleavage pathway and kinetic analysis as discussed below.
Prior to heating, trace amounts of several early cleavage products were observed in
purified Het2-6 solutions (Figure 3-1, 0 hours; minor peaks eluting at 35-40 min).The
method employed for purification of Het2-6 from E. coli yielded the protein in solution at pH
2 (Chapter 2). This means that the concatemer is exposed to the cleavage pH during
handling at lower temperature prior to deliberate initiation of cleavage by heating, resulting
in low levels of cleavage and the formation of early cleavage products.
On heating a solution of Het2-6 at pH 2 to 70 °C, dimer and trimer combinations of EDP-11
and RDP-4 were observed in the first several hours. From 4 h, these species decreased in
concentration, in parallel with the emergence of monomer helix species with or without
linkers. Intact Het2-6 was no longer detected after 10 h, and at 24 h cleavage was largely
complete, with 93% of species present at this time point corresponding to monomeric
peptides with or without linker attached.
108
Table 3-1. Species observed by LCMS and their potential and assigned identities.
Observed
molecular weight
Potential species
Assigned species
2626.4
R-
R-
2711.3
R
R
26.4
3440.1
R-L, L-R
L-R
30.5
3823.2
L-E-L
L-E-L
30.9
3209.5
E-L, L-E
L-E
31.7
2595.8
E
E
32.4
2842.1
M-E
M-E
2480.7
E-
E-
6631.6
L-R-L-E, R-L-E-L,
L-E-L-R, E-L-R-L
L-R-L-E
34.1
2726.4
M-E-
M-E-
34.8
6631.6
L-R-L-E, R-L-E-L,
L-E-L-R, E-L-R-L
L-E-L-R
35.3
6877.9
M-E-L-R-L
M-E-L-R-L
35.8
6017.9
E-L-R, R-L-E
E-L-R
36.5
6264.2
M-E-L-R
M-E-L-R
37.6
10053.7
L-R-L-E-L-R,
R-L-E-L-R-L
L-R-L-E-L-R
40.8
10069.5
M-E-L-R-L-E-L
M-E-L-R-L-E-L
9455.8
M-E-L-R-L-E
M-E-L-R-L-E
9209.5
E-L-R-L-E
E-L-R-L-E
12877.9
Het2-6
Het2-6
Elution time
24.8
33.5
42.2
43.0
109
In addition to EDP-11 and RDP-4, monomer products in which the C-terminal aspartate
had been lost were observed at longer incubation times. The resulting products were
designated as EP-11 (E-) and RP-4 (R-). EP-11 and RP-4 appear to be formed via a
second cleavage event N-terminal to aspartate, following initial cleavage between
aspartate and proline. Cleavage of the bond C-terminal to aspartate is preferred under
acidic conditions, with cleavage at aspartate-proline further preferred due to the basicity of
proline (Figure 3-2A). However, cleavage N-terminal to aspartate is also possible, and this
can account for loss of the final aspartate of EDP-11 and RDP-4 (Figure 3-2B).5, 6
Initial cleavage of the Het2-6 backbone appeared to occur C-terminal to aspartate. All
species observed at early time points were able to be assigned to products of cleavage
between aspartate and proline. Furthermore, no species were observed that corresponded
to a C-terminal fragment of Het2-6 with an N-terminal aspartate residue, which would be
expected to arise if initial cleavage occurred N-terminal to aspartate. Furthermore, Cterminally truncated species were observed only late in the cleavage process, consistent
with cleavage N-terminal to aspartate occurring less readily than the expected C-terminal
cleavage. Incubation of synthetic EDP-11 and RDP-4 at pH 2 and 70 °C also gives a rate
of aspartate loss similar to that seen during Het2-6 cleavage (not shown). The loss of Cterminal aspartate is undesirable, as a 1:1 mixture of the truncated peptides have lower
solubility than the parent EDP-11 and RDP-4 species (Section 3.3.2).
Examination of the intermediates observed at early time points during cleavage showed
that not all aspartate-proline sites are equally susceptible to cleavage. In intact Het2-6
there are seven chemically comparable cleavage sites. Early intermediates were identified
corresponding to cleavage at three major sites, giving product pairs M-E/L-R-L-E-L-R, ME-L-R/L-E-L-R and M-E-L-R-L-E/L-R. In each case, cleavage occurred at the C-terminus
of an α-helix, leaving a linker group attached to the N-terminus of the following helix.
Smaller amounts (<1% of total species) of intermediates corresponding to cleavage at ME-L-R-L/E-L-R and M-E-L-R-L-E-L/R were also observed, but primary cleavage at the two
remaining aspartate-proline sites was not detected.
These initial observations yielded important clues as to cleavage selectivity in Het2-6, and
also aided in identifying several intermediates. MS alone was unable to distinguish
between the four different species comprising EDP-11 and RDP-4 with two linkers
attached (E-L-R-L, L-R-L-E, R-L-E-L and L-E-L-R, MW 6632) that could arise by cleavage
of Het2-6. However, it is possible to exclude two of these species (E-L-R-L and R-L-E-L)
110
from the model due to the absence of obligate precursors. Neither R-L-E-L-R nor L-R-L-EL (required to give R-L-E-L) was observed, and M-E-L-R-L (required to give E-L-R-L) was
present only in trace amounts. Two separate species with mass-to-charge ratios
corresponding to MW 6632 were observed using LCMS. Of these, one arose rapidly in the
first hours of cleavage, while formation of the second species showed an initial lag period.
The first species was assigned as L-E-L-R, as it can be generated by a single cleavage
event (M-E-L-R/L-E-L-R), while the second species was assigned as L-R-L-E, which
requires two cleavage events to be formed (M-E/L-R-L-E/L-R). In both cases, cleavage
occurs at the C-terminus of an α-helix.
The absence of several other species is also revealing. Among potential late intermediates
in the cleavage pathway, neither M-E-L nor L-R-L was detected, which require a cleavage
event at the N-terminus of an α-helix. Based on these observations, in further cases where
it was not possible to distinguish between isomeric pairs (R-L-E / E-L-R, L-E / E-L and L-R
/ R-L), the polypeptide was assigned as the product that would arise from cleavage at the
C-terminus of an α-helix. At the same time, the preference for C-terminal cleavage was not
absolute. Formation of L-E-L was observed at low concentrations, requiring cleavage at ME-L-R/L-E-L and/or L-E-L/R. In addition, while M-E-L-R-L-E and E-L-R-L-E were not able
to be separately quantified, loss of the methionine-aspartate dipeptide by cleavage at the
N-terminus of an E α-helix is a necessary element in overall cleavage and this step has
accordingly been modelled into an overall kinetic analysis of cleavage.
Figure 3-2. Sequential acid hydrolysis pathways A) C-terminal to aspartate at a DP site and B) Nterminal to aspartate, leading to loss of the C-terminal aspartate residue.
111
Based on the observed and non-observed intermediates, a minimal scheme for cleavage
of Het2-6 to products was proposed (Figure 3-3). This scheme was used to model the
concentrations of cleavage intermediates with time and obtain values for individual rate
constants, assuming sequential first-order reactions. Data for experimentally determined
and model concentrations are given in Figure 3-4. Least-squares fitting of this model to
experimental data also enabled refinement of the proposed pathways by removal of
potential cleavage events that did not enhance the fit. Limitations in the data, in particular
difficulties in obtaining integrated concentrations for several low-concentration
intermediates, mean that there is uncertainty in the values of individual rate constants.
Nevertheless, the overall fit provides interpretable results. Initial cleavage occurs primarily
at the C-termini of peptide helices (M-E/L-R-L-E-L-R, M-E-L-R/L-E-L-R and M-E-L-R-LE/L-R), a preference which may be a result of helix dipole effects. Cleavage rates are
similar at these three sites despite differences in the amino acid N-terminal to aspartate,
with first-order rate constants of 0.100-0.120 h-1. Some cleavage occurs by reaction at two
additional sites (M-E-L-R-L/E-L-R and M-E-L-R-L-E-L/R), with lower rate constants of
0.050 and 0.016 h-1 respectively.
Second-round cleavage can also partly be accounted for by cleavage at helix C-termini, as
seen from reactions at M-E-L-R/L (0.498 h-1), M-E-L-R/L-E-L (0.142 h-1), L-R/L-E-L-R
(0.171 h-1) and L-R-L-E/L-R (0.090 h-1). However, it becomes necessary to account for
cleavage at L-E-L/R (0.154 h-1), as well as the loss of N-terminal MD at M/E-L-R-L-E
(0.324 h-1) and M/E-L-R (0.103 h-1). Subsequent steps give rise to lower molecular weight
intermediates, including monomer helices with short groups attached at the helix Nterminus (L-R, L-E and M-E). For L-E and L-R, it is possible to model loss of the linker
group at 0.144 h-1 and 0.058 h-1 respectively, with a subsequent slower loss of C-terminal
aspartate. For M-E, it is necessary to model an additional pathway by which the C-terminal
aspartate is lost first (forming M-E-), followed by loss of MD to give E- (0.136 h-1). The loss
of C-terminal aspartate from M-E, E and R occurs at similar rates for all three species
(0.01-0.04 h-1), suggesting the local sequence has little effect on terminal aspartate loss.
Furthermore, the slower rate of these secondary cleavage events compared to initial
cleavage C-terminal to aspartate, is consistent with the expected preferential cleavage at
the aspartate-proline peptide bond.
112
Figure 3-3. Proposed minimal cleavage pathway for Het2-6. M represents the N-terminal MD
dipeptide, E is EDP-11, L is linker and R is RDP-4, while E- represents EP-11 and R- represents RP4. Line thicknesses are approximately proportional to the magnitude of first-order rate constants
derived from modelling.
To the author’s knowledge, this work represents the first attempt to quantitatively analyse
the acid cleavage of a pure protein containing distinct aspartate sites. Although the local
sequence at each site is similar, it was found that particular sites were preferentially
cleaved, apparently due to secondary structure effects. This was particularly striking for
initial cleavage of intact Het2-6, where similar cleavage rates were observed at each of
three sites located at the C-terminus of an α-helix, while cleavage at the four sites Nterminal to an α-helix was either slower or undetected. It appears that this is a result of
helix dipole effects, possibly mediated by an effect on the pKa of the aspartate residues. At
the same time, after initial cleavage, it is probable that at least some cleavage reactions
occur in peptides liberated from the coiled coil, where local unfolding of α-helices and/or
113
Figure 3-4. Time course of Het2-6 cleavage. Separate panels show species present at A) high, B)
intermediate and C) low concentrations. Symbols indicate species concentrations determined by
LCMS, while solid lines indicate data from the model fit. (M)-E-L-R-L-E represents a combination
of the species M-E-L-R-L-E and E-L-R-L-E due to the inability to separately quantify these species
(as discussed in text).
the effects of local chemical context begin to be important. For example, linker cleavage
from L-E (0.144 h-1) was found to occur faster than for L-R (0.058 h-1), implying an effect of
local charge. The most rapid cleavage was found to occur for intermediates generated by
a prior N-terminal cleavage (E-L-R/L-E, 1.93 h-1; E/L-R-L-E, 2.08 h-1; L-E/L, 1.13 h-1). No
reason for this difference is apparent, but it is possible that local unravelling of uncapped
helix ends may play a role.
114
3.3.2 Product characterisation
The initial design of this concatemer expression system was focused on the production of
the anionic peptide EDP-11. EDP-11 was to be separated from the cationic partner RDP-4
by pH precipitation following cleavage; however, it was found that EDP-11 and RDP-4
formed a stable heteromeric complex. Following cleavage, or when using an equal mixture
of synthetically produced EDP-11 and RDP-4, the peptides exhibited co-precipitation at the
pI predicted for a heteromeric complex in which the peptides were incorporated at a 1:1
ratio, similar to intact Het2-6. The co-precipitation of EDP-11 and RDP-4 made separation
by pH precipitation impossible. LCMS analysis showed that reversed-phase
chromatography was able to isolate the product peptides (Section 3.3.1). However, the
bioproduction method described in this and the preceding chapter was centred on easily
scalable methods and therefore precluded utilising chromatography to separate the
product peptides. Given that RDP-4 was based on an amphiphilic helical design similar to
Figure 3-5. Effect of aspartate loss on molecular charge. Predicted molecular charge for 2:2
complexes of EDP-11/RDP-4 (solid line) and EP-11/RP-4 (dashed)
115
EDP-11, the decision was made to continue characterisation as a mixture of the two
peptides with the hope that this would still possess surfactant functionality.
Het2-6 cleavage products showed high solubility at acidic pH at all cleavage times
assessed. To determine the solubility of cleavage products at higher pH, samples were
taken at varied cleavage times and titrated to alkaline pH. Cleaved products showed
decreased solubility at alkaline pH when cleavage was allowed to proceed for longer than
9 h at 70 °C. This was undesirable for many potential surfactant applications where
exposure to near neutral or alkaline pH may be required.7 This decreased alkaline
solubility correlated with the accumulation of end-truncated EP-11 and RP-4 peptides
(Figure 3-4A). The possibility that lower solubility was a result of lower molecular charge
under alkaline conditions was investigated. Figure 3-5 shows the effect of terminal
aspartate loss on predicted molecular charge for coiled coils comprising a 2:2 ratio of
either EDP-11 and RDP-4 or EP-11 and RP-4. The predicted charge per helix for a 2:2
complex of EDP-11 and RDP-4 is -1.94 at pH 9, whereas the charge for a 2:2 complex of
EP-11 and RP-4 is -0.94, which is lower but is still close to the charge of ±1 per helix
expected to be required for solubility (Chapter 2). However, the GRAVY value for EP11/RP-4 is 0.610, while for EDP-11/RDP-4 the value is 0.449, at the threshold of values
considered to define integral membrane proteins.8, 9 It appears that the hydrophobicity of
the peptide complex causes insolubility when the terminal aspartate is lost. At the 9 h
timepoint, cleavage of Het2-6 is 95% complete, with 58% of material present as single
peptides with or without linkers attached as determined by LCMS (Section 3.3.1). Cleaved
products at 9 h were expected to possess desirable surfactant properties. When carrying
out preparative cleavage of Het2-6 at 70 °C to give material for further characterisation,
heating was terminated at 9 h to minimise accumulation of over-cleaved products.
3.3.3 Secondary structure
ECD was used to assess the secondary structure of cleaved Het2-6 at acid and alkaline
pH (Figure 3-6) and compare it to intact Het2-6 (Chapter 2). The spectrum for both intact
and cleaved Het2-6 shows a double minimum at 208 and 222 nm and maximum at 192
nm, which is characteristic of an α-helical fold. At pH 3, the mean residue ellipticity at 222
nm was -28,200 deg cm2 dmol-1 for cleaved Het2-6 and -29,600 deg cm2 dmol-1 for intact
Het2-6. At pH 8, ellipticity values were -24,500 and -26,600 deg cm2 dmol-1 for cleaved
and intact Het2-6, respectively. This shows both intact and cleaved Het2-6 to have a
predominantly α-helical secondary structure, with cleaved Het2-6 having a slightly lower
116
Figure 3-6. Electronic circular dichroism spectra of cleaved Het2-6 at pH 3 (solid) and pH 8
(dashed).
helical content.10 Part of the decrease in helical structure following cleavage may be due to
end-fraying of unlinked helices and suggests cleavage of Het2-6 may partially destabilise
coiled-coil assembly.11
3.3.4 Thermodynamic stability
Given the unusually high thermodynamic stability of intact Het2-6 (Chapter 2) it was of
interest to investigate the stability of cleaved Het2-6 in more detail. Characterisation of
intact Het2-6 showed that a temperature of 90 °C in the presence of G.HCl was required to
effect complete unfolding. Denaturation of cleaved Het2-6 was conducted under the same
conditions as denaturation of intact Het2-6 to enable comparison with the intact protein.
Figure 3-7 shows the G.HCl induced denaturation of secondary structure of both intact
(reproduced from Chapter 2 for comparison) and cleaved Het2-6 at 90 °C and pH 8.
Cleaved Het2-6 precipitated at 90 °C in the absence of G.HCl; however addition of 1 M
G.HCl restored solubility. Unfolding of cleaved Het2-6 (at 0.4 mg mL-1) occurs
117
progressively above 2 M G.HCl, with 50% unfolding achieved at 4.7 M G.HCl, while
unfolding of the intact protein does not commence until the G.HCl concentration reaches
5.0 M, and is only 50% complete at 7.0 M G.HCl. The precipitation of cleaved Het2-6 at 90
°C in the absence of denaturant suggests the cleaved products are able to rearrange and
aggregate, unlike intact Het2-6 (Chapter 2). This suggests that at least part of the unusual
stability to high temperature observed during purification of Het2-6 is due to the linkers
constraining rearrangement of the coiled coil and preventing aggregation.
Analysis of the stability of cleaved Het2-6 from ECD data utilised a model originally
developed by Fairman et al.,12 assuming in the first instance that the cleaved protein
retains a tetramer configuration. The fitting equation is given as:
[𝜃]𝑜𝑏𝑠
[𝑚𝑜𝑛]
= [𝜃]𝑚𝑜𝑛 ×
+ [𝜃]𝑡𝑒𝑡 ×
[𝑃]𝑡𝑜𝑡
𝑒
�
4 × [𝑚𝑜𝑛]4
∆𝐺 (𝐻2 𝑂)
−𝑅𝑇
�
×
−𝑚[𝐷]
�
�
𝑒 −𝑅𝑇
1
[𝑃]𝑡𝑜𝑡
(3-1)
where [θ]obs is the observed mean residue ellipticity at 222 nm, [θ]mon is the mean residue
ellipticity of the monomer state, [θ]tet is the mean residue ellipticity of the tetramer state,
[mon] is the concentration of free monomer as either EDP-11 or RDP-4, [P]tot is the total
concentration of peptide, ΔG(H2O) is the unfolding free energy in the absence of denaturant
and m is an empirical constant accounting for the dependence of folding free energy on
the G.HCl concentration. Values for [θ]mon and [θ]tet were chosen based on observed
ellipticity values at high and low G.HCl concentration respectively. Data were plotted as
fraction folded (Ff):
𝐹𝑓 =
[𝜃]𝑜𝑏𝑠 − [𝜃]𝑚𝑜𝑛
[𝜃]𝑡𝑒𝑡 − [𝜃]𝑚𝑜𝑛
(3-2)
118
Figure 3-7. G.HCl-dependent denaturation of intact and cleaved Het2-6. Denaturation of intact (●)
and cleaved Het2-6 (○) was carried out at 90 °C and pH 8. The fits are to a three-state unfolding
transition for intact Het2-6 (reproduced from Chapter 2) and disassembly of a tetrameric coiled coil
for cleaved Het2-6.
Least-squares fitting was carried out to obtain values for ΔG(H2O) and m given estimated
values of [θ]mon and [θ]tet. A free energy of unfolding of 32.8 kcal mol-1 was determined for
cleaved Het2-6, taking a value of -20,300 deg cm2 dmol-1 for [θ]tet and -4,600 deg cm2
dmol-1 for [θ]mon at 90 °C and pH 8. The estimated value of [θ]tet at 90 °C is approximately
18% lower than the experimentally determined value of [θ]222 for cleaved Het2-6 at 20 °C
(Section 3.3.3). This is similar to the decrease in helicity seen on heating intact Het2-6 in
the absence of G.HCl (Chapter 2). While this suggests a degree of unfolding, addition of
more than 2 M G.HCl is required to induce further denaturation. The free energy of
unfolding obtained in this fit is lower than the 38.9 kcal mol-1 determined for intact Het2-6
under the same conditions (Chapter 2), but still represents a remarkable degree of stability
for a designed coiled-coil system.12-14 In addition, the relatively small change afforded by
the cleavage of linker regions suggests that Het2-6 thermodynamic stability derives more
119
from the hydrophobic association of amphipathic helices, than the fact that the helices
have been constrained into a single polypeptide chain.
Fitting also obtained an m value of 2.44 kcal mol-1 M-1 for cleaved Het2-6, comparable to
values obtained for native proteins of similar size,15 but much lower than the m value of
5.65 kcal mol-1 M-1 for intact Het2-6. As discussed in Chapter 2, it is not possible to obtain
good estimates of the surface area exposed on unfolding from m values obtained at 90
°C.16, 17 However, the smaller value of m for unfolding of cleaved Het2-6 compared to
intact Het2-6 suggests that there is a smaller change in exposed surface area on unfolding
of the cleaved concatemer. This suggests that the structure of cleaved Het2-6 is more
open than that of the intact concatemer, likely due to the removal of helix end-capping by
the linker regions.
3.3.5 Gel electrophoresis
Gel electrophoresis was used an alternate method to study Het2-6 cleavage. Analysis of
Het2-6 cleavage by gel electrophoresis revealed unusual staining patterns for the
cleavage products (Figure 3-8). Conventional Coomassie blue and silver staining tests for
intact and cleaved Het2-6, as well as synthetic EDP-11 and RDP-4, are shown in Figure
3-8;A-B. Intact Het2-6 appears as a band with an apparent molecular weight of 8 kDa
(Lane 2), lower than expected due to incomplete denaturation under standard sample
preparation conditions (Chapter 2). Samples of cleaved Het2-6 (Lane 3), showed several
bands that migrated similarly to Het2-6, along with a reverse-stacked band near 6 kDa that
matched the mobility of authentic RDP-4 (Lane 5). The additional bands at ca. 8 kDa are
likely intermediate cleavage species, several of which have molecular weights of between
6 and 10 kDa (Section 3.3.1). No band is observed for EDP-11 (Lane 4).
The unusual mobility of RDP-4 (Lane 5, actual MW 2826) may be due either to the high
isoelectric point of the peptide, or to co-assembly with sodium dodecyl sulfate into a folded
structure under the conditions of electrophoresis. RDP-4 band intensity is weak under
silver staining, relative to Coomassie blue staining, which may be attributed both to high
affinity of the Coomassie dye for basic residues,18 and a relative lack of chemistries
supporting silver reduction in RDP-4.19-21 The lack of staining of EDP-11 by Coomasie blue
is likely due to this peptide possessing no basic residues to promote staining,18 while the
absence of visualisation by silver staining is likely due to a lack of interactions between
stain molecules and the anionic peptide.22
120
EDP-11 was also unable to be visualized using zinc-imidazole staining23 (not shown).
While exploring alternate methods for protein staining in SDS-PAGE gels using transition
metals,23, 24 a novel staining method was developed based on peptide precipitation in the
presence of lanthanum chloride. Soaking of SDS-PAGE gels in a lanthanum chloride
solution gave a semi-opaque background. Synthetic EDP-11 was visualised as a white
band in these gels due to local precipitation of a peptide-La3+ complex due to the high
carboxylate content of this peptide (Figure 3-8C, Lane 4). EDP-11 is visible at
approximately 2.5 kDa, in agreement with the actual molecular weight of 2596 Da. A
similar faint band can be observed for cleaved Het2-6 (Lane 3), suggesting that cleaved
Het2-6 is dissociated to EDP-11 and RDP-4 monomers. This is in contrast to intact Het2-6,
which shows anomalous electrophoretic mobility, proposed to be due to incomplete
concatemer unfolding by heating in sample loading buffer (Chapter 2). The observation of
monomeric peptides following cleavage suggests that removal of concatemer linkers
destabilises coiled-coil assembly sufficiently for dissociation of monomers under the
conditions used for SDS-PAGE sample preparation.
In gels soaked in a lanthanum chloride solution, RDP-4, Het2-6 and acid-cleavage
products apart from EDP-11, are visualised as negatively stained bands due to a lack of
Figure 3-8. Gel electrophoresis of Het2-6 and cleavage products. A) Coomassie blue R-250 stain;
B) silver stained gel after partial destaining and C) lanthanum chloride stain. Lanes are: 1, MW
standard; 2, intact Het2-6; 3, cleaved Het2-6; 4, EDP-11 and 5, RDP-4.
121
precipitation of these species in the presence of La3+. Additionally, RDP-4, Het2-6 and
acid-cleavage products apart from EDP-11 could be weakly visualised in unstained gels,
apparently as a result of complexation with sample dye components. Post-staining of
lanthanum chloride stained gels with Coomassie blue dye was also possible and gave
similar results to when gels were stained using this method immediately.
3.3.6 Surfactant function
Peptide EDP-11 was designed as a renewable surfactant for practical applications. While
RDP-4 was initially intended only as an expression partner, it was based on a similar
amphiphilic helical design to EDP-11 and was also expected to possess surfactant
properties. It was thus of importance to investigate the surface activity of the peptides at
fluid interfaces. Air-water interfacial tension measurements were carried out on cleaved
and intact Het2-6. Results were compared to those of the previously characterised αhelical surfactant peptide AM125 as an example of a designed peptide surfactant. The
globular protein bovine serum albumin (BSA), similar to several other native proteins, has
previously been shown to be surface active and was included in this study as a reference
point for a native protein with no design for surfactancy.26
Figure 3-9 shows the changes in IFT over time at the air-water interface for solutions of
intact and cleaved Het2-6, AM1 and BSA. BSA is slow to adsorb, reaching an equilibrium
surface pressure of only 16 mN m-1 and requiring 290 sec to achieve 50% of this value.
Intact and cleaved Het2-6 adsorb more readily, with both showing equilibrium surface
pressures of 29 mN m-1. Cleaved Het2-6 was faster to adsorb to the interface, taking only
15 sec to achieve 50% of equilibrium surface pressure, while intact Het2-6 took 30 sec to
achieve the same proportionate change in IFT. This suggests cleavage of Het2-6 and the
resulting lower coiled-coil stability (Section 3.3.4) may enhance the rate of adsorption at
the interface. The lowering of IFT by intact and cleaved Het2-6 is consistent with the
design of peptide sequences with surfactant properties. Reference peptide AM1, which is
largely unstructured in solution at the concentrations used here,27 gave an equilibrium
surface pressure of 25.2 mN m-1, with most of the change occurring before the first
measurement point. This is consistent with the unstructured state of AM1 enabling rapid
adsorption to the interface, while the thermodynamically stable assembled state of intact
and cleaved Het2-6 (Section 3.3.4) may retard adsorption due to the requirement for
disassembly.
122
To investigate the effects of peptide coiled-coil assembly on interfacial adsorption,
interfacial tension studies were carried out on partly-denatured peptides. Addition of 4 M
G.HCl to intact and cleaved Het2-6 slightly lowered equilibrium surface pressure (intact
Het2-6, 24.8 mN m-1; cleaved Het2-6, 25.7 mN m-1; not shown), presumably by solubilising
hydrophobic portions of the peptide structure in aqueous solution. At the same time,
addition of G.HCl brought the time required to achieve 50% of these equilibrium surface
pressures below the dead time of the apparatus. These results are similar to those
obtained for AM1, supporting the argument that it is the liberation of surface-active
monomers from stable complexes that limits the rate of association with the air-water
interface.28 This suggests that the high thermodynamic stability (Section 3.3.4) of the EDP11/RDP-4 coiled coil acts to retard the rate of interfacial adsorption of these peptides.
As a final test of peptide surfactant functionality, an oil-in-water emulsion was prepared
using cleaved Het2-6 (2.5 mg mL-1). Emulsion preparation used a two-step process with
initial preparation of a coarse emulsion followed by passage through a high-pressure
Figure 3-9. Interfacial activity of Het2-6 (red line), cleaved Het2-6 (black line), BSA (green line)
and AM1 (blue line).
123
homogeniser. A droplet size of 200 nm and polydispersity index of 0.27 was determined
for the emulsion as initially prepared, containing 5% (v/v) of an industrial lubricating oil.
When heated to 90 °C for 3 h, the emulsion showed only a slight increase in droplet size to
210 nm, while the polydispersity index increased to 0.35. The resulting emulsion is a
candidate for a renewable, low-toxicity industrial coolant or lubricant that could be
separated into oil and water phases once spent,29 allowing potential reuse of the oil phase.
3.4 Conclusions
This chapter described the acid-catalysed cleavage of the designed heteroconcatemer
Het2-6 to monomer surfactant peptides. By identifying and quantifying the intermediate
species present over time it was possible to carry out kinetic analysis of the cleavage
pathway. Notably, this showed previously unobserved preferences in cleavage across
chemically similar sequences. Specifically, cleavage occurred preferentially at the helix Ctermini, potentially due to effects of peptide secondary structure. Also observed during
analysis of Het2-6 cleavage were monomeric peptides which had lost C-terminal aspartate
residues. This appeared to occur in a secondary cleavage event and resulted in decreased
solubility of product peptides at alkaline pH.
Cleaved peptides assembled into a remarkably thermodynamically stable coiled coil, with
stabilities exceeding those seen in most designer coiled-coil systems. While the initial
design of this expression system included the separation of target anionic peptide from
cationic expression partner, the high heteromeric coiled-coil stability prevented separation
of these species by selective precipitation. A solution of intact or cleaved Het2-6 lowered
interfacial tension at the air-water interface. However, the high stability of the coiled-coil
structure retarded interfacial adsorption, with more rapid association at the interface
achieved by partial chemical denaturation of peptide assembly. Even in the absence of
denaturant, it was possible to use cleaved Het2-6 to prepare a thermostable oil-in-water
emulsion with droplet sizes in the nanometre range, demonstrating the functionality of
these bio-produced self-assembling peptides.
Chapters 2 and 3 report the design and implementation of a bioproduction method for a
strongly anionic peptide surfactant through incorporation into a charge-compensated
coiled-coil heteroconcatemer. This heteroconcatemer expressed at high levels in E. coli
and showed extreme thermodynamic stability, facilitating purification using
chromatography-free pH and heat precipitation steps. Heat- and acid- mediated cleavage
of the concatemer was used to produce monomeric peptides, however isolation of the
124
anionic surfactant peptide using pH precipitation was not possible. Intact and cleaved
concatemer was demonstrated to possess surfactant functionality. Recovery and
processing of the expressed concatemer utilised inexpensive and scalable methods to
increase the potential for future large-scale peptide production. These methods may be
adaptable to future bioproduction approaches, especially those targeting highly charged
sequences. Furthermore, while these results demonstrate a step towards the
bioproduction of designer peptides, there are several future studies that may be able to
improve upon this initial bioproduction process.
3.5 Future directions
3.5.1 Design parameters
The design of peptide and concatemer sequences in Chapter 2 makes use of several
parameters for predicting sequence properties. These were predicted molecular charge as
a function of pH, GRAVY and instability index. To the candidates knowledge there have
been no systematic investigations into the use of these parameters in peptide sequence
design. While the sequences reported in this thesis provide some reference values for
these parameters, a comprehensive investigation of the relationships between these
parameters and polypeptide properties was not the target of this research. A beneficial
objective for future research may therefore be more detailed studies of the effects of
variations of molecular charge, GRAVY and instability index on sequence properties, to
better define target values for future designs. Any such study would require careful design
of new polypeptide sequences to minimise changes to other physicochemical properties
such as residue helix propensity and patterning, to allow isolation of the effects of varying
each parameter.
Of particular interest for future study may be the relationship between the calculated
instability index for designed sequences and in vivo stability. While minimization of the
instability index was used in the design of sequences in Chapter 2, the physicochemical
basis for the index is not well understood. Furthermore, in the absence of further studies, it
is not possible to determine whether minimisation of the instability index alone was
instrumental in producing the high proteolytic stability observed for Het2-6. Future studies
may involve the investigation of a series of sequences in which the instability index is
intentionally varied, by substitution of dipeptides with residue pairs having similar
physicochemical properties, but either higher or lower associated DIWV values. Such
studies would involve monitoring the rates of proteolysis of polypeptides with varied
125
instability indices. If calculated instability index values were observed to predict proteolytic
stability of designed sequences, this would further validate the use of the index in future
design processes.
In designing sequences for such a study, maintaining other factors such as sequence
GRAVY and net charge, as well as residue patterning and helix propensity will also be key
considerations. For example, if such a study were conducted on sequences derived from
EDP-11 or Het2-6, appropriate sequence modifications may involve replacement of
leucine with isoleucine in the tripeptides glutamate-leucine-glutamate. While this
substitution maintains similar residue hydropathy and helix propensity,9, 30, 31 the
isoleucine-glutamate and glutamate-isoleucine dipeptides both have high DIWVs.32 The
incorporation of one, or multiple such substitutions would allow a series of sequences with
varied instability indices to be produced, while maintaining other sequence parameters.
3.5.2 Bioproduction
The failed expression of concatemers Het2-5 and Het2-7 indicate that the design strategy
described in Chapter 2 is not in itself sufficient to give highly-expressed sequences. The
successful expression of Het2-6 and failure of other concatemers may be due to a variety
of factors such as hydrophobic packing in the coiled-coil concatemers or net molecular
charge. While predictions of stable folding based on the FoldIndex (Chapter 2) suggested
that excess molecular charge was not the key factor in poor expression of Het2-5 and
Het2-7, it is not possible to deduce the exact reasons for expression failure from the data
in hand. Future experiments may further probe several of the potential factors involved.
Studies of the effects of net molecular charge on concatemer expression will be beneficial
for the design of future bioproduction constructs, especially for those targeting expression
of highly charged peptides. Modification of the sequence of Het2-6 to vary net molecular
charge while maintaining other design considerations would allow investigation of the
effects of this parameter on a design already demonstrated to express at high levels. By
observing the levels of expression of sequences with varied net charge, it may be possible
to determine an optimal target level of charge for future designs. Such studies may involve
the replacement of the serine of EDP-11 with glutamate, or the replacement of one or
more of the arginine residues of RDP-4 with alanine residues. However, such substitutions
will also affect other design considerations such as GRAVY values and residue helix
propensity that may complicate results.
126
Additionally, the increased molecular charge of Het2-5 and Het2-7 compared to the
charge-compensated Het2-6 is due to the incorporation of a higher proportion of anionic
peptide EDP-11. It may therefore be of interest to investigate the expression of a
hexameric charge-compensated heteroconcatemer, composed of three copies of each
EDP-11 and RDP-4. Such a heteroconcatemer would possess a level of charge closer to
Het2-6 than the highly-charged Het2-5 and Het2-7, and provide more information about
whether the lack of expression of Het2-5 and Het2-7 resulted from poor hydrophobic
packing, or excess anionic charge. More detailed information about the effects of net
molecular charge and concatemer coiled-coil assembly on expression will aid in future
design of bioproduction constructs.
It may also be possible in future studies to determine a crystal structure for Het2-6.
Computer modelling of the folding of the Het2-6 coiled coil may also be possible. A better
understanding of the folded structure of Het2-6 may provide insights into the packing of
residues within the hydrophobic core, as well as allow conclusions to be drawn about the
origins of the concatemers extremely high thermodynamic stability.
3.5.3 Het2-6 redesign
Chapters 2 and 3 describe the design and implementation of a successful bioproduction
method for the highly anionic surfactant peptide EDP-11. However, several unfavourable
properties of expressed peptides were observed which may be accounted for in future
designs. This future design process will likely be a recursive procedure, in which
observations from successfully expressed sequences are used to influence design
considerations.
For example, the loss of terminal aspartate residues from EDP-11 and RDP-4 lowered
resulting complex solubility and required cleavage to be halted prior to completion (Section
3.3.2). However, if these secondary cleavage events are accounted for during the design
process it may be possible to produce sequences which maintain solubility following the
loss of terminal aspartate residues. This would then allow for cleavage to be continued to
completion to give homogenous surfactant peptide products.
The decreased solubility of the EP-11/RP-4 complex compared to the EDP-11/RDP-4
complex was attributed to increased hydrophobicity following loss of terminal aspartate as
reflected by a higher calculated GRAVY value (Section 3.3.2). A potential redesign
approach to overcome this would be to replace one or more non-polar alanine residues
127
with the more polar serine.9 The placement of serine residues would need to consider
factors such as residue patterning and avoiding dipeptide combinations with high DIWVs,
however several potential positions consistent with these considerations are possible
(Table 3-2). Of the potential substitutions, those in the anionic peptide may be more
beneficial in future designs, as these would increase the solubility of the target surfactant
peptide even if it is isolated from the cationic expression partner. The solubility of such
redesigned sequences could be assessed using synthetically produced peptides prior to
attempting bioproduction.
Table 3-2. Potential redesign of EDP-11 and RDP-4
Peptide
Sequence
EDP-11
PG IAELEAE LSAVEAE LEAILAE LD
RDP-4
PG IRALARA IRALARA VRALIRA VRD
Potential sequence redesign of Het2-6 constituent peptides EDP-11 and RDP-4. Bold indicates
potential residues for replacement with serine to increase solubility following loss of terminal
aspartate residues. Underline indicates potential residues for replacement with aspartate for internal
cleavage sites.
Both the lack of ability to separate EDP-11 and RDP-4 via pH precipitation (Section 3.3.2)
and results from interfacial tension studies (Section 3.3.6) indicate that the highly stable
coiled-coil assembly, which is beneficial during expression and chromatography-free
purification, is detrimental to the functionality of this surfactant peptide system following
cleavage. It may therefore be beneficial in future designs to include elements that promote
disassembly of the peptide coiled coil to allow selective precipitation and more rapid
interfacial adsorption.
The ability of the concatemer to fold into a stable coiled coil is required for expression and
during downstream processing. Destabilising elements would therefore be required to only
influence assembly following expression and recovery. This may be accomplished by
including an aspartate residue within the cationic peptide to give an internal cleavage site,
thereby destabilising the heteromeric coiled coil following cleavage by decreasing cationic
peptide length.12, 13 Similar to the redesign approaches discussed above, any such
redesign would need to consider factors such as residue patterning and avoiding
dipeptides with high DIWVs. Several alanine residues close to the centre of RDP-4 may be
appropriate for insertion of such an internal cleavage site (Table 3-2).
128
Due to the shorter length of fragments of the cleaved cationic peptide, heteromeric coiledcoil formation would likely be disfavoured. The anionic peptide may therefore be
selectively precipitated at acidic pH, providing a simple method to isolate the desired
anionic peptide surfactant. Even if selective precipitation is found to be impractical or
undesirable, internal cleavage of the cationic peptide will result in destabilised heteromeric
coiled-coil assembly which will result in more rapid interfacial adsorption of the surfactant
peptides. The properties of such redesigned peptides may be assessed using synthetically
produced peptides prior to attempting bioproduction. The additional anionic residue in
each cationic helix will also increase the net charge on the heteroconcatemer. The effect
this has on expression will also need to be assessed.
129
3.6 References
1.
Breslow, R.; Guo, T. Surface-Tension Measurements Show That Chaotropic
Salting-in Denaturants Are Not Just Water-Structure Breakers. Proc. Natl. Acad. Sci. U. S.
A. 1990, 87, 167-169.
2.
Dexter, A. F. Interfacial and Emulsifying Properties of Designed Beta-Strand
Peptides. Langmuir 2010, 26, 17997-18007.
3.
Dimitrijev-Dwyer, M.; He, L.; James, M.; Nelson, A.; Wang, L.; Middelberg, A. P. J.
The Effects of Acid Hydrolysis on Protein Biosurfactant Molecular, Interfacial, and Foam
Properties: Ph Responsive Protein Hydrolysates. Soft Matter 2012, 8, 5131-5139.
4.
Inglis, A. S. Cleavage at Aspartic-Acid. Methods Enzymol. 1983, 91, 324-332.
5.
Li, A. Q.; Sowder, R. C.; Henderson, L. E.; Moore, S. P.; Garfinkel, D. J.; Fisher, R.
J. Chemical Cleavage at Aspartyl Residues for Protein Identification. Anal. Chem. 2001,
73, 5395-5402.
6.
Zhang, S. F.; Basile, F. Site-Specific Pyrolysis-Induced Cleavage at Aspartic Acid
Residue in Peptides and Proteins. J. Proteome Res. 2007, 6, 1700-1704.
7.
Banat, I. M.; Makkar, R. S.; Cameotra, S. S. Potential Commercial Applications of
Microbial Surfactants. Appl. Microbiol. Biotechnol. 2000, 53, 495-508.
8.
Lobry, J. R. Influence of Genomic G+C Content on Average Amino-Acid
Composition of Proteins from 59 Bacterial Species. Gene 1997, 205, 309-316.
9.
Kyte, J.; Doolittle, R. F. A Simple Method for Displaying the Hydropathic Character
of a Protein. J. Mol. Biol. 1982, 157, 105-132.
10.
Chen, Y. H.; Yang, J. T.; Chau, K. H. Determination of Helix and Beta-Form of
Proteins in Aqueous-Solution by Circular-Dichroism. Biochemistry 1974, 13, 3350-3359.
11.
Zhou, N. E.; Kay, C. M.; Hodges, R. S. Synthetic Model Proteins - Positional Effects
of Interchain Hydrophobic Interactions on Stability of 2-Stranded Alpha-Helical CoiledCoils. J. Biol. Chem. 1992, 267, 2664-2670.
12.
Fairman, R.; Chao, H. G.; Mueller, L.; Lavoie, T. B.; Shen, L. Y.; Novotny, J.;
Matsueda, G. R. Characterization of a New Four-Chain Coiled-Coil: Influence of Chain
Length on Stability. Protein Sci. 1995, 4, 1457-1469.
13.
Litowski, J. R.; Hodges, R. S. Designing Heterodimeric Two-Stranded Alpha-Helical
Coiled-Coils: The Effect of Chain Length on Protein Folding, Stability and Specificity. J.
Pept. Res. 2001, 58, 477-492.
14.
Gibney, B. R.; Johansson, J. S.; Rabanal, F.; Skalicky, J. J.; Wand, A. J.; Dutton, P.
L. Global Topology & Stability and Local Structure & Dynamics in a Synthetic Spin-Labeled
Four-Helix Bundle Protein. Biochemistry 1997, 36, 2798-2806.
15.
Myers, J. K.; Pace, C. N.; Scholtz, J. M. Denaturant M-Values and Heat-Capacity
Changes - Relation to Changes in Accessible Surface-Areas of Protein Unfolding. Protein
Sci. 1995, 4, 2138-2148.
16.
Zweifel, M. E.; Barrick, D. Relationships between the Temperature Dependence of
Solvent Denaturation and the Denaturant Dependence of Protein Stability Curves.
Biophys. Chem. 2002, 101, 221-237.
17.
Scholtz, J. M.; Grimsley, G. R.; Pace, C. N., Solvent Denaturation of Proteins and
Interpretations of the M Value. In Methods in Enzymology: Biothermodynamics, Pt B,
Johnson, M. L.; Holt, J. M.; Ackers, G. K., Eds. Elsevier Academic Press Inc: San Diego,
2009; Vol. 466, pp 549-565.
18.
Congdon, R. W.; Muth, G. W.; Splittgerber, A. G. The Binding Interaction of
Coomassie Blue with Proteins. Anal. Biochem. 1993, 213, 407-413.
19.
Syrovy, I.; Hodny, Z. Staining and Quantification of Proteins Separated by
Polyacrylamide-Gel Electrophoresis. J. Chromatogr.-Biomed. Appl. 1991, 569, 175-196.
20.
Williams, L. R. Staining Nucleic Acids and Proteins in Electrophoresis Gels.
Biotech. Histochem. 2001, 76, 127-132.
130
21.
Merril, C. R.; Pratt, M. E. A Silver Stain for the Rapid Quantitative Detection of
Proteins or Nucleic-Acids on Membranes or Thin-Layer Plates. Anal. Biochem. 1986, 156,
96-110.
22.
Goldberg, H. A.; Warner, K. J. The Staining of Acidic Proteins on Polyacrylamide
Gels: Enhanced Sensitivity and Stability of ''Stains-All'' Staining in Combination with Silver
Nitrate. Anal. Biochem. 1997, 251, 227-233.
23.
Hardy, E.; Castellanos-Serra, L. R. "Reverse-Staining" of Biomolecules in
Electrophoresis Gels: Analytical and Micropreparative Applications. Anal. Biochem. 2004,
328, 1-13.
24.
Lee, C.; Levin, A.; Branton, D. Copper Staining - a 5-Minute Protein Stain for
Sodium Dodecyl-Sulfate Polyacrylamide Gels. Anal. Biochem. 1987, 166, 308-312.
25.
Dexter, A. F.; Malcolm, A. S.; Middelberg, A. P. J. Reversible Active Switching of
the Mechanical Properties of a Peptide Film at a Fluid-Fluid Interface. Nat. Mater. 2006, 5,
502-506.
26.
Pereira, L. G. C.; Theodoly, O.; Blanch, H. W.; Radke, C. J. Dilatational Rheology of
Bsa Conformers at the Air/Water Interface. Langmuir 2003, 19, 2349-2356.
27.
Dexter, A. F.; Middelberg, A. P. J. Switchable Peptide Surfactants with Designed
Metal Binding Capacity. J. Phys. Chem. C 2007, 111, 10484-10492.
28.
Middelberg, A. P. J.; Radke, C. J.; Blanch, H. W. Peptide Interfacial Adsorption Is
Kinetically Limited by the Thermodynamic Stability of Self Association. Proc. Natl. Acad.
Sci. U. S. A. 2000, 97, 5054-5059.
29.
Dexter, A. F. Controlling Coalescence and/or Phase Separation of an Emulsion
Stabilized by an Ionic Surfactant Comprises Adding to the Emulsion a Chaotropic
Counterion. WO2011116412-A1.
30.
O'Neil, K. T.; Degrado, W. F. A Thermodynamic Scale for the Helix-Forming
Tendencies of the Commonly Occurring Amino Acids. Science 1990, 250, 646-651.
31.
Blaber, M.; Zhang, X. J.; Matthews, B. W. Structural Basis of Amino-Acid AlphaHelix Propensity. Science 1993, 260, 1637-1640.
32.
Guruprasad, K.; Reddy, B. V. B.; Pandit, M. W. Correlation between Stability of a
Protein and Its Dipeptide Composition - a Novel Approach for Predicting in Vivo Stability of
a Protein from Its Primary Sequence. Protein Eng. 1990, 4, 155-161.
131
4 Characterisation of a pH-responsive coiled-coil peptide
hydrogel
4.1 Introduction
This chapter details the characterisation of the 21-residue α-helical peptide AFD19, which
was designed as part of ongoing work on peptide surfactants. This peptide was designed
using similar methods to those reported in Chapter 2 for the design of amphiphilic peptides
EDP-11 and RDP-4. This process targeted residues of high helix propensity, as well as
predicted sequence solubility and protease resistance. Characterisation of AFD19
unexpectedly showed the peptide to form self-healing hydrogels at low weight fractions as
a function of pH.
As discussed in Chapter 1, hydrogels are promising materials for use in the biomedical
area where they show utility in applications such as tissue engineering,1, 2 drug delivery2, 3
and wound healing.4, 5 Hydrogels formed by self-assembling peptides are of particular
interest.6, 7 However, to date the majority of reported peptide hydrogels have been based
on β-sheet structures,8-12 for which there remain concerns about clinical application.13 One
alternative to these β-sheet systems are α-helical peptide-based hydrogels, however thus
far few have been reported.14-16 The investigation of AFD19 self-assembly therefore
represented an opportunity to further develop the field and characterise a system that
avoids these undesirable β-sheet elements.
Hydrogel formation by AFD19 was pH-dependent, whereby low pH gave free flowing
solutions, gelation occurred as the pH approached 6 and the peptide precipitated at pH 7.
Characterisation via electronic circular dichroism indicated AFD19 coiled-coil assembly to
occur with a degree of cooperativity of ~ 6, while electron microscopy showed an
entangled network of thin fibrils, with diameters consistent with a higher-order coiled-coil
as the cross-sectional unit. Characterisation of AFD19 self-assembly and comparison of
peptide sequence with previously reported fibril-forming α-helical peptides allowed the
proposal of a model for AFD19 fibril and hydrogel formation. In this model, the self-similar
heptad repeats of the peptide allow permissive helix offsetting. These offset helices
provide end-to-end overlaps resulting in staggered coiled-coil fibril formation. Physical
cross-link formation between fibrils and consequent gelation is controlled by the level of
molecular charge on the peptide whereby highly-charged peptide states gave free-flowing
132
solutions, gelation occurred at a peptide charge of ±1, and the peptide precipitated as the
charge approached zero.
While AFD19 formed hydrogels at pH 6 and precipitated at physiological pH, targeted
modification of this peptide based on the proposed model of self-assembly gave peptide
AFD36, which formed hydrogels under conditions of physiological pH and salt. Results
from small-angle X-ray scattering showed assembly of AFD36 to give fibrils similar to
those observed for AFD19 by electron microscopy. Hydrogels formed by both AFD19 and
AFD36 were extremely thermostable and the constituent coiled coils showed high
thermodynamic stability. The use of bicarbonate as a transient buffer enabled the
preparation of homogenous materials from this single-peptide gelling system. The ability to
form hydrogels under physiological conditions gives these materials potential for use in
biomedical applications
4.2 Experimental
4.2.1 Materials
All reagents used were of the highest grade available. Water was purified using an Elga
Purelab Classic system (Veolia Water, Saint Maurice, France) and had a resistivity of
>18.2 MΩ.cm. Guanidine hydrochloride (G.HCl) (≥99%), 2-(N-morpholino)ethanesulfonate
hydrate (MES; ≥99.5%) and sodium bicarbonate (≥99.5%) were purchased from SigmaAldrich. Glassware was acid-cleaned as previously described.17
Peptide AFD19 (trifluoroacetate salt, Ac-LKELAKV LHELAKL VSEALHA-CONH2, FW
2354) was custom synthesised and RP-HPLC purified by Genscript (Piscataway, New
Jersey). Peptide AFD36 (trifluoroacetate salt, Ac-LKELAKV LHELAKL VKEALHA-CONH2,
FW 2395) was custom synthesised and RP-HPLC purified by Genscript (Piscataway, New
Jersey) or Peptide 2.0 (Chantilly, VA). The final purity was >95% in each case. The
peptide content of the solid was determined by quantitative amino acid analysis (Australian
Proteome Analysis Facility, Sydney, Australia).
Similar to the Het2-6 concentration determination described in Chapter 2, the
concentration of AFD36 was in some cases estimated by measurement of ellipticity at 222
nm using a mean residue ellipticity of -27,500 deg cm2 dmol-1 at a peptide concentration of
close to 1 mM in 10 mM MES buffer, pH 6.0.
133
4.2.2 Electronic circular dichroism
ECD measurements were collected and mean residue ellipticity calculated as described in
Chapter 2.
AFD19 solutions for studies on the effect of peptide concentration or solution temperature
on self-assembly were prepared directly in 10 mN or 1 mN HCl or buffered in 10 mM Na+
acetate (pKa = 4.76) at pH 4.0 or 5.0 or 10 mM Na+ 2-(N-morpholino)ethanesulfonate
(MES; pKa = 6.27) pH 6.0. NaCl was added to adjust the final ionic strength to 0.01.
Samples for studying the effect of peptide concentration on self-assembly were prepared
by serial dilution in the appropriate buffer from a concentrated peptide solution.
All studies of the effect of temperature on peptide self-assembly were carried out in the
temperature range of 20-90 °C at a heating rate of 1.5 °C min-1. AFD19 samples were
prepared at 200 μM peptide in buffers as above. AFD36 samples were prepared at 0.84
mM peptide in 10 mM MES pH 7.0.
Peptide solutions for chemical denaturation studies were prepared at 250 µM peptide in 10
mM Na+ acetate pH 5.0 (AFD19) or 10 mM MES pH 6.0 (AFD36) containing 0-8 M G.HCl,
and were equilibrated at 20 or 50 °C before recording ellipticity at 222 nm.
For thermodynamic analysis, data were subjected to least-squares fitting using Microsoft
Excel Solver.
4.2.3 Peptide charge predictions
The net charge on the peptide was calculated as described in Chapter 2.
4.2.4 AFD19 gel formation
AFD19 samples for photographing at varied pH values and determination of hydrogel
melting temperature were prepared by titrating unbuffered solutions of 1.2 mM (0.3 %
(w/v)) peptide with strong acid or base. Hydrogels at pH 6.0 and 10.7 were prepared by
titrating peptide solutions from pH 3.0 and 11.0 respectively. Aggregated peptide at pH 7.5
was prepared by neutralisation of an acidic solution.
4.2.5 Electron microscopy
For negative staining transmission electron microscopy (TEM), self-buffered AFD19
solutions (0.2 mg mL-1) were prepared at integral pH values from 3.0 to 7.0. A drop of
peptide solution was placed on a formvar-coated 100 mesh copper TEM support grid
134
(ProSciTech, Thuringowa, Australia) and allowed to stand for 1 minute before blotting.
Each sample was stained for 1 minute by addition of 2% (w/v) ammonium molybdate
solution, previously adjusted to the same pH as the initial peptide solution using 1 N HCl or
NH4OH. Samples were viewed on a Jeol-1010 instrument (JEOL, Tokyo, Japan) under an
accelerating voltage of 80 kV and imaged using a Veleta CCD camera (Olympus Soft
Imaging Solutions). Image processing used iTEM software (Olympus).
For cryo-TEM, peptide samples at 1 mg mL-1 were plunge frozen on C-flat holey carbon
grids (EMS, Hatfield, PA, USA) into liquid ethane using an FEI Vitrobot Mark 3 (FEI,
Hillsboro, OR, USA). Optimal blot time was 6 s. Frozen/vitrified samples were viewed on a
Technai F30 TEM (FEI) fitted with a Direct Electron LC-1100 4k camera (San Diego, CA,
USA) using the low-dose mode of the SerialEM image acquisition software.18 Images were
captured unbinned with an electron dose of 20 Å-2 at magnification 39,000×, 50 Å-2 at
78,000× and 44 Å-2 at 93,000×.
For the measurement of fibril diameters, images were analysed using imageJ software
(National Institutes of Health, USA), with twenty measurements averaged for each pH
value.
4.2.6 Plate reader assays
Spectrophotometric determination of pH used an Infinite M200 plate reader (Tecan,
Männedorf, Switzerland) with 48-well plates and sample volumes of 200 μL. The acidity
constant of phenol red in Dulbecco’s modified Eagle’s medium (DMEM) was determined
by titration, with absorbance monitored at 433 nm. Reference A433 readings for fully
protonated and fully deprotonated phenol red were obtained by addition of 10 mN HCl and
10 mN NaOH, respectively. A pKa of 7.70 was obtained as the y-intercept of a plot of pH
vs. log((Aacid – A433)/(A433 – Abase)).
For gelling tests, 100 μL of an AFD36 stock solution (8.4 mM, pH ~ 3) was mixed in a 48well plate with 100 μL of an NaHCO3 solution containing 120 μM phenol red. UV-visible
spectra (700-350 nm) were recorded at 3 min intervals over 60 min and single wavelength
readings were extracted for analysis. Spectra showing an isosbestic point at 477 nm were
used for pH calculations, while samples showing light scattering were excluded. For
greater sensitivity, the pH of gelling samples was determined at 558 nm, with pH
calculated as 7.70 + (A558 – Aacid)/(Abase – A558).
135
4.2.7 Small-angle X-ray scattering
Small-angle X-ray scattering (SAXS) data were collected using an Anton Paar SAXSess
instrument using a monochromated Cu sealed tube operating at 40 kV and 50 mA as a
source, a line focus (Kratky) geometry and a CCD detector at a fixed sample-to-detector
distance. Peptide samples were prepared at 5 mg mL-1 in 10 mM Na+ acetate (pH 4.0 or
5.0) or MES (pH 6.0 or 7.0) with addition of NaCl to maintain an ionic strength of 0.01.
Samples were housed in a silica capillary maintained at 20 °C during the measurements.
SAXS data were measured over a range of q = 0.007 to 0.2 Å-1 (where q is the momentum
transfer vector) and the data were reduced by standard procedures to remove the dark
current and background measured from the buffer, accounting for the relative transmission
of the sample and background. The data were placed on an absolute intensity scale using
the known cross section of water as a primary standard.19 The data presented represent
the average of 1800 exposures collected for 10 sec each to ensure that there were no
changes occurring during exposure to the X-ray beam.
Analysis of the X-ray data was carried out using the NCNR analysis macros in Igor Pro.
For the data obtained here, a flexible cylinder model was used to determine both the mean
radius and persistence length of the fibrils as a function of pH. In these models the
electron density for the buffer was calculated to be 9.45 × 10-6 Å-2, while that for AFD36
was estimated to be 1.22 × 10-5 Å-2 based on chemical composition.
4.2.8 AFD36 gel melting
A ball drop test was conducted using a stainless steel ball (6 mm diameter; 1.0 g) placed
on the surface of an AFD36 gel heated in a water bath. The gel was prepared at 4.2 mM
AFD36 in 10 mM MES and 22.5 mM NaHCO3 (final concentrations), and was allowed to
stand overnight before the heating test to allow loss of CO2 and complete setting. The gel
was heated from 30 to 80 °C in 10 °C steps, and was observed and photographed after
being held for 15 min at each temperature. A separate test was conducted on an AFD36
gel prepared in phosphate-buffered saline (140 mM NaCl, 5 mM KCl, 1.5 mM Na+
phosphate, 22.5 mM NaHCO3, final concentrations) at the same peptide concentration with
heating from 30 to 90 °C. Gel melting was recorded at the temperature where the ball
dropped completely through the gel.
136
Figure 4-1. Helical wheel representations of; A) canonical 18 position helical wheel representation
of a facially amphiphilic α-helix associated with a hydrophobic interface (boxed); B) the peptide
AFD19 originally designed as a peptide surfactant with expanded hydrophobic face (boxed).
4.3 Results and Discussion
4.3.1 Peptide Design
The peptide AFD19 was designed as part of ongoing work on α-helical surfactant
peptides.20-22 The design of this peptide used similar principles to the design of surfactant
peptides EDP-11 and RDP-4 of Chapter 2, as well as the designs of previously reported αhelical surfactants AM1 and AFD4 discussed in Chapter 1. This approach utilises a
canonical α-helix structure, with 3.6 residues per turn resulting in an 18 position repeat. In
previous designs, hydrophobic residues were selectively placed at positions 1, 4, 8, 11, 15
and 18, with the remaining positions populated by hydrophilic residues to give an
amphiphilic helical structure (Figure 4-1A). In the design of AFD19, the hydrophobic
component was expanded to include positions 5, 7, 12 and 14 with the aim of increasing
the affinity of the surfactant peptide for the interface.
Residue selection was based on considerations of high helix propensity,23, 24 similar to the
peptide design process described in Chapter 2. For hydrophobic positions, residues
alanine and leucine were favoured due to their high helix propensity, with a smaller
component of valine residues. Similarly for hydrophilic positions, glutamate was selected
over aspartate as an anionic residue due to its high helix propensity. Cationic lysine
residues were used to balance the charge contributions from glutamate residues to avoid a
sequence with high net charge. The inclusion of two histidine residues (pKa ~ 6.1) at
positions 8 and 20 was driven by a desire to possess control over molecular charge and
137
potentially surfactant functionality near physiological pH, as well as the potential for
intermolecular cross-link formation similar to peptide AFD4.21 Serine was preferred over
glutamine or asparagine as a neutral polar residue due to side-chain chemical stability
considerations.25 Similar to the design of EDP-11, RDP-4 and heteroconcatemer
sequences in Chapter 2, minimisation of the instability index was used to optimise
proteolytic stability,26 while GRAVY27 values of less than 0.6 were targeted to provide
solubility. This resulted in a final sequence for AFD19 of (Ac-LKELAKV LHELAKL
VSEALHA-CONH2, FW 2354) that displays an amphiphilic character (Figure 4-1B), had a
predicted molecular charge of +0.23 at pH 7, a pI of 8.3, an instability index of 6.57 and a
GRAVY value of 0.429. This predicted AFD19 to be soluble at high and low pH values,
and stable to proteolysis.
4.3.2 Peptide characterisation
The solution properties of synthetically produced AFD19 in water were initially assessed
as a function of pH. As-received peptide dissolved in water to give a free-flowing solution
with a pH close to 3.0 due to residual trifluoroacetic acid (TFA) from the chromatography
step utilised in purification. Unexpectedly, AFD19 formed optically clear gels at 0.1-0.5%
(w/v) as the pH approached 6.0 on addition of NaOH. Gel strength and rate of formation
strongly depended on peptide concentration. Mechanical disruption of the hydrogel by
pipetting or vigorous stirring gave a liquid state that re-gelled on a similar time scale to the
original gelation, consistent with the formation of physical cross-links between
components. Samples lyophilized from pH 6.0 did not re-gel on hydration, consistent with
the formation of physically cross-linked hydrogels that do not form by swelling. Solutions of
AD19 at pH 5.0 thickened but did not gel, while above pH 7.0, peptide precipitated from
solution. Reacidification of gelled samples caused rapid gel dissolution, as did the addition
of excess NaOH to give pH ~ 12. Precipitate obtained at pH 7.0 redissolved on the
addition of excess HCl or NaOH. Given the few α-helical peptide hydrogels reported to
date (Chapter 1), the serendipitous observation of gel formation by AFD19 prompted
further investigation of the self-assembly of this peptide.
4.3.3 AFD19 secondary structure
Titration of an acidic 1.75 mM (0.41% (w/v)) AFD19 solution to pH 5.9 resulted in gelation
within 1 minute. ECD was used to probe the peptide secondary structure under the initial
solution (pH 3.4) and final gelled (pH 5.9) conditions (Figure 4-2). Both gelled and ungelled
AFD19 samples showed characteristic α-helical spectra with double minima at 208 and
138
Figure 4-2. ECD spectra of self-buffered AFD19 (1.75 mM) at 20 °C in gelled (solid line, pH
5.9) and ungelled (dashed, pH 3.4) states.
222 nm and a maximum at 192 nm with very little change between the two conditions. The
mean residue ellipticities at 222 nm for gelled and ungelled samples were -27,800 and 30,500 deg cm2 dmol-1 respectively. These values indicate AFD19 to be in a predominantly
helical structure in both states,28 consistent with the design of an α-helical peptide. The
similarity in helical content between these states indicated that hydrogel formation was not
solely dependent on folding of the peptide into α-helices.
As discussed in Chapter 1, the folding of an amphiphilic peptide such as AFD19 to give a
soluble α-helical product, as opposed to adsorption of the peptide at an interface or being
unfolded in solution, was expected to involve the formation of a coiled-coil structure. In
such an arrangement, the hydrophobic faces of the peptides associate, resulting in the
helices wrapping around one another in a left-handed coil. The positioning of residues in
such a supercoil involves a smaller periodicity of 3.5 residues per helix turn compared to
the 18-position repeat used to design AFD19 (Section 4.3.1) and is represented using a
classical heptad-based helical wheel with positions a, b, c, d, e, f and g (Figure 4-3).
139
Figure 4-3. Heptad-based helical wheel projection of peptide AFD19. Positions a, d, e and g are
occupied by non-polar residues. Positions b, c and f are occupied by hydrophilic residues with
self-similar heptad repeats.
As discussed in Chapter 1, in the design of coiled coil forming peptides, positions a and d
are generally reserved for hydrophobic residues while the remaining positions are
populated with hydrophilic residues.29 Specifically, positions e and g are often populated
with charged residues that are involved in electrostatic helix-helix interactions.29-31 In the
design of AFD19, the hydrophobic component of the peptide was increased to include
positions 5, 7, 12 and 14 of the original design (Section 4.3.1) in an effort to enhance
peptide adsorption at the interface. One consequence of this is that when arranged in a
coiled coil, positions e and g of AFD19 are populated with the non-polar amino acids
leucine, valine and alanine. As discussed in Chapter 1, such an expansion of the
hydrophobic face of helical peptides has been associated with an increased coiled-coil
oligomerization state.32-34
The helical structure of AFD19 and gel formation at low weight percentages suggested
that this peptide most likely assembled into fibrillar coiled coils rather than globular
structures as the basis of the gel. Furthermore, minimal change in the relative mean
residue ellipticities at 222 and 208 nm was seen upon gelation. This is in contrast to
140
observations of thickened α-helical fibrils where lateral association of helices often results
in large changes to this ratio, likely due to light scattering.14, 35, 36 The absence of this
spectral change and the optical transparency of the hydrogels supported the formation of
unbundled coiled-coil fibrils. The self-assembly of AFD19 was further investigated to better
characterise the assembled state of AFD19 and the mode of hydrogel formation.
4.3.4 Characterisation of coiled-coil assembly
To further investigate the self-assembly of AFD19, ECD data were collected across a
range of peptide concentrations at the non-gelling pH values of 2.0, 3.0, 4.0 and 5.0 at a
buffer ionic strength of 0.01 (Figure 4-4). Mean residue ellipticity at 222 nm ([θ]222) was
used as a measure of helical secondary structure, with smaller magnitude values
indicating less helical structure.
Figure 4-4. Electronic circular dichroism spectroscopy study of the self-assembly of AFD19 into
α-helical structures as a function of peptide concentration at pH values of 2.0 (○), 3.0 (▲), 4.0
(□) and 5.0 (▼). Lines are a fitted two-state dissociation model described in text.
141
At acidic pH and low peptide concentration, AFD19 is fairly unstructured in solution as
indicated by [θ]222 values close to zero. As the peptide concentration is increased, the
value of [θ]222 decreases sharply, reflecting peptide assembly into coiled coils. As the pH is
increased from 2.0 to 5.0 the midpoints of the folding curve shift to lower concentrations.
This indicates the peptide-peptide affinity is increased as gelling pH is approached. Also
notable from these results at higher pH, is the increasing magnitude of minimum [θ]222
values seen at low peptide concentrations. This indicates that as the gelling pH is
approached, the monomeric peptide is able to adopt a partly helical arrangement, likely
due to changes in the charged state of the peptide and potentially the pH dependent helix
propensity of ionisable residues.37 The [θ]222 values at higher peptide concentrations are
also seen to vary with pH, possibly indicating subtle structural differences in assembled
coiled coils. In particular, at pH 2, assembled AFD19 has a much smaller magnitude [θ]222
than at higher pH values, indicating that as glutamate residues become protonated at
acidic pH, electrostatic repulsion may partially destabilise coiled-coil assembly due to the
high level of positive charge on the peptide.
The cooperativity of AFD19 self-assembly was assessed with the aim of determining the
oligomerization state of the coiled coil. ECD data was analysed using a two-state model
based on an approach by Fairman et al.,38 adapted for a coiled coil of unknown
oligomerization state (n-mer). In this model, n peptide monomers (mon) with a low helix
content assemble into helical n-mers in an equilibrium governed by a dissociation constant
Kd:
𝐾𝑑 =
[𝑚𝑜𝑛]𝑛
[𝑛−𝑚𝑒𝑟]
(4-1)
For each pH, fraction folded (Ff) was calculated at each concentration from observed
values of [θ]222; [θ]obs:
[𝜃]𝑜𝑏𝑠 −[𝜃]𝑚𝑖𝑛
𝐹𝑓 = [𝜃]
𝑚𝑎𝑥 − [𝜃]𝑚𝑖𝑛
(4-2)
Where [θ]min and [θ]max are the mean residue ellipticity values for the unstructured and fully
folded peptide states respectively, and were chosen using an iterative approach to achieve
the best fit to the collected data.
Using the self-assembly model, the fraction folded can also be expressed as:
𝐹𝑓 =
𝑛 [𝑛−𝑚𝑒𝑟]
[𝑚𝑜𝑛]+ 𝑛 [𝑛−𝑚𝑒𝑟]
(4-3)
142
Figure 4-5. Linear least-squares fits of AFD19 assembly using Equation 4-5 at pH values of 2.0
(○), 3.0 (▲), 4.0 (□) and 5.0 (▼).
The degree and strength of self-association may be determined from the concentration of
monomer and n-mer in solution:
log[𝑛 − 𝑚𝑒𝑟] = 𝑛 log[𝑚𝑜𝑛] − log[𝐾𝑑 ]
(4-4)
log 𝑛[𝑛 − 𝑚𝑒𝑟] = 𝑛 log[𝑚𝑜𝑛] − log 𝐾𝑑 + log 𝑛
(4-5)
or equivalently:
Equation 4-5 is in terms of n[n-mer], which corresponds to the concentration of peptide in
the assembled state and allows determination of the values of n and Kd from a linear leastsquares fit to data points at peptide concentrations in the assembly transition region
(Figure 4-5). The values obtained by this approach are given in Table 4-1. These results
indicate AFD19 assembly under non-gelling conditions to proceed with a cooperativity of n
~ 6, suggesting a hexameric coiled-coil structure.
143
From the determined values of n and Kd it is possible to calculate the thermodynamic
stability of the peptide coiled coil. The stability of AFD19 complexes increases with pH,
varying from 4.5 kcal mol-1 per monomer at pH 2.0, to 5.9 kcal mol-1 per monomer at pH
5.0. The AFD19 samples for this concentration dependence study were prepared as
dilution series and the use of a similar method for gelling peptide solutions was not
feasible. This thereby precluded the study of AFD19 self-assembly at gelling pH by the
same methods. However, based on the observed trend of increasing stability, it may be
expected that gelling conditions would give even higher stability.
Table 4-1. Parameters used to fit ECD data for AFD19 self-assembly under non-gelling conditions
pH
n
Kd
ΔGmona
[θ]minb
[θ]maxb
2.0
6.0
1.3 × 10-20
-4.5
-3.5
-20.8
3.0
6.0
2.5 × 10-21
-4.7
-6.2
-26.9
4.0
6.1
2.6 × 10-25
-5.5
-12.9
-25.5
5.0
5.8
3.0 × 10-26
-5.9
-15.7
-24.3
a
binding energy per monomer in kcal mol-1
b
103 deg cm2 dmol-1 at 222 nm
As discussed in Chapter 1 and in Section 4.3.3, the expansion of the hydrophobic face of
amphiphilic α-helices is associated with the formation of higher-order coiled coils such as
pentamers,16, 39 hexamers34 and heptamers.33, 40 Given the hydrophobic face of AFD19
was expanded to include positions e and g of the heptad, the hexameric structure
proposed here is therefore reasonable. However, the data analysis procedure used here to
examine the self-assembly of AFD19 may be too insensitive to changes in oligomerization
state of higher-order coiled coils to unequivocally determine the assembled state. For
example, in the case of AFD19, values of n of ~ 5-7 were obtained with a similar goodness
of fit by varying the chosen values for [θ]max.
The results of characterising coiled-coil assembly were therefore consistent with AFD19
forming a higher-order coiled coil of approximately a hexamer, consistent with the known
effects of expanding the peptide hydrophobic face. However, more detailed studies are
required to confirm the oligomeric state of AFD19 and better understand the mode of selfassembly as a function of pH. This may involve the use of techniques such as
sedimentation equilibrium ultracentrifugation in the future.
144
4.3.5 AFD19 gelling as a function of charge
The dependence of peptide gel formation on solution pH indicated the important role
molecular charge plays in gelation. At acidic pH, AFD19 solutions have low viscosity that
then increases upon titration to pH 5 where a thickened solution is observed. Upon titration
of AFD19 solutions to pH 6, gelation occurs in a concentration-dependent manner. While
no detailed study of charge dependence of gelation appears to have been made in αhelical peptides, there have been several studies of the effects of charge and solubility for
β-sheet peptide hydrogelators, as discussed in Chapter 1.
The formation of physical hydrogels involves a balance between component solubility and
self-association. For example, Aggeli et al.41 have shown that gelation of β-sheet peptides
proceeds as a function of solvent quality in both aqueous and organic solvents for a series
of peptides of varying hydrophilicity. The same group also reported that one unit of positive
or negative charge per 11-residue P11 peptide permitted fibril and hydrogel formation in
unbuffered solutions, while higher charged states were required in higher ionic strength
cell culture media.11, 42 This charge dependence allows pH control of gelation. Separate
work has also shown pH-dependent gelation of a number of β-sheet peptides under
conditions of moderate ionic strength.43, 44 Several charged peptides have also been
shown to undergo gelation in response to solution ionic strength,45-47 while other related
sequences display gel formation as peptide net charge is lowered to near zero.45, 48, 49 The
precise charge dependence of gel formation varies across these β-sheet peptide systems
depending on peptide sequence and solution conditions. However, the balancing of
solubility due to molecular charge with the strength of self-association is a common theme
throughout these reports.
The pH-dependent gelation of AFD19 indicated molecular charge to play a role in
determining peptide self-assembly. To further investigate this, the predicted molecular
charge for AFD19 was calculated across the pH range 2-12, based on an assumption of
independent, non-interacting ionisable sites (Figure 4-6). Figure 4-6 also shows images of
AFD19 solutions prepared at pH values 3.0, 6.0, 7.5, 10.7 and 11.5 to illustrate
macroscopic properties as a function of molecular charge. At low pH, where solutions are
free flowing, the predicted charge on AFD19 is very positive (+4.9 at pH 3.0). As the pH is
raised this charge decreases to approximately +1 at pH 6 where gelation occurs. If the pH
is raised further to pH 7.5, where the predicted charge is approaching 0, a precipitate
forms that can be redissolved at either high or low pH.
145
Figure 4-6. Predicted per peptide molecular charge of AFD19 and observed phase changes in
self-buffered AFD19 (1.2 mM; 0.3% (w/v)) solutions as a function of pH.
To investigate whether the level rather than type of charge was dominant in controlling
gelation, a free-flowing AFD19 pH 11.5 solution with a predicted per-peptide charge of -2.5
was titrated with HCl until gelation occurred over several minutes at pH 10.7 where the
predicted charge approaches -1. These results indicate that AFD19 gelation occurs at a
charge of ±1 and is inhibited when the peptide is highly charged in either basic or acidic
conditions, most likely due to charge-charge repulsion of peptide assemblies preventing
association.
4.3.6 Fibril formation
Formation of low weight percentage peptide hydrogels is generally contingent on the
formation of fibrillar structures for both β-sheet43, 50 and α-helical based14, 15 systems.
Negative staining transmission electron microscopy (TEM) was therefore employed as a
means to visualise the assembled state of AFD19. TEM of dried peptide samples utilised
pH adjusted ammonium molybdate staining solutions and images are presented in Figure
4-7;A-E. The low peptide concentration of 0.2 mg mL-1 was utilised to enable sample
146
preparation without gel formation and enable visualisation of assembly at pH 7.0 where
AFD19 is only sparingly soluble.
Under acidic conditions AFD19 formed thin fibrils of only limited length (Figure 4-7;A-B).
Measured fibril diameters at pH 3.0 and 4.0 were 2.9 ± 0.8 nm and 3.7 ± 0.8 nm
respectively. As the pH is raised to 5.0, longer fibrils are observed, with concomitant
formation of a bundled structure (Figure 4-7C). This resulted in average fibril diameters of
6.7 ± 2.4 nm, however individual thinner fibrils are still visible. Bundling is further increased
at pH 6.0 and 7.0 (Figure 4-7;D-E) with average fibril diameters of approximately 20 nm.
The fibril diameters observed at acidic pH are comparable to the previously reported
diameters of higher-order coiled coils.33, 40, 51 The fragmented structure observed by
negative staining TEM at acidic pH may be due to peptide assembly being less stable at
higher molecular charge (Section 4.3.5). The bundling of fibrils as the pH approaches
neutrality likely results from a reduction in molecular charge coupled with the drying
process used to prepare samples. Given the ECD spectrum of gelled AFD19 does not
Figure 4-7. A-E) Negative staining TEM images of 0.2 mg mL-1 AFD19 at pH 3.0, 4.0, 5.0, 6.0
and 7.0 respectively (scale bars 100 nm); F) Cryo-TEM image of 1.0 mg mL-1 AFD19 gel at pH
6 (scale bar 50 nm).
147
display changes associated with fibril bundling (Figure 4-2), it seems likely that this
bundled state does not exist in solution but rather forms during sample preparation.
Cryo-TEM was utilised as an alternate method to visualise self-assembled structures
formed by AFD19. This method allows visualisation of plunge-frozen hydrated samples
and thus avoids drying and staining steps that are potential sources of artefacts. In a
gelled sample at 0.1% (w/v) peptide and pH 6, an entangled network of hydrated fibrils
with an unbundled morphology can be seen (Figure 4-7F). Measurement of fibril diameters
gives 7.1 ± 1.1 nm, which while much smaller than the bundled fibrils seen using
conventional TEM at equivalent pH, is still larger than may be expected for a hexameric
coiled coil. Measurement inaccuracies may be due to the low contrast present in these
organic molecules limiting resolution. Most notable from this cryo-TEM experiment is the
observation of an entangled but unbundled fibrillar structure, which is in agreement with
ECD predictions of unbundled fibrils.
4.3.7 Thermal stability
To better understand the mechanism of AFD19 assembly and gel formation, the helical
secondary structure of AFD19 was monitored as a function of temperature at different pH
values (Figure 4-8). While hydrophobic interactions are strengthened at elevated
temperatures,52, 53 hydrogen bonding tends to be weakened,54, 55 which allows inferences
to be drawn about the stabilising interactions of peptide hydrogel systems in which selfassembly is promoted by either chilling14, 56 or heating.14, 43
Heat destabilised AFD19 helical structure most effectively at acidic pH, with temperature of
50% denaturation (Tm) values of ~ 45 °C at pH 2.0 and ~ 59 °C at pH 3.0. At pH values of
4.0 and above, α-helix thermostability was greatly increased, with Tm values in excess of
100 °C. Indeed, at pH 4-6 more than 65% of initial secondary structure was preserved at
90 °C, consistent with thermostable hydrophobic interactions stabilising coiled-coil
assembly. Cooling of peptide samples gave recovery of almost all helical structure,
demonstrating the denaturation to be reversible. The pH dependence of stability is likely
due to the high molecular charge at acidic pH partially destabilising coiled-coil assembly.
This destabilising effect is then lessened as the pH is increased and peptide molecular
charge is partially neutralised (Section 4.3.5).
148
Figure 4-8. Secondary structure of AFD19 as assessed at 222 nm using ECD as a function of
temperature during heating (solid line) and cooling (dashed) at different pH values. The
concentration of peptide was 200 µM and the buffer ionic strength was maintained at 0.01.
Solution pH as indicated by colour.
Compared to helical secondary structure, the macroscopic gel structure was less stable. A
hydrogel prepared at 1.2 mM AFD19 and pH 6.0 showed a gel-sol conversion on heating
to 90 °C, which is well below the temperature required to denature the constituent helices.
These results together reinforce the distinction between coiled-coil self-assembly and
hydrogel formation. While the former is likely to be predominantly stabilised by
hydrophobic interactions, gel melting suggests that hydrogen bonding may play a role in
cross-linking between peptide fibrils.
4.3.8 Mechanism of gelation
By considering the data collected on AFD19 self-assembly, it becomes possible to
propose a mechanism for fibril assembly and hydrogel formation. Figure 4-9 illustrates a
149
Figure 4-9. Proposed scheme for AFD19 self-assembly into fibrils that form physically crosslinked hydrogels. Stage a involves the assembly of monomeric peptides into staggered coiled-coil
fibrils. Stage b then involves the formation of a three-dimensional network of fibrils where
association is controlled by peptide charge as a function of pH. Horizontal lines within each
peptide mark the ends of heptads.
two stage process for the assembly of AFD19 to form low weight percentage hydrogels as
a function of pH.
Monomeric AFD19 adopts an increasing level of helical secondary structure as pH is
raised towards gelling conditions and molecular charge decreases (Section 4.3.4). Stage a
involves the self-assembly of these monomeric peptides into staggered coiled-coil fibrils,
involving association of helices through the expanded hydrophobic face. The concentration
dependence of self-assembly suggests a cooperativity of assembly of ~ 6 (Section 4.3.4).
The formation of extended fibrils (as observed in Section 4.3.6) would involve the
assembly of many orders of magnitude more peptide monomers than this determined n
value. It was therefore thought that the concentration dependence study is predominantly
probing an initial coiled-coil nucleation event that involves ~ 6 peptide monomers. This
suggests the formation of an approximately hexameric coiled coil with staggered helices,
which is consistent with the increase in oligomerization state previously associated with
expansion of the hydrophobic core from positions a and d to include positions e and g of
the heptad repeat.32-34 Fibril diameters observed via electron microscopy are consistent
with the formation of a higher-order coiled coil, while an oligomerization state larger than a
heptamer is thought to be unlikely as to the candidates knowledge, these have not been
observed in self-assembling peptide systems.32 These results therefore suggest an
150
oligomerization state of between a pentamer and a heptamer, however confirmation of
oligomerization state will require more detailed structural studies such as sedimentation
equilibrium ultracentrifugation due to limitations inherent in the self-assembly modelling
and TEM studies.
The helix orientation within AFD19 coiled coils has not been determined; however
pentamer,39 hexamer34 and heptamer33 structures with expanded hydrophobic cores have
been reported to form parallel coiled coils. Furthermore, α-helices in which e and g sites
are populated by oppositely-charged residues are generally expected to form parallel
coiled coils due to charge-charge repulsion in the antiparallel arrangement.29 While the e
and g sites in AFD19 are occupied by non-polar residues, similar interactions may occur
between the b and c flanking positions, making it likely that these peptides assume a
parallel arrangement.57
The coiled-coil structure of AFD19 is shown to be extremely stable to thermal
denaturation, most likely as a consequence of the hydrophobic interactions driving
assembly (Section 4.3.7). The α-helical form of AFD19 is quite stable as judged by the
energetics of self-assembly (ΔG = -4.5 to -5.9 kcal mol-1, Section 4.3.4). However the
measured thermodynamic parameters cannot be assigned to the formation of a discrete
coiled coil, as TEM studies showed fibril formation under conditions similar to those
employed for ECD data collection (Section 4.3.6). The fitted values of ΔG may therefore
correspond to the formation of staggered coiled-coil fibrils rather than discrete coiled coils,
as illustrated in Stage a, a point that will be returned to later.
Several approaches to specify the formation of coiled-coil fibrils through the introduction of
forced helix offsets have been reported by the Woolfson,14, 36, 56, 58-63 Conticello64-66 and
Fairman67, 68 groups (Chapter 1). The peptide AFD19 lacks these design elements, but
possesses some similarities with the peptides reported by Kojima et al.,69 Takei et al.,70
Potekhin et al.,16 Melnik et al.71 and Dong et al.15 to form coiled-coil fibrils in the absence of
any design for forced helix offsets. As described in Chapter 1, these peptides all possess
self-similar heptad repeats. While AFD19 has a lower degree of self-similarity than many
of these peptides, it possesses all non-polar residues at positions a, d, e and g, anionic
glutamate residues at position c and predominantly cationic residues at positions b and f.
This gives it a comparable degree of self-similarity to a peptide reported by Dong et al.15 to
form fibrils and transparent gels. Based on the similarities between these sequences, it
appears likely that fibril formation occurs by permissive, rather than specified, offsetting of
151
facially amphiphilic helices. One consequence of these self-similar heptad repeats is that
helices offset by a full heptad would have similar net intermolecular interactions to an inregister arrangement. Similarly to the specified offset helices of the Woolfson,14, 36, 56, 58-63
Conticello64-66 and Fairman67, 68 groups’ designs, these out-of-register helices could then
provide end-to-end overlaps leading to staggered coiled-coil fibril formation (Figure 4-9).
Furthermore, crystal structures of parallel coiled coils formed by GCN4 peptide variants
have shown that non-polar residue placement at peptide e and g positions (as in AFD19)
can result in helix offsetting due to non-classical packing of residues within the
hydrophobic core.32 For example, peptide GCN4-pVe has valine residues at all e positions
and formed tetrameric coiled coils with three-residue offsets between adjacent helices.72
Additionally, peptide GCN4-pAA has alanine residues at both e and g positions and
formed a heptameric coiled coil with single-residue offsets between adjacent helices.33 If
such an effect were involved in AFD19 assembly, then this offsetting may assist in driving
fibril formation by providing “sticky ends” for end-to-end helix overlaps. The prediction of
such interactions in the hydrophobic core of coiled coils are not straightforward and further
experiments are required to determine the packing of residues within the core of AFD19
fibrils.32
More detailed studies are required to determine the exact structure of AFD19 coiled-coil
fibrils in terms of oligomerization state and helix offsetting. This is a non-trivial process as
such fibril forming peptides are difficult to characterise using conventional techniques such
as X-ray crystallography or solution nuclear magnetic resonance spectroscopy. However,
in the future, further analysis by techniques such as higher resolution electron microscopy,
X-ray fibre diffraction or sedimentation equilibrium ultracentrifugation may be able to better
characterise peptide assembly.34, 40, 73 Given the data in hand it appears likely that fibril
formation occurs by the permissive offsetting of helices due to self-similar heptad repeats,
potentially assisted by preferential packing of hydrophobic residues at positions a, d, e and
g.
The similar intermolecular interactions predicted for offset and in-register helices with selfsimilar heptad repeats have implications for analysis of peptide self-assembly. The
determined thermodynamic parameters (Section 4.3.4) will correspond primarily to the
formation of a coiled coil of offset helices and be comparable to values determined for an
in-register coiled coil. This is supported by the determined cooperativity of assembly (n ~
6), and is consistent with the proposal (above) that the concentration dependence study is
152
predominantly probing an initial nucleation event, as further assembly to coiled-coil fibrils
will not involve changes in peptide-peptide intermolecular interactions. This also suggests
that extended fibril formation is not readily observed by ECD.
Despite the absence of a strong driving force for helix offsetting, it appears that the
majority of peptide in solution is involved in extended fibril formation even at moderate
concentrations, as indicated by electron microscopy and the formation of hydrogels at low
peptide concentrations. Head-to-tail association of α-helices has been observed in several
crystal structures,74, 75 stabilised by hydrogen bonding and possibly also by interactions
between helix dipoles. The proposed parallel arrangement of helices within AFD19 coiledcoil fibrils would yield summed α-helix dipoles that may stabilise the head-to-tail
association of helices, thereby increasing the strength of fibril assembly.
Given the non-specified nature of helix offsetting, there are several potential combinations
of offset helices that would result in staggered coiled-coil fibril formation. Adjacent helices
may be offset by either one or two heptads, of which Figure 4-9 illustrates one possible
combination. However, many other possibilities may be envisaged and it appears unlikely
that a single configuration would be present across all fibrils. If the packing of hydrophobic
residues within the core is involved in specifying the formation of offset helices, a more
ordered arrangement of helix offsetting may be present. However, in the absence of
further experimental data this cannot be confirmed. Fibril-forming peptides based on selfsimilar heptad repeats avoid the thermodynamically destabilising63, 76 placement of polar
residues in the coiled-coil core. This facilitates development of gels composed of short
single peptides, as opposed to paired peptides or longer sequences that may be
expensive to synthesise. As exact repeats are not required for fibril assembly, this also
allows the incorporation of specific residues to control molecular charge (as is described in
Section 4.3.9). It also allows for diversity of residue composition, which may be useful in
the contexts of peptide bioproduction or functionalization for enhanced cellular
interactions.
Stage b of hydrogel formation is physical cross-link formation between coiled-coil fibrils to
create a three-dimensional network that can immobilise large volumes of water.
Visualisation of hydrated fibrils in a gelled AFD19 sample using cryo-TEM (Section 4.3.6)
indicated that individual fibrils are able to associate laterally and may wrap around each
other for distances of up to 50-100 nm. The formation of physical cross-links between
these fibrils results in network formation and gelation. It is likely fibril-fibril interactions are
153
inhibited by electrostatic repulsion at higher molecular charge, thereby generating the
observed pH dependence of gel formation. These fibrillar networks were observed in
increasing amounts as the pH was adjusted from 4.0 to 6.0, reflecting the decreasing
peptide charge (Section 4.3.5). This is then followed by precipitation as the charge is
neutralised to a point too low to provide solubility, illustrating the fine balance of molecular
charge effects on fibril solubility and fibril-fibril association. Observations of AFD19
hydrogel formation as a function of pH indicate that a charge of ±1 per 21-residue peptide
is required for fibril-fibril interactions to overcome electrostatic repulsion without
precipitation (Section 4.3.5). The character of fibril-fibril interactions for AFD19 fibrils has
not been explicitly determined; however they are likely to be a combination of hydrophobic
interactions and hydrogen bonding. This is consistent with the gel melting temperature of ~
90°C, which while indicating high thermostability of fibril-fibril interactions, is still a lower
temperature than that required for helix unfolding (Section 4.3.7). This indicates that fibrilfibril interactions are less stable to elevated temperature than the hydrophobic interactions
within the coiled coil.
This proposed model of AFD19 coiled-coil fibril formation is consistent with observations of
fibril formation via electron microscopy, hydrogel formation as a function of solution pH, as
well as the characterisation of peptide self-assembly via ECD. Furthermore, the role of
molecular charge in determining gelling conditions allows directed modification of peptide
sequence to obtain desired gelling properties as is discussed in Section 4.3.9.
4.3.9 Peptide re-design for physiological gelling
As the peptide AFD19 forms hydrogels at pH 6.0 and precipitates at pH 7.0, it is not usable
in applications where exposure to physiological conditions will occur. However, with the
model of assembly proposed in Section 4.3.8 in hand, it was possible to redesign this
peptide sequence with the aim of generating a peptide that gels under physiological
conditions.
Changes to the sequence of AFD19 were minimised in an attempt to preserve peptide
self-assembly. Given the dependence of hydrogel formation on peptide charge proposed
in Section 4.3.5, the serine at position 16 was replaced with a lysine to increase the
predicted peptide charge to +1 at physiological pH. This generated the new sequence
AFD36 (Ac-LKELAKV LHELAKL VKEALHA-CONH2, FW 2395, Figure 4-10). This
substitution is consistent with the placement of residues for formation of an amphiphilic
helix, as well as the self-similar heptad repeats proposed to be required for fibril formation
154
Figure 4-10. Comparison of heptad based helical wheel projections of peptides AFD19 and
AFD36 with the serine-lysine substitution highlighted in red.
(Section 4.3.8). The sequence of AFD36 possessed a calculated GRAVY value of 0.281,
predicting higher solubility than AFD19. The instability index of AFD36 is -3.98, which
predicts a sequence more stable to proteolysis than AFD19 and much lower than the value
of 40 proposed to divide proteolytically stable and unstable sequences.26 The predicted
charge per peptide of AFD36 is shown in comparison to AFD19 in Figure 4-11. AFD19 is
predicted to possess +1 charge at pH 6.0, where gel formation is observed, followed by
zero charge at pH 7.0, which is in line with the observed precipitation at this pH. In
comparison, the predicted charge on AFD36 only approaches +1 as the pH is raised
above pH 7.0 and does not decrease to zero until pH 10.3. It was therefore expected for
AFD36 to be compatible with gel formation at physiological pH.
4.3.10
Self-assembly and gel formation
Similarly to AFD19, as-received AFD36 peptide dissolved in water to give an acidic
solution. However, upon titration of a 0.1-0.5% (w/v) peptide solution with NaOH, AFD36
gave a free-flowing solution at pH 6.0 with gelation occurring at pH 7.0, where the
molecular charge is predicted to approach +1, and precipitation not observed until pH
10.3. Similarly to AFD19, if a highly alkaline solution of AFD36 is taken and the pH
adjusted down, gelation was seen to occur where the molecular charge approaches -1,
155
Figure 4-11. Predicted charge per peptide of AFD19 (dotted) and AFD36 (solid line) as a
function of pH.
which for AFD36 is close to pH 11. As found with AFD19, mechanical disruption of AFD36
hydrogels gave a liquid state that then re-gelled on a similar time scale to the original
gelation.
As expected, AFD36 adopted a helical conformation as judged by ECD, with a double
minimum at 208 and 222 nm and a maximum at 192 nm characteristic of α-helix formation
observed for both ungelled (pH 3.1) and gelled (pH 7.0) states (Figure 4-12). In the gelled
state the mean residue ellipticity at 222 nm is -30,100 deg cm2 dmol-1, indicating AFD36 to
be predominantly helical under these conditions.28 This is similar to the level of helicity of
AFD19 in gelled and ungelled states (Figure 4-1). In the ungelled state, the mean residue
ellipticity at 222 nm is -25,700 deg cm2 dmol-1, indicating marginally lower α-helical
content. This possibly reflects destabilisation of the coiled-coil structure by electrostatic
repulsion due to the higher charge of +6 present on AFD36 at pH 3.1 compared to the +5
for AFD19 at pH 3.0. These spectra also show no change in [θ]222/[θ]208 ratio upon gel
formation, supporting the formation of unbundled fibrils as discussed in Section 4.3.3.
156
Figure 4-12. ECD spectra of self-buffered AFD36 (1.6 mM) at 20 °C in gelled (solid line, pH 7.0)
and non-gelled (dashed, pH 3.1) states.
These results indicate successful redesign of the gelling peptide AFD19 to give a peptide
that is capable of gelation at physiologically pH. Similarly to AFD19, gelation of AFD36
occurs at a predicted charge of ±1, is reversible with pH, and demonstrates self-healing
after mechanical disruption. The self-assembly and hydrogel formation of peptide AFD36
at the predicted pH further supports the model proposed in Section 4.3.8 for the assembly
of the α-helical peptides to form staggered coiled-coil fibrils which then form physical
cross-links as a function of molecular charge. The tolerance of the self-assembling
peptides for modifications such as that described here may allow future targeted sequence
modifications as desired for individual applications. The gelation of AFD36 under
physiological conditions makes this peptide a potential candidate for biomaterial
applications, specifically ones where injection of a hydrogel under physiological conditions
is required.
157
4.3.11
Investigation of fibril structure
Small-angle X-ray scattering (SAXS) was used to further investigate the fibril selfassembly of AFD36. Samples were prepared at pH 4.0-6.0 (non-gelled) and 7.0 (gelled) in
an effort to provide structural information for a range of assembled states. Figure 4-13
shows X-ray scattering data for each pH with fits to a flexible-cylinder model. Fibril
formation is observed at all pH values assessed, even though gelation is not seen below
pH 7.0. This supports the proposal in Section 4.3.8 that monomeric peptides assemble to
form staggered coiled-coil fibrils, with gelation controlled by fibril-fibril interactions as a
function of molecular charge.
Table 4-2. Diameters of AFD36 fibrils at different pH values obtained by SAXS fitted to a flexible
cylinder model
pH
a
Fibril
diametera
Persistence
lengtha
4.0
3.91 ± 0.01
10.6 ± 1.4
5.0
3.82 ± 0.02
10.7 ± 2.1
6.0
3.79 ± 0.01
14.0 ± 0.3
7.0
3.78 ± 0.01
12.4 ± 3.5
nm
Fitting of X-ray scattering data allowed determination of fibril diameters and persistence
lengths across the entire pH range (Table 4-2). Fibril diameter is similar across the pH
range with values of 3.8-3.9 nm. This is comparable to the widths of short AFD19 fibrils
observed at acidic pH using negative staining TEM, although much smaller than the
bundled fibrils observed at increased pH (Section 4.3.6). This indicates the formation of
unbundled AFD36 fibrils in the pH range 4.0 to 7.0. Assuming that assembly of AFD19
follows a similar pathway to AFD36, this supports the idea that the thickened fibrils
observed at pH 4.0-6.0 in negative staining TEM are an artefact of sample preparation and
staining, with fibrils instead existing unbundled in solution as visualised in cryo-TEM
(Figure 4-7). For comparison, the hSAF peptides reported by Banwell et al.14 that were
designed to form coiled-coil fibrils with forced offsets showed bundled fibrils (as judged by
electron microscopy and ECD; discussed in Chapter 1) as well as opaque gels. It appears
that the molecular charge on AFD36 at the pH values tested here prevents fibril bundling
through electrostatic repulsion.
158
Figure 4-13. Small-angle X-ray scattering by AFD36 as a function of pH. Data were collected
for 0.5% (w/v) peptide at pH 4.0 (○), 5.0 (□), 6.0 (◇) and 7.0 (△). Solid lines show fits to a
flexible-cylinder model. Data sets above pH 4.0 are displaced vertically progressively by a factor
of ten for clarity.
The diameter of AFD36 fibrils may be compared with those of other higher-order coiled
coils for which 3D structures are available from X-ray crystallography or nuclear magnetic
resonance spectroscopy. The data give approximate diameters of 3.1 nm for the
heptameric designed peptide GCN4-pAA (2HY6),33 3.3 nm for the hexameric designed
peptide CC-Hex (3R48),34 3.0 nm for the pentameric M2 segment of the acetylcholine
receptor (1EQ8)77 and 2.9 nm for the pentameric cartilage oligomeric matrix protein
(COMP, 1VDF).51 Redesign of GCN4-pAA has recently yielded peptide 7HSAP1, which
forms fibrils with a proposed heptamer structure and a diameter of 3.0 nm as measured by
X-ray diffraction.40
The modelled diameter of AFD36 fibrils, at 3.8-3.9 nm is larger than these reference
systems. The diameter of an α-helix is not constant but instead dependent on residue
composition. The larger observed fibril diameter may in part reflect the higher content of
larger aliphatic residues leucine and valine in the AFD36 core than are present in the
159
comparison systems. However, the diameter is still consistent with a fibril structure
comprising unbundled higher-order coiled coils of at least a pentamer cross-section.
Size exclusion chromatography (SEC) was attempted under acidic conditions in an effort
to more precisely determine the minimum oligomer state of the peptide. At lower pH, high
molecular charge was hoped to prevent fibril assembly. Due to instrument limitations the
lowest pH able to be assessed was 4.0, and solutions of AFD36 were thickened at the
concentrations required for SEC. This suggested the peptide to be in an assembled state,
and when applied to the size exclusion column the peptide was seen to be retained on the
column, preventing determination of AFD36 oligomerisation state. The inability to
determine conditions that give coiled-coil assembly in the absence of fibril formation
hinders the use of SEC in more accurately determining the oligomerization state of the
peptide coiled coil. Future studies involving alternate techniques such as sedimentation
equilibrium ultracentrifugation may be better suited for such studies, and allow
experimentation under more acidic conditions.
Comparison of the determined persistence lengths of AFD36 fibrils with other biomolecular
systems shows them to be relatively flexible. In the pH range 4.0 (non-gelled) to 7.0
(gelled), AFD36 fibrils show relatively constant persistence lengths of 10-14 nm, indicating
that fibril flexibility varies little with the non-gelled to gelled transition. As comparison, the
persistence length of double-stranded DNA, with a comparable hydrodynamic diameter of
3 nm, is a much larger 50-100 nm,78 while fibrils of 1-6 nm diameter formed from the milk
protein β-lactoglobulin have persistence lengths of 90 nm to 4 µm depending on β-sheet
content.79 Additionally, the fibrils formed by the designer β-sheet peptide P11-2 have been
reported to have persistence lengths of up to 70 µm at a diameter of 8 nm.80 When
compared with these other fibril forming biomolecules it can be seen that AFD36 fibrils are
relatively flexible even given their small diameter, which may relate to the non-specific
character of hydrophobic interactions between subunits in the fibril. This is consistent with
the entangled network of bending fibrils of AFD19 observed using electron microscopy in
Section 4.3.6.
One feature common to higher-order coiled coils is the presence of a hydrophobically lined
central pore.33, 34, 39, 40 Fitting of SAXS data was also attempted to a hollow-cylinder model,
however this was unsuccessful, possibly due to the data not being across a broad enough
range of q to distinguish such a structure.40 This may be overcome in the future by
conducting experiments over a larger scattering range as well as enhancing the X-ray
160
scattering contrast by loading hydrophobic molecules with a larger X-ray scattering cross
section into the coiled-coil core.
4.3.12
Bicarbonate controlled pH adjustment for gelation
The pH-responsive gelling of the AFD peptides provides a simple and reversible method
for gel formation; however it also poses the disadvantage of difficulties in producing a
homogenous material. Poor mixing due to local gelation can be observed on the addition
of acid or base to a peptide solution containing pH indicator dyes (Figure 4-14; inset). On
NaOH addition to a stirred peptide solution containing phenol red, a basic gel forms locally
(crimson) that is surrounded by acidic medium (yellow). Equilibration occurs only slowly,
leading to a non-uniform structure and lower gel strength; in some cases, gel formation
may fail completely.
Figure 4-14. pH changes over time for AFD36 gels prepared on titration with different
concentrations of bicarbonate. Solid line, 20.0 mM NaHCO3; dotted line, 22.5 mM NaHCO3;
dashed line, 25 mM NaHCO3 (Final concentrations). Inset: local gelling and pH inhomogeneity on
addition of NaOH to acidic solution of AFD36 containing phenol red. The peptide concentration in
each case was 4.2 mM.
161
One method to overcome this problem involves slow pH adjustment based on transient
buffering by carbon dioxide. This slower rate of gelling provides a handling window for
gelation of concentrated peptide solutions, while also allowing uniform mixing to maximise
gel homogeneity and strength. This is achieved by titration of an acidic peptide solution
with the weak base sodium bicarbonate, leading to the formation of carbonic acid, which
itself dissociates to give a slightly acidic solution pH. However, carbonic acid is in
equilibrium with atmospheric carbon dioxide and is gradually lost from the gelling solution
(Equation 4-6), leading to a slow pH rise that can be observed using an indicator dye.
𝐻𝐶𝑂3 − + 𝐻 + �
� 𝐻2 𝐶𝑂3 ⟷ 𝐶𝑂2 + 𝐻2 𝑂
𝑝𝐾
𝑎
6.3
(4-6)
UV-visible spectrophotometry and phenol red were used to track pH changes in solutions
of AFD36 titrated with different amounts of sodium bicarbonate (Figure 4-14). Samples
were prepared in the wells of a 48-well plate as would be utilised for cell culture
experimentation. As would occur on titration with a strong base, the final gel pH is
dependent on the amount of base added. Following bicarbonate addition, pH change is
seen to follow an initial step change (prior to measurements commencing) followed by a
slow rise, with equilibrium in a 200 µL volume with an exposed surface area of ~ 1 cm2
taking more than an hour.
Clear gels are obtained at pH values up to 8.2, however at more alkaline pH light
scattering is observed, consistent with fibril bundling as the peptide molecular charge
approaches zero and solubility is lowered. Gel formation occurs before the pH change is
complete, however the slow rate of pH change afforded with this method allows a handling
window of several minutes for the peptide solutions prior to gelation. This property may be
useful in a clinical setting where gelling solutions may initially be handled and applied in
liquid form before assuming their soft-solid state. Overnight exposure of the gels to a 5%
CO2 atmosphere in a mammalian cell incubator lowered the pH by approximately 0.4-0.5
pH units (not shown), but appropriate concentrations of bicarbonate could be chosen to
allow the preparation of gels at physiological pH after equilibration in an incubator.
Additionally, gel preparation using 10 mM MES-bicarbonate buffer allowed sample
characterisation (Section 4.3.13) in the absence of interference by salt.
As discussed in Chapter 1, the majority of peptide hydrogels reported to date are formed
on the addition of salt to a peptide solution to increase ionic strength and trigger assembly.
This approach appears to result in sufficiently homogenous materials for characterisation
and further experimentation. However, there are several examples of hydrogels formed on
162
adjusting the pH of peptide, most commonly Fmoc-dipeptide, solutions.81-86 As discussed
here, the addition of inorganic acid or base to trigger gelation of these systems often gives
inhomogeneous materials due to gelation occurring on a shorter timescale than pH
equilibration across the solution. Ongoing research has led to the development of a
number of approaches to overcome this, the majority of which involve the slow
homogenous acidification of a peptide solution. For example, Adams et al.81 incorporated
glucono-δ-lactone in the peptide solution which gradually hydrolysed to gluconic acid and
lowered solution pH, thereby triggering homogenous gelation. Another approach by
Raeburn et al.82 involved the incorporation of a photoacid generator, allowing gelation to
be triggered by exposure to UV light and consequent acidification. More complex
approaches to lower pH have also been developed which involve the use of enzymatic
processes,83 as well as the binding of sugar to a boronic acid.84 The approach described in
this thesis where bicarbonate is used to adjust pH is most similar to the approach of
Adams et al., whereby a simple organic molecule, which alters pH over time, induces
homogenous gelation. Importantly, while previously reported approaches are directed
towards acidification of a solution, the approach used here induces a slow increase in pH
which may be useful in many systems.
4.3.13
Thermal denaturation of AFD36
Thermal denaturation studies were conducted to determine the heat stability of selfassembled AFD36 structures. This was done in terms of both primary (coiled coil) and
secondary (gel) assembly as assessed via ECD and gel melting respectively. Figure 4-15
shows the effects of temperature on AFD36 helical structure, as measured by ECD at 222
nm. At 20 °C, the mean residue ellipticity at 222 nm is -30,100 deg cm2 dmol-1 for a gelled
sample at pH 7.0. This decreases to -25,500 deg cm2 dmol-1 upon heating to 90 °C,
indicating that similarly to AFD19 (Section 4.3.7), the Tm of AFD36 helical structure at
gelling pH is well in excess of 100 °C. Upon cooling most, but not all, of the original helical
content is recovered, giving a final mean residue ellipticity of -28,800 deg cm2 dmol-1. The
extreme stability of helical structure to thermal denaturation shown for AFD19 and AFD36
is similar to that seen for other higher-order peptide coiled coils, with the CC-Hex
hexamer34 shown to also have a melting temperature > 100 °C and the GCN4-pAA
heptamer33 reported to have a melting temperature of ~ 95 °C (although at lower
concentrations than used here). These higher-order coiled-coil forming peptides possess
expanded hydrophobic faces compared to lower-order coiled-coil forming peptides. This
163
increased level of thermostable hydrophobic interactions52 driving assembly is likely
responsible for the observed trend in high stability to thermal denaturation.
Compared to helical content, the macroscopic gel structure of AFD36 is less stable. A ball
drop test showed a gel-sol transition temperature of ~ 80 °C for an AFD36 gel prepared in
10 mM MES-bicarbonate buffer. This increased to ~ 90 °C for a gel prepared in
phosphate-buffered saline at higher ionic strength, which is similar to the stability of AFD19
hydrogels described previously (Section 4.3.7). The dependence of gel melting
temperature on ionic strength may be due to both enhanced hydrophobic interactions and
weakened repulsive electrostatic interactions at higher ionic strength. The disparity
between ECD and gel-sol transition temperature is consistent with the idea that the gel-sol
transition may occur by loss of the physical cross-links between fibrils, rather than
denaturation of the helical structure. Alternatively, the slight decrease in helix content seen
at elevated temperature may lead to shortening of the fibril segments and eventual loss of
Figure 4-15. Thermal stability of secondary structure for gelled AFD36. Thick line, heating
curve; thin line, cooling curve. Inset: Determination of gel melting temperature via a ball drop
test with photos taken at 30 °C (upper left) and 80 °C (lower right). Arrow indicates melting
temperature. The peptide concentration was 4.2 mM in the gel.
164
the overall network structure. In either case, the melting temperature for the AFD36 gel lies
well above physiological temperatures, which is promising for biomedical applications.
4.3.14
Guanidinium denaturation of AFD36
G.HCl denaturation studies of AFD19 and AFD36 were used to quantify the stability of
these two peptide systems. Both peptides were studied at concentrations slightly below
gelling conditions and approximately 1 pH unit bellow gelling pH (AFD19 at pH 5.0 and
AFD36 at pH 6.0). Under these conditions the peptides were expected to be assembled
and possess similar levels of charge. Figure 4-16 shows the effect of added G.HCl on
mean residue ellipticity for 250 µM AFD19 and AFD36 at 20 °C. Both peptides are seen to
be highly resistant to chemical denaturation, with a G.HCl concentration of approximately
5.3 M required for 50% denaturation of AFD36, while AFD19 required an even higher
concentration of 6.0 M. To assess the effect of temperature, a similar G.HCl denaturation
study of AFD36 was conducted at 50 °C. This showed a G.HCl concentration of
approximately 5.8 M was required for 50% denaturation at this temperature.
Thermodynamic analysis of peptide self-assembly in the presence of G.HCl utilised a
similar method to that described in Chapter 3, and was based on an approach by Fairman
et al.38 that was adapted for higher-order coiled-coil assembly. This allowed the fitting of
ECD data to monomer-oligomer models as a function of G.HCl concentration and
calculation of parameters ΔG(H2O) and m.
Assembly of monomers to higher-order oligomers (n-mers) in the absence of denaturant
was treated as a two-state system in which Kd is the equilibrium constant for disassembly
of coiled-coil n-mers to n unstructured peptide monomers (mon):
𝐾𝑑 =
[𝑚𝑜𝑛]𝑛
[𝑛−𝑚𝑒𝑟]
(4-7)
and ΔG is the associated positive free energy change:
Δ𝐺 = −𝑅𝑇𝑙𝑛𝐾𝑑
(4-8)
∆𝐺 = ∆𝐺 (𝐻2 𝑂) − 𝑚[𝐺. 𝐻𝐶𝑙]
(4-9)
The presence of G.HCl was taken to modify the free energy of peptide self-assembly in a
linear fashion:
where m acts to decrease the free energy penalty of coiled coil disassembly, thereby
promoting denaturation. A larger value of m represents a larger effect of G.HCl on peptide
folding.
165
Figure 4-16. Guanidinium chloride denaturation of AFD19 (○) and AFD36 (▲) at 20 °C. Lines
represent fits to the data obtained as described in text.
For electronic circular dichroism, the contribution of monomers and n-mers to the ellipticity
at a single wavelength (in this case, 222 nm) is given by:
[𝜃]𝑜𝑏𝑠 = [𝜃]𝑚𝑜𝑛 ×
[𝑚𝑜𝑛]
𝑛 × [𝑚𝑜𝑛]𝑛
+ [𝜃]𝑛−𝑚𝑒𝑟 ×
[𝑃]𝑡𝑜𝑡
𝐾𝑑 × [P]𝑡𝑜𝑡
(4-10)
where [P]tot is the total concentration of peptide in solution, [θ]mon is the mean residue
ellipticity of the monomer and [θ]n-mer the corresponding value for the n-mer. Values for
[θ]mon and [θ]n-mer were chosen based on observed ellipticity values at low and high G.HCl
concentration respectively. For chemical denaturation studies with G.HCl, in which the free
energy of self-assembly is different at each denaturant concentration, Equation 4-10 was
modified to give:
[𝜃]𝑜𝑏𝑠 = [𝜃]𝑚𝑜𝑛 ×
[𝑚𝑜𝑛]
+ [𝜃]𝑛−𝑚𝑒𝑟 ×
[𝑃]𝑡𝑜𝑡
𝑛 × [𝑚𝑜𝑛]𝑛
Δ𝐺 (𝐻2 𝑂)
𝑒(
−𝑅𝑇
)
−𝑚 [G.HCl]
)
−𝑅𝑇
× 𝑒(
×
1
[𝑃]𝑡𝑜𝑡
(4-11)
166
For test values of Kd and m for each value of n, values for the free monomer concentration
[mon] are obtained by numerical fitting to the mass balance equation for the total peptide
in solution, allowing assessment of the least-squares fit to the data:
[𝑃]𝑡𝑜𝑡 = [𝑚𝑜𝑛] +
Table 4-3 gives the values of ΔG
(H2O)
𝑒
(
𝑛 × [𝑚𝑜𝑛]𝑛
Δ𝐺 (𝐻2 𝑂)
)
−𝑅𝑇
×𝑒
(
−𝑚 [G.HCl]
)
−𝑅𝑇
(4-12)
, m, [θ]mon and [θ]n-mer obtained for fitting of AFD19
and AFD36 experimental data to this model. Fitting of AFD19 denaturation at 20 °C
assumed a hexameric oligomerization state based on previous results (Section 4.3.4).
Fitting of AFD36 data collected at 20 °C used pentamer, hexamer and heptamer models
as the coiled-coil cross-section, as well as a dodecamer based on end-to-end association
of two hexamers. This allowed the effect of assumed oligomerization state on determined
values to be assessed. As similar results on a per monomer basis were seen for each
model, fitting of AFD36 denaturation at 50 °C used only a hexamer model for comparison
to data collected at 20 °C.
Modelling of AFD19 denaturation using a hexamer model gave a ΔG(H2O) value of 9.5 kcal
mol-1 per monomer. This is larger than the 5.9 kcal mol-1 per monomer previously
determined in Section 4.3.4 via a dilution method. Notably, in the dilution study the
monomeric peptide retained high helix content. In this experiment, G.HCl denatured the
oligomeric AFD19 to an unstructured monomer, as indicated by the low helical content at
high denaturant concentration. The difference of 3.6 kcal mol-1 likely represents at least in
part the free energy of unfolding of a monomeric α-helix to a random coil, which has also
been studied in model polyalanine systems.87 However, it is also possible that effects
relating to ionic strength may interfere with modelling G.HCl denaturation and cause an
overestimation of α-helix stability. Indeed, an increase in helicity at up to 2 M G.HCl is
observed for both AFD19 and AFD36, with AFD36 samples showing induced gelation in
the presence of 2-6 M G.HCl. This suggests an initial stabilisation effect of added salt,
likely due to increased strength of hydrophobic interactions. Above 2 M G.HCl,
denaturation of secondary structure occurs in both peptides.
167
Table 4-3. Fitting parameters for modelling G.HCl denaturation of AFD19 and AFD36
Peptide
pH
Ta
ΔG(H2O)b
mc
nd
[θ]mone
[θ]n-mere
AFD19
5.0
20
9.5
0.88
6
0.4
-28.5
AFD36
6.0
20
9.1
0.95
5
0.9
-28.0
AFD36
6.0
20
9.3
0.95
6
0.9
-28.0
AFD36
6.0
20
9.2
0.92
7
0.9
-28.0
AFD36
6.0
20
9.5
0.90
12
0.9
-28.0
AFD36
6.0
50
10.6
1.04
6
0.0
-27.0
a
°C
b
kcal mol-1 per monomer
c
kcal mol-1 (M G.HCl)-1 per monomer
d
oligomer state assumed in fitting
e
103 deg cm2 dmol-1 at 222 nm
The stability of AFD19 is high compared to values reported for peptide coiled coils also
determined using G.HCl denaturation.38, 88 Values of 3.4-5.6 kcal mol-1 per monomer were
reported for a series of dimeric coiled coils of three to four heptads in length.88 A tetrameric
coiled coil of the same length as AFD19 was reported to have a stability of 3.9 kcal mol-1
per monomer,38 while a related tetrameric coiled coil was reported to have a stability of 5.6
kcal mol-1 per monomer as determined using thermal denaturation.89 The high stability of
AFD19 assembly likely reflects both the selection of amino acids with high helix propensity
during peptide design, as well as a strong hydrophobic driving force for coiled-coil
assembly. The per-helix value of ΔG for AFD19 approaches the value determined for the
extremely thermodynamically stable Het2-6 in Chapter 2, although in that case,
denaturation was conducted at 90 °C.
The stability of AFD36 was found to vary only slightly, from 9.1 to 9.5 kcal mol-1 per
monomer when modelled using pentamer, hexamer, heptamer and dodecamer models,
indicating that the per monomer stability values are relatively insensitive to assumed
oligomerization state. This enabled comparison of determined stability values with AFD19
and previously reported systems without explicitly determining the coiled-coil
oligomerization state. As AFD19 was previously modelled as a hexamer, the hexamer
model state was used to compare self-assembly across different conditions. AFD36 at pH
168
6.0 has a stability of approximately 9.3 kcal mol-1 per monomer when analysed using a
hexamer model. This is slightly lower than AFD19 in a similar charge state at pH 5.0, but
still high compared to previously reported coiled coils of similar peptide length.38, 88, 89
The lower stability of AFD36 is surprising, given that the serine-to-lysine mutation should
favour helix formation due to the higher helix propensity of lysine.23 However, the
difference is small (0.2 kcal mol-1 per monomer) and may relate to stabilisation of the
unfolded state by interactions involving lysine, or slight differences in the pKa values of
amino acid sidechains between the two peptides, due to interactions between charged
groups or changes in side chain burial,90 leading to differences in the actual molecular
charge. Raising the temperature from 20 to 50 °C increased the stability of AFD36, with an
increase in G.HCl concentration required for 50% denaturation of secondary structure from
5.3 to 5.8 M. This corresponds to an increase in stability from 9.3 to 10.6 kcal mol-1 per
monomer for AFD36 on increasing the temperature form 20 to 50 °C. This is consistent
with the strengthening of hydrophobic association with temperature,52 and highlights the
key role of hydrophobic interactions in stabilising the coiled coil.
As well as the ΔG of unfolding, the m value, or the dependence of free energy of unfolding
on denaturant concentration may be determined from the fitting of chemical denaturation
data. Similar to the stability values, the per-monomer m value varies little with assumed
oligomerization state, with values from 0.90-0.95 kcal mol-1 (M G.HCl)-1 per monomer. The
m values reported here are larger than those previously reported for designed coiled coils
of 0.68 kcal mol-1 (M G.HCl)-1 per monomer for a 28-residue tetramer coiled coil38 and
between 0.56 and 0.81 kcal mol-1 (M G.HCl)-1 per monomer for a series of 21-residue
dimer coiled coils.88 In fact the per monomer m value for the 21-residue AFD36 is closer to
that reported for a 35-residue tetrameric coiled coil.38 The m values are also larger than
reported for native proteins of similar molecular weight,91 and on a per helix basis
approach the anomalously large value determined for Het2-6 in Chapter 2. As discussed in
Chapters 2 and 3, these values are also used to analyse the change in ASA on
unfolding.91 However, similar to Het2-6, applying literature correlations to the large
calculated m values for AFD19 and AFD36 gives a value for the change in ASA wich is not
physically meaningful. As discussed in Chapter 2, the correlations across different proteins
are relatively scattered91 and outlier values for m have previously been observed.92, 93
Moreover, the correlations between m values and change in ASA were determined from
series of reference proteins, and how these values may translate to peptide systems has
not as yet been reported. Modelling of AFD36 denaturation at 50 °C gives an even larger
169
m value than that already discussed. However as discussed in Chapter 2, the m values for
reference proteins have been determined at room temperature and extrapolation of this
relationship to elevated temperatures relies on assumptions about protein-denaturant
interactions that may not be justifiable.94
4.4 Conclusions
This chapter describes characterisation of the self-assembly of α-helical peptide AFD19
that, while designed as part of ongoing work on peptide surfactants, was serendipitously
observed to form hydrogels as a function of solution pH. Peptide hydrogels formed at low
weight percentages and were self-healing following mechanical disruption. Investigation of
peptide self-assembly via ECD indicated AFD19 to form a higher-order coiled coil, and
electron microscopy showed the formation of a network of unbundled fibrils with diameters
consistent with this proposal. Self-assembly as a function of pH was consistent with coiledcoil fibrils forming physical cross-links resulting in gel formation at a molecular charge of
approximately ±1. Precipitation was observed at pH values where molecular charge
approached zero, while solution pH values where the molecular charge was high gave
free-flowing solutions. Comparison with reported fibril-forming peptides allowed proposal of
a fibril-formation pathway involving permissive offsetting of helices due to self-similar
heptad repeats, thereby forming end-to-end overlaps and staggered coiled-coil fibrils.
While AFD19 displayed gel formation at non-physiological pH, targeted redesign of this
peptide based on the proposed model of self-assembly gave peptide AFD36, which forms
hydrogels under conditions of physiological pH and salt, making it potentially applicable in
biomedical applications. Characterisation of AFD36 assembly using SAXS showed fibril
formation similar to that shown for AFD19 using electron microscopy. Hydrogels formed by
both peptides were extremely thermostable, and chemical denaturation experiments
showed the constituent coiled coils to possess high thermodynamic stability. The use of
bicarbonate as a transient buffer enabled the preparation of homogenous materials. These
peptide hydrogels are materials of interest for applications such as tissue engineering and
drug delivery, and are further investigated in Chapter 5.
170
4.5 References
1.
Van Vlierberghe, S.; Dubruel, P.; Schacht, E. Biopolymer-Based Hydrogels as
Scaffolds for Tissue Engineering Applications: A Review. Biomacromolecules 2011, 12,
1387-408.
2.
Hoffman, A. S. Hydrogels for Biomedical Applications. Adv. Drug Delivery. Rev.
2002, 54, 3-12.
3.
Vashist, A.; Vashist, A.; Gupta, Y. K.; Ahmad, S. Recent Advances in Hydrogel
Based Drug Delivery Systems for the Human Body. J. Mater. Chem. B 2014, 2, 147-166.
4.
Xie, Y.; Rizzi, S. C.; Dawson, R.; Lynam, E.; Richards, S.; Leavesley, D. I.; Upton,
Z. Development of a Three-Dimensional Human Skin Equivalent Wound Model for
Investigating Novel Wound Healing Therapies. Tissue Eng., Part C 2010, 16, 1111-1123.
5.
Li, H. N.; Yang, J.; Hu, X. N.; Liang, J.; Fan, Y. J.; Zhang, X. D. Superabsorbent
Polysaccharide Hydrogels Based on Pullulan Derivate as Antibacterial Release Wound
Dressing. J. Biomed. Mater. Res., Part A 2011, 98A, 31-39.
6.
Matson, J. B.; Stupp, S. I. Self-Assembling Peptide Scaffolds for Regenerative
Medicine. Chem. Commun. 2012, 48, 26-33.
7.
Maude, S.; Ingham, E.; Aggeli, A. Biomimetic Self-Assembling Peptides as
Scaffolds for Soft Tissue Engineering. Nanomed. 2013, 8, 823-847.
8.
Holmes, T. C.; de Lacalle, S.; Su, X.; Liu, G. S.; Rich, A.; Zhang, S. G. Extensive
Neurite Outgrowth and Active Synapse Formation on Self-Assembling Peptide Scaffolds.
Proc. Natl. Acad. Sci. U. S. A. 2000, 97, 6728-6733.
9.
Zhang, S. G.; Holmes, T. C.; Dipersio, C. M.; Hynes, R. O.; Su, X.; Rich, A. SelfComplementary Oligopeptide Matrices Support Mammalian-Cell Attachment. Biomaterials
1995, 16, 1385-1393.
10.
Collier, J. H.; Messersmith, P. B. Enzymatic Modification of Self-Assembled Peptide
Structures with Tissue Transglutaminase. Bioconjug. Chem. 2003, 14, 748-755.
11.
Aggeli, A.; Bell, M.; Carrick, L. M.; Fishwick, C. W. G.; Harding, R.; Mawer, P. J.;
Radford, S. E.; Strong, A. E.; Boden, N. Ph as a Trigger of Peptide Beta-Sheet SelfAssembly and Reversible Switching between Nematic and Isotropic Phases. J. Am. Chem.
Soc. 2003, 125, 9619-9628.
12.
Schneider, J. P.; Pochan, D. J.; Ozbas, B.; Rajagopal, K.; Pakstis, L.; Kretsinger, J.
Responsive Hydrogels from the Intramolecular Folding and Self-Assembly of a Designed
Peptide. J. Am. Chem. Soc. 2002, 124, 15030-15037.
13.
Westermark, P.; Lundmark, K.; Westermark, G. T. Fibrils from Designed NonAmyloid-Related Synthetic Peptides Induce Aa-Amyloidosis During Inflammation in an
Animal Model. PLoS One 2009, 4, e6041.
14.
Banwell, E. F.; Abelardo, E. S.; Adams, D. J.; Birchall, M. A.; Corrigan, A.; Donald,
A. M.; Kirkland, M.; Serpell, L. C.; Butler, M. F.; Woolfson, D. N. Rational Design and
Application of Responsive Alpha-Helical Peptide Hydrogels. Nat. Mater. 2009, 8, 596-600.
15.
Dong, H.; Paramonov, S. E.; Hartgerink, J. D. Self-Assembly of Alpha-Helical
Coiled Coil Nanofibers. J. Am. Chem. Soc. 2008, 130, 13691-13695.
16.
Potekhin, S. A.; Melnik, T. N.; Popov, V.; Lanina, N. F.; Vazina, A. A.; Rigler, P.;
Verdini, A. S.; Corradin, G.; Kajava, A. V. De Novo Design of Fibrils Made of Short AlphaHelical Coiled Coil Peptides. Chem. Biol. 2001, 8, 1025-1032.
17.
Dexter, A. F. Interfacial and Emulsifying Properties of Designed Β-Strand Peptides.
Langmuir 2010, 26, 17997-18007.
18.
Mastronarde, D. N. Automated Electron Microscope Tomography Using Robust
Prediction of Specimen Movements. J. Struct. Biol. 2005, 152, 36-51.
19.
Dreiss, C. A.; Jack, K. S.; Parker, A. P. On the Absolute Calibration of Bench-Top
Small-Angle X-Ray Scattering Instruments: A Comparison of Different Standard Methods.
J. Appl. Crystallogr. 2006, 39, 32-38.
171
20.
Dexter, A. F.; Malcolm, A. S.; Middelberg, A. P. J. Reversible Active Switching of
the Mechanical Properties of a Peptide Film at a Fluid-Fluid Interface. Nat. Mater. 2006, 5,
502-506.
21.
Dexter, A. F.; Middelberg, A. P. J. Switchable Peptide Surfactants with Designed
Metal Binding Capacity. J. Phys. Chem. C 2007, 111, 10484-10492.
22.
Malcolm, A. S.; Dexter, A. F.; Katakdhond, J. A.; Karakashev, S. I.; Nguyen, A. V.;
Middelberg, A. P. J. Tuneable Control of Interfacial Rheology and Emulsion Coalescence.
ChemPhysChem 2009, 10, 778-781.
23.
O'Neil, K. T.; Degrado, W. F. A Thermodynamic Scale for the Helix-Forming
Tendencies of the Commonly Occurring Amino Acids. Science 1990, 250, 646-651.
24.
Blaber, M.; Zhang, X. J.; Matthews, B. W. Structural Basis of Amino-Acid AlphaHelix Propensity. Science 1993, 260, 1637-1640.
25.
Reubsaet, J. L. E.; Beijnen, J. H.; Bult, A.; van Maanen, R. J.; Marchal, J. A. D.;
Underberg, W. J. M. Analytical Techniques Used to Study the Degradation of Proteins and
Peptides: Chemical Instability. J. Pharm. Biomed. Anal. 1998, 17, 955-978.
26.
Guruprasad, K.; Reddy, B. V. B.; Pandit, M. W. Correlation between Stability of a
Protein and Its Dipeptide Composition - a Novel Approach for Predicting in Vivo Stability of
a Protein from Its Primary Sequence. Protein Eng. 1990, 4, 155-161.
27.
Kyte, J.; Doolittle, R. F. A Simple Method for Displaying the Hydropathic Character
of a Protein. J. Mol. Biol. 1982, 157, 105-132.
28.
Scholtz, J. M.; Qian, H.; York, E. J.; Stewart, J. M.; Baldwin, R. L. Parameters of
Helix-Coil Transition Theory for Alanine-Based Peptides of Varying Chain Lengths in
Water. Biopolymers 1991, 31, 1463-1470.
29.
Apostolovic, B.; Danial, M.; Klok, H. A. Coiled Coils: Attractive Protein Folding
Motifs for the Fabrication of Self-Assembled, Responsive and Bioactive Materials. Chem.
Soc. Rev. 2010, 39, 3541-3575.
30.
Armstrong, C. T.; Boyle, A. L.; Bromley, E. H. C.; Mahmoud, Z. N.; Smith, L.;
Thomson, A. R.; Woolfson, D. N. Rational Design of Peptide-Based Building Blocks for
Nanoscience and Synthetic Biology. Faraday Discuss. 2009, 143, 305-17.
31.
Marti, D. N.; Jelesarov, I.; Bosshard, H. R. Interhelical Ion Pairing in Coiled Coils:
Solution Structure of a Heterodimeric Leucine Zipper and Determination of Pk(a) Values of
Glu Side Chains. Biochemistry 2000, 39, 12804-12818.
32.
Woolfson, D. N.; Bartlett, G. J.; Bruning, M.; Thomson, A. R. New Currency for Old
Rope: From Coiled-Coil Assemblies to Alpha-Helical Barrels. Curr. Opin. Struct. Biol.
2012, 22, 432-441.
33.
Liu, J.; Zheng, Q.; Deng, Y. Q.; Cheng, C. S.; Kallenbach, N. R.; Lu, M. A SevenHelix Coiled Coil. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 15457-15462.
34.
Zaccai, N. R.; Chi, B.; Thomson, A. R.; Boyle, A. L.; Bartlett, G. J.; Bruning, M.;
Linden, N.; Sessions, R. B.; Booth, P. J.; Brady, R. L.; Woolfson, D. N. A De Novo Peptide
Hexamer with a Mutable Channel. Nat. Chem. Biol. 2011, 7, 935-941.
35.
Frost, D. W. H.; Yip, C. M.; Chakrabartty, A. Reversible Assembly of Helical
Filaments by De Novo Designed Minimalist Peptides. Biopolymers 2005, 80, 26-33.
36.
Ryadnov, M. G.; Ceyhan, B.; Niemeyer, C. M.; Woolfson, D. N. "Belt and Braces": A
Peptide-Based Linker System of De Novo Design. J. Am. Chem. Soc. 2003, 125, 93889394.
37.
Myers, J. K.; Pace, C. N.; Scholtz, J. M. Helix Propensities Are Identical in Proteins
and Peptides. Biochemistry 1997, 36, 10923-10929.
38.
Fairman, R.; Chao, H. G.; Mueller, L.; Lavoie, T. B.; Shen, L. Y.; Novotny, J.;
Matsueda, G. R. Characterization of a New Four-Chain Coiled-Coil: Influence of Chain
Length on Stability. Protein Sci. 1995, 4, 1457-1469.
172
39.
Liu, J.; Zheng, Q.; Deng, Y. Q.; Kallenbach, N. R.; Lu, M. Conformational Transition
between Four and Five-Stranded Phenylalanine Zippers Determined by a Local Packing
Interaction. J. Mol. Biol. 2006, 361, 168-179.
40.
Xu, C. F.; Liu, R.; Mehta, A. K.; Guerrero-Ferreira, R. C.; Wright, E. R.; DuninHorkawicz, S.; Morris, K.; Serpell, L. C.; Zuo, X. B.; Wall, J. S.; Conticello, V. P. Rational
Design of Helical Nanotubes from Self-Assembly of Coiled-Coil Lock Washers. J. Am.
Chem. Soc. 2013, 135, 15565-15578.
41.
Aggeli, A.; Bell, M.; Boden, N.; Keen, J. N.; Knowles, P. F.; McLeish, T. C. B.;
Pitkeathly, M.; Radford, S. E. Responsive Gels Formed by the Spontaneous SelfAssembly of Peptides into Polymeric Beta-Sheet Tapes. Nature 1997, 386, 259-262.
42.
Carrick, L. M.; Aggeli, A.; Boden, N.; Fisher, J.; Ingham, E.; Waigh, T. A. Effect of
Ionic Strength on the Self-Assembly, Morphology and Gelation of Ph Responsive BetaSheet Tape-Forming Peptides. Tetrahedron 2007, 63, 7457-7467.
43.
Rajagopal, K.; Lamm, M. S.; Haines-Butterick, L. A.; Pochan, D. J.; Schneider, J. P.
Tuning the Ph Responsiveness of Beta-Hairpin Peptide Folding, Self-Assembly, and
Hydrogel Material Formation. Biomacromolecules 2009, 10, 2619-2625.
44.
Nagayasu, A.; Yokoi, H.; Minaguchi, J. A.; Hosaka, Y. Z.; Ueda, H.; Takehana, K.
Efficacy of Self-Assembled Hydrogels Composed of Positively or Negatively Charged
Peptides as Scaffolds for Cell Culture. J. Biomater. Appl. 2012, 26, 651-665.
45.
Caplan, M. R.; Schwartzfarb, E. M.; Zhang, S. G.; Kamm, R. D.; Lauffenburger, D.
A. Control of Self-Assembling Oligopeptide Matrix Formation through Systematic Variation
of Amino Acid Sequence. Biomaterials 2002, 23, 219-227.
46.
Measey, T. J.; Schweitzer-Stenner, R.; Sa, V.; Kornev, K. Anomalous
Conformational Instability and Hydrogel Formation of a Cationic Class of Self-Assembling
Oligopeptides. Macromolecules 2010, 43, 7800-7806.
47.
Aulisa, L.; Dong, H.; Hartgerink, J. D. Self-Assembly of Multidomain Peptides:
Sequence Variation Allows Control over Cross-Linking and Viscoelasticity.
Biomacromolecules 2009, 10, 2694-2698.
48.
Kisiday, J.; Jin, M.; Kurz, B.; Hung, H.; Semino, C.; Zhang, S.; Grodzinsky, A. J.
Self-Assembling Peptide Hydrogel Fosters Chondrocyte Extracellular Matrix Production
and Cell Division: Implications for Cartilage Tissue Repair. Proc. Natl. Acad. Sci. U. S. A.
2002, 99, 9996-10001.
49.
Caplan, M. R.; Moore, P. N.; Zhang, S. G.; Kamm, R. D.; Lauffenburger, D. A. SelfAssembly of a Beta-Sheet Protein Governed by Relief of Electrostatic Repulsion Relative
to Van Der Waals Attraction. Biomacromolecules 2000, 1, 627-631.
50.
Maude, S.; Miles, D. E.; Felton, S. H.; Ingram, J.; Carrick, L. M.; Wilcox, R. K.;
Ingham, E.; Aggeli, A. De Novo Designed Positively Charged Tape-Forming Peptides:
Self-Assembly and Gelation in Physiological Solutions and Their Evaluation as 3d Matrices
for Cell Growth. Soft Matter 2011, 7, 8085-8099.
51.
Malashkevich, V. N.; Kammerer, R. A.; Efimov, V. P.; Schulthess, T.; Engel, J. The
Crystal Structure of a Five-Stranded Coiled Coil in Comp: A Prototype Ion Channel?
Science 1996, 274, 761-765.
52.
Privalov, P. L.; Gill, S. J. Stability of Protein-Structure and Hydrophobic Interaction.
Adv. Protein Chem. 1988, 39, 191-234.
53.
Schellman, J. A. Temperature, Stability, and the Hydrophobic Interaction. Biophys.
J. 1997, 73, 2960-2964.
54.
Cordier, F.; Grzesiek, S. Temperature-Dependence of Protein Hydrogen Bond
Properties as Studied by High-Resolution Nmr. J. Mol. Biol. 2002, 317, 739-752.
55.
Grzesiek, S.; Cordier, F.; Jaravine, V.; Barfield, M. Insights into Biomolecular
Hydrogen Bonds from Hydrogen Bond Scalar Couplings. Prog. Nucl. Magn. Reson.
Spectrosc. 2004, 45, 275-300.
173
56.
Papapostolou, D.; Smith, A. M.; Atkins, E. D. T.; Oliver, S. J.; Ryadnov, M. G.;
Serpell, L. C.; Woolfson, D. N. Engineering Nanoscale Order into a Designed Protein
Fiber. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 10853-10858.
57.
Vu, C.; Robblee, J.; Werner, K. M.; Fairman, R. Effects of Charged Amino Acids at
B and C Heptad Positions on Specificity and Stability of Four-Chain Coiled Coils. Protein
Sci. 2001, 10, 631-637.
58.
Ryadnov, M. G.; Woolfson, D. N. Introducing Branches into a Self-Assembling
Peptide Fiber. Angew. Chem., Int. Ed. 2003, 42, 3021-3023.
59.
Ryadnov, M. G.; Woolfson, D. N. Engineering the Morphology of a Selfassembling
Protein Fibre. Nat. Mater. 2003, 2, 329-332.
60.
Pandya, M. J.; Spooner, G. M.; Sunde, M.; Thorpe, J. R.; Rodger, A.; Woolfson, D.
N. Sticky-End Assembly of a Designed Peptide Fiber Provides Insight into Protein
Fibrillogenesis. Biochemistry 2000, 39, 8728-8734.
61.
Gribbon, C.; Channon, K. J.; Zhang, W. J.; Banwell, E. F.; Bromley, E. H. C.;
Chaudhuri, J. B.; Oreffo, R. O. C.; Woolfson, D. N. Magicwand: A Single, Designed
Peptide That Assembles to Stable, Ordered Alpha-Helical Fibers. Biochemistry 2008, 47,
10365-10371.
62.
Papapostolou, D.; Bromley, E. H. C.; Bano, C.; Woolfson, D. N. Electrostatic
Control of Thickness and Stiffness in a Designed Protein Fiber. J. Am. Chem. Soc. 2008,
130, 5124-5130.
63.
Smith, A. M.; Banwell, E. F.; Edwards, W. R.; Pandya, M. J.; Woolfson, D. N.
Engineering Increased Stability into Self-Assembled Protein Fibers. Adv. Funct. Mater.
2006, 16, 1022-1030.
64.
Zimenkov, Y.; Conticello, V. P.; Guo, L.; Thiyagarajan, P. Rational Design of a
Nanoscale Helical Scaffold Derived from Self-Assembly of a Dimeric Coiled Coil Motif.
Tetrahedron 2004, 60, 7237-7246.
65.
Zimenkov, Y.; Dublin, S. N.; Ni, R.; Tu, R. S.; Breedveld, V.; Apkarian, R. P.;
Conticello, V. P. Rational Design of a Reversible Ph-Responsive Switch for Peptide SelfAssembly. J. Am. Chem. Soc. 2006, 128, 6770-6771.
66.
Dublin, S. N.; Conticello, V. P. Design of a Selective Metal Ion Switch for SelfAssembly of Peptide-Based Fibrils. J. Am. Chem. Soc. 2008, 130, 49-+.
67.
Tsang, B. P.; Bretscher, H. S.; Kokona, B.; Manning, R. S.; Fairman, R.
Thermodynamic Analysis of Self-Assembly in Coiled-Coil Biomaterials. Biochemistry 2011,
50, 8548-8558.
68.
Wagner, D. E.; Phillips, C. L.; Ali, W. M.; Nybakken, G. E.; Crawford, E. D.; Schwab,
A. D.; Smith, W. F.; Fairman, R. Toward the Development of Peptide Nanofilaments and
Nanoropes as Smart Materials. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 12656-12661.
69.
Kojima, S.; Kuriki, Y.; Yoshida, T.; Yazaki, K.; Miura, K. Fibril Formation by an
Amphipathic Alpha-Helix-Forming Polypeptide Produced by Gene Engineering. Proc. Jpn.
Acad. Ser. B Phys. Biol. Sci. 1997, 73, 7-11.
70.
Takei, T.; Hasegawa, K.; Imada, K.; Namba, K.; Tsumoto, K.; Kuriki, Y.; Yoshino,
M.; Yazaki, K.; Kojima, S.; Takei, T.; Ueda, T.; Miura, K. Effects of Chain Length of an
Amphipathic Polypeptide Carrying the Repeated Amino Acid Sequence (Letlaka)(N) on
Alpha-Helix and Fibrous Assembly Formation. Biochemistry 2013, 52, 2810-2820.
71.
Melnik, T. N.; Villard, V.; Vasiliev, V.; Corradin, G.; Kajava, A. V.; Potekhin, S. A.
Shift of Fibril-Forming Ability of the Designed Alpha-Helical Coiled-Coil Peptides into the
Physiological Ph Region. Protein Eng. 2003, 16, 1125-1130.
72.
Liu, J.; Deng, Y. Q.; Zheng, Q.; Cheng, C. S.; Kallenbach, N. R.; Lu, M. A Parallel
Coiled-Coil Tetramer with Offset Helices. Biochemistry 2006, 45, 15224-15231.
73.
Sharp, T. H.; Bruning, M.; Mantell, J.; Sessions, R. B.; Thomson, A. R.; Zaccai, N.
R.; Brady, R. L.; Verkade, P.; Woolfson, D. N. Cryo-Transmission Electron Microscopy
174
Structure of a Gigadalton Peptide Fiber of De Novo Design. Proc. Natl. Acad. Sci. U. S. A.
2012, 109, 13266-13271.
74.
Ogihara, N. L.; Weiss, M. S.; Degrado, W. F.; Eisenberg, D. The Crystal Structure
of the Designed Trimeric Coiled Coil Coil-V(a)L(D): Implications for Engineering Crystals
and Supramolecular Assemblies. Protein Sci. 1997, 6, 80-88.
75.
Prive, G. G.; Anderson, D. H.; Wesson, L.; Cascio, D.; Eisenberg, D. Packed
Protein Bilayers in the 0.90 Angstrom Resolution Structure of a Designed Alpha Helical
Bundle. Protein Sci. 1999, 8, 1400-1409.
76.
Harbury, P. B.; Zhang, T.; Kim, P. S.; Alber, T. A Switch between Two-, Three-, and
Four-Stranded Coiled Coils in Gcn4 Leucine Zipper Mutants. Science 1993, 262, 14011407.
77.
Opella, S. J.; Marassi, F. M.; Gesell, J. J.; Valente, A. P.; Kim, Y.; Oblatt-Montal, M.;
Montal, M. Structures of the M2 Channel-Lining Segments from Nicotinic Acetylcholine
and Nmda Receptors by Nmr Spectroscopy. Nat. Struct. Biol. 1999, 6, 374-379.
78.
Lu, Y.; Weers, B.; Stellwagen, N. C. DNA Persistence Length Revisited.
Biopolymers 2002, 61, 261-275.
79.
vandenAkker, C. C.; Engel, M. F. M.; Velikov, K. P.; Bonn, M.; Koenderink, G. H.
Morphology and Persistence Length of Amyloid Fibrils Are Correlated to Peptide Molecular
Structure. J. Am. Chem. Soc. 133, 18030-18033.
80.
Aggeli, A.; Nyrkova, I. A.; Bell, M.; Harding, R.; Carrick, L.; McLeish, T. C. B.;
Semenov, A. N.; Boden, N. Hierarchical Self-Assembly of Chiral Rod-Like Molecules as a
Model for Peptide Beta-Sheet Tapes, Ribbons, Fibrils, and Fibers. Proc. Natl. Acad. Sci.
U. S. A. 2001, 98, 11857-11862.
81.
Adams, D. J.; Butler, M. F.; Frith, W. J.; Kirkland, M.; Mullen, L.; Sanderson, P. A
New Method for Maintaining Homogeneity During Liquid-Hydrogel Transitions Using Low
Molecular Weight Hydrogelators. Soft Matter 2009, 5, 1856-1862.
82.
Raeburn, J.; McDonald, T. O.; Adams, D. J. Dipeptide Hydrogelation Triggered Via
Ultraviolet Light. Chem. Commun. 2012, 48, 9355-9357.
83.
Xu, X. D.; Lin, B. B.; Feng, J.; Wang, Y.; Cheng, S. X.; Zhang, X. Z.; Zhuo, R. X.
Biological Glucose Metabolism Regulated Peptide Self-Assembly as a Simple Visual
Biosensor for Glucose Detection. Macromol. Rapid Commun. 2012, 33, 426-431.
84.
Grigoriou, S.; Johnson, E. K.; Chen, L.; Adams, D. J.; James, T. D.; Cameron, P. J.
Dipeptide Hydrogel Formation Triggered by Boronic Acid-Sugar Recognition. Soft Matter
2012, 8, 6788-6791.
85.
Wang, H.; Yang, Z.; Adams, D. J. Controlling Peptidebased Hydrogelation.
Materials Today 2012, 15, 500-507.
86.
Pochan, D. J.; Schneider, J. P.; Kretsinger, J.; Ozbas, B.; Rajagopal, K.; Haines, L.
Thermally Reversible Hydrogels Via Intramolecular Folding and Consequent SelfAssembly of a De Novo Designed Peptide. J. Am. Chem. Soc. 2003, 125, 11802-11803.
87.
Jas, G. S.; Kuczera, K. Equilibrium Structure and Folding of a Helix-Forming
Peptide: Circular Dichroism Measurements and Replica-Exchange Molecular Dynamics
Simulations. Biophys. J. 2004, 87, 3786-3798.
88.
Litowski, J. R.; Hodges, R. S. Designing Heterodimeric Two-Stranded Alpha-Helical
Coiled-Coils - Effects of Hydrophobicity and Alpha-Helical Propensity on Protein Folding,
Stability, and Specificity. J. Biol. Chem. 2002, 277, 37272-37279.
89.
Fairman, R.; Chao, H. G.; Lavoie, T. B.; Villafranca, J. J.; Matsueda, G. R.;
Novotny, J. Design of Heterotetrameric Coiled Coils: Evidence for Increased Stabilization
by Glu(-)-Lys(+) Ion Pair Interactions. Biochemistry 1996, 35, 2824-2829.
90.
Harms, M. J.; Castañeda, C. A.; Schlessman, J. L.; Sue, G. R.; Isom, D. G.;
Cannon, B. R.; García-Moreno E, B. The Pka Values of Acidic and Basic Residues Buried
at the Same Internal Location in a Protein Are Governed by Different Factors. J. Mol. Biol.
2009, 389, 34-47.
175
91.
Myers, J. K.; Pace, C. N.; Scholtz, J. M. Denaturant M-Values and Heat Capacity
Changes - Relation to Changes in Accessible Surface Areas of Protein Unfolding. Protein
Sci. 1995, 4, 2138-2148.
92.
Shortle, D.; Meeker, A. K. Mutant Forms of Staphylococcal Nuclease with Altered
Patterns of Guanidine Hydrochloride and Urea Denaturation. Proteins: Struct., Funct.,
Genet. 1986, 1, 81-89.
93.
Zhang, T.; Bertelsen, E.; Benvegnu, D.; Alber, T. Circular Permutation of T4Lysozyme. Biochemistry 1993, 32, 12311-12318.
94.
Zweifel, M. E.; Barrick, D. Relationships between the Temperature Dependence of
Solvent Denaturation and the Denaturant Dependence of Protein Stability Curves.
Biophys. Chem. 2002, 101, 221-237.
176
5 α-Helical peptide hydrogels for mammalian cell growth and
drug delivery
5.1 Introduction
Chapter 4 detailed characterisation of the self-assembly of peptide AFD19, which formed a
hydrogel at weight fractions below 0.1%. Gelation occurred as a function of solution pH,
with gel formation observed at a peptide charge of ±1 and precipitation occurring where
charge approached zero. While AFD19 precipitated at physiological pH, redesign of this
sequence gave the peptide AFD36, whsich was capable of gel formation under
physiological conditions of pH and salt, giving this peptide hydrogel potential utility in
tissue engineering and wound healing applications.1, 2 Only one α-helical peptide hydrogel
has previously been characterised for its ability to support cell growth,3 and it was
therefore of interest to assess AFD36 hydrogels as cellular scaffold materials.
Characterisation of AFD19 and AFD36 showed the peptides to assemble into coiled-coil
fibrils of approximately 5-7 helices in cross-section. As discussed in Chapter 1, higherorder coiled coils such as these are predicted to possess a hollow internal channel; as
observed in heptamer,4, 5 hexamer,6 pentamer7 and tetramer7, 8 structures. The
hydrophobically lined channels of coiled coils may potentially be used for encapsulation
and delivery of hydrophobic drugs.9 Loading of such molecules into the coiled-coil fibrils of
a hydrogel may allow for slow release of therapeutics from the hydrogel material in
applications such as wound healing and drug delivery.
This chapter reports investigation of hydrogels formed by AFD36 and the subsequent
peptide design AFD49 as cellular scaffolds and drug delivery vehicles. Rheological
characterisation showed AFD36 hydrogels to possess soft-solid properties, similar in
modulus to soft tissues of the body. Treatment of the peptide to remove chemical
contaminants was required for peptide hydrogels to support the growth of mouse fibroblast
NIH/3T3 cells. Fibroblasts proliferated on AFD36 hydrogels at levels similar to tissue
culture polystyrene controls. As fibroblasts proliferated to higher density, some detachment
of dense cellular aggregates from the hydrogel surfaces occurred.
The peptide AFD49 was then designed based on the self-assembly model proposed in
Chapter 4. AFD49 underwent hydrogel formation and supported fibroblast growth at similar
levels to AFD36. The effects of hydrogel equilibration prior to cell seeding and
177
requirements for serum components for cellular attachment to the hydrogels were also
investigated. AFD49 hydrogels also supported the proliferation of embedded fibroblasts
over several weeks, demonstrating these peptide hydrogels potential as 3D tissue
engineering matrices. The growth of induced pluripotent stem cells (iPSCs) was
characterised on AFD49 hydrogels to test the applicability of peptide gels as substrates for
alternate cell types. However non-functionalized AFD49 hydrogels showed poor iPSC
attachment and required coating with Matrigel to achieve iPSC attachment and
proliferation. Finally, the ability of peptide hydrogels to incorporate and release the
hydrophobic drug all-trans retinoic acid was demonstrated. AFD36 and AFD49 peptide
hydrogels are promising candidates for future tissue engineering, wound healing and drug
delivery applications.
5.2 Experimental
5.2.1 Materials
All reagents used were of the highest grade available. Water was purified using an Elga
Purelab Classic and had a resistivity of >18.2 MΩ.cm. Sodium bicarbonate (≥99.5%) was
purchased from Sigma-Aldrich. Peptides AFD36 (acetate salt, Ac-LKELAKV LHELAKL
VKEALHA-CONH2, FW 2395) and AFD49 (acetate salt, Ac-LKELAKL LHELAKL
LHELAKL-CONH2, FW 2465) were synthesised and purified by Biomatik (Wilmington, DE).
The final purity was >95% in each case.
5.2.2 Gel rheology
Rheological measurements were carried out at 23 °C on a Thermo Scientific Haake MARS
III stress-controlled rheometer using parallel plates (35 mm titanium) at a gap of 500 μm;
the top plate had emery paper attached to limit slip. To prepare a gel, 500 μL 2.8 mM
AFD36 pH 3 was mixed with 500 μL 25 mM NaHCO3, 2 × Dulbecco’s Modified Eagle
Medium (DMEM; Gibco, Life Technologies, Carlsbad, CA) and loaded onto the lower plate
using a micropipette, after which the upper plate was lowered. To limit evaporation, drops
of water were placed in the solvent trap around the outer edge of the lower plate and a
hood was lowered to cover the plates; the water drops were not in contact with the sample.
A transient oscillatory test was used to capture development of structure as the pH rose
with loss of CO2 from the sample, permitting gelation. The angular frequency was set to
6.28 rad/s for quick measurements and the oscillatory shear stress was set to 0.1 Pa,
which gave strain less than 0.005. Relative to the final gel, an oscillatory shear stress of
178
0.1 Pa was well within the linear viscoelastic region. The oscillation time test was run for
3600 seconds and measurements commenced within 5 minutes of peptide mixing.
An oscillatory stress sweep at an angular frequency of 6.28 rad/s was then run on the
same sample, to determine the linear viscoelastic region. Using an oscillatory shear stress
of 0.1 Pa, an oscillatory frequency sweep was also run from 0.29 to 19.9 rad/s.
5.2.3 Ultraviolet-visible absorption spectroscopy
Ultraviolet-visible absorption spectroscopy spectra were recorded on an Agilent Cary 4000
spectrometer in 1-10 mm quartz cuvettes.
5.2.4 Electronic circular dichroism
ECD measurements were collected and mean residue ellipticity calculated as described in
Chapter 2.
5.2.5 Peptide charge predictions
The net charge on the peptide was calculated as described in Chapter 2.
5.2.6 AFD36 treatment for cell culture
To remove chemical contaminants, solutions of AFD36 were stirred with 8 M urea and 3
mg mL-1 activated charcoal for 24 hours. Charcoal was removed by centrifugation at
20,000 g for 45 minutes and filtering (0.2 μm). The solution was then dialysed against 0.1
M NaCl (24 h) and water (2 × 24 h). Recovered peptide (2.8 mM) was adjusted to pH 3
and sterile-filtered (0.20 μm) then stored at -80 °C until use.
Samples of treated peptide were lyophilized and the peptide content of the solid was
determined by quantitative amino acid analysis (Australian Proteome Analysis Facility,
Sydney). Similar to the Het2-6 concentration determination described in Chapter 2, the
concentration of AFD36 was in some cases estimated by measurement of ellipticity at 222
nm using a mean residue ellipticity of -27,500 deg cm2 dmol-1 at a peptide concentration of
close to 1 mM in 10 mM MES buffer, pH 6.0.
5.2.7 AFD49 treatment for cell culture
AFD49 peptide solutions were dialysed against either 0.1 M NaCl pH 3 (24 h) then 1 mN
HCl (8 h then 16 h; Method 1) or 0.1 g L-1 activated charcoal in water (3 × 24 h; Method 2).
Recovered peptide (7.5 mM in both cases) was adjusted to pH 3 and sterile-filtered (0.20
μm) then stored at -80 °C until use.
179
Samples of treated peptide were lyophilized and the peptide content of the solid was
determined by quantitative amino acid analysis (Australian Proteome Analysis Facility,
Sydney). The concentration of AFD49 was in some cases estimated by measurement of
ellipticity at 222 nm using a mean residue ellipticity of -27,500 deg cm2 dmol-1 at a peptide
concentration of close to 1 mM in 10 mM MES buffer, pH 6.0.
5.2.8 Protease digestion of peptide hydrogel
Trypsin cleavage samples were prepared at a final composition of 0.5 mM AFD36, 10 mM
4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), 0.035% (w/v) trypsin (from
Trypsin-EDTA (Gibco, Life Technologies)) at pH 7.1 and incubated at 37 °C.
5.2.9 Fibroblast cell culture
NIH/3T3 cells (CRL-1658, ATCC) were maintained in Dulbecco’s modified Eagle’s medium
(DMEM) supplemented with 10% (v/v) newborn calf serum (except where indicated;
Ausgenex, Australia). Cultures were kept at 37 °C in a humidified 5% CO2 incubator.
To prepare gels for cell culture, 50 μL acidic peptide solution was mixed thoroughly with 50
μL of either 25 mM NaHCO3 (for 2.8 mM AFD36), 10 mM NaHCO3 (for 2 mM AFD49), 20
mM NaHCO3 (for 4 mM AFD49), or 35 mM NaHCO3 (for 7.5 mM AFD49), 2 × DMEM
(initially bicarbonate free) in individual wells of a 48-well tissue culture plate. Samples were
allowed to gel for 1 h at RT then transferred to an incubator and equilibrated in 500 μL
media as indicated in text before seeding with NIH/3T3 cells at 104 cells per well. Control
wells contained NIH/3T3 cells seeded onto tissue culture polystyrene (TCPS) alone that
was not equilibrated with media except where indicated in text. Fibroblasts were seeded
on either hydrogel or TCPS surfaces in DMEM supplemented with 10% (v/v) newborn calf
serum except where indicated in text.
To prepare layered AFD49 gels for 3D cell culture, lower gels were cast and seeded with
NIH/3T3 cells as above with no pre-soaking. Cells were also seeded onto bare TCPS for
controls. After 1 day, equal volumes of warmed (37 °C) acidic 4 mM AFD49 and 20 mM
NaHCO3, 2 × DMEM (initially bicarbonate free) were mixed thoroughly and 100 μL of this
gelling solution was immediately added on top of cells on either TCPS or AFD49 gels. This
was then allowed to gel in an incubator for 30 minutes before complete media was added.
Cell growth was quantified using AlamarBlue (Life Technologies) at times indicated in text.
Fluorescence readings used a Tecan Infinite M200 plate reader with λex 560 nm and λem
590 nm. Cell-free wells were prepared either with or without hydrogels and fluorescence
180
values for these control wells containing medium were subtracted from the values of the
relevant cell-containing wells. Statistical analyses used Student’s t-test, with differences
considered significant for p values of <0.05.
5.2.10
Cytotoxicity of free peptide assay
NIH/3T3 cells were seeded into wells of a 24-well plate at 4 x 104 cells per well in 1 mL of
media and allowed to adhere for 24 hours. At this point the media was removed and was
replaced with 1 mL of media which had been incubated with either bare TCPS or 3 mM
AFD49 hydrogels (prepared as above; n = 3 for each) for 24 hours in wells of a 24-well
plate. The cells were then allowed to proliferate for 48 hours before growth was quantified
using AlamarBlue (as above).
5.2.11
Induced pluripotent stem cell culture
Human induced pluripotent stem cells (iPSCs) were generated as wild-type nonviral cells
by episomal reprograming.10 The cells were transfected using the EoS lentiviral vector11 to
express enhanced green fluorescent protein (EGFP) as a reporter for pluripotency and
adapted to single-cell culture. Cells were maintained on Matrigel (BD Biosciences, San
Diego, CA) coated TCPS in conditioned DMEM/F12 culture medium prepared as
described in Xu et al.12 and supplemented with 20% KnockOut Serum Replacement, 0.1
mM nonessential amino acids, 2 mM GlutaMAX, 0.1 mM β-mercaptoethanol and 85 ng
mL-1 human basic fibroblast growth factor (all from Invitrogen, Life Technologies, Carlsbad,
CA). Cultures were kept at 37 °C in a humidified 5% CO2 incubator. Rho-associated
kinase (ROCK) inhibitor (Y27632; ATCC) was included in the culture media at 10 µM for
the first 24 hours following cell passaging to increase dissociated iPSC survival rates and
help maintain pluripotency.13, 14 Subsequent media changes occurred every day and did
not include ROCK inhibitor.
Peptide hydrogels were prepared as for fibroblast cell culture in wells of a 48-well plate
and then equilibrated in serum-free DMEM/F12 media as indicated in text. TCPS and
peptide hydrogel surfaces were coated with Matrigel (BD Biosciences, Massachusetts)
where indicated by soaking for 1-2 days with 150 µL of Matrigel prepared at 0.3 mg mL-1
final protein concentration in serum-free DMEM/F12. Phase contrast and fluorescence
images were collected each day following seeding.
181
5.2.12
Recombinant vitronectin fragment
The recombinant N-terminal somatomedin B (SMB) domain of vitronectin contained amino
acids 1-54 of mature vitronectin and a C-terminal six histidine tag for purification (H2NMDQESCKGRCTEGFNVDKKCQCDELCSYYQSCCTDYTAECKPQVTRGDVFTMLEHHH
HHH-COOH). This protein fragment was produced in E. coli and recovered as per
previously described methods.15 Recombinant vitronectin SMB domain was mixed into
acidic AFD49 stocks at concentrations as indicated in text prior to gel formation.
5.2.13
Microscopy
Phase-contrast and fluorescence images were collected utilising an Olympus IX50
microscope with standard 4',6-diamidino-2-phenylindole (DAPI), Tetramethylrhodamine
(TRITC) and Fluorescein isothiocyanate (FITC) filter sets. Cell images were taken at 4 × or
10 × magnification. Images were processed utilising ImageJ software.
Confocal microscopy was conducted on a Zeiss LSM 710 microscope. Fluorescein
imaging used a 488 nm laser for excitation and λem 493 – 523 nm, while propidium iodide
imaging used a 514 nm laser for excitation and λem 548 – 666 nm. Z-series of equidistant
x-y scans at 5 µm (Figure 5-16) and 3 µm (Figure 5-18) intervals were acquired and
processed utilising Zeiss Zen 2012 software.
Live-dead staining methods were adapted from the methods of Jones and Seft (1985) and
the protocol of Marker Gene Technologies (Eugene, Oregon) live:dead/cytotoxicity assay
kit (product M0795). Briefly, in a 48-well plate, cells were rinsed 2 × with 500 μL phosphate
buffered saline (without calcium or magnesium; PBS; Lonza) then incubated with 150 μL of
2 μM fluorescein diacetate (FDA; Sigma Aldrich), 4 μM (Section 5.3.4) or 1.5 μM (other
experiments) propidium iodide (PI; Sigma Aldrich) in PBS for imaging.
To estimate the proportion of cells adopting spread morphologies on TCPS or hydrogel
surfaces, cells were visually assessed and categorised as either rounded or spread. Cells
of each type were counted with >150 cells assessed per sample type.
5.2.14
All-trans retinoic acid analysis
All-trans retinoic acid (ATRA) was added to peptide solutions by pipetting small volumes of
an ATRA stock in dimethyl sulfoxide (DMSO) into the lid of an Eppendorf tube that
contained AFD49 solution without mixing the two solutions. The Eppendorf tube was then
closed and vortexed for 60 seconds to achieve rapid and even mixing. Final DMSO
182
content was 1% (v/v). The solution was then centrifuged at 16,000 g and the supernatant
recovered.
ECD spectra of ATRA-AFD49 solutions were recorded from 450-250 nm in a 1 mm quartz
cuvette.
To prepare gels (n=3) for ATRA release experiments, 50 μL acidic ATRA-AFD49 solution
(2 mM peptide) was mixed thoroughly with 50 μL of 10 mM NaHCO3, 2 × DMEM (initially
bicarbonate free) in wells of a 48-well plate. Samples were allowed to gel for 30 min at RT
and then 500 μL cell culture media (DMEM supplemented with 10% (v/v) newborn calf
serum) was added on top and the gels were incubated at 37 °C. Media was changed daily.
Samples for retinoic acid measurements were diluted 1:10 with 0.1% formic acid (FA) in
acetonitrile for LCMS analysis. LCMS used a Waters 2695 separations module with a
Waters 2489 UV detector (Waters Corporation, Milford, MA). Samples were applied to a
Kinetex 2.6 µm C18 100 Å column (Phenomenex) at a flow rate of 1 mL min-1, using a
linear gradient of 60-100% (v/v) buffer B over 30 min. Buffer A was 0.1% (v/v) FA and
buffer B was 0.1% (v/v) FA in acetonitrile. Eluate was injected directly into a Quattro micro
API tandem quadrupole system in positive ion mode. Retinoic acid species were quantified
by integration of absorption peak areas (monitored at 350 nm) using standards of all-trans
retinoic acid and 13-cis retinoic acid.
5.3 Results and Discussion
5.3.1 Mechanical characterisation of an AFD36 hydrogel
As discussed in Chapter 1, it has been shown that cell growth and function are impacted
by the mechanical properties of the underlying substrate.17-20 It was therefore appealing to
characterise the mechanical properties of AFD36 hydrogels.
Oscillatory shear rheology was used to characterise the mechanical properties of AFD36
hydrogels. As discussed in Section 4.3.12, peptide hydrogels prepared using addition of
NaOH showed non-uniform gelation. Preliminary rheological studies of peptide hydrogels
prepared using NaOH addition showed poor reproducibility. This was likely due to rapid
local gelation on base addition, while the remaining peptide solution remained ungelled.
This is consistent with rapid local network formation on pH adjustment by addition of
NaOH. Hydrogels prepared in this manner were not investigated further due to this inability
to produce a homogenous material for meaningful analysis.
183
An even change in pH and consequent homogenous gelation occurred on addition of
bicarbonate in Dulbecco’s modified Eagle medium (DMEM) to an acidic solution of AFD36,
with similar kinetics to when bicarbonate alone was used (Chapter 4, Section 4.3.12). This
allowed meaningful analysis of the mechanical properties of a homogenous peptide
hydrogel as would be used as a cellular scaffold. The storage (G′) and loss (G″) moduli of
a gelling AFD36 solution (0.35 % (w/v)) were recorded using oscillatory shear rheology as
the pH rose with loss of CO2 from the sample following bicarbonate addition, allowing
observation of the rate of network formation (Figure 5-1A).
A viscoelastic solid (G′ > G″) was formed in the time taken to mix and load the sample and
structure development then followed approximately first order kinetics. After 60 minutes
the value of G′ was 350 Pa, indicating mechanical properties similar to soft tissues such as
brain and lymph node.21 Few hydrogels based on coiled coils have been reported and only
the hSAF peptides designed by the Woolfson group have been investigated using
rheology.3 Hydrogels formed by hSAF peptides had higher mechanical strength than the
AFD36 gel characterised here, however the hSAF gels were prepared at almost 3-fold
higher peptide concentration, which would greatly increase gel strength. Comparison of
AFD36 hydrogels with previously reported RAD16-I and MAX1 β-sheet peptide hydrogels
showed comparable strength at similar or lower weight concentrations (Table 5-1).22-24
After one hour, the gel formed showed moderate frequency dependence of G′ and G″
(Figure 5-1B). However, no cross-over point is observed, with G′ being greater than G″ by
at least one order of magnitude for all frequencies tested. The frequency dependence of G′
and G″ shown here is similar to that previously reported for hydrogels formed by selfassembling β-sheet peptides.23, 25 This dependency likely relates to the semi fluid-like
properties of these physically cross-linked gels. At higher frequencies there is less time for
the network to rearange and relax, which results in a more rigid material.
At the completion of testing, visual observations of pH-dependent colour change showed
the peptide sample to be at lower than physiological pH, likely due to limitations to the loss
of CO2 at the plate edge. The results obtained are thus likely to represent a low-end
estimate of the rate of gelation and the final gel strength.
184
Figure 5-1. Rheological characterisation of a 1.4 mM (0.35 % (w/v)) AFD36 hydrogel A) over
time during gelation (x-axis is time from measurement initialization), and B) as a function of
frequency (two samples). In both plots G′; triangles, G″; squares.
185
Table 5-1. Table of previously reported self-assembling peptide hydrogels referenced in this chapter
Peptide
Class
hSAFAAA3
α-helical
hSAFAAA-W3
α-helical
Peptide
G′b
Cell typec
0.9
1500
-
0.9
-
concentration
a
PC12 rat adrenal
pheochromocytoma
2D or 3D
culture
Recognition motifd
-
-
2D
-
Unmodified as
RADA16-123
β-sheet
0.5, 1, 2
12, 37, 60
Murine neuronal stem cells
3D
well as RGD and
two bone marrow
homing peptides
MAX122
β-sheet
MAX125
β-sheet
MAX126
0.7, 1, 1.4, 2
250, 1300,
1600, 4000
-
-
-
-
-
-
2 (20, 150 and 400
100, 300,
mM NaCl)
3000
β-sheet
2
10000
NIH/3T3 murine fibroblasts
2D
-
MAX7CNB27
β-sheet
2
1000
NIH/3T3 murine fibroblasts
2D
-
MAX824
β-sheet
0.5
500
C3H10t1/2 murine stem cells
3D
186
Fmoc-RGD28
29
PAs
Fmoc
Alkyl-tail
PA
-
-
Human adult dermal
fibroblasts
3D
RGD
3D
RGD
3D
-
3D
-
NIH/3T3 murine fibroblasts,
N/A
100
MDA-MB-231 human breast
carcinoma
P11-430
β-sheet
3
-
P11-8/12/1631
β-sheet
2
-
Q1132
β-sheet
4.5
-
Primary human dermal
fibroblasts
L929 murine fibroblasts
C3H10t1/2 murine stem cells,
NIH/3T3 murine fibroblasts
3D
Unmodified and
RGD
a
Reported peptide concentration used in % w/v
b
Storage modulus reported for peptide hydrogel in Pa (where G′ was not reported in text, values are estimated from plotted rheological curves)
c
Cell type reported to be successfully cultured on or in the peptide hydrogel
d
Recognition motif incorporated within the peptide sequence where relevant
187
5.3.2 Chemical contaminants
As discussed in Chapter 1, peptide hydrogels are promising materials for use as cellular
scaffolds. It was therefore of interest to characterise mammalian cell growth on AFD36
hydrogels. Initial experiments using hydrogels prepared from AFD36 peptide received as a
TFA salt showed these gels were unable to support mammalian cell growth, with high
levels of cell death observed (not shown). It was thought that this may be due to the
toxicity of the residual TFA present in the solid.31, 33 In an attempt to overcome this issue,
peptide AFD36 was purchased as an acetate salt. The level of TFA remaining in the
peptide solid was 0.16% (w/w) (determined by the manufacturer), which once made up as
a hydrogel is below the concentration reported to be detrimental to cell growth.31
However, hydrogels prepared from AFD36 peptide received as an acetate salt were also
unable to support mammalian cell growth (not shown). Ultraviolet-visible spectroscopy of
as-received peptide indicated the presence of strongly absorbing species, with maxima at
262, 285 and 296 nm (Figure 5-2). These spectra resembled those of the
Fluorenylmethoxycarbonyl protecting group,34, 35 which is an aromatic protecting group
used during solid-phase peptide synthesis. The presence of this contaminant in received
peptide was suprising, given that the peptide had been purified by reversed-phase
chromatography. The presence of this contaminant in purified peptide material may reflect
binding of the aromatic molecule by the peptide, or introduction of the contaminant postpurification. Similar chemical contaminants were observed in several batches of peptides
of related sequence from different suppliers, suggesting an inability to fully purify the
relatively hydrophobic AFD peptides from chemical contaminants using standard
chromatography methods.
188
Figure 5-2. Absorbance spectra of AFD36 solutions at steps during purification; as-received
peptide (solid line), peptide following urea and charcoal treatment (dotted), dialysed product
(dashed). Spectra normalised to 1 mM peptide. All spectra collected in 10 mm path length cuvette.
In an attempt to remove chemical contaminants, AFD36 peptide was mixed with a
suspension of activated charcoal in the hope that this would bind aromatic compounds.
Urea was also added at 8 M to destabilise the coiled coil to allow removal of contaminants
bound in the hydrophobic core. A decrease in absorbing species was observed when
charcoal was removed via centrifugation and filtering (Figure 5-2). The peptide solution
was then dialysed against 0.1 M NaCl, to exchange any remaining TFA counterions,
followed by water. The recovered peptide contained much lower levels of absorbing
contaminants, with A260 values of ca. 10% of initial (Figure 5-2). This purification method
gave peptide that was capable of supporting mammalian cell growth once gelled (Section
5.3.3).
While previous studies have highlighted the effects of residual TFA in synthetically
produced peptides,25, 26 to the candidates knowledge, no studies have reported the
presence of other chemical contaminants in peptide solids that effect cellular growth. The
persistence of chemical contaminants in peptide material may relate to strong interactions
189
between the hydrophobic peptides and contaminant aromatic molecules, preventing
separation via chromatography. This binding may be enhanced if the peptide is not fully
denatured during chromatographic purification and encapsulates the hydrophobic chemical
contaminants. Future studies evaluating the use of peptide hydrogels as cellular scaffolds
will need to consider the presence of such contaminants.
5.3.3 Cell Growth on AFD36
AFD36 hydrogels prepared by mixing an acidic peptide solution with bicarbonate in DMEM
were stable to repeated addition and removal of cell culture media, as required for tissue
culture experimentation. Furthermore, the cell culture media above the gel provided
additional buffering capacity and maintained gelling conditions. Common to many reported
studies of cell growth on peptide hydrogels is equilibration of gels with culture media prior
to cell seeding.26, 27, 36-38 To assess the ability of AFD36 hydrogels to support mammalian
cell growth, NIH/3T3 cells were cultured for 4 days on either uncoated tissue culture
polystyrene (TCPS) or AFD36 gels pre-equilibrated for 2 × 2 days in serum-free DMEM.
Cell growth was quantified by AlamarBlue assay at each day, with cell images recorded at
day 3 (Figure 5-3). The fibroblasts proliferated well on AFD36 hydrogels, with cell growth
at 68% of TCPS levels at day 4. This level of 3T3/NIH fibroblast growth is similar to that
previously reported on MAX β-hairpin peptide gels, although in these systems no pretreatment of the peptide was required and hydrogels were only equilibrated with serum
free media overnight.26, 27 Table 5-1 lists several hydrogels formed by self-assembling
peptides which have been used for cell culture including the types of cells successfully
cultured for comparison.
The fibroblasts formed spread morphologies on AFD36 gels, similar to the control TCPS
surface. Adhesion and spreading may be due to non-specific protein adsorption from
serum onto the surface of the hydrogels, providing cellular attachment points.39 This is
further investigated in Section 5.3.10. The proliferation of fibroblasts at levels similar to
TCPS indicate that the AFD36 gels are not strongly toxic towards NIH/3T3 fibroblasts and
suggests that the peptide gels may be suitable biomaterials for tissue engineering
applications.
190
Figure 5-3. Growth of NIH/3T3 fibroblasts on AFD36 gels and TCPS controls. Representative
images of cells cultured on A) TCPS control surface or B) AFD36 hydrogel at 3 days (scale bars
100 µm). C) Cell growth measured via AlamarBlue fluorescence assay on TCPS (black bars) and
AFD36 hydrogels (grey bars) at 1-4 days ±SD (n=3).
* p values ≤ 0.05, relative to the control, as determined by Student’s T-test. Peptide gels were 1.4
mM AFD36 final.
191
5.3.4 Long term growth on AFD36
To assess the AFD36 peptide hydrogel’s potential as a substrate for long-term cell growth,
NIH/3T3 cells were cultured on either TCPS or AFD36 hydrogels pre-equilibrated for 6
days (one media change at day 2) in serum-free DMEM. Cell growth and morphology was
observed via phase contrast microscopy at days 1, 7, 10, 14 and 18 (Figure 5-4).
At 1 day fibroblasts were attached and had adopted a spread morphology on both TCPS
and peptide hydrogels. At 7 days, fibroblasts grown on TCPS had reached confluence and
higher magnification images of these fibroblasts were not collected at longer time points.
At 7 days, fibroblasts grown on AFD36 hydrogels had begun to form localised islands of
higher cell density. This “clumping” of fibroblasts on hydrogels increased with time, forming
clusters at day 10 and dense balls at day 14. At days 14 and 18 it can be seen that in
addition to the high-density cell aggregates, fibroblasts were proliferating and re-covering
the gel surface that fibroblasts had previously receded from. The apparent decrease in
number of cell aggregates at day 18 is likely due to loss of detached aggregates during
media changes.
Live-dead staining at day 21 was used to assess whether the observed clumping of cells
corresponded with cell death (Figure 5-5). The washing procedure for staining resulted in
the loss of some less attached fibroblast clusters on peptide hydrogels. Few dead cells
were seen on the TCPS surface, however several dead floating cells were observed (not
shown). This may be due to a loss of TCPS surface adhesion following cell death. On
peptide hydrogels some cell death was seen in the dense cell aggregates; however the
majority of fibroblasts in these masses were alive, suggesting these are not necrotic
masses of cells. The presence of dead cells on hydrogel surfaces after washing also
indicates that with the exception of the larger cell-agregates, cells may be partially
embedded in the hydrogel material and resist detachment.
192
Figure 5-4. Representative images of NIH/3T3 fibroblasts grown on AFD36 hydrogels and TCPS
controls. Days cultured as indicated (scale bars 100 µm). Peptide gels were 1.4 mM AFD36 final.
193
Figure 5-5. Live-dead staining of NIH/3T3 fibroblasts grown for 21 days on A) TCPS or B-D)
AFD36 hydrogels (scale bars 100 µm).
Previous experimentation has shown that culture of fibroblasts inside collagen40 and
peptide28 gels caused contraction of these soft physically cross-linked materials. While no
macroscopic contraction of peptide hydrogels was seen in this experiment, it may be
possible that contraction of surface material due to cytoskeletal forces could result in the
clustering of cells observed here. Such an effect would be enhanced as fibroblasts
proliferate and cell density increases, resulting in increased numbers of cell-cell
interactions leading to cell clustering. Additionally, fibroblasts may initially grow well on
hydrogels, but later deposit their own ECM and form better attachments to this material
and each other than to the hydrogel surface, resulting in detachment of high-density
clusters of cells. The embedding of cells within a hydrogel matrix to give cells a threedimensional environment may overcome cellular detachment.
194
Characterisation of NIH/3T3 fibroblast growth on peptide hydrogels has generally been
reported over shorter time periods of less than 1 week.26, 27, 29 The progressive aggregation
and detachment of fibroblasts shown here may therefore relate to effects only seen at
longer culture times, when cells have begun to remodel the hydrogel surface or produce
their own ECM. There have also been few reports of fibroblast growth on peptide
hydrogels at high cell density.26 In previous studies of high cell-density culture on peptide
hydrogels no cellular aggregation and detachment was reported. However the mechanical
properties of the peptide hydrogel in this case was much higher than for AFD36 hydrogels
(Section 5.3.1), which may have prevented the contraction of surface material leading to
cell clustering. The generality of this cell clustering and progressive detachment will need
to be determined for alternate cell lines and hydrogel systems.
5.3.5 Trypsin digestion
NIH/3T3 fibroblast cells grown on AFD36 hydrogels were not detached by standard trypsin
dissociation methods. One potential reason for this is the trypsin enzyme binding to
cleavage sites within the peptide hydrogel and slowly digesting the peptide scaffold.
Trypsin cleaves polypeptides C-terminal to lysine and arginine residues,41 and given the
presence of four lysine residues within AFD36 may be expected to digest this peptide.
During the design of the AFD peptides in Chapter 4, optimisation of the sequence
instability index was used in an attempt to increase resistance to protease digestion.42 It
was therefore of interest to assess trypsin digestion of the peptide hydrogel. A mixture of
AFD36 and a commercial trypsin preparation was made at a gelling concentration of
peptide and a final trypsin concentration similar to those recommended for cell dissociation
by the manufacturer. The mixture gelled and was incubated at 37 °C. Gel breakdown
occurred after ca. 18 hours. The helical secondary structure of AFD36 during digestion
was monitored using ECD. Figure 5-6 shows the change in helical structure as measured
at 222 nm over the first 2 hours. The helical secondary structure of AFD36 is degraded
over time, with 63% remaining after 2 hours and only 10% at 24 hours (not shown).
Decreasing helical structure shows trypsin to be capable of digesting AFD36, although
loss of gel structure required much longer than the gels would be exposed to in normal cell
culture experiments. Furthermore, in this digestion experiment the trypsin was mixed
evenly through the gel, whereas in tissue culture experiments only the surface of the gel
would initially be exposed to enzymatic degradation. The observed lack of cell detachment
on treatment with trypsin may therefore be due the enzyme digesting AFD36 rather than
195
Figure 5-6. Digestion of a 0.5 mM (0.12% (w/v)) AFD36 gel by trypsin. Helical secondary
structure monitored via ECD at 222 nm.
dissociating cells. Additionally, serum proteins and trypsin inhibitor from the culture media
may remain in the hydrogel volume during washing steps and further decrease the efficacy
of trypsin in detaching cells for recovery. The lack of cellular detachment from peptide
hydrogels on treatment with trypsin indicates that the peptide hydrogels are not amenable
to cell culture where cells are recovered via trypsinization. However, the use of alternate
dissociation agents or dissolution of the hydrogel via pH adjustment may be applicable in
some circumstances. Furthermore, AFD36 peptide hydrogels may be useful as matrices
for toxicological assays or in vivo applications where the recovery of cells is not required.
Further experiments are required to determine the susceptibility of peptide hydrogels to
digestion by proteases secreted by mammalian cells. However, if the gels are susceptible
to digestion at low levels, as appears likely, this may be an advantage in applications
where cells are injected into the body encapsulated in a peptide hydrogel. In this case
digestion of the gel over time by secreted proteases43 would allow for progressive
replacement with native ECM, potentially resulting in enhanced tissue regrowth.44 Of note
196
is that a high protease concentration was used in this experiment, and therefore
degradation of AFD36 hydrogels may be expected to be slower in in vivo environments
than is observed here.
The design process used for peptide AFD36 included sequence optimisation for in vivo
stability using the instability index. It may be of interest in the future to compare the
proteolytic susceptibility of hydrogels formed by alternate sequences that assemble
similarly but possess different instability index values. This would allow for validation of the
use of such an index in designing peptides for in vivo applications. Furthermore, if the
instability indices of designed peptide sequences were shown to be predictors for the rate
of hydrogel degradation, this may offer a way to design hydrogels with pre-determined
degradation rates as required for clinical applications.
5.3.6 Peptide design
The peptide AFD49 (Figure 5-7) was designed based on the self-assembly model
proposed in Chapter 4, to further explore the relationship between sequence and selfassembly for gel formation. This design maintains the same charged residue composition
as AFD36 of four lysine, three glutamate and two histidine residues. This is predicted to
Figure 5-7. Comparison of heptad based helical wheel projections of peptides AFD36 and
AFD49. Sequence changes are highlighted in red.
197
give a charge of +1 at pH 7.0,45 which is associated with gel formation (Chapter 4). In
AFD49 the charged residues are arranged to have a higher degree of heptad repetition
than AFD36, with only glutamate at position c, lysine at position f and a mixture of lysine
and histidine residues at position b.
The hydrophobic residue composition of AFD49 included a higher proportion of leucine
residues compared to AFD36 and AFD19. This choice was driven by the hydrophobicity
and high helix propensity of leucine.46, 47 The hydrophobic residues were arranged to
maintain the expanded hydrophobic face for coiled-coil assembly while providing a higherdegree of heptad repetition, with only leucine residues in positions a, d and g, and alanine
residues at position e. The high degree of heptad repetition in AFD49 may reduce
differences in peptide-peptide interactions between in-register and offset helices,
potentially stabilising out-of-register arrangements and enhancing fibril formation
(discussed in Chapter 4; Section 4.3.8). The final sequence of AFD49 (Ac-LKELAKL
LHELAKL LHELAKL-CONH2, FW 2465), possesses a GRAVY value of 0.338, which is
higher than AFD36 but lower than AFD19 and predicted sequence solubility. The instability
index of AFD49 is -6.65, which is lower than both AFD19 and AFD36 and places it well
below the threshold value of 40 thought to divide proteolytically stable from unstable
polypeptides.42
5.3.7 AFD49 characterisation
Similarly to AFD19 and AFD36, as-received AFD49 peptide dissolved in water to give an
acidic solution. Titration of a 0.1-0.5% (w/v) solution of AFD49 with NaOH gave similar
results to those of AFD36, with gelation occurring at pH 7.0, where the molecular charge is
predicted to approach +1, and precipitation observed above pH 10, where the predicted
molecular charge approaches zero. ECD of AFD49 in both ungelled (pH 2.8) and gelled
(pH 7.0) states shows a spectrum with a double minimum at 208 and 222 nm and a
maximum at 192 nm indicating α-helix formation (Figure 5-8). The mean residue ellipticity
at 222 nm for gelled and ungelled samples were -33,300 and -31,900 deg cm2 dmol-1
respectively. These values indicate AFD49 to be in an almost fully helical conformation in
both states.48 The slightly lower helicity at acidic pH likely reflects increased electrostatic
repulsion destabilising helix folding due to higher molecular charge under acidic
conditions. The observed hydrogel formation and secondary structure of AFD49 is
consistent with this peptide self-assembling similarly to peptide AFD36. This is consistent
with the self-assembly model proposed in Chapter 4, whereby amphiphilic α-helices with
198
Figure 5-8. ECD spectra of self-buffered AFD49 (1.6 mM) at 20 °C in gelled (solid line, pH
7.0) and non-gelled (dashed, pH 2.8) states.
self-similar heptad repeats assemble into staggered coiled coils due to permissive helix
offsetting. The success of this new peptide design also demonstrates the tolerance of the
self-assembling system for modification of gelling sequences. This potentially enables
future designs to incorporate sequence modifications for specific property or functionality
requirements.
In contrast to AFD36 (Section 5.3.2), as-received AFD49 peptide did not contain strongly
absorbing species (Figure 5-9). Chemical contaminants were therefore thought to be at
lower levels in this peptide solid. Similar to the AFD36 solid used for cell culture
experiments (Section 5.3.3), this peptide was purchased as an acetate salt and the level of
TFA remaining in the peptide solid was 0.2% (w/w) (determined by the manufacturer),
which once made up as a hydrogel is below the concentration reported to be detrimental to
cell growth.31 Two methods were assessed for their ability to remove any remaining
contaminants from AFD49. Peptide solutions were dialysed against either 100 mM NaCl
199
Figure 5-9. Absorbance spectra of AFD49 stock solutions before (solid line) and after dialysis
using Method 1 (dotted) and Method 2 (dashed). Spectra normalized to 1 mM peptide and 10 mm
path length.
pH 3 (24 h) then 1 mN HCl (8 h then 16 h; Method 1) or 1 g L-1 activated charcoal in water
(3 × 24 h; Method 2; Figure 5-9).
To compare the ability of the two methods to give peptide material able to support
mammalian cell growth, NIH/3T3 fibroblasts were cultured for 3 days on either TCPS or
AFD49 gels prepared from peptide recovered after treatment with Method 1 or 2 and preequilibrated for 16 hours in cell culture media. At day 3 cell growth was quantified by
AlamarBlue assay and cell images were taken (Figure 5-10). No statistical difference in
cell growth was seen between hydrogel samples, with cell growth at 35-38 % of TCPS
levels. This indicated material recovered from both methods was similarly able to support
cell growth, potentially in part due to lower contaminant levels in the initial solid, and were
therefore both utilised in subsequent experimentation. The fibroblasts adopted spread
morphologies on all AFD49 gels, similar to the control TCPS surface. However some
clusters of cells were observed on the AFD49 gels, similar to those observed on AFD36
200
Figure 5-10. Growth of NIH/3T3 fibroblasts on AFD49 gels and TCPS controls. A) Cell growth
measured via AlamarBlue fluorescence assay on TCPS and AFD49 hydrogels at 3 days ±SD
(n=3). Representative images of cells cultured on B) TCPS; or AFD49 hydrogels prepared from
peptide dialysed using either C) Method 1 or D) Method 2 (discussed in text) (scale bars 100
µm). Peptide gels were 3.75 mM AFD49 final.
gels (Section 5.3.4). The level of cell growth on AFD49 hydrogels compared to TCPS
controls was lower than that seen for AFD36 hydrogels (Section 5.3.3). However the
shorter hydrogel equilibration in cell culture media prior to cell seeding on AFD49 may
have affected the rate of cell growth, as is investigated in Section 5.3.9.
Similar experiments to those comparing the batches of treated AFD49 were used to
investigate the growth of 3T3/NIH fibroblast on hydrogels of varied final AFD49 peptide
concentration (not shown). No significant difference was seen between final peptide
concentrations of 1-2 mM. This suggests that at the peptide concentrations investigated
here, the differences in mechanical properties and residual contaminant concentrations
are not sufficient to influence fibroblast proliferation.
201
5.3.8 Cytoxicity of peptide released from hydrogels
One potential reason for the lower level of proliferation seen on peptide hydrogels
compared to TCPS alone is the potential continual release of “free peptide” from the
dynamic network of peptide fibrils. If this low level of unassembled peptide in the media
were cytotoxic to the cells it may partially explain the lower level of proliferation seen on
hydrogels compared to TCPS alone. To investigate this, NIH/3T3 cells were seeded in
empty wells and allowed to adhere to TCPS. After 1 day the media was removed and then
replaced with cell culture media which had been incubated with either TCPS or AFD49
hydrogel (n = 3 for each) for 1 day. Cell growth was quantified by AlamarBlue assay after 2
days. These results showed no statistical difference in growth between cells growing in
media incubated with TCPS or AFD49 hydrogels (not shown). This indicates that if free
peptide is slowly released from peptide hydrogels, it is not cytotoxic to NIH/3T3 cells under
the conditions tested here. While futher studies will be required to investigate the effects of
free peptide at varied concentrations, different growth and incubation times, and for
different cell types, this preliminary experiment is promising for future application of AFD49
hydyrogels as cell scaffold materials and therapeutic delivery vehicles. Alternate reasons
for the lower level of growth on hydrogels compared to TCPS may relate to the inherent
propensity for NIH/3T3 fibroblasts to proliferate on soft materials compared to TCPS, a
factor which would need to be explored for alternate cell lines.
5.3.9 Effect of length of equilibration time for AFD49 gels
Given the lower level of cell growth observed on AFD49 gels when pre-equilibrated for
only 16 hours (Section 5.3.6) compared to growth on AFD36 gels when pre-equilibrated for
2 × 2 days (Section 5.3.3), it was of interest to further assess the effect of the length of gel
pre-soaking time. To investigate this, NIH/3T3 cells were cultured for 3 days on either
uncoated TCPS or AFD49 gels that were either washed for 1 hour or pre-equilibrated for 1
or 2 × 1 days in cell culture media prior to cell seeding. Cell growth was quantified by
AlamarBlue assay at day 3 (Figure 5-11).The fibroblasts proliferated more on hydrogels
that had been equilibrated for longer in cell culture media. At 3 days, growth on AFD49
hydrogels equilibrated for 2 × 1 days is not significantly different from levels on TCPS and
more than double that on gels that were seeded after only washing the surface. These
results show the importance of pre-soaking time for peptide hydrogels.
202
There are two potential reasons for the observed increased proliferation following preequilibration of hydrogels with media. Although AFD49 was received as an acetate salt, no
absorbing species were observable via absorption spectroscopy and the peptide was
further treated by dialysis (Section 5.3.6); there is still the potential for chemical
contaminants to remain in the peptide solution at trace levels and reduce proliferation as
the gels degrade. If contaminants are able to diffuse out of the gel and affect cell growth, it
may be expected that they could be removed via pre-soaking with cell culture media.
While the results of Section 5.3.8 suggest that media incubated with hydrogels is not itself
Figure 5-11. NIH/3T3 fibroblast growth on TCPS and AFD49 gels equilibrated in cell culture
media for varying times as indicated, measured via AlamarBlue fluorescence assay at 3 days ±SD
(n=3). Peptide gels were 3.75 mM AFD49 final.
* p values ≤ 0.05, as determined by Student’s T-test.
203
cytotoxic to fibroblasts, factors such as the proximity of cells to the contaminants being
leached from the hydrogel and the length of time hydrogels are pre-soaked may play a
role. Alternatively, the presence of serum in the equilibration media may enhance
subsequent cell growth by increasing the levels of surface adsorbed proteins, or by the
accumulation of metabolic factors such as growth factors within the gels. This is further
investigated in Section 5.3.10.
5.3.10
Presence of serum in pre-soaking media
The results of Section 5.3.9 indicated the benefits to cell growth of equilibrating AFD49
hydrogels with cell culture media prior to cell seeding. To separate the effects of
contaminant removal and accumulation of components from media serum, NIH/3T3 cells
were cultured for 3 days on either TCPS or AFD49 gels that were pre-equilibrated in
DMEM with or without serum for 3 × 1 days. Growth on these surfaces was compared to
control TCPS and AFD49 gels that were not pre-equilibrated. Cell images were recorded
and cell growth was quantified by AlamarBlue assay at day 3 (Figure 5-12).
Cell growth on AFD49 gels pre-equilibrated with serum-free media was more than double
that on gels that were not pre-equilibrated, with growth increased by a further 12% when
serum was included in the soaking media. On TCPS the effect of equilibrating with serumfree media was not statistically significant; however the inclusion of serum in the soaking
media increased growth by 32%. Fibroblasts adopted spread morphologies on both
surfaces for all conditions tested. While the surfaces not pre-equilibrated with serumcontaining media will not initially be coated with serum proteins, protein may adsorb to the
surface from the growth media after cell seeding and provide cellular attachment sites. Cell
growth on AFD49 gels pre-equilibrated for 3 × 1 days with serum-free media was 67% of
that on non-equilibrated TCPS at 3 days. This is comparable to growth on AFD36
204
Figure 5-12. Growth of NIH/3T3 fibroblasts on AFD49 gels and TCPS controls using different
equilibration conditions. Representative images of cells cultured for 3 days on A) AFD49
hydrogels or B) TCPS control surfaces pre-equilibrated with media as indicated (scale bars 100
µm). C) Cell growth measured via AlamarBlue fluorescence assay on AFD49 hydrogels (black
bars) and TCPS (grey bars) at 3 days ±SD (n=3). Peptide gels were 3.75 mM AFD49 final.
* p values ≤ 0.05, relative to samples of same substrate material, as determined by Student’s Ttest.
205
hydrogels pre-equilibrated using similar conditions (Section 5.3.3), indicating the two
peptide hydrogels to support similar levels of fibroblast growth.
The increase in fibroblast proliferation on AFD49 hydrogels pre-equilibrated with serumfree media indicates that soaking may further remove residual chemical contaminants. As
the gels are initially cast containing DMEM, it is not expected for components of this basal
media to diffuse into the gel and accumulate. The increase in cellular proliferation on both
TCPS and AFD49 hydrogels following pre-equilibration with serum-containing media
compared to serum-free media likely reflects the benefits to early cell growth of the preadsorption of serum proteins to the surfaces. In the case of the hydrogels this may be
further enhanced by the accumulation of metabolic factors from the serum within the
hydrogel volume.
The benefits to cell growth demonstrated by equilibration of hydrogels in serum-free media
suggest that low levels of chemical contaminant remain in the peptide material after
dialysis. The ability to remove these contaminants by equilibrating peptide hydrogels is
somewhat surprising, given that equilibration media is only at a five-fold excess compared
to gel volume, while dialysis is conducted at an approximately hundred-fold larger excess
of external dialysis volume to peptide solution. This may reflect differences in solubility of
the contaminant under the conditions used for dialysis compared to equilibration due to
solution pH or ionic strength. Furthermore, equilibration is conducted at 37 °C, which may
increase contaminant solubility compared to room temperature dialysis conditions. Another
contributing factor may be slow diffusion of material from peptide solutions under gelling
conditions during dialysis due to a lower surface area to volume ratio than when gels are
cast in wells for cell culture experiments. Future studies may be able to overcome
contaminant issues by investigating alternate dialysis conditions. This may include further
investigating the effects of dialysis solution pH, ionic strength and temperature, as well as
varying peptide solution surface area to volume ratios by using different dialysis cassette
volumes. The results of Sections 5.3.9 and 5.3.10 demonstrate the importance of
appropriate selection of pre-equilibration conditions for both peptide hydrogels and TCPS
controls. These conditions may vary across different cell lines and applications but should
be further explored and optimised for each case.
5.3.11
Fibroblast attachment without serum
Given the lack of cellular recognition motifs in the AFD49 peptide, it was thought that
fibroblast attachment may be mediated via serum proteins adsorbed to the hydrogel
206
surface. To investigate the requirement for serum proteins for fibroblast attachment,
NIH/3T3 fibroblasts were cultured for 1 day on either TCPS or AFD49 hydrogel surfaces
that had been pre-equilibrated for 1 day in either serum-free or serum-containing media.
On each type of surface, fibroblasts were seeded in either serum-free or serum-containing
media and images of cell attachment were recorded at 1 day (Figure 5-13).
As fibroblasts do not detach from hydrogel surfaces even on cell death (Section 5.3.4) it
was not possible to assess the level of attachment to different surfaces by counting
detached cells that were washed off. In the absence of this metric, the level of cellular
spreading was used as a measurement of cell-substrate interactions. To assess the level
of cellular spreading on each surface type, fibroblasts were visually assessed and
classified as having either a rounded or spread morphology. Comparison of the numbers
of each type allowed for an estimate of the levels of cells adopting spread morphologies on
each surface.
Approximately 90 % of fibroblasts adopted a spread morphology on AFD49 or TCPS
surfaces when serum was included in the cell-seeding media, regardless of preequilibration conditions (Figure 5-13A). However, when serum-free media was used for
cell-seeding, fibroblast spreading on TCPS surfaces was only 60%, regardless of preequilibration conditions. On AFD49 hydrogels pre-equilibrated in the absence of serum,
only 20% of cells adopted spread morphologies when serum-free medium was also used
for cell-seeding. However, pre-equilibration with serum-containing media increased the
level of spreading to 90%, even when serum-free media was used for cell seeding.
The higher level of cell spreading on surfaces in serum-containing media illustrates the
benefits of the presence of serum components for cell-substrate interactions on peptide
hydrogels and TCPS, likely due to the lack of cellular recognition motifs on both bare
surfaces. In the case of fibroblasts seeded on peptide hydrogels in the absence of serum,
the pre-equilibration of gels with serum-containing media was sufficient for fibroblast
spreading to occur at levels similar to controls where cells were seeded in serumcontaining media. This implies that adsorption of serum components to peptide hydrogels
may decorate these surfaces with sufficient cellular recognition sequences to achieve cell
spreading in the absence of specific integrin sequences within the peptides themselves.
207
Figure 5-13. The effects of serum on spreading of NIH/3T3 fibroblasts on TCPS or AFD49
surfaces. Representative images of cells after 24 hours when they were added in A) serumcontaining media, or B) serum-free media to surfaces pre-equilibrated as indicated. (scale bars
200 µm). Peptide gels were 1 mM AFD49 final.
208
Cell spreading may also be enhanced by accumulation of other serum components within
the hydrogel volume that increase fibroblast interactions with the substrate. The similar
levels of spreading on TCPS surfaces when cells were seeded in serum-free media,
regardless of pre-equilibration conditions, indicates that for these surfaces, the presence of
serum components in the seeding media is more important than in equilibration media.
This may reflect the influence of other soluble serum components in promoting
attachment.
5.3.12
Fibroblast attachment in the presence of free RGD
As discussed in Chapter 1, arginine-glycine-aspartic acid (RGD) is an oligopeptide
sequence found across many ECM components such as fibronectin,49 laminin,50 collagen51
and vitronectin.52 It is implicated in cellular recognition and attachment through integrin
binding. The inclusion of free cyclic-RGD peptides in growth media is expected to inhibit
cellular recognition and binding of proteins containing this sequence.37, 38 To investigate
whether the attachment of fibroblasts to hydrogel surfaces proceeded through integrin
recognition of RGD sequences, NIH/3T3 fibroblasts were grown in serum-containing
media for 1 day in the presence or absence of 100 μM free cyclic-RGD
(c[RGDfK(Biotin)])53, 54 on TCPS or AFD49 hydrogels pre-equilibrated for 1 day in serumfree media. Cell images were recorded after 1 day (Figure 5-14). Fibroblast attachment
and spreading occurred on all surfaces, irrespective of the presence of cyclic-RGD. On
AFD49 hydrogels in the presence of cyclic-RGD, the cellular morphology was slightly less
extended, potentially relating to a lower level of attachment and spreading. Interestingly,
the fibroblasts incubated with cyclic-RGD did not proliferate and adopted rounded
morphologies and underwent cell death in the days after initial attachment. Previous
studies have implicated soluble RGD-containing peptides in cell-cycle arrest and apoptosis
via integrin signalling pathways, an effect that may be active here.55, 56 The lack of
attachment inhibition is likely due to alternate integrin recognition sites being involved in
fibroblast adhesion on serum-adsorbed surfaces57 and is consistent with the absence of
integrin recognition sequences within the AFD49 peptide. Future studies may involve using
varied concentrations of cyclic-RGD to determine the concentration dependence of
attachment inhibition and cellular proliferation.
209
Figure 5-14. Effect of free cyclic-RGD on NIH/3T3 fibroblast adhesion to TCPS and AFD49.
Fibroblasts incubated on TCPS in A) culture media, B) 100 µM cyclic-RGD in culture media, or
on AFD49 in C) culture media, D) 100 µM cyclic-RGD in culture media. (scale bars 100 µm).
Peptide gels were 2 mM AFD49 final.
5.3.13
Effects of hydrogel pre-equilibration on fibroblast attachment and
proliferation
The results of Sections 5.3.9-5.3.12 demonstrate the importance of equilibration of peptide
hydrogels prior to cell seeding. The benefits of equilibration to proliferation is cumulative,
with longer pre-equilibration times resulting in increased proliferation (Section 5.3.9). The
exact reason for enhanced proliferation has not yet been determined; however it appears
to be a combination of the removal of residual chemical contaminants and the deposition
of serum components on or in the hydrogels (Section 5.3.10).
Fibroblast attachment to peptide hydrogels was dependent on the presence of serum
components either in the pre-equilibration media or in the cell-seeding media (Section
5.3.10), and was not affected by the presence of free cyclic-RGD peptides (Section
5.3.12). This is consistent with the lack of designed integrin recognition sites within the
210
hydrogel forming peptides, and involvement of adsorbed serum proteins in providing
cellular attachment points. As discussed in Chapter 1, many peptide hydrogel scaffolds
require functionalization with cell recognition motifs to support cellular attachment, and to
the candidates knowledge only the MAX peptide system has been reported to support
cellular attachment in the absence of recognition motifs or adsorbed serum proteins,26
although the mechanism for this is not clear. Cellular attachment in the absence of serum
proteins is not a requirement for most tissue engineering applications, as serum will
typically be supplied in the cell culture media. AFD36 and AFD49 hydrogels are therefore
potential cell scaffold materials for these applications.
To the candidates knowledge no detailed reports have been made on the effects of preequilibration of peptide hydrogels on cellular proliferation. However, there is the potential
for the effects of pre-equilibration conditions on proliferation rates to be common to a
range of hydrogel systems, especially those in which chemical contaminants are present.
Accounting for these effects during material preparation will thus be required to allow direct
comparison between cell proliferation on different hydrogel systems. Furthermore, if AFD
peptide hydrogels are utilised in future in vivo applications, the effects of pre-treating
peptide material will need to be assessed.
5.3.14
Culture of encapsulated fibroblasts
As discussed in Chapter 1, the culture of cells within a 3D environment can provide
models that better mimic the functions of living tissue.58 The effects of encapsulating cells
within a 3D peptide hydrogel were therefore of interest. As a first test, fibroblasts were
mixed directly into a gelling AFD49 peptide solution. However, encapsulated cells did not
proliferate (not shown), likely due to the transient acidic conditions that occur during
gelation causing cell death. The growth of fibroblasts inside layered AFD49 hydrogels was
then tested as an alternate method to reduce the exposure of cells to acidic conditions.
One benefit of this layered approach is that cells are initially found in a single plane.
Therefore cells visualised outside of this plane give an indication of cellular migration into
the peptide scaffold.
NIH/3T3 fibroblasts were allowed to adhere overnight on either TCPS or AFD49 hydrogels.
Fibroblasts were then either allowed to continue growing in 2D, or had a gelling peptide
solution cast on top to encapsulate the cells in a 3D environment. Cell images were taken
at 4 and 6 days after seeding (Figure 5-15). Fibroblasts on TCPS surfaces that were not
covered by a peptide hydrogel proliferated and reached confluence by 6 days (Figure
211
5-15A). Those on 2D AFD49 hydrogels also proliferated, with some dense clusters forming
by 6 days (Figure 5-15C). Fibroblasts on TCPS with AFD49 gels cast on top underwent a
high degree of cell death, with only small local areas of viable cells (Figure 5-15B). The
remaining fibroblasts were viable and continued to proliferate, however growth was slow
and they maintained a flat morphology, with cells appearing less raised than those on
TCPS alone, possible due to the presence of gel material above them. The fibroblasts
encapsulated inside layered AFD49 hydrogels proliferated, although at a much lower level
than those on 2D hydrogel surfaces (Figure 5-15D). At 6 days some of the encapsulated
cells were also visualised in slightly different focal planes, indicating some vertical
distribution of cells. The encapsulated cells also adopted a more spindle-like morphology,
with some cellular protrusions forming.
The high degree of cell death on TCPS surfaces with hydrogels cast on top may be a
consequence of the mildly-acidic gelling solution causing cell death and detachment. The
lack of extensive cell death observed inside the layered hydrogels upon encapsulation
may be due to the base hydrogel providing sufficient buffering capacity to protect the
fibroblasts from the acidic environment during gelation. The slower growth of fibroblasts
encapsulated within hydrogels is likely a consequence of limited nutrient diffusion through
the gel compared to on a flat exposed surface. Furthermore, for proliferation to occur, the
surrounding hydrogel may need to be degraded or moved to make space for new cells.
At 7 days after seeding, confocal microscopy and live-dead staining was utilised to image
the fibroblasts encapsulated in peptide hydrogels in 3D (Figure 5-16). The fibroblasts had
formed small local clusters of spindle-shaped cells with some extending long processes
into the hydrogel. The local clusters may be due to proliferation of individual cells forming
clusters, or alternately, migrating cells may contact each other and adhere to form clusters.
Side-on views of the fibroblasts show a general slope to the plane in which the cells are
found. This is likely due to the uneven surface of the initial bottom hydrogel due to
meniscus effects when it was cast. Side-on views of the fibroblasts also show some
spreading of cells in the z-axis by ca. 20-30 µm, which is larger than the observed average
fibroblast diameter of ~ 10 µm, potentially indicating the fibroblasts are beginning to grow
into the 3D environment.
212
Figure 5-15. Effects of hydrogel encapsulation on NIH/3T3 fibroblast growth. NIH/3T3 fibroblasts
cultured A) on TCPS, B) on TCPS with AFD49 hydrogel cast on top, C) on AFD49 hydrogel, D) on
AFD49 hydrogel with second gel cast on top. (scale bars 100 µm) Peptide concentration was 2.1
mM in the base gels and 2 mM in the top layered gels.
213
The morphology of fibroblasts embedded in AFD49 hydrogels is similar to those of primary
fibroblasts cultured in 3D environments inside degradable PEG40 and peptide28 hydrogels.
Previous research has indicated the necessity for cell-mediated proteolytic degradation of
chemically cross-linked hydrogel networks to allow remodelling of fibroblasts’ extracellular
environment to create sufficient space for cellular proliferation.40, 59, 60 However these
studies were completed at higher concentrations of scaffold material than used in the
peptide hydrogels reported here. While proteolytic degradation (Section 5.3.5) of the
peptide network may be involved in cellular proliferation, the low solid content and noncovalent nature of the crosslinks of AFD49 hydrogels likely provide a sufficiently low
density and deformable network for proliferation to occur even in the absence of proteolytic
action.
Due to the slow growth of encapsulated fibroblasts it was of interest to observe cell growth
over longer times. Cell images were recorded at days 11, 19 and 28 (Figure 5-17).
Fibroblasts proliferated over time, with increasingly large clusters of cells forming with
large numbers of long processes. At these longer times, more fibroblasts were in different
focal planes, indicating growth in 3D. The growth rate of fibroblasts encapsulated in
AFD36 hydrogels appears qualitatively similar to rates previously reported for
encapsulated fibroblasts in β-sheet peptide30, 31 and synthetic polymer40 hydrogels.
However the growth rate appeared slower than that reported for cells encapsulated within
peptide microgels that had a larger surface area to volume ratio that may increase nutrient
diffusion to the cells.32
Confocal microscopy and live-dead staining at day 42 were used to visualise the long-term
growth of fibroblasts encapsulated inside AFD49 hydrogels in 3D (Figure 5-18). At this
time point, fibroblasts had proliferated to give a large network of cells within the gel. The
number of cells with spindle morphologies had decreased compared to earlier times,
potentially due to cells forming larger assemblies where cell-cell interactions limit the
potential for long process formation. While the majority of the fibroblasts are found in the
plane of the initial hydrogel surface, side-on views show cells to have extended up to ca.
160 µm into the gel. This indicates that while fibroblasts encapsulated within AFD49
hydrogels proliferate slower than on a 2D hydrogel surface, they form 3D networks
extending into the peptide hydrogel. This makes this material a promising candidate for
future use in areas such as wound healing and therapeutic models where the cell
microenvironment and the ability of cells to invade the hydrogel material may be of
importance.
214
Figure 5-16. Confocal microscopy of live-dead staining of NIH/3T3 fibroblasts in layered AFD49
hydrogels at day 7. Images A) top-down, B) and C) side-on. (green, live; red, dead). Peptide
concentration was 2.1 mM in the base gels and 2 mM in the top layered gels.
215
Figure 5-17. Representative images of NIH/3T3 fibroblast growth encapsulated in AFD49
hydrogels at longer times. (scale bars 100 µm). Peptide concentration was 2.1 mM in the base gels
and 2 mM in the top layered gels.
216
Figure 5-18. Confocal microscopy of live-dead staining of NIH/3T3 fibroblasts in layered
AFD49 hydrogels at 42 days. Images A) top-down, B) and C) side-on. (green, live; red, dead).
Peptide concentration was 2.1 mM in the base gels and 2 mM in the top layered gels.
217
5.3.15
Induced pluripotent stem cell growth on AFD49 gels
There has been recent interest shown in induced pluripotent stem cells (iPSCs) for
applications such as toxicity assays61 and regenerative medicine.62 Culture of iPSCs often
utilises a feeder-cell layer to assist in cell proliferation and maintaining their pluripotent
state.63 Recent advances have enabled the replacement of feeder-cell layers with
substitute extracellular matrices such as MatrigelTM.12, 63 While this greatly simplifies the
culture of iPSCs, there still remain issues with these naturally-derived materials such as
batch-to-batch variability, risk of viral contamination and a lack of the ability to modulate
their characteristics.58, 63, 64 Given the demonstrated ability of AFD49 hydrogels to support
mammalian cell growth (Sections 5.3.3-5.3.14), it was of interest to assess the ability of
this defined and non-naturally-derived material to support the culture of iPSCs.
Previous studies have demonstrated the functionalization of surfaces with recombinant
vitronectin to be sufficient for the attachment and proliferation of human embryonic stem
cells (hESCs).65 More recent work has shown that surfaces functionalised with the Nterminal somatomedin B (SMB) domain of vitronectin, which contains an RGD sequence,
are capable of supporting hESC attachment and maintenance in long-term culture.15 In an
attempt to functionalise AFD49 hydrogels for enhanced iPSC attachment and proliferation,
the SMB vitronectin fragment was included in hydrogels at the time of gelling at final
concentrations of 10, 20 and 50 ug mL-1. If evenly distributed throughout the hydrogels this
would result in spacing between RGD ligands of 105, 84 and 62 nm respectively, which is
within the range reported to be required for the attachment of human mesenchymal stem
cells66 and primary human fibroblasts.67 Non-specific incorporation of the vitronectin
fragment into peptide fibrils during gelation was hoped to be sufficient to give
functionalization for cellular attachment. The iPSC line utilised in this experiment was
transformed using the EOS lentiviral vector11 and maintained in single cell culture. The
reporter system engineered into this vector means that viable pluripotent cells express
enhanced green fluorescent protein (EGFP), allowing for simple visual assessment of the
maintenance of viable and undifferentiated cells.
To assess the ability of AFD49 hydrogels to support iPSC growth and the effect of
vitronectin fragment incorporation, iPSCs were cultured for two days on either Matrigel
coated TCPS or AFD49 gels with varied amounts of vitronectin fragment that were preequilibrated for two days in serum-free media. Images of iPSCs on all surfaces were taken
each day (Figure 5-19). Similar to experiments with fibroblasts, rounded iPSCs on
hydrogels were resistant to being washed off and were likely partially embedded in the soft
218
material. Similar to Section 5.3.11, the number of cells adopting a spread morphology was
used to assess cell-substrate interactions. At day 1, ~ 40% of iPSCs on Matrigel coated
TCPS had adopted spread morphologies. On gels of AFD49 alone this value was only ~
20%. Incorporation of the vitronectin fragment did not increase the level of attachment of
iPSCs on peptide hydrogels with ~ 10% of cells adopting spread morphologies on all
vitronectin fragment containing gels. The cells expressing EGFP generally also had spread
morphologies, indicating that the rounded cells are likely undergoing cell death. At day 2
the iPSCs on the Matrigel coated TCPS surface had proliferated, with EGFP expressing
cells forming adherent clusters. iPSCs on all peptide hydrogels adopted rounded
morphologies and the number of EGFP expressing cells decreased, likely indicating cell
death.
The low level of cell spreading followed by cell death on all peptide hydrogels indicated
AFD49 hydrogels are not suitable for the culture of iPSCs, even with the incorporation of
the vitronectin fragment. The initial spread morphology of some iPSCs on peptide
hydrogels at day 1, followed by iPSCs rounding up and undergoing cell death is
unexpected, but may be due to the cells taking some time to respond to the new substrate
following their previous culture on Matrigel coated TCPS. Alternately, the soft peptide
hydrogel material may be too weak to support the cytoskeletal forces of iPSCs, and
following initial attachment, the iPSCs then contract to form rounded cells that do not
further proliferate and in some cases completely detach from the surface. In an attempt to
increase mechanical strength of the hydrogels, a separate experiment assessed iPSC
spreading and proliferation on AFD49 hydrogels where peptide concentration was varied
from 1-3.75 mM. This did not increase cellular spreading at longer than 1 day (unpublished
data), indicating that altering the mechanical properties within the range tested did not
enhance the ability of the peptide hydrogels to support iPSC culture.
Another possibility is that the peptide itself is cytotoxic to the iPSC cells, and causes cell
death on longer exposure, however the lack of cytotoxicity to fibroblasts in previous
experiments makes this unlikely. The removal of Rho-associated kinase (ROCK) inhibitor
from the culture media after 24 hours was also a potential influence in the higher degree of
cell death during the second day of culture; however experiments in which the ROCK
inhibitor was maintained in the media past the first 24 hours showed no decrease in cell
death (not shown).
219
The lower level of cellular spreading on hydrogels containing the vitronectin fragment
compared to unfunctionalised hydrogels was unexpected. This may be due to the
vitronectin fragment not being incorporated into the fibrils during self-assembly and
remaining free in the gel volume. As discussed for the free cyclic-RGD peptide (Section
5.3.12), free vitronectin fragment may bind to receptors on the iPSCs and lower cellular
attachment to other surface adsorbed proteins. While the hydrogels were equilibrated with
media prior to cell seeding, the soaking media was at a five-fold excess to gel volume, and
therefore there still remains the potential for low levels of this protein fragment to remain
within the gel volume and effect cellular attachment.
220
221
Figure 5-19. Growth of iPSCs on AFD49 hydrogels and Matrigel coated TCPS controls. A)
Matrigel coated TCPS, B) AFD49 hydrogels, C) AFD49 hydrogels cast with 10 µg mL-1
vitronectin fragment, D) AFD49 hydrogels cast with 20 µg mL-1 vitronectin fragment, E) AFD49
hydrogels cast with 50 µg mL-1 vitronectin fragment. Each fluorescence image (green, EGFP
fluorescence) is of the same location as the phase contrast image above. (scale bars 100 µm).
Hydrogels were 2 mM AFD49 final.
222
5.3.16
iPSC growth on Matrigel coated hydrogels
Preliminary experiments investigating the ability of AFD49 hydrogels to support iPSC
growth showed a low level of cells initially adopting spread morphologies followed by
iPSCs adopting rounded morphologies and in some cases detaching and undergoing cell
death. The presence of the SMB vitronectin fragment in the hydrogels did not enhance
iPSC spreading (Section 5.3.15). From initial experiments it was unclear whether iPSC
detachment and death resulted from one of several factors such as potential cytotoxicity,
unfavourable mechanical properties or a lack of appropriate cellular recognition motifs. As
a test of the effects of functionalization, iPSC growth was assessed on AFD49 gels coated
with Matrigel.
iPSCs were cultured for 6 days on Matrigel coated TCPS and AFD49 gels with images
taken at days 1, 2, 4 and 6 (Figure 5-20). At day 1 iPSCs were well attached with spread
morphologies and only limited numbers of rounded cells on both AFD49 and TCPS
surfaces. At day 2 the iPSCs had proliferated to form sheets of cells on both surfaces. On
TCPS these sheets were better spread than on hydrogel surfaces, where the iPSCs had
formed areas of high cell density with sparse areas between. The majority of iPSCs on
both TCPS and AFD49 surfaces maintained EGFP expression, indicating viability and
pluripotency. However, small numbers of iPSCs in areas of high cell density lost EGFP
expression and were likely undergoing differentiation. At day 4, iPSCs on TCPS had
reached confluence, with a small number of cells forming cell aggregates and detaching
from the surface at day 6. The majority of cells at days 4 and 6 on TCPS maintained
expression of EGFP, indicating viability and pluripotency. At day 4, iPSCs cultured on
AFD49 had formed dense cell masses that began to detach from the hydrogel surface. At
day 6 iPSCs had detached from the hydrogel surface and formed rounded cellular
aggregates that had diameters of 150-250 µm and resembled spheroids of hESCs and
iPSCs in suspension culture.68 Interestingly, EGFP expression was maintained by iPSCs
in these spheroids, suggesting that iPSCs in the spheroids may still be viable and
pluripotent.
223
Figure 5-20. Growth of iPSCs on Matrigel coated surfaces. A) and B) TCPS, C) and D) AFD49 hydrogels. Each fluorescence image (B and D; Green,
EGFP fluorescence) is of the same location as the phase contrast image above (A and C). (scale bars 100 µm). Hydrogels were 2 mM AFD49 final.
224
Matrigel coating of AFD49 hydrogels is sufficient to promote iPSC attachment, spreading
and proliferation over several days. This suggests that AFD49 hydrogels are not directly
cytotoxic to iPSCs, and furthermore the mechanical properties of these gels are able to
support iPSC cytoskeletal forces, at least at low cell density. Matrigel functionalization of
hydrogels is not optimal for future work as it does not avoid this naturally-sourced material.
However, future experimentation may elucidate appropriate recognition motifs for
functionalization of AFD49 hydrogels to achieve a defined matrix capable of supporting
iPSC culture. The detachment of iPSCs at days 4 and 6 appears mechanistically similar to
the detachment of fibroblasts noted previously (Sections 0). This may relate to the iPSCs
forming strong cell-cell attachments as the density of cells increases, which eventually
overcome the attachment to the hydrogel surface. This may be coupled with iPSCs
generating their own ECM over time, which detaches from the hydrogel surface.
Additionally, AFD49 gels likely possess similar proteolytic degradability (Section 5.3.5) and
low mechanical strength (Section 4.4.14) to AFD36 gels, which may allow for cells to
progressively remodel and weaken the hydrogel surface, leading to detachment. Further
experiments may focus on increasing the mechanical strength of peptide hydrogels
through either covalent cross-linking or increasing peptide concentration. Embedding of
iPSCs inside an appropriately functionalised peptide hydrogel may also be of interest, as
this would prevent iPSCs from detaching from hydrogel surfaces.
5.3.17
Drug loading into peptide hydrogels
Vitamin A and its derivatives such as all-trans retinoic acid (ATRA; Figure 5-21A) play
important roles in both embryonic development and adult cell differentiation.69 The
influence of ATRA and its most common isomers 9-cis retinoic acid and 13-cis retinoic acid
(Figure 5-21B) on cell maturation has resulted in its use in treatment of a number of clinical
disorders such as certain leukaemias70 and acne.71 It is also of interest for stem-cell
research, as it affects the differentiation of pluripotent cell lines.72, 73 Due to the
hydrophobic and polyenic structure of retinoic acids, they possess low solubility in
aqueous solutions and undergo isomerization under exposure to light, making them
difficult to work with.74, 75 As discussed in Chapter 1, a potential delivery vehicle for
hydrophobic molecules such as retinoic acids is the hydrophobic cavity formed in the core
of higher-order coiled coils.9 For example, Guo et al.76 have shown the pentameric coiledcoil domain of cartilage oligomerization matrix protein to bind ATRA within its hydrophobic
core. It was therefore of interest to assess the ability of AFD49 to bind ATRA with the
225
Figure 5-21. Chemical structure of A) ATRA and B) 13-cis retinoic acid.
potential for peptide hydrogels containing this retinoid to act as a slow-release reservoir for
cell culture and possible clinical applications.
LCMS allowed quantification of both all-trans retinoic acid and the most common isomer
13-cis retinoic acid. The solubility of ATRA was measured to be ~ 5-10 µM in water or
phosphate-buffered saline alone, while 290 µM ATRA was dissolved in an aqueous
solution of 7.5 mM AFD49. This demonstrated the ability of AFD49 to solubilise this
hydrophobic molecule.
ATRA is non-chiral, and therefore will not generally exhibit a spectrum via circular
dichroism. However, when bound in a chiral environment such as a protein binding site,
induced circular dichroism may be observed at wavelengths close to the chromophore’s
absorption maxima of 350 nm (in ethanol).77, 78 ECD spectra of a solution of ATRA
solubilised by AFD49 were collected over 250-450 nm (Figure 5-22). Peptide secondary
structure is generally measured over 190-260 nm, and in the wavelength range used for
measurement of ATRA shows minimal contribution. Additionally, under the solution
conditions used for this experiment AFD49 is in a fully helical state (Section 5.3.7), and no
change in peptide secondary structure was observed on ATRA addition.
226
ATRA solubilised by AFD49 gave a weak spectrum with maxima at 283 and 385 nm and a
minimum at 329 nm, indicating some interaction with the peptide that did not change over
time when the sample was kept in the dark. Upon exposure of the solution to room light for
15 minutes the spectrum changed, with maxima at 293 and 372 nm and minima at 335
and 427 nm. LCMS analysis showed a small amount of isomerization of ATRA, with 0.6 %
of detected retinoic acid now in the 13-cis form. Exposure of ATRA samples to light also
gave decreasing total amounts of retinoic acid isomers, indicating some degradation to
alternate products not detected by the method used here. Continued exposure of the
ATRA-AFD49 solution to room light gave further changes to the ECD spectra. After 17
hours the spectra showed a single maximum at 307 nm and a minimum at 397 nm. LCMS
analysis at this point in time showed 68% of the initial retinoic acid to remain in the alltrans state, with 4% of the initial retinoic acid now present as the 13-cis isomer. Extremely
low concentrations of 9-cis retinoic acid and an unassigned isomer were also detected.
After 4 days of light exposure only ca. 1% of initial retinoic acid is present as either ATRA
Figure 5-22. ECD of ATRA in AFD49 solution on exposure to room light. ATRA initial
concentration was 290 µM, AFD49 concentration was 7.5 mM. Spectra collected at times as
indicated by colour legend.
227
or 13-cis retinoic acid. No circular dichroism signal corresponding to retinoic acid was
observed. This indicated near complete degradation of the retinoic acid under prolonged
exposure to room light. The initial observation of increasing circular dichroism signal on
exposure of ATRA to light indicates that it may be an isomer or other degradation product
of ATRA that binds within a chiral environment of AFD49. Further studies using techniques
such as nuclear magnetic resonance spectroscopy may determine the identity and precise
binding location of this species within AFD49 coiled coils. Notably, ATRA was solubilised
by AFD49, indicating it may be useful for drug-delivery applications.
5.3.18
ATRA release from peptide hydrogels
The release of ATRA from AFD49 hydrogels was investigated under conditions similar to
those that would be used for tissue culture experiments. Gelation of an AFD49-ATRA
solution on addition of bicarbonate containing 2 × DMEM proceeded with similar kinetics to
when ATRA was omitted. No attempt was made to characterise the rheological properties
of ATRA containing AFD49 hydrogels due to difficulties in handling this light sensitive
compound. However, qualitatively the gels behaved similarly to hydrogels formed with the
same concentration of peptide in the absence of ATRA.
1 mM AFD49 hydrogels containing 29 µM ATRA were incubated in the dark at 37 °C in a
48-well plate with culture medium for 3 days with media changes each day. The amount of
ATRA in the recovered media was quantified at each day. After 1 day ATRA was
measured in the soaking media at 4 µM, equating to 64% of the total loaded ATRA. At 2
days only 1 µM ATRA was present in the soaking media, giving a cumulative release of
82%. This increased to 87% at day 3, and no further release was observed when media
was sampled on later days. At each day the concentration of retinoic acid isomers other
than ATRA were too low to quantify using the LCMS method employed here.
The total recovery of only 87% of loaded ATRA may correspond to degradation to low
levels of isomers and other products not detected by the LCMS method employed here.79,
80
Alternately, some ATRA may remain bound in the hydrophobic core of AFD49 coiled-coil
fibrils. This experiment indicated successful loading and release of ATRA from AFD49
hydrogels. However, the uncontrolled and rapid release of ATRA over the first 24 hours
may be faster than is desirable for many applications relating to cell differentiation and
potential cancer therapies where it is commonly used in the nM to low µM range over
several days.73, 81, 82 Lowering the loaded concentration of ATRA may achieve the desired
228
low therapeutic concentrations over longer times, however the initial burst release seen in
this trial may be unfavourable for some applications.
The ability of hydrophobic molecules to bind within the coiled-coil core is dependent on
factors such as their charge, molecular size, shape and hydrophobicity.9 Therefore, while
this initial experiment demonstrates the ability for AFD49 hydrogels to be loaded with
hydrophobic drugs for later release, alternate drug molecules may be investigated that
could show enhanced binding and slower release rates. Alternately, it may be possible to
redesign the peptide structure to better accommodate various hydrophobic molecules. In
either case, more detailed structural information of the packing of hydrophobic molecules
within coiled-coil fibrils will be required to allow directed peptide modifications and
hydrophobic molecule selections.
5.4 Conclusions
In this chapter α-helical peptide hydrogels were investigated as cellular scaffolds and drug
delivery vehicles. Oscillatory shear rheology results showed AFD36 hydrogels to possess
soft-solid properties similar to soft tissues and previously reported β-sheet peptide
hydrogels. This suggested the hydrogels to be suitable as matrices for mammalian cell
growth. However, hydrogels formed from as-received peptide material were unable to
support cell growth and caused cell death. Peptide solutions contained absorbing species
proposed to be contaminants remaining from peptide synthesis. A purification method was
developed that allowed removal of these contaminants and gave AFD36 peptide material
able to support NIH/3T3 fibroblast growth at levels similar to TCPS. This highlights the
potential for contaminants present in peptide materials to affect biological experiments, a
factor that may be more widespread than is currently recognised.
The long-term growth of NIH/3T3 fibroblast cells on AFD36 hydrogels showed cellular
aggregation with eventual detachment that was attributed to cell-cell interactions
overcoming cell-hydrogel attachments at higher cell densities. AFD36 was susceptible to
trypsin digestion, with progressive loss of secondary structure over time and a loss of gel
structure after 18 hours. While protease susceptibility will need to be further investigated
using secreted proteases, low levels of proteolytic susceptibility of the hydrogel may be an
advantage in wound-healing applications where gradual degradation of the gel and
replacement with native ECM may enhance tissue regeneration.
229
The peptide AFD49 was then designed based on the self-assembly model proposed for
AFD19 and AFD36. This peptide showed hydrogel formation under similar conditions to
AFD36 and supported NIH/3T3 fibroblast growth. The pre-equilibration of AFD49
hydrogels in cell culture media enhanced subsequent cell growth. This was attributed to a
combination of the removal of residual contaminants that are detrimental to cell growth,
and the deposition of serum components on or in the hydrogel. Extremely low fibroblast
attachment to hydrogel surfaces occurred in the absence of serum, indicating that
attachment to this material, which lacks cellular recognition motifs, is likely mediated by
adsorbed serum components. The proliferation of fibroblasts encapsulated within AFD49
hydrogels demonstrated these materials to be suitable as 3D cell scaffold matrices. This
indicates AFD peptide hydrogels may have potential in wound healing applications, as well
as demonstrating potential utility in applications such as toxicological assays, where the
cellular microenvironment can affect drug susceptibilities.
The culture of iPSCs on AFD49 hydrogels was investigated to determine the ability of
these gels to support the growth of alternate cell types. Initial experimentation showed
both unfunctionalised peptide hydrogels and hydrogels incorporating the SMB vitronectin
domain to be unable to support iPSC attachment and proliferation. However, coating of
gels with Matrigel gave sufficient functionalization for iPSC attachment and growth over
several days. These experiments demonstrate that when appropriately functionalised, the
peptide hydrogels may function as a substrate for iPSCs. However future experimentation
will be required to determine appropriate specific recognition motifs to allow iPSC culture
on a completely defined substrate. Similar to the results seen with 3T3/NIH fibroblasts,
iPSCs showed detachment from peptide hydrogel surfaces at longer times. Increasing the
mechanical strength of peptide hydrogels may overcome this cellular detachment at higher
cell densities by providing a more rigid substrate.
ATRA was solubilised by AFD49, demonstrating the potential for hydrophobic drug loading
and delivery. The release of ATRA after loading into peptide hydrogels occurred rapidly
over the first two days. This is likely too fast for most applications, however varying drug
concentration may be able to influence release rates. Further screening of potential
hydrophobic drugs as well as possibly redesigning hydrogel forming peptides will
potentially increase the future utility of α-helical peptide hydrogel drug delivery systems.
The results of this chapter indicate that the peptide hydrogels developed in Chapters 4 and
5 support mammalian cell growth and solubilise hydrophobic molecules, thereby
230
demonstrating their potential in applications such as tissue engineering, wound healing
and drug delivery. There are several directions for future research that may be of benefit
for further developing this class of α-helical peptide hydrogel that are now discussed.
5.5 Future directions
5.5.1 Characterisation of fibril forming peptides
As discussed in Chapter 4, future studies of the self-assembling AFD peptides may focus
on further characterising the assembled state of AFD19, AFD36 and AFD49 peptides. This
may focus on better determining the oligomerization state of the coiled coil formed by
these peptides using techniques such as sedimentation equilibrium ultracentrifugation.
Characterisation of coiled-coil fibrils by techniques such as higher resolution electron
microscopy and X-ray scattering studies may also serve to better define the peptides
assembled state and enhance the understanding of these peptides self-assembly. This
may then allow for a more targeted peptide design process for future applications.
5.1.1 Hydrogel forming peptides as cell scaffold materials
The investigation of mammalian cell growth on peptide hydrogels reported in this thesis
demonstrated the suitability of these materials as cell scaffolds. However, a full exploration
of the many factors involved in the use of the peptide hydrogels as defined matrices for
cell growth will require further studies. Future studies will likely involve investigation of the
effects of hydrogel mechanical properties, as well as approaches for hydrogel
functionalization to support more demanding cell types such as iPSCs.
Chapter 5 demonstrated hydrogels formed by the peptides AFD36 and AFD49 to support
the growth of NIH/3T3 fibroblasts. However at higher cell densities fibroblasts began to
detach from the gel surface, likely in part due to the gels weak mechanical properties.
Future studies may investigate methods for increasing the mechanical strength of peptide
hydrogels. As discussed in Chapter 1, this may be achieved by increasing the peptide
concentration to give a denser fibril network. Further rheological studies would be required
to determine the precise relationship between peptide concentration and hydrogel
mechanical properties. Alternately, chemical cross-linking of peptide hydrogels may be
investigated to give a more rigid material. This could be accomplished by the addition of
cross-linking agents such as glutaraldehyde, or the incorporation of cysteine residues to
form disulfide cross-links. While chemical cross-linking may increase hydrogel mechanical
strength and potentially enhance the utility of peptide hydrogels as tissue culture
231
substrates, these cross-links will impair the self-healing properties of peptide hydrogels. In
applications where it is desirable for hydrogel material to be shear-thinned for injection,
such chemical cross-links would therefore have to be formed in situ. Control over the
mechanical properties of peptide hydrogels will be of use in future applications, and
studies such as those outlined here will be required to assess the variety of methods
available.
The mechanism involved in cellular detachment from peptide hydrogels at higher cell
densities may be of interest for future studies. For example, if the cells on the surface of
hydrogels are producing their own ECM, which then detaches from the hydrogel surface,
this may be visualised by immunostaining of ECM components.38 Such progressive
replacement of peptide hydrogel with native ECM may be beneficial for wound healing
applications, where native tissue would eventually replace the hydrogel. Additionally, a
better understanding of the cellular behaviour on hydrogel surfaces may be obtained if
cellular migration and detachment were visualised using time-lapse microscopy studies.
This would provide a better understanding of the processes by which cells migrate on, and
eventually detach from, hydrogel surfaces and potentially provide more general insights
into cellular behaviour on such soft substrates.
Layered AFD49 hydrogels were demonstrated to support the growth of encapsulated
fibroblasts (Chapter 5). However, directly mixing cells into a gelling peptide solution gave
high cell death, likely due to the transient acidic conditions. Future studies may determine
improved methods for encapsulating cells within peptide hydrogels. This may involve
titrating the starting peptide solution to a less acidic pH prior to inducing gelation and
seeding cells, or increasing the bicarbonate concentration to achieve faster gelation and
lessen the exposure of cells to acidic conditions. Alternately, it may be possible to mix cells
into sheer-thinned peptide hydrogels, however obtaining a homogenous distribution may
be challenging. Future work may explore these potential approaches for obtaining a
homogenous distribution of cells through a hydrogel matrix.
Non-functionalized AFD49 hydrogels were unable to support the growth of iPSCs (Chapter
5). While coating the hydrogels with Matrigel was sufficient to provide iPSC attachment
and proliferation, this is not optimal for future work as this does not avoid the naturallyderived Matrigel. Future research may determine methods for peptide functionalization to
give a defined substrate capable of supporting iPSC attachment and growth. This could be
accomplished by the incorporation of an integrin recognition sequence containing peptide
232
into peptide hydrogels. The design of such a peptide may involve appending a recognition
sequence via a flexible linker to the termini of a gel forming peptide to allow the recognition
sequence to extend laterally from fibrils, similar to the structure proposed by Potekhin et
al..83 An alternate approach to functionalization may involve covalent attachment of
integrin-recognition sequence peptides to side-chains of hydrogel forming peptides. This
may use similar chemistries to those used by the Woolfson group to synthesize branched
fibril-forming peptides.84 This would allow for a variety of alternate integrin recognition
sequences to be attached to a single hydrogel forming peptide without the need for
peptide redesign. However, such an approach would be be incompatible with potential
future bioproduction of hydrogel forming peptides. Future work may explore these two
approaches for the functionalization of hydrogel forming peptides to better support the
attachment and proliferation of more demanding cell types such as iPSCs.
5.5.2 Hydrogel forming peptides for therapeutic delivery
Results reported in Chapter 5 demonstrated the ability of AFD49 hydrogels to incorporate
and release the hydrophobic therapeutic ATRA. While this proof of concept system
showed rapid release of loaded therapeutic, future studies may investigate alternate
hydrophobic molecules that may display slower release rates. It would also be beneficial to
use techniques such as nuclear magnetic resonance spectroscopy to better determine the
molecular structure of the AFD49-ATRA complex. This would enable targeted redesign of
the peptide to alter interactions with this therapeutic. Future studies may also investigate
the loading and release of therapeutic biomolecules such as antibodies from peptide
hydrogels. While the results presented in this thesis demonstrate the potential for use of
AFD49 hydrogels as therapeutic delivery vehicles, studies such as those outlined here will
be required in the future to tailor release rates and expand the use of these materials to
other therapeutics.
5.5.3 Bioproduction of hydrogel forming peptides
The results reported in Chapter 5 show potential utility of AFD49 hydrogels as both cell
scaffold materials and therapeutic delivery vehicles. As discussed in Chapter 1.4, highvalue products such as pharmaceuticals and materials for biomedical applications may be
feasibly produced using high-cost processes such as solid-phase peptide synthesis. With
this in mind there is less of a requirement for the bioproduction of hydrogel forming
peptides in comparison to surfactant peptides that find utility in large-scale applications.
However, in the future it may be beneficial to reduce the costs associated with producing
233
these materials. This may be achieved using a bioproduction approach similar to that
described in Chapters 2 and 3, applied to hydrogel forming peptides. A concatemer
approach based on the peptide AFD49 will be presented here; although many similar
approaches are easily envisioned and it is still difficult to predict which will be most
successful. Notably, given that the target peptide AFD49 does not possess a high level of
charge under physiological conditions, it is likely that a homoconcatemer approach rather
than a heteroconcatemer (as used in Chapter 2) may be appropriate. Furthermore, it is
proposed that by joining the peptides at their termini with linkers as well as constraining
coiled-coil assembly to an antiparallel arrangement, gel formation during expression may
be avoided and will only occur after concatemer cleavage.
A concatemer approach similar to that described in Chapters 2 and 3 will utilise acid
cleavage at aspartate-proline dipeptides to release product peptides following expression.
As described in Chapter 3, if cleavage is allowed to continue to completion, removal of Cterminal aspartate residues may be achieved. This could be accounted for in the design of
a bioproduction method for a hydrogel forming peptide. One notable difference between
synthetically produced and expressed peptides is capped vs uncapped termini
respectively, which will effect the level of charge present on the peptide. A target peptide
for expression based on AFD49 may therefore be uAFD49 (H2N-P LKELAKL LHELAKL
LHELAKL-COOH, FW 2521). This peptide possesses the same helix-foming core as
AFD49 with an N-terminal proline residue for cleavage as well as uncapped termini. This
peptide sequence possesses a GRAVY value of 0.250 and an instability index of -5.89,
suggesting a soluble and stable sequence, similar to AFD49. The predicted charge on
uAFD49 approaches +1 as the pH is raised above 7.0, in line with requirements for gel
formation under physiological conditions. The predicted pH for 0 net charge for uAFD49 is
9.0, which while lower than AFD49, is still well outside the physiological range. Future
experiments will be required to determine the gelation behaviour of this peptide, however it
may be expected that such a peptide would undergo gelation similarly to the previous
peptides AFD36 and AFD49.
A linker sequence (L) similar to the one successfully used in Het2-6 (PGRGMD) could be
used to link repeats of uAFD49 within a concatemer structure. In this case, it may be
beneficial to substitute the cationic arginine with an anionic glutamate residue (PGEGMD)
to increase negative charge at physiological pH and avoid pI values close to physiological
conditions. Also similar to the approach used in Chapter 2, an N-terminal methionineaspartate dipeptide (M) would be required to initiate expression. This would result in a
234
concatemer with sequence (M-(uAFD49)-(L-(uAFD49))n) where n+1 is the total number of
uAFD49 repeats. As described in Chapter 1, the number of peptide repeats is a factor
which will likely need empirical optimization. Given the higher-order coiled coil proposed to
be formed by AFD49, it seems likely that a concatemer of approximately 6 (n=5) repeats
may be appropriate. For example, the sequence (M-(uAFD49)-(L-(uAFD49))5) has a
predicted pI of 5.5, which while higher than Het2-6, is comparable to the majority of native
bacterial proteins (Chapter 2). This sequence also has a GRAVY value of -0.159 and an
instability index of -4.26, suggesting a soluble and stable sequence. The predicted charge
at pH 7.0 of this sequence is -9.7, which is higher than Het2-6. However the computed
Foldindex value for this sequence is +0.127, suggesting a stably folded sequence. A
series of concatemers where the n value is varied between 4 and 6 may be investigated to
determine the optimal value for expression.
Following expression, heat and acid-mediated cleavage of the aspartate-proline dipeptides
and subsequent removal of C-terminal aspartate residues could be used to produce the
desired product peptide. Following cleavage it is likely that a chromatography approach
may be required to recover product peptides. However, given the high-value potential
applications of these materials this is likely to be a feasible approach.
As stated above it is still difficult to accurately predict the success of bioproduction
strategies due to the level of complexity involved. However a bioproduction strategy similar
to that outlined here is one potential future route to achieving bioproduction of hydrogel
forming designer peptides.
235
5.6 References
1.
Maude, S.; Ingham, E.; Aggeli, A. Biomimetic Self-Assembling Peptides as
Scaffolds for Soft Tissue Engineering. Nanomed. 2013, 8, 823-847.
2.
Matson, J. B.; Stupp, S. I. Self-Assembling Peptide Scaffolds for Regenerative
Medicine. Chem. Commun. 2012, 48, 26-33.
3.
Banwell, E. F.; Abelardo, E. S.; Adams, D. J.; Birchall, M. A.; Corrigan, A.; Donald,
A. M.; Kirkland, M.; Serpell, L. C.; Butler, M. F.; Woolfson, D. N. Rational Design and
Application of Responsive Alpha-Helical Peptide Hydrogels. Nat. Mater. 2009, 8, 596-600.
4.
Liu, J.; Zheng, Q.; Deng, Y. Q.; Cheng, C. S.; Kallenbach, N. R.; Lu, M. A SevenHelix Coiled Coil. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 15457-15462.
5.
Xu, C. F.; Liu, R.; Mehta, A. K.; Guerrero-Ferreira, R. C.; Wright, E. R.; DuninHorkawicz, S.; Morris, K.; Serpell, L. C.; Zuo, X. B.; Wall, J. S.; Conticello, V. P. Rational
Design of Helical Nanotubes from Self-Assembly of Coiled-Coil Lock Washers. J. Am.
Chem. Soc. 2013, 135, 15565-15578.
6.
Zaccai, N. R.; Chi, B.; Thomson, A. R.; Boyle, A. L.; Bartlett, G. J.; Bruning, M.;
Linden, N.; Sessions, R. B.; Booth, P. J.; Brady, R. L.; Woolfson, D. N. A De Novo Peptide
Hexamer with a Mutable Channel. Nat. Chem. Biol. 2011, 7, 935-941.
7.
Liu, J.; Zheng, Q.; Deng, Y. Q.; Kallenbach, N. R.; Lu, M. Conformational Transition
between Four and Five-Stranded Phenylalanine Zippers Determined by a Local Packing
Interaction. J. Mol. Biol. 2006, 361, 168-179.
8.
Harbury, P. B.; Zhang, T.; Kim, P. S.; Alber, T. A Switch between Two-, Three-, and
Four-Stranded Coiled Coils in Gcn4 Leucine Zipper Mutants. Science 1993, 262, 14011407.
9.
McFarlane, A. A.; Orriss, G. L.; Stetefeld, J. The Use of Coiled-Coil Proteins in Drug
Delivery Systems. Eur. J. Pharmacol. 2009, 625, 101-107.
10.
Briggs, J. A.; Sun, J.; Shepherd, J.; Ovchinnikov, D. A.; Chung, T. L.; Nayler, S. P.;
Kao, L. P.; Morrow, C. A.; Thakar, N. Y.; Soo, S. Y.; Peura, T.; Grimmond, S.; Wolvetang,
E. J. Integration-Free Induced Pluripotent Stem Cells Model Genetic and Neural
Developmental Features of Down Syndrome Etiology. Stem Cells 2013, 31, 467-478.
11.
Hotta, A.; Cheung, A. Y. L.; Farra, N.; Vijayaragavan, K.; Seguin, C. A.; Draper, J.
S.; Pasceri, P.; Maksakova, I. A.; Mager, D. L.; Rossant, J.; Bhatia, M.; Ellis, J. Isolation of
Human Ips Cells Using Eos Lentiviral Vectors to Select for Pluripotency. Nat. Meth. 2009,
6, 370-376.
12.
Xu, C. H.; Inokuma, M. S.; Denham, J.; Golds, K.; Kundu, P.; Gold, J. D.;
Carpenter, M. K. Feeder-Free Growth of Undifferentiated Human Embryonic Stem Cells.
Nat. Biotechnol. 2001, 19, 971-974.
13.
Watanabe, K.; Ueno, M.; Kamiya, D.; Nishiyama, A.; Matsumura, M.; Wataya, T.;
Takahashi, J. B.; Nishikawa, S.; Nishikawa, S.; Muguruma, K.; Sasai, Y. A Rock Inhibitor
Permits Survival of Dissociated Human Embryonic Stem Cells. Nat. Biotechnol. 2007, 25,
681-686.
14.
Mollamohammadi, S.; Taei, A.; Pakzad, M.; Totonchi, M.; Seifinejad, A.; Masoudi,
N.; Baharvand, H. A Simple and Efficient Cryopreservation Method for Feeder-Free
Dissociated Human Induced Pluripotent Stem Cells and Human Embryonic Stem Cells.
Hum. Reprod. 2009, 24, 2468-2476.
15.
Prowse, A. B. J.; Doran, M. R.; Cooper-White, J. J.; Chong, F.; Munro, T. R.;
Fitzpatrick, J.; Chung, T. L.; Haylock, D. N.; Gray, P. P.; Wolvetang, E. J. Long Term
Culture of Human Embryonic Stem Cells on Recombinant Vitronectin in Ascorbate Free
Media. Biomaterials 2010, 31, 8281-8288.
16.
Jones, K. H.; Senft, J. A. An Improved Method to Determine Cell Viability by
Simultaneous Staining with Fluorescein Diacetate Propidium Iodide. J. Histochem.
Cytochem. 1985, 33, 77-79.
236
17.
Peyton, S. R.; Ghajar, C. M.; Khatiwala, C. B.; Putnam, A. J. The Emergence of
Ecm Mechanics and Cytoskeletal Tension as Important Regulators of Cell Function. Cell
Biochem. Biophys. 2007, 47, 300-320.
18.
Sieminski, A. L.; Was, A. S.; Kim, G.; Gong, H.; Kamm, R. D. The Stiffness of
Three-Dimensional Ionic Self-Assembling Peptide Gels Affects the Extent of Capillary-Like
Network Formation. Cell Biochem. Biophys. 2007, 49, 73-83.
19.
Stevenson, M. D.; Piristine, H.; Hogrebe, N. J.; Nocera, T. M.; Boehm, M. W.; Reen,
R. K.; Koelling, K. W.; Agarwal, G.; Sarang-Sieminski, A. L.; Gooch, K. J. A SelfAssembling Peptide Matrix Used to Control Stiffness and Binding Site Density Supports
the Formation of Microvascular Networks in Three Dimensions. Acta Biomater. 2013, 9,
7651-7661.
20.
Rowlands, A. S.; George, P. A.; Cooper-White, J. J. Directing Osteogenic and
Myogenic Differentiation of Mscs: Interplay of Stiffness and Adhesive Ligand Presentation.
Am. J. Physiol. Cell Physiol. 2008, 295, C1037-C1044.
21.
Collier, J. H.; Rudra, J. S.; Gasiorowski, J. Z.; Jung, J. P. Multi-Component
Extracellular Matrices Based on Peptide Self-Assembly. Chem. Soc. Rev. 2010, 39, 34133424.
22.
Ozbas, B.; Rajagopal, K.; Schneider, J. P.; Pochan, D. J. Semiflexible Chain
Networks Formed Via Self-Assembly of Beta-Hairpin Molecules. Phys. Rev. Lett. 2004, 93,
268106
23.
Cunha, C.; Panseri, S.; Villa, O.; Silva, D.; Gelain, F. 3d Culture of Adult Mouse
Neural Stem Cells within Functionalized Self-Assembling Peptide Scaffolds. Int. J.
Nanomedicine 2011, 6, 943-955.
24.
Haines-Butterick, L.; Rajagopal, K.; Branco, M.; Salick, D.; Rughani, R.; Pilarz, M.;
Lamm, M. S.; Pochan, D. J.; Schneider, J. P. Controlling Hydrogelation Kinetics by Peptide
Design for Three-Dimensional Encapsulation and Injectable Delivery of Cells. Proc. Natl.
Acad. Sci. U. S. A. 2007, 104, 7791-7796.
25.
Ozbas, B.; Kretsinger, J.; Rajagopal, K.; Schneider, J. P.; Pochan, D. J. SaltTriggered Peptide Folding and Consequent Self-Assembly into Hydrogels with Tunable
Modulus. Macromolecules 2004, 37, 7331-7337.
26.
Kretsinger, J. K.; Haines, L. A.; Ozbas, B.; Pochan, D. J.; Schneider, J. P.
Cytocompatibility of Self-Assembled Ss-Hairpin Peptide Hydrogel Surfaces. Biomaterials
2005, 26, 5177-5186.
27.
Haines, L. A.; Rajagopal, K.; Ozbas, B.; Salick, D. A.; Pochan, D. J.; Schneider, J.
P. Light-Activated Hydrogel Formation Via the Triggered Folding and Self-Assembly of a
Designed Peptide. J. Am. Chem. Soc. 2005, 127, 17025-17029.
28.
Zhou, M.; Smith, A. M.; Das, A. K.; Hodson, N. W.; Collins, R. F.; Ulijn, R. V.;
Gough, J. E. Self-Assembled Peptide-Based Hydrogels as Scaffolds for AnchorageDependent Cells. Biomaterials 2009, 30, 2523-2530.
29.
Storrie, H.; Guler, M. O.; Abu-Amara, S. N.; Volberg, T.; Rao, M.; Geiger, B.; Stupp,
S. I. Supramolecular Crafting of Cell Adhesion. Biomaterials 2007, 28, 4608-4618.
30.
Kyle, S.; Felton, S. H.; McPherson, M. J.; Aggeli, A.; Ingham, E. Rational Molecular
Design of Complementary Self-Assembling Peptide Hydrogels. Adv. Healthcare Mater.
2012, 1, 640-645.
31.
Maude, S.; Miles, D. E.; Felton, S. H.; Ingram, J.; Carrick, L. M.; Wilcox, R. K.;
Ingham, E.; Aggeli, A. De Novo Designed Positively Charged Tape-Forming Peptides:
Self-Assembly and Gelation in Physiological Solutions and Their Evaluation as 3d Matrices
for Cell Growth. Soft Matter 2011, 7, 8085-8099.
32.
Tian, Y. F.; Devgun, J. M.; Collier, J. H. Fibrillized Peptide Microgels for Cell
Encapsulation and 3d Cell Culture. Soft Matter 2011, 7, 6005-6011.
237
33.
Cornish, J.; Callon, K. E.; Lin, C. Q. X.; Xiao, C. L.; Mulvey, T. B.; Cooper, G. J. S.;
Reid, I. R. Trifluoroacetate, a Contaminant in Purified Proteins, Inhibits Proliferation of
Osteoblasts and Chondrocytes. Am. J. Physiol. Endocrinol. Metab. 1999, 277, E779-E783.
34.
Ryan, D. M.; Doran, T. M.; Nilsson, B. L. Stabilizing Self-Assembled Fmoc-F5-Phe
Hydrogels by Co-Assembly with Peg-Functionalized Monomers. Chem. Commun. 2011,
47, 475-477.
35.
Thygesen, M. B.; Sorensen, K. K.; Clo, E.; Jensen, K. J. Direct Chemoselective
Synthesis of Glyconanoparticles from Unprotected Reducing Glycans and Glycopeptide
Aldehydes. Chem. Commun. 2009, 6367-6369.
36.
Giano, M. C.; Pochan, D. J.; Schneider, J. P. Controlled Biodegradation of SelfAssembling Beta-Hairpin Peptide Hydrogels by Proteolysis with Matrix Metalloproteinase13. Biomaterials 2011, 32, 6471-6477.
37.
Horii, A.; Wang, X. M.; Gelain, F.; Zhang, S. G. Biological Designer Self-Assembling
Peptide Nanofiber Scaffolds Significantly Enhance Osteoblast Proliferation, Differentiation
and 3-D Migration. PLoS One 2007, 2, e190.
38.
Kumada, Y.; Zhang, S. G. Significant Type I and Type Iii Collagen Production from
Human Periodontal Ligament Fibroblasts in 3d Peptide Scaffolds without Extra Growth
Factors. PLoS One 2010, 5, e10305.
39.
Cao, Y.; Croll, T. I.; Lees, J. G.; Tuch, B. E.; Cooper-White, J. J. Scaffolds, Stem
Cells, and Tissue Engineering: A Potent Combination! Aust. J. Chem. 2005, 58, 691-703.
40.
Bott, K.; Upton, Z.; Schrobback, K.; Ehrbar, M.; Hubbell, J. A.; Lutolf, M. P.; Rizzi, S.
C. The Effect of Matrix Characteristics on Fibroblast Proliferation in 3d Gels. Biomaterials
2010, 31, 8454-8464.
41.
Olsen, J. V.; Ong, S. E.; Mann, M. Trypsin Cleaves Exclusively C-Terminal to
Arginine and Lysine Residues. Mol. Cell. Proteomics 2004, 3, 608-614.
42.
Guruprasad, K.; Reddy, B. V. B.; Pandit, M. W. Correlation between Stability of a
Protein and Its Dipeptide Composition - a Novel Approach for Predicting in Vivo Stability of
a Protein from Its Primary Sequence. Protein Eng. 1990, 4, 155-161.
43.
Toriseva, M.; Kahari, V. M. Proteinases in Cutaneous Wound Healing. Cell. Mol.
Life Sci. 2009, 66, 203-224.
44.
Nicodemus, G. D.; Bryant, S. J. Cell Encapsulation in Biodegradable Hydrogels for
Tissue Engineering Applications. Tissue Eng. Part B-Rev 2008, 14, 149-165.
45.
Fletcher, N. L.; Lockett, C. V.; Dexter, A. F. A Ph-Responsive Coiled-Coil Peptide
Hydrogel. Soft Matter 2011, 7, 10210-10218.
46.
Kyte, J.; Doolittle, R. F. A Simple Method for Displaying the Hydropathic Character
of a Protein. J. Mol. Biol. 1982, 157, 105-132.
47.
O'Neil, K. T.; Degrado, W. F. A Thermodynamic Scale for the Helix-Forming
Tendencies of the Commonly Occurring Amino Acids. Science 1990, 250, 646-651.
48.
Scholtz, J. M.; Qian, H.; York, E. J.; Stewart, J. M.; Baldwin, R. L. Parameters of
Helix-Coil Transition Theory for Alanine-Based Peptides of Varying Chain Lengths in
Water. Biopolymers 1991, 31, 1463-1470.
49.
Pierschbacher, M. D.; Ruoslahti, E. Cell Attachment Activity of Fibronectin Can Be
Duplicated by Small Synthetic Fragments of the Molecule. Nature 1984, 309, 30-33.
50.
Aumailley, M.; Gerl, M.; Sonnenberg, A.; Deutzmann, R.; Timpl, R. Identification of
the Arg-Gly-Asp Sequence in Laminin-a Chain as a Latent Cell-Binding Site Being
Exposed in Fragment P1. FEBS Lett. 1990, 262, 82-86.
51.
Staatz, W. D.; Fok, K. F.; Zutter, M. M.; Adams, S. P.; Rodriguez, B. A.; Santoro, S.
A. Identification of a Tetrapeptide Recognition Sequence for the Alpha-2-Beta-1-Integrin in
Collagen. J. Biol. Chem. 1991, 266, 7363-7367.
52.
Smith, J. W.; Cheresh, D. A. The Arg-Gly-Asp Binding Domain of the Vitronectin
Receptor - Photoaffinity Cross-Linking Implicates Amino-Acid Residues-61-203 of the
Beta-Subunit. J. Biol. Chem. 1988, 263, 18726-18731.
238
53.
Pierschbacher, M. D.; Ruoslahti, E. Influence of Stereochemistry of the Sequence
Arg-Gly-Asp-Xaa on Binding-Specificity in Cell-Adhesion. J. Biol. Chem. 1987, 262, 1729417298.
54.
Boturyn, D.; Coll, J. L.; Garanger, E.; Favrot, M. C.; Dumy, P. Template Assembled
Cyclopeptides as Multimeric System for Integrin Targeting and Endocytosis. J. Am. Chem.
Soc. 2004, 126, 5730-5739.
55.
Madden, H. L.; Henke, C. A. Induction of Lung Fibroblast Apoptosis by Soluble
Fibronectin Peptides. Am. J. Respir. Crit. Care. Med. 2000, 162, 1553-1560.
56.
Buckley, C. D.; Pilling, D.; Henriquez, N. V.; Parsonage, G.; Threlfall, K.; ScheelToellner, D.; Simmons, D. L.; Albar, A. N.; Lord, J. M.; Salmon, M. Rgd Peptides Induce
Apoptosis by Direct Caspase-3 Activation. Nature 1999, 397, 534-539.
57.
Humphries, J. D.; Byron, A.; Humphries, M. J. Integrin Ligands at a Glance. J. Cell
Sci. 2006, 119, 3901-3903.
58.
Pampaloni, F.; Reynaud, E. G.; Stelzer, E. H. K. The Third Dimension Bridges the
Gap between Cell Culture and Live Tissue. Nat. Rev. Mol. Cell Biol. 2007, 8, 839-845.
59.
Lutolf, M. P.; Lauer-Fields, J. L.; Schmoekel, H. G.; Metters, A. T.; Weber, F. E.;
Fields, G. B.; Hubbell, J. A. Synthetic Matrix Metalloproteinase-Sensitive Hydrogels for the
Conduction of Tissue Regeneration: Engineering Cell-Invasion Characteristics. Proc. Natl.
Acad. Sci. U. S. A. 2003, 100, 5413-8.
60.
Rizzi, S. C.; Ehrbar, M.; Halstenberg, S.; Raeber, G. P.; Schmoekel, H. G.;
Hagenmuller, H.; Muller, R.; Weber, F. E.; Hubbell, J. A. Recombinant Protein-Co-Peg
Networks as Cell-Adhesive and Proteolytically Degradable Hydrogel Matrixes. Part Ii:
Biofunctional Characteristics. Biomacromolecules 2006, 7, 3019-29.
61.
Mori, H.; Hara, M. Cultured Stem Cells as Tools for Toxicological Assays. J. Biosci.
Bioeng. 2013, 116, 647-652.
62.
Svendsen, C. N. Back to the Future: How Human Induced Pluripotent Stem Cells
Will Transform Regenerative Medicine. Hum. Mol. Genet. 2013, 22, R32-R38.
63.
Villa-Diaz, L. G.; Ross, A. M.; Lahann, J.; Krebsbach, P. H. Concise Review: The
Evolution of Human Pluripotent Stem Cell Culture: From Feeder Cells to Synthetic
Coatings. Stem Cells 2013, 31, 1-7.
64.
Rosso, F.; Marino, G.; Giordano, A.; Barbarisi, M.; Parmeggiani, D.; Barbarisi, A.
Smart Materials as Scaffolds for Tissue Engineering. J. Cell. Physiol. 2005, 203, 465-470.
65.
Braam, S. R.; Zeinstra, L.; Litjens, S.; Ward-van Oostwaard, D.; van den Brink, S.;
van Laake, L.; Lebrin, F.; Kats, P.; Hochstenbach, R.; Passier, R.; Sonnenberg, A.;
Mummery, C. L. Recombinant Vitronectin Is a Functionally Defined Substrate That
Supports Human Embryonic Stem Cell Self-Renewal Via Alpha V Beta 5 Integrin. Stem
Cells 2008, 26, 2257-2265.
66.
Frith, J. E.; Mills, R. J.; Cooper-White, J. J. Lateral Spacing of Adhesion Peptides
Influences Human Mesenchymal Stem Cell Behaviour. J. Cell Sci. 2012, 125, 317-327.
67.
Massia, S. P.; Hubbell, J. A. An Rgd Spacing of 440nm Is Sufficient for Integrin
Alpha-V-Beta-3-Mediated Fibroblast Spreading and 140nm for Focal Contact and Stress
Fiber Formation. J. Cell Biol. 1991, 114, 1089-1100.
68.
Amit, M.; Chebath, J.; Margulets, V.; Laevsky, I.; Miropolsky, Y.; Shariki, K.; Peri,
M.; Blais, I.; Slutsky, G.; Revel, M.; Itskovitz-Eldor, J. Suspension Culture of
Undifferentiated Human Embryonic and Induced Pluripotent Stem Cells. Stem Cell Rev.
Rep. 2010, 6, 248-259.
69.
Ross, S. A.; McCaffery, P. J.; Drager, U. C.; De Luca, L. M. Retinoids in Embryonal
Development. Physiol. Rev. 2000, 80, 1021-1054.
70.
Siddikuzzaman; Guruvayoorappan, C.; Grace, V. M. B. All Trans Retinoic Acid and
Cancer. Immunopharmacol. Immunotoxicol. 2011, 33, 241-249.
239
71.
Thielitz, A.; Abdel-Naser, M. B.; Fluhr, J. W.; Zouboulis, C. C.; Gollnick, H. Topical
Retinoids in Acne - an Evidence-Based Overview. J. Dtsch. Dermatol. Ges. 2008, 6, 10231031.
72.
Gudas, L. J. Retinoids Induce Stem Cell Differentiation Via Epigenetic Changes.
Semin. Cell Dev. Biol. 2013, 24, 701-705.
73.
Kashyap, V.; Gudas, L. J. Epigenetic Regulatory Mechanisms Distinguish Retinoic
Acid-Mediated Transcriptional Responses in Stem Cells and Fibroblasts. J. Biol. Chem.
2010, 285, 14534-48.
74.
Ioele, G.; Cione, E.; Risoli, A.; Genchi, G.; Ragno, G. Accelerated Photostability
Study of Tretinoin and Isotretinoin in Liposome Formulations. Int. J. Pharm. 2005, 293,
251-260.
75.
Regazzi, M. B.; Iacona, I.; Gervasutti, C.; Lazzarino, M.; Toma, S. Clinical
Pharmacokinetics of Tretinoin. Clin. Pharmacokinet. 1997, 32, 382-402.
76.
Guo, Y.; Bozic, D.; Malashkevich, V. N.; Kammerer, R. A.; Schulthess, T.; Enger, J.
All-Trans Retinol, Vitamin D and Other Hydrophobic Compounds Bind in the Axial Pore of
the Five-Stranded Coiled-Coil Domain of Cartilage Oligomeric Matrix Protein. EMBO J.
1998, 17, 5265-5272.
77.
Karnaukhova, E. Interactions of Human Serum Albumin with Retinoic Acid, Retinal
and Retinyl Acetate. Biochem. Pharmacol. 2007, 73, 901-910.
78.
Zsila, F.; Bikadi, Z.; Simonyi, M. Retinoic Acid Binding Properties of the Lipocalin
Member Beta-Lactoglobulin Studied by Circular Dichroism, Electronic Absorption
Spectroscopy and Molecular Modeling Methods. Biochem. Pharmacol. 2002, 64, 16511660.
79.
Kane, M. A.; Folias, A. E.; Wang, C.; Napoli, J. L. Quantitative Profiling of
Endogenous Retinoic Acid in Vivo and in Vitro by Tandem Mass Spectrometry. Anal.
Chem. 2008, 80, 1702-1708.
80.
Wu, X. Q.; Hu, J. Y.; Jia, A.; Peng, H.; Wu, S. M.; Dong, Z. M. Determination and
Occurrence of Retinoic Acids and Their 4-Oxo Metabolites in Liaodong Bay, China, and Its
Adjacent Rivers. Environ. Toxicol. Chem. 2010, 29, 2491-2497.
81.
Crowe, D. L.; Kim, R.; Chandraratna, R. A. S. Retinoic Acid Differentially Regulates
Cancer Cell Proliferation Via Dose-Dependent Modulation of the Mitogen-Activated Protein
Kinase Pathway. Mol. Cancer Res. 2003, 1, 532-540.
82.
Van heusden, J.; Wouters, W.; Ramaekers, F. C. S.; Krekels, M.; Dillen, L.;
Borgers, M.; Smets, G. All-Trans-Retinoic Acid Metabolites Significantly Inhibit the
Proliferation of Mcf-7 Human Breast Cancer Cells in Vitro. Br. J. Cancer 1998, 77, 26-32.
83.
Potekhin, S. A.; Melnik, T. N.; Popov, V.; Lanina, N. F.; Vazina, A. A.; Rigler, P.;
Verdini, A. S.; Corradin, G.; Kajava, A. V. De Novo Design of Fibrils Made of Short AlphaHelical Coiled Coil Peptides. Chem. Biol. 2001, 8, 1025-1032.
84.
Ryadnov, M. G.; Woolfson, D. N. Introducing Branches into a Self-Assembling
Peptide Fiber. Angew. Chem., Int. Ed. 2003, 42, 3021-3023.
240