Self-defensive antibacterial layer-by

Biomaterials 45 (2015) 64e71
Contents lists available at ScienceDirect
Biomaterials
journal homepage: www.elsevier.com/locate/biomaterials
Self-defensive antibacterial layer-by-layer hydrogel coatings
with pH-triggered hydrophobicity
Yiming Lu a, Yong Wu b, Jing Liang b, Matthew R. Libera b, Svetlana A. Sukhishvili a, *
a
b
Department of Chemistry, Chemical Biology and Biomedical Engineering, Hoboken, NJ 07030, USA
Department of Chemical Engineering and Materials Science, Stevens Institute of Technology, Hoboken, NJ 07030, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 11 October 2014
Accepted 20 December 2014
Available online
We report on negatively charged layer-by-layer (LbL) hydrogel films, which turn hydrophobic and
bactericidal in response to bacteria-induced acidification of the medium. Single-component hydrogel
thin films, abbreviated as PaAALbLs, consisting of chemically crosslinked poly(2-alkylacrylic acids)
(PaAAs) with varying hydrophobicity [polymethacrylic acid (PMAA), poly(2-ethylacrylic acid) (PEAA),
poly(2-n-propylacrylic acid) (PPAA) or poly(2-n-butylacrylic acid) (PBAA)]. With increasing polyacid
hydrophobicity, the hydrogel films showed a decrease in water uptake and an increase in elastic
modulus. Both parameters were strongly dependent on pH. At pH 7.4, hydrogels of higher hydrophobicity
were more resistant to colonization by Staphylococcus epidermidis, with the PBAA coating showing
almost negligible colonization. As the medium became more acidic due to bacterial proliferation, the
more hydrophobic PEAALbL, PPAALbL and PBAALbL hydrogels became dehydrated and killed bacteria upon
contact with the surface. The killing efficiency was strongly enhanced by the polymer hydrophobicity.
The films remained cytocompatible with human osteoblasts, as indicated by the MTS assay and live/dead
staining. Our approach exploits bacteria-responsive properties of the coating itself without the
involvement of potentially toxic cationic polymers or the release of antimicrobial agents. These coatings
thus demonstrate a novel approach to the antibacterial protection of tissue-contacting biomedical-device
surfaces.
© 2014 Elsevier Ltd. All rights reserved.
Keywords:
Bacterial adhesion
Surface modification
Antibacterial
Hydrogel
1. Introduction
Because of the growing clinical importance of biomaterialsassociated infection [1], significant research has recently focused
on various ways to modify implant surfaces using coatings that
prevent bacterial adhesion, kill bacteria on contact, or enable
controlled elution of antimicrobial compounds [2]. Many such
coatings involve highly hydrated polymers, such as poly(ethylene
glycol) (PEG) [3]. Such coatings exhibit antifouling properties and
significantly reduce both protein adsorption and bacterial adhesion
due to the high entropy and strong affinity of PEG for water. In
contrast, superhydrophobic surfaces also display antifouling properties and can inhibit microbial adhesion [4]. However, for many
tissue-contacting biomedical devices, the potential of these
continuous antifouling coatings is somewhat limited since they
* Corresponding author.
E-mail address: ssukhish@stevens.edu (S.A. Sukhishvili).
http://dx.doi.org/10.1016/j.biomaterials.2014.12.048
0142-9612/© 2014 Elsevier Ltd. All rights reserved.
indiscriminately suppress interactions not only with bacteria but
also with mammalian cells [5].
Other antibacterial coatings have been designed with various
molecular architectures. Among these, for example, are selfassembled monolayers [6], polymer brushes [7], and LangmuireBlodgett films [8]. Furthermore, polycations, including chitosan [9],
are known to be antibacterial, and this property is attributed to
their positively charged chains, which can efficiently disrupt
negatively charged bacterial walls. A polycation-based approach
has been developed by Klibanov and co-workers, who showed that
surfaces coated with poly(N-alkyl-pyridinium) or N-alkyl-polyethyleneimine efficiently kill gram-positive and gram-negative
bacteria upon contact [10].
Another approach exploits the versatility of the layer-by-layer
(LbL) technique to create polyelectrolyte multilayer coatings of
diverse chemical compositions, architectures, and functions [11].
LbL films include a powerful portfolio of multifunctional antibacterial and antibiofilm coatings [12], and they can exhibit significant
antibacterial activity due to their polycationic constituents [13,14],
dual anti-adhesion and antibacterial activity achieved by the
Y. Lu et al. / Biomaterials 45 (2015) 64e71
simultaneous incorporation of hydrophilic polymers and polycations [13,15], or dual contact killing and antimicrobial release
achieved by the inclusion, for example, of polycations and silverion-releasing species [16]. Despite the ability of such films to provide for controlled antimicrobial release [17,18], continuous-elution
strategies face several challenges. Among them is the fact that the
antimicrobial supply in continuously eluting coatings will ultimately deplete. Since they elute antimicrobials whether those antimicrobials are needed or not, continuously eluting coatings
furthermore contribute to the selection of antibiotic-resistant
bacteria [19] and can lead to undesirable cytotoxicity effects.
Therefore, alternative strategies have recently been developed that
create antibacterial coatings which are responsive to various
environmental triggers, including pH [20].
We have recently developed the concept of bacteria-triggered
delivery of antibacterial agents from LbL coatings. These coatings
release antibacterial agents in response to the local acidification
from natural bacterial metabolic processes [20]. In contrast to
continuously eluting coatings, these new coatings are selfdefensive because they release their antimicrobial payload only
when and where bacteria interact with the surface. Another type of
self-defensive coating has been developed by G. Cado et al. This
coating responds to enzymatic molecules secreted by growing
biofilms [21] rather than to pH. Here, we continue to exploit local
acidification as a trigger to activate antibacterial functions. However, in contrast to our previous work where bacteria-induced pH
variations trigger the release of antibacterial agents [20], here we
exploit a new concept where the polymer matrix itself becomes
antibacterial in response to bacterial growth. These new biologically active coatings do not include any antibacterial compounds
such as small-molecule antibiotics. They are based on thin-film
hydrogels made from a family of polyanionic poly(2-alkylacrylic
acids) (PaAAs) whose hydrophobicity varies depending on the
length of their alkyl side chain. The net negative charge and relatively large water uptake by these coatings under normal physiological conditions (pH 7.4) make these PaAA films non-toxic to
osteoblasts. At the same time, the more hydrophobic members of
this family are selectively toxic to staphylococcal bacteria because
of pH-triggered changes in coating charge and water content.
Previous studies to design drug-delivery systems have exploited
the capacity of dissolved hydrophobic PaAA homopolymers and
copolymers to become membrane-disruptive in acidic endosomal
and tumor environments [22]. However, the potential of this class
of polymers as antibacterial agents has not yet been explored. We
thus study the pH-dependent antibacterial properties of a series of
poly(2-alkylacrylic acid) polymers with varying hydrophobicity
when these polymers are either dissolved in aqueous solutions or
are included within single-component layer-by-layer hydrogel
thin-film coatings, which we designate here as PaAALbL. We have
recently reported on similar PaAALbL hydrogel coatings which can
control the wettability of micro-structured substrates and detect
small-molecule adsorption [23]. The present work focuses on the
capability of these coatings to selectively control adhesion and
viability of bacteria and mammalian cells on planar surfaces.
65
experiments. Undoped silicon substrates of 525 ± 25 mm thickness were purchased
from Virginia Semiconductor Inc.
2.2. Methods
2.2.1. Polymer synthesis and characterization
Poly(2-ethylacrylic acid) (PEAA) (Mw ~ 79,000 g/mol, polydispersity index (PDI)
2.52), poly(2-n-propylacrylic acid) (PPAA) (Mw ~ 110,000 g/mol, PDI 2.18), and
poly(2-n-butylacrylic acid) (PBAA) (Mw ~ 138,000 g/mol, PDI 2.41) (Scheme 1) were
synthesized by free-radical polymerization using a procedure similar to that
described elsewhere [24]. Briefly, 1 g of 2-alkylacrylic acid monomers and 0.2 mL
DMF were mixed in Schlenk vials prior the addition of 3 mg of AIBN. The vials were
flame-sealed under vacuum and submerged in an oil bath at 60 C after 3 freezeethaw cycles. The polymerization was stopped after 24 h by breaking the vial to
allow oxygen access, followed by immersion of the vial in liquid nitrogen for 2 min to
stop the reaction. The molecular weights of the PaAAs were determined by GPC at a
flow rate of 1 mL/min and 60 C using 0.01 mM LiBr DMF as solvent and polystyrene
standards for column calibration.
2.2.2. Construction of surface hydrogels
Silicon substrates (1 cm 1 cm in size) were cleaned by 12 h exposure to UV
light followed by treatment with piranha solution (1:3 v/v H2O2:H2SO4) at 80 C.
Prior to multilayer construction, a bilayer of PMAA and BPEI, designated as (PMAA/
BPEI)2, was deposited and crosslinked [20]. PaAA/PVPON multilayers were then
deposited by sequential adsorption of PaAA and PVPON from 5 mL of 0.5 mg/mL
PaAA and PVPON solutions in methanol, except for PMAA/PVPON films, which were
deposited from 0.01 M phosphate buffer solutions at pH 2. The deposition time per
polymer layer was 10 min. In between the PaAA and PVPON deposition steps, the
wafers were rinsed for 30 s with a corresponding solvent (0.01 M phosphate buffer at
pH 2 for PMAA/PVPON films, and methanol for all other PaAA/PVPON multilayers).
All films were dried with nitrogen gas flow after completion of multilayer assembly.
After EDC (3 mg/mL) activation at pH 5.2, the PaAA components of the multilayers
were crosslinked by AADH (3 mg/mL) for 3 h. To release uncrosslinked PVPON, the
substrates were immersed in pH 7.5 buffer (for PMAALbL) or in a mixture of pH 6
0.005 M phosphate buffer and DMSO (1:5 v/v) (for PEAALbL, PPAALbL and PBAALbL) for
30 min. The substrates were then washed with water and methanol and dried.
2.2.3. Ellipsometry
The thickness of the dry PaAALbL hydrogels was measured by phase-modulated
ellipsometry with an incident angle of 65 assuming a constant film refractive index
of 1.5. The optical properties of the substrates and the native oxide layer were
determined prior to polymer deposition. Measurements of PaAALbL hydrogel
swelling were performed using a custom-made cylindrical flow-through glass cell.
For wet film measurements, the cell was filled with 0.01 M phosphate buffer at
various pH values, and thickness and refractive-index measurements were taken
after the hydrogels were allowed to equilibrate for 20 min.
2.2.4. FTIR spectroscopy
FTIR spectra of PaAALbL hydrogels were acquired using a Tensor 27 FTIR spectrometer (Bruker Optik GmbH). All samples were pre-titrated by phosphate buffer
solutions and dried from solutions at varied pH.
2.2.5. Contact angle (CA) measurements
Static СAs were measured under ambient conditions (~25 C, relative humidity
~23%) using a Contact Angle and Surface Tension Meter CAM 101 (KVS Instrument
Inc.). Prior to a CA measurement at a certain pH value, samples were pretreated by
0.01 M phosphate buffer solution with the same pH as used in the CA measurements.
Measurements were repeated 3 times at different positions of the samples.
2. Materials and methods
2.1. Materials
Polymethacrylic acid (PMAA) (Mw ~ 56,000 g/mol), branched poly(ethylene
imine) (BPEI) (Mw ~ 101,000 g/mol), polyvinylpyrrolidone (PVPON) (Mw ~ 2500 g/
mol), monobasic sodium phosphate, 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC), adipic acid dihydrazide (AADH), azobisisobutyronitrile
(AIBN), hydrochloric acid, sodium hydroxide, methanol, N,N-dimethylformamide
(DMF) and dimethyl sulfoxide (DMSO) were purchased from Sigma. LIVE/DEAD®
BacLight™ Bacterial Viability Kit was purchased from Life Technologies Corp. Millipore (Milli-Q) type I water with a resistivity of 18.2 MU was used in all
Scheme 1. Chemical structures of poly(2-alkylacrylic acids) (PaAAs). The abbreviations, PMAA, PEAA, PPAA and PBAA stand for polymethacrylic acid, poly(2-ethylacrylic
acid), poly(2-n-propylacrylic acid), and poly(2-n-butylacrylic acid), respectively.
66
Y. Lu et al. / Biomaterials 45 (2015) 64e71
2.2.6. Zeta-potential measurements
Zeta potentials of bacterial cells were measured using Zetasizer Nano-ZS
equipment (Malvern Instruments, Ltd). These measurements were performed
with a 106 CFU/mL bacterial concentration in 0.05 M phosphate buffer at 37 C.
2.2.7. Atomic force microscopy (AFM)
AFM was used to characterize the stiffness of hydrated PaAALbL hydrogels at both
low and high pH using an NSCRIPTOR™ dip pen nanolithography system (Nanoink).
Aspire™ CCS conical contact mode short cantilevers (NanoScience Instruments, Inc.)
with a force constant of 0.1 N/m were used for nanoindentation. Under dry conditions, the cantilevers were calibrated using the thermal fluctuation technique. For
wet measurements, cantilevers were calibrated in 0.05 M phosphate buffer solutions
with 0.2 M NaCl. All force curves were collected in situ using a cantilever velocity of
0.01 mm/s. Force curves were analyzed based on the conical geometric shape of the
cantilever. Young's moduli of PaAALbL hydrogels measured by nanoindentation, Eind,
were calculated using the following equation [25]:
F¼
8Eind tan ad2
3p
(
1 þ 1:7795
2 tan a d
d2
þ 16ð1:7795Þ2 tan a 2
h
p2 h
)
humidity. The complete medium was replaced every 2 days, and confluent flasks
were subcultured using trypsin/EDTA. The cells were then seeded in 24-well culture
plates (10,000 cells/cm2) onto PaAALbL hydrogels in 2 mL of medium. The medium
was refreshed every 2 days. Each experiment was performed in quintuple (n ¼ 5) for
each group of substrates. Cell viability was assessed by a colorimetric assay based on
the conversion of the tetrazolium salt into a soluble formazan dye using a Cell Titer
96 Aqueous assay (MTS, Promega). After 1, 3, and 7 days, the culture medium was
removed from each well, and the specimens were transferred to new cell culture
plates. 100 mL of pre-warmed MTS solution with 1 mL culture medium was added to
each well with continuous culture for 3 h. 200 mL of supernatant from each well was
then transferred to a 96-well plate, and the absorbance was measured at 490 nm.
Four specimens for each group were tested, and each test was repeated three times.
The cytotoxicity of the PaAALbL hydrogel thin films was also evaluated using a
live/dead cell viability assay kit (Invitrogen) after 1 and 7 days. The cells were stained
based on membrane integrity and intracellular esterase activity per the manufacturer's protocols. Briefly, cell-seeded substrates were incubated at room temperature for 30 min in PBS containing 2 mM of calcein AM (green) and 4 mM of ethidium
homodimer-1 (red). Cells stained green (live) and red (dead) were imaged using a
Nikon C1 confocal microscope.
where F is the applied force, d is the indentation depth, a is the half angle (15 ) of the
conical cantilever, and h is the sample height. This equation takes into account the
effects of the incompressible underlying substrate on the measurement of the film's
elastic properties.
3. Results and discussion
2.2.8. Antibacterial activity assay
Staphylococcus epidermidis (ATCC 35984) was incubated in tryptic soy broth
(TSB) with 6 mg/mL yeast extract and 8 mg/mL glucose (TSB) for 18 h. The resulting
biofilm was vortexed, sonicated, and filtered (5 mm) in order to create a near singlecell suspension. To test the antibacterial of PaAAs in solution, 1 mL of suspension
(2 106 CFU/mL) was mixed with 1 mL of PaAA dissolved in TSB (2.5 mg/mL) at pH
7.5. The bacteria-polymer solutions were then incubated for 24 h at 37 C in 12-well
culture plates. Bacterial suspensions (1 106 CFU/mL) in TSB, solutions of PaAAs
(1.25 mg/mL), and pure TSB solution were used as controls. The final bacterial
concentrations were determined using calibrated optical densities corrected for the
absorbance characteristic of the control samples.
To test the antibacterial activities of PaAALbL hydrogel thin films, all Si wafers
(with or without coatings) were exposed to UV light for 5 min prior to bacterial
inoculation. Then 2 mL of bacterial suspension (1 106 CFU/mL) was seeded onto
each PaAALbL hydrogel thin film followed by incubation for various time periods (6,
10, 12, 14, 16, 24 h) at 37 C in 12-well culture plates. Samples were then rinsed with
phosphate buffered saline (PBS) twice before fluorescence labeling. Using a live/
dead bacterial viability test kit (Invitrogen), stained specimens were characterized
using a Nikon C1 confocal laser scanning microscope (CLSM) system.
To determine the bacterial coverage on PaAALbL hydrogels, 2D CLSM images of
live/dead stained bacteria were analyzed using ImageJ software. Bacterial coverage
was defined as the ratio: Abact.,total/Asurface, where Abact.,total is the surface area
covered by bacteria, both live and dead, and Asurface is the area of the entire surface.
The fraction of dead bacteria was determined as the ratio of the area covered by
dead bacteria, Abact.,dead, to the total area covered by both dead and live bacteria,
Atotal.
Because PEAA, PPAA and PBAA are not soluble in aqueous acidic
media, hydrogen-bonded multilayer films of hydrophobic PEAA,
PPAA, or PBAA with polyvinylpyrrolidone (PVPON) were deposited
using methanol as the solvent. PMAA/PVPON films were deposited
from aqueous solutions at pH 2. In all cases, the film thickness increases linearly as the number of deposited bilayers increases, but
individual bilayer thickness depends on the type of PaAA. For PMAA
(hydrophilic), the PMAA/PVPON bilayer thickness is ~5.3 nm, while
the bilayer thicknesses of films involving the other three PaAAs
(hydrophobic) are only ~3.6 nm. Because of the coupling of thinfilm mechanical properties to the underlying solid substrates
[26], we constructed films all having the same thickness of
54 ± 1 nm by depositing 10 bilayers of PMAA/PVPON or 15 bilayers
of the more hydrophobic PaAA/PVPON films (Figure S1A). After
crosslinking of the self-assembled films and the subsequent pHinduced release of uncrosslinked PVPON, the film thicknesses
decrease by ~35% to ~35 nm when dried (Figure S1A). FTIR spectra
of dry hydrogen-bonded PMAA/PVPON multilayers and dry
PMAALbL hydrogel thin films show a complete release of PVPON,
which is indicated by the significant decrease of the amide band
intensity at ~1650 cm1 (Figure S1B). The remaining absorbance at
~1650 cm1 is due to amide bonds introduced through covalent
attachment of AADH crosslinker to the PaAA. FTIR spectra of the
other crosslinked PaAALbL films demonstrate the same (23e26%)
fractional contribution of amide band intensities to the total
2.2.9. Cell culture and cell viability assays
Human fetal osteoblasts (hFOB 1.19, ATCC CRL-11372) were cultured in DMEM
containing 10% FBS and 1% penicillin/streptomycin at 37 C in 5% CO2 and 95%
3.1. PaAALbL thin-film hydrogels: synthesis and properties
Fig. 1. (A) FTIR spectra of PaAALbL hydrogel films. The bands centered at 1550 cm1 and 1700 cm1 represent carboxylate (>COO) and carboxylic (>COOH) groups, respectively. The
absorbance band at 1650 cm1 corresponds to amide (eNHeCOe) groups introduced via PaAA crosslinking. (B) Representative curve fitting of a PEAALbL FTIR spectrum.
Y. Lu et al. / Biomaterials 45 (2015) 64e71
absorbance in the 1500e1800 cm1 region, which also includes
a eCOO band at 1550 cm1 and a eCOOH band at ~1700 cm1
(Fig. 1). This suggests that all PaAALbL films have similar crosslinking
degrees and, consequently, that hydrogel swelling is determined
primarily by the solubility of the polymer chains between
crosslinks.
The wetting and swelling properties of the various hydrogel
films depend strongly on pH. The static CAs decrease (Fig. 2A) and
the swelling increases (Fig. 2B) with increasing pH because of the
emergence of negative charges in the PaAA chains. The pH of the
wettability transition, pHtr, defined as the pH at the transition halfheight, depends on polyacid hydrophobicity. It increases from 5.8 to
6, 6.4, and 7.8 for the PMAALbL, PEAALbL, PPAALbL and PBAALbL
hydrogels, respectively. These transition values roughly correlate
with the pKa values of 6.2, 6.5, 6.8 and 7.2 for PMAA, PEAA, PPAA
67
and PBAA, respectively. Compared to previous pHtr data for PaAALbL
films in low-salt solutions [23], the transition pHs in salt solutions
are smaller by ~0.3 pH units and reflect the screening of chargeecharge interactions by the salt. The PMAALbL hydrogel, whose
constituent PMAA chains remain hydrated over the entire pH region, demonstrate the lowest static CAs and the largest swelling
among the family of PaAALbL hydrogels. In contrast, the more hydrophobic PEAALbL, PPAALbL and PBAALbL hydrogels, whose constituent chains dehydrate and collapsed at pH values below 5.2, 5.6
and 7, respectively, show less swelling and higher contact angles. In
an acidic environment, the most hydrophobic of the PaAALbL films
(PBAALbL) contains as little as 8% water.
Nanoindentation was used to correlate the PaAALbL film swelling
with changes in the surface mechanical properties (Fig. 3). The
hydrophilic PMAALbL coating, which exhibits the highest water
uptake of the various hydrogel films, exhibits a modulus six orders
of magnitude lower than that of Si wafers. Importantly, its modulus
decreases by almost 3-fold between pH 3 and 8. A similar decrease
in modulus has been reported for more rigid electrostaticallyassembled multilayers with increased hydration [26] and for
microgels that swell in response to external stimuli (pH and temperature) [27]. Because of the large changes in the swelling and the
stronger electrostatic repulsion at higher pH, the modulus strongly
depends on pH in the entire polymer system studied here. Moreover, Eind increases with PaAA hydrophobicity, with the most hydrophobic hydrogel (PBAALbL) having the highest modulus at pH 3
(Eind ~ 338 kPa). The value of Eind was as low as ~72 kPa for the
PMAALbL hydrogel at pH 8, which is within the range reported for
crosslinked poly(L-lysine)/hyaluronan LbL films (30e150 kPa) [28]
but lower than that reported for crosslinked PAH/PAA [29]. The
dependence of the elastic modulus on polyacid hydrophobicity
reflects the impact of the reduced water content and the increased
number of polymerepolymer contacts within the more hydrophobic PaAALbL films.
3.2. Antibacterial activity of PaAAs in solution
Prior to exploring the antibacterial activity of PaAA chains
immobilized within LbL hydrogels, we studied the interactions of
PaAA chains and bacteria in solution. S. epidermidis was chosen for
Fig. 2. (A) Static CAs of PaAALbL-coated silicon wafers as a function of pH. (B) pH
dependence of the swelling ratios of PaAALbL surface hydrogels, defined as the ratio of
wet film to dry film thicknesses, as measured by in situ ellipsometry. All experiments
were done in 0.01 M phosphate buffer solutions containing 0.2 M NaCl. The error bars
represent the standard deviation of the contact angle data, obtained by measurements
at three different spots on each of three identically prepared samples.
Fig. 3. Young's moduli measured by nanoindentation, Eind, of PaAALbL coatings
deposited on Si wafers and exposed to 0.2 M NaCl PBS solutions at pH 3 and pH 8.
Representative force curves for all LbL-coated samples and control Si wafers are shown
in Figure S2 in Supporting Information. The error bars represent standard deviations
based on five repeated measurements.
68
Y. Lu et al. / Biomaterials 45 (2015) 64e71
Fig. 4. (A) Time evolution of pH during culturing of S. epidermidis (initial concentration 1 106 CFU/mL) in TSB. (B) Bacterial concentration and pH of S. epidermidis TSB culture
solutions containing 1.25 mg/mL of various PaAA acids after 24 h of bacteria culture. The shaded color gradient illustrates increased polyacid hydrophobicity. (For interpretation of
the references to color in this figure legend, the reader is referred to the web version of this article.)
these experiments, because it is frequently implicated in many
biomaterials-associated infections. As a result of metabolic activity,
these bacteria secret lactic acid during growth, causing acidification
of the immediate environment [30]. Fig. 4A shows that growing
bacteria, in the absence of any dissolved PaAA, cause a significant
drop in solution pH. The pH plateau for t < 10 h is due to the
buffering effect of TSB. This is followed by a steep decrease in pH
from 7.5 to ~5 over the period of 10 h < t < 20 h (shaded area in
Fig. 4A). After 24 h, the solution pH decreases to 4.8 while the
concentration of bacteria increases by almost 300%.
In contrast, adding PaAA to the culture medium substantially
reduces the concentration of bacteria measured after 24 h of culture (Fig. 4B). And, with fewer bacteria, the pH decrease is
concomitantly less. These effects increase as the hydrophobicity of
the PaAA increases. We hypothesize that this bacteria-induced pH
drop drives conformational coil-to-globule transitions in the polyacid chains, reveals hydrophobic moieties within the more hydrophobic PaAAs, and triggers the penetration of PaAA segments
through the bacterial wall, ultimately resulting in bacterial death.
The data in Fig. 4B corroborate this hypothesis. The bacterial concentration in suspension after 24 h of culture is largest for the
PMAA, and there is a systematic decrease in the bacterial concentration when cultured in the presence of PaAAs with increased
hydrophobicity. Correspondingly, suspensions cultured with the
more hydrophobic PaAAs maintain higher pH values, because these
solutions grow a smaller number of media-acidifying bacteria. The
strongest inhibition is observed with the most hydrophobic PaAA
(PBAA).
These results indicate that a new interaction mechanism between the PaAAs and the bacteria is activated at decreased pH
values, which leads to the inhibition of bacterial growth. The
mechanism is probably similar to that described for interactions
between PaAAs and phospholipid membranes, which involves the
insertion of hydrophobic segments of PaAA into the membrane,
membrane reorganization, and membrane permeabilization
[31,32]. Note that these interactions occur despite the negative
change of S. epidermidis between pH 5 and pH 7.5 (zeta
potential ~ 24 mV, Figure S5). To the best of our knowledge, this is
the first study of interactions between bacteria and hydrophobic
pH-sensitive anionic polymers, with prior reports exploring the
antibacterial activity of cationic polymers [33].
study PaAAs thin films when these polymers are immobilized
within LbL coatings. Previous studies indicate that when bioactive
polyelectrolyte chains are included within traditional LbL coatings,
charge compensation and ionic pairing neutralize the cytotoxicity
of polyamines [34]. We hypothesize, however, that segmental
3.3. Antibacterial activity of PaAALbL-coated substrates
Fig. 5. (A) pH evolution in S. epidermidis suspensions (initial concentration
1 106 CFU/mL) in TSB above uncoated or coated Si wafers. (B) Fractional bacterial
coverage of PaAALbL surfaces at different culture times. The data and standard deviations are calculated from 27 CLSM images of bacteria-covered hydrogels taken from
four repeated experiments. Representative CLSM images are shown in Figure S3 and
S4.
While these experiments indicate that anionic polymers with
pH-switchable hydrophobicity exhibit effective antimicrobial
properties when dissolved in solution, the main focus here is to
Y. Lu et al. / Biomaterials 45 (2015) 64e71
mobility within LbL hydrogels can enable the penetration of hydrophobic polymer segments into the bacterial membrane and
therefore introduce antibacterial functionality into an LbL coating.
The effect of surface-attached hydrogel thin films on bacterial
growth and viability were evaluated by culturing bacteria in suspensions brought in contact with either bare Si wafers (control) or
Si wafers coated with PMAALbL, PEAALbL, PPAALbL or PBAALbL
hydrogel thin films (Fig. 5A). In all cases, bacterial growth in solution decreases the pH of the TSB medium. Furthermore, the time
evolution of solution pH (Fig. 5A) is similar in all suspensions,
regardless of the presence or the absence of the PaAALbL coatings on
the substrates. This finding indicates that bacteria/substrate interactions do not significantly affect bacterial growth in the surrounding solution. At the same time, the pH changes in the medium
induce deswelling and hydrophobicity transitions in the PaAALbL
coatings (see Fig. 2). These transitions have a strong impact on the
bacteria/substrate interactions. Fig. 5B illustrates the time dependence of the bacterial surface coverage for the various types of
PaAALbL coatings. Interestingly, the bacterial colonization of the
more hydrophilic and softer PMAALbL coatings is indistinguishable
from that of the coating-free Si control surface. In contrast, the
more hydrophobic and stiffer PaAALbL coatings resist bacterial
colonization. In fact, the most hydrophobic coating (PBAALbL) remains largely bacteria-free until 12e14 h into the 24 h culture
period. While the literature reports conflicting observations on the
effect of substrate stiffness on bacterial coverage [28,35], our results
are in agreement with work by N. Saha et al. [28] who also reported
more rapid bacterial growth on softer LbL films.
A steep onset in the bacterial colonization of PEAALbL, PPAALbL
and PBAALbL hydrogels occurs for times longer than 12 h when the
pH decreases below ~6.3. This relatively abrupt increase in bacterial
surface coverage reflects an important property of our hydrophobic
hydrogel coatings: they reveal their hydrophobic moieties in an
acidic environment, which, in turn, dramatically alters their interactions with bacteria. While bacterial colonization increases at
the onset of the hydrophilicehydrophobic transition (Fig. 5), live/
dead staining experiments show that a significant fraction of these
69
bacteria are dead (Fig. 6). The fraction of dead bacteria increases
with increasing hydrogel hydrophobicity and becomes as high as
0.65 for the case of the PBAALbL coatings (Fig. 6E). In contrast, live
bacteria almost exclusively populate surfaces of hydrophilic
PMAALbL films (Fig. 6A) and control Si wafers (data not shown).
Moreover, none of the coatings show a bacteria-killing ability for
bacteria culture times shorter than 12 h, when the pH does not drop
below 7.0 and the coatings remain highly hydrated. The bactericidal
efficiency of PaAALbL coatings turns on only as pH decreases and
reveals the coating hydrophobicity.
3.3.1. Cytocompatibility of PaAALbL-coated substrates
A critical objective in the development of bacteria-resistant
tissue-contacting implants is to endow the coatings with specific
activity against bacteria without compromising their ability to
support the adhesion and proliferation of tissue cells [36]. We thus
used human fetal osteoblasts to assess the cytocompatibility of the
various thin-film hydrogel coatings. Fig. 7 shows the results of MTS
proliferation and viability tests. Osteoblast proliferation occurs on
all of the PaAALbL coatings, but proliferation on the hydrophilic and
more swollen PMAALbL surfaces is ~50% higher than that on the
more hydrophobic PaAALbL coatings. The three hydrogels with pHswitchable hydrophobicity (PEAALbL, PBAALbL and PBAALbL)
demonstrate similar osteoblast growth properties. In particular,
while it has been established that softer, more hydrated films are
typically more cytophobic [37], here, in agreement with other
earlier report [38], we find that osteoblast proliferation is favored
by more hydrophilic surfaces.
In addition, Fig. 7A shows HFOB 1.19 cells, visualized by livedead staining, on silicon wafers and on PaAALbL hydrogels. All
proliferating cells appear as green (Fig. 7A), indicating that the
hydrogels do not induce cellular death.
Overall, the data in Fig. 7A and B illustrate that all types of
hydrogels studied here are non-toxic to osteoblasts. This finding
might not be surprising in the case of PMAALbL, PEAALbL and
PPAALbL hydrogels, since these polyacids do not disrupt lipid bilayer
or cellular membranes at pH 7.4 [39]. However, the compatibility of
Fig. 6. (AeD) Comparison of the total bacteria (black symbols) and dead bacteria (red symbols, shaded areas) coverage for various PaAALbL coatings. Live/dead bacteria staining
images are shown in Figure S3 and S4. (E) Fraction of dead bacteria as a function of culture time. The error bars represent the standard deviations of bacterial coverage data,
analyzed by ImageJ software on 9 live/dead stained, bacterial CLSM images. Description of how to calculate bacterial coverage data can be found in the Materials and Methods
section. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
70
Y. Lu et al. / Biomaterials 45 (2015) 64e71
Fig. 7. (A) CLSM images of HFOB 1.19 cell attachment and proliferation on Si wafers and PaAALbL hydrogels at day 1 and day 7. The scale bar is 100 mm. The cells were live/dead
stained prior to imaging, but showed exclusively live (green) cells, with no observable dead (red) cells. (B) Mammalian cell metabolic activity studied by the MTS assay of osteoblast
proliferation on control Si wafers and wafers covered by PaAALbL hydrogels. Statistically significant differences from the control Si samples and PMAALbL-coated substrates were
determined using a student's t-test (p < 0.05) are marked by stars. The magnitude and error bars represent the average and standard deviation, respectively, from 9 different
experiments. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
the PBAALbL hydrogels with human osteoblasts is an important new
finding, since free PBAA homopolymer in solution causes red blood
cell lysis even at pH 7.4 [39]. While the diminished cytotoxicity of
polycations has been shown to be a result of strong binding between polycations and polyanions within an LbL film [34], here we
have demonstrated that cytotoxicity of polymer chains can be
hampered by their immobilization within a surface network.
Obviously, the immobilization of polyacid chains within hydrogels
significantly hinders the availability of hydrophobic functional
groups for membrane interactions at pH 7.4, rendering these
hydrogel coatings non-toxic to osteoblasts, which do not acidify the
immediate environment. Being included within a hydrogel nevertheless preserves sufficient segmental mobility that enables these
polymers to become bactericidal in acidic environments when the
coatings turn hydrophobic and become more dehydrated.
4. Conclusions
Here, we have developed the self-defensive concept for
bacteria-responsive hydrogel thin films that do not incorporate
antimicrobial agents for triggered release. These films instead act
via a pH-induced mechanism of antibacterial activity inherent
within the thin film itself. We studied a series of four LbL hydrogel
thin films, which enable the systematic variation of their pHdependent hydrophobicity. The pH-induced hydrophilic-to-hydrophobic transitions of these coatings appear to be the central
enabler underlying their bactericidal properties. The results of
these experiments suggest that the antibacterial mechanism involves the insertion of hydrophobic segments from the polymer
networks into bacterial walls, which can lead to wall damage and
bacterial death. However, regardless of their hydrophobicity, the
Y. Lu et al. / Biomaterials 45 (2015) 64e71
coatings are noncytotoxic to osteoblasts. We suggest that the
design of antibiotic-free bacteria-triggered films presents a promising path towards the development of bacteria-responsive surface
coatings that may prevent biofilm growth on tissue-contacting
biomedical devices.
Acknowledgments
This work was supported by the Office of Innovation and
Entrepreneurship at Stevens Institute of Technology and by the
Army Research Office through grant #W911NF-12-1-0331.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.biomaterials.2014.12.048.
References
[1] Busscher HJ, van der Mei HC, Subbiahdoss G, Jutte PC, van den Dungen JJ,
Zaat SA, et al. Biomaterial-associated infection: locating the finish line in the
race for the surface. Sci Transl Med 2012;4:153rv10.
[2] Pavlukhina S, Sukhishvili S. Polymer assemblies for controlled delivery of
bioactive molecules from surfaces. Adv Drug Deliv Rev 2011;63:822e36.
[3] Campoccia D, Montanaro L, Arciola CR. A review of the biomaterials technologies for infection-resistant surfaces. Biomaterials 2013;34:8533e54.
[4] Genzer J, Efimenko K. Recent developments in superhydrophobic surfaces and
their relevance to marine fouling: a review. Biofouling 2006;22:339e60.
[5] Wang Y, da Silva Domingues JF, Subbiahdoss G, van der Mei HC, Busscher HJ,
Libera M. Conditions of lateral surface confinement that promote tissue-cell
integration and inhibit biofilm growth. Biomaterials 2014;35:5446e52.
quet A, Berjeaud J-M, et al.
[6] Humblot V, Yala J-F, Thebault P, Boukerma K, He
The antibacterial activity of Magainin I immobilized onto mixed thiols selfassembled monolayers. Biomaterials 2009;30:3503e12.
[7] Murata H, Koepsel RR, Matyjaszewski K, Russell AJ. Permanent, non-leaching
antibacterial surfacesd2: how high density cationic surfaces kill bacterial
cells. Biomaterials 2007;28:4870e9.
[8] Helttunen K, Moridi N, Shahgaldian P, Nissinen M. Resorcinarene bis-crown
silver complexes and their application as antibacterial Langmuir-Blodgett
films. Org Biomol Chem 2012;10:2019e25.
[9] Huang M, Khor E, Lee-Yong L. Uptake and cytotoxicity of chitosan molecules
and nanoparticles: effects of molecular weight and degree of deacetylation.
Pharm Res 2004;21:344e53.
[10] Schaer TP, Stewart S, Hsu BB, Klibanov AM. Hydrophobic polycationic coatings
that inhibit biofilms and support bone healing during infection. Biomaterials
2012;33:1245e54.
[11] Decher G, Hong JD, Schmitt J. Buildup of ultrathin multilayer films by a selfassembly process: III. Consecutively alternating adsorption of anionic and
cationic polyelectrolytes on charged surfaces. Thin Solid Films
1992;210e211(Part 2):831e5.
[12] Lichter JA, Van Vliet KJ, Rubner MF. Design of antibacterial surfaces and interfaces: polyelectrolyte multilayers as a multifunctional platform. Macromolecules 2009;42:8573e86.
[13] Fu J, Ji J, Yuan W, Shen J. Construction of anti-adhesive and antibacterial
multilayer films via layer-by-layer assembly of heparin and chitosan. Biomaterials 2005;26:6684e92.
[14] Lichter JA, Rubner MF. Polyelectrolyte multilayers with intrinsic antimicrobial
functionality: the importance of mobile polycations. Langmuir 2009;25:
7686e94.
[15] Boulmedais F, Frisch B, Etienne O, Lavalle P, Picart C, Ogier J, et al. Polyelectrolyte multilayer films with pegylated polypeptides as a new type of
anti-microbial protection for biomaterials. Biomaterials 2004;25:2003e11.
[16] Li Z, Lee D, Sheng X, Cohen RE, Rubner MF. Two-level antibacterial coating
with both release-killing and contact-killing capabilities. Langmuir 2006;22:
9820e3.
71
[17] Chuang HF, Smith RC, Hammond PT. Polyelectrolyte multilayers for tunable
release of antibiotics. Biomacromolecules 2008;9:1660e8.
[18] Moskowitz JS, Blaisse MR, Samuel RE, Hsu H-P, Harris MB, Martin SD, et al. The
effectiveness of the controlled release of gentamicin from polyelectrolyte
multilayers in the treatment of Staphylococcus aureus infection in a rabbit
bone model. Biomaterials 2010;31:6019e30.
[19] Stewart PS, William Costerton J. Antibiotic resistance of bacteria in biofilms.
Lancet 2001;358:135e8.
[20] Pavlukhina S, Lu Y, Patimetha A, Libera M, Sukhishvili S. Polymer multilayers
with pH-triggered release of antibacterial agents. Biomacromolecules
2010;11:3448e56.
[21] Cado G, Aslam R, Seon L, Garnier T, Fabre R, Parat A, et al. Self-defensive
biomaterial coating against bacteria and yeasts: polysaccharide multilayer
film with embedded antimicrobial peptide. Adv Funct Mater 2013;23:
4801e9.
[22] Murthy N, Campbell J, Fausto N, Hoffman AS, Stayton PS. Design and synthesis
of pH-responsive polymeric carriers that target uptake and enhance the
intracellular delivery of oligonucleotides. J Control Release 2003;89:365e74.
[23] Lu Y, Sarshar MA, Du K, Chou T, Choi C-H, Sukhishvili SA. Large-amplitude,
reversible, pH-triggered wetting transitions enabled by layer-by-layer films.
ACS Appl Mater Inter 2013;5:12617e23.
[24] You H, Tirrell DA. Radical copolymerization of 2-ethylacrylic acid and methacrylic acid. J Polym Sci Part A Polym Chem 1990;28:3155e63.
[25] Gavara N, Chadwick RS. Determination of the elastic moduli of thin samples
and adherent cells using conical atomic force microscope tips. Nat Nano
2012;7:733e6.
[26] Lehaf AM, Hariri HH, Schlenoff JB. Homogeneity, modulus, and viscoelasticity
of polyelectrolyte multilayers by nanoindentation: refining the buildup
mechanism. Langmuir 2012;28:6348e55.
[27] Wiedemair J, Serpe MJ, Kim J, Masson J-F, Lyon LA, Mizaikoff B, et al. In-situ
AFM studies of the phase-transition behavior of single thermoresponsive
hydrogel particles. Langmuir 2006;23:130e7.
[28] Saha N, Monge C, Dulong V, Picart C, Glinel K. Influence of polyelectrolyte film
stiffness on bacterial growth. Biomacromolecules 2013;14:520e8.
[29] Lehaf AM, Moussallem MD, Schlenoff JB. Correlating the compliance and
permeability of photo-cross-linked polyelectrolyte multilayers. Langmuir
2011;27:4756e63.
[30] Handke LD, Rogers KL, Olson ME, Somerville GA, Jerrells TJ, Rupp ME, et al.
Staphylococcus epidermidis saeR is an effector of anaerobic growth and a
mediator of acute inflammation. Infect Immun 2008;76:141e52.
[31] Yessine M-A, Leroux J-C. Membrane-destabilizing polyanions: interaction
with lipid bilayers and endosomal escape of biomacromolecules. Adv Drug
Deliv Rev 2004;56:999e1021.
[32] Jones RA, Cheung CY, Black FE, Zia JK, Stayton PS, Hoffman AS, et al. Poly(2alkylacrylic acid) polymers deliver molecules to the cytosol by pH-sensitive
disruption of endosomal vesicles. Biochem J 2003;372:65e75.
[33] Rosenfeld Y, Lev N, Shai Y. Effect of the hydrophobicity to net positive charge
ratio on antibacterial and anti-endotoxin activities of structurally similar
antimicrobial peptides. Biochemistry 2010;49:853e61.
[34] Martinez JS, Keller TCS, Schlenoff JB. Cytotoxicity of free versus multilayered
polyelectrolytes. Biomacromolecules 2011;12:4063e70.
[35] Lichter JA, Thompson MT, Delgadillo M, Nishikawa T, Rubner MF, Van Vliet KJ.
Substrata mechanical stiffness can regulate adhesion of viable bacteria. Biomacromolecules 2008;9:1571e8.
[36] Wang Y, Libera M, Busscher HJ, van der Mei HC. Mono-functional to multifunctional coatings: the changing paradigm to control biofilm formation on
biomedical implants. In: Simoes M, Mergulho F, editors. Biofilms in bioengineering. Commack: Nova Science; 2012.
[37] Mendelsohn JD, Yang SY, Hiller JA, Hochbaum AI, Rubner MF. Rational design
of cytophilic and cytophobic polyelectrolyte multilayer thin films. Biomacromolecules 2002;4:96e106.
[38] Lim JY, Liu X, Vogler EA, Donahue HJ. Systematic variation in osteoblast
adhesion and phenotype with substratum surface characteristics. J Biomed
Mater Res A 2004;68A:504e12.
[39] Murthy N, Chang I, Stayton P, Hoffman A. pH-sensitive hemolysis by random
copolymers of alkyl acrylates and acrylic acid. Macromol Symp 2001;172:
49e56.