Terpyridine Platinum(II) - Andrew H.

Open Access Article
The authors, the publisher, and the right
holders grant the right to use, reproduce,
and disseminate the work in digital form
to all users.
Terpyridine Platinum(II) Complexes Inhibit Cysteine
Proteases by Binding to Active-site Cysteine
Abstract
Platinum(II) complexes have been demonstrated to form covalent bonds with sulfur-donating
ligands (in glutathione, metallothionein and other sulfur-containing biomolecules) or coordination bonds with nitrogen-donating ligands (such as histidine and guanine). To investigate
how these compounds interact with cysteine proteases, we chose terpyridine platinum(II)
(TP-Pt(II)) complexes as a model system. By using X-ray crystallography, we demonstrated
that TP-Pt(II) formed a covalent bond with the catalytic cysteine residue in pyroglutamyl
peptidase I. Moreover, by using MALDI (matrix-assisted laser desorption/ionization) and
TOF-TOF (time of flight) mass spectrometry, we elucidated that the TP-Pt(II) complex
formed a covalent bond with the active-site cysteine residue in two other types of cysteine
protease. Taken together, the results unequivocally showed that TP-Pt(II) complexes can
selectively bind to the active site of most cysteine proteases. Our findings here can be useful
in the design of new anti-cancer, anti-parasite or anti-virus platinum(II) compounds.
Key words: Platinum(II); Enzyme activity; Cysteine protease inhibitors; Crystal structure.
Journal of Biomolecular Structure &
Dynamics, ISSN 0739-1102
Volume 29, Issue Number 2, (2011)
©Adenine Press (2011)
Yan-Chung Loa,b
Wen-Chi Suc
Tzu-Ping Kob,d
Nai-Chen Wangb,d
Andrew H.-J. Wanga,b,d*
aDepartment
and Institute of
Pharmacology, National
Yang-Ming University,
Taipei 112, Taiwan
bInstitute
of Biological Chemistry,
cInstitute
of Molecular Biology, and
dNational
Core Facility of High-
Throughput Protein Crystallography,
Academia Sinica, Taipei 115, Taiwan
Introduction
Metal-based drugs (containing for example, Pt, Pd, Au, Ag, Fe, Mo, Ru, Zn, or Sn)
are widely investigated for their anti-cancer, anti-parasitic and anti-fungal activities as well as for inhibition and stabilization of several biological systems (1-11).
The first metal-based anti-cancer drug is a platinum(II) complex, namely cisplatin
(cis-diamminedichloroplatinum(II)). Cisplatin and other platinum(II) coordination complexes are commonly used for the treatment of testicular and ovarian cancers. The molecular mechanism of cisplatin cytotoxicity is by binding to the N-7
positions of two adjacent guanine bases to disrupt DNA functions leading to cell
death (12). Other platinum(II) complexes, like TP-Pt(II) chloride (Figure 1), also
coordinate to guanine N-7 (13). The effect of the platinum(II) complexes on DNA
has been well studied; however, direct observation of how platinum(II) complexes
affect protein function is rare (14-17). From previous results, the platinum(II) complexes were found to easily interact with cysteine, methionine and histidine residue
of proteins (14-17).
Abbreviations: PGP I: Bacillus amyloliquefaciens pyroglutamyl-peptidase I; BP-Pt(II): Bipyridine
platinum(II); Cisplatin: Cis-diamminedichloroplatinum(II); Carboplatin: Cis-diammine-1,1-cyclobutanedicarboxylate platinum(II); CVB3 3Cpro: Coxsackievrius stain B3 3Cpro; DHB: 2,5-Dihydroxybenzoic
acid; DTT: Dithiothreitol; GspA: N-teminal amidase of glutathionyl-spermidine synthetase/amidase;
GspSA: Glutathionyl-spermidine synthetase/amidase; hTrxR: Human thioredoxin reductase;
IPTG: Isopropyl β-D-1-thiogalactopyranoside; IC50: Inhibitory concentration at 50% inhibition;
MALDI: Matrix-assisted laser desorption/ionization; PDB: Protein Data Bank; PGL: Pyrogultaminal;
RMSD: Root mean square deviation; SeC: Selenocysteine; TCEP: Tris(2-carboxyethyl)phosphine;
TGR: Thioredoxin/glutathione reductase; TP-Pt(II): (2,2′:6′,2′′-terpyridine)platinum(II); TryR: Trypanothione reductase; HTP-TP-Pt(II) (4-Hydroxylthiophenolato) (2,2′:6′,2′′-terpyridine)platinum(II);
NAP-TP-Pt(II) (N-Acetyl-4-aminothiophenolato) (2,2′:6′,2′′-terpyridine)platinum(II).
*Phone: 1886-2-2788-1981
Fax: 1886-2-2788-2043
E-mail: ahjwang@gate.sinica.edu.tw
267
268
Lo et al.
H3N
Pt
NH 3
Cl
N
1. Cisplatin
Cl
Cl
Pt
+
N
N
N
Cl
Pt
N
S
4. Bipyridine Platinum(II)
H3N
Cl
Pt
Cl
6. HTP-TP-Pt(II)
NH 3
2. Transplatin
O
H3N
H3N
Pt
OH
O
O
3. Carboplatin
O
+
N
N
Pt
N
Cl
5. Terpyridine Platinum(II)
+
N
N
Pt
N
S
7. NAP-TP-Pt(II)
O
N
H
Figure 1: The platinum(II) complexes used in this study. 1. Cisplatin, cis-diamminedichloroplatinum(II);
2. Transplatin, trans-diamminedichloroplatinum(II); 3 Carboplatin, cis-diammine-1,1-cyclobutanedicarboxylate platinum(II); 4. BP-Pt(II), bipyridine platinum(II) dichloride; 5. TP-Pt(II), terpyridine
platimun(II) chloride; 6. HTP-TP-Pt(II), (4-hydroxylthiophenolato) (2,2′:6′,2′′-terpyridine) platinum(II)
and 7. NAP-TP-Pt(II), (N-acetyl-4-aminothiophenolato) (2,2′:6′,2′′-terpyridine) platinum(II).
Recently, the interactions between platinum(II) complexes (e.g., cisplatin, TP-Pt(II),
etc.) and cysteine proteases (e.g., papain, cathepsin B, caspases, etc.) have been
investigated (18-20). Cisplatin was shown to inactivate human caspase 3 and ultimately inhibited proteins (such as TNF-α, TRAIL, and FasL) that induce cell death
in vitro (18). TP-Pt(II) chloride was proposed to label the active-site Cys25 of papain
based on kinetic assay (19). However, the platinum(II) amine complexes did not
show significant inhibitory activity against cathepsin B (20). How do platinum(II)
complexes interact with cysteine proteases remains unclear.
Our previous study showed that TP-Pt(II) complexes are effective inhibitors of
human thioredoxin reductase 1 (hTrxR1) (21). The active site of hTrxR1 is composed of Cys59, Cys64, His472, Glu477, Cys497, SeC498. His472 and Glu477
which act as acid-base catalysts, facilitating the electron and proton transfers. TPPt(II) complexes were found to bind to the essential pair of C497-SeC498 in the
active site.
The catalytic triad of the cysteine protease (e.g., papain, cathepsin B, etc.) contains
active-site cysteine, along with a histidine and a glutamate or asparagine. Because
cysteine proteases and hTrxR1 contain similar active-site residues (cysteine or
selenocysteine, histidine and a glutamate or asparagine), we hypothesize that
platinum(II) complexes may interact with many cysteine proteases, and chose three
of them, namely the 3C-protease (3Cpro) from human coxsackievrius strain B3, the
glutathionyl-spermidine synthetase/amidase (GspSA) from Escherichia coli, and
the pyroglutamyl-peptidase I (PGP I) from Bacillus amyloliquefaciens, to address
whether the platinum(II) complexes can specifically interact with cysteine proteases and inhibit their functions.
Cysteine proteases play important roles in a number of biological processes and
can be classified into three groups according to their structural folds (see Table SI,
Figure S1) (22). 3Cpro, a cysteine protease with a chymotrypsin-like fold, is required
for the proteolytic processing of large polyproteins and is essential for viral replication and maturation, making it a target for antiviral drug design (23). GspSA plays
an important role in redox regulation and its N-terminal amidase domain (GspA),
a papain-like fold cysteine protease (24), is capable of hydrolyzing trypanothione
Table I
Diffraction data, refinement, and model statistics for PGP I, TP-Pt(II)-PGP I and TP-Pt(II)-SsPTP
structures.
Crystals
PGP I
TPT-PGP I
TPT-SsPTP
Data collection
Space group
Unit cell a, b, c (Å)
β
Resolution (Å)
No. of observations
Unique reflections
Redundancy
Completeness (%)
Average I/σ(I)
Rmerge (%)
C2
C2
P41
289.9/45.4/68.1
290.3/45.5/68.0
72.8/72.8/32.3
91.3
91.5
25.022.01 (2.0422.01) 30.021.66 (1.7221.66) 25.021.9 (1.9321.90)
250120
402080
64337
59251 (2926)
102303 (9960)
13525 (646)
4.2 (4.0)
3.9 (3.8)
4.8 (4.6)
99.6 (98.6)
97.2 (95.7)
99.4 (100)
32.9 (8.4)
24.1 (6.0)
22.7 (6.2)
3.5 (13.8)
4.5 (21.2)
5.1 (32.5)
Refinement
Positive reflections
Rwork (95% data)
Rfree (5% data)
R.m.s.d bond distance (Å)
R.m.s.d bond angle (°)
Ramachandran plot (% residues)
Core dihedral angles (%)
Allowed areas (%)
Other areas (%)
58470 (2721)
0.174 (0.183)
0.209 (0.237)
0.019
1.84
99739 (8880)
0.174 (0.211)
0.203 (0.248)
0.023
2.01
13107 (583)
0.183 (0.191)
0.242 (0.265)
0.023
2.06
91.4
8.0
0.6
91.4
8.0
0.6
94.9
3.8
1.3
18.0 (6345)
35.3 (958)
39.1 (95, 5)
22.9 (1284)
42.5 (141)
71.4 (19, 1)
Average B (Å2) (no. of atoms and no. of ligands)
Protein
Water
TPT
21.8 (6341)
29.4 (436)
–
PGP I and TPT-SsPTP data were collected using in-house Rigaku X-ray source.
TPT-PGP I data were collected using Taiwan beamline BL12B2 at the SPring-8 in Japan.
Values within parentheses represent data in the highest resolution shell.
Rwork5 Σ Fobs  2 Fcalc /Σ Fobs  was calculated for all observed data. No sigma cut-off was used.
Rfree was calculated for 5% of randomly chosen unique reflections that were excluded from the
refinement.
or glutathionylspermidine back to spermidine and glutathione. PGP I, a cysteine
peptidase, can remove the N-terminal pyroglutamyl residue which protects oligopeptides and proteins from being digested by aminopeptidases. PGP I may abort the
biological activities of N-terminal pyroglutamyl signal pepetides and proteins, such
as neurotensin, luteinizing hormone releasing hormone and thyrotropin releasing
hormone (25, 26).
Here, the inhibition assays offered a clue to the interactions between platinum(II)
complexes and cysteine protease. Furthermore, detailed features of these interactions were unraveled by using X-ray crystallographic and MALDI TOF-TOF
spectrometric analyses. It is likely that the TP-Pt(II) complexes inhibit the cysteine
proteases by binding to the cysteine residue of the catalytic triad in the active site.
Materials and Methods
The chemicals used in these studies were purchased from Alfa Aesar® and SigmaAldrich. The TP-Pt(II) complexes, CVB3 3Cpro, GspA and SsPTP were prepared as
shown in our previous studies (21, 23, 24, and 27).
Cloning, Expression, Purification and Activity Measurement of PGP I
The PGP I gene was PCR-amplified from Bacillus amyloliquefaciens (ATCC 49763)
genomic DNA and cloned into pET23a (Novagen), generating pET23a-PGP I-(His)6.
269
TP-Pt(II) Complexes Inhibit
Cysteine Proteases
270
Lo et al.
The PGP I gene was verified by sequencing and found to have two altered residues,
M58I and A202V as compared to the published PGP I (28). These two mutations
do not appear to affect the structure. Cultured Ecos 21 (Yeastern Biotech Co.) was
used for overexpression of PGP I, to an A600 of 0.6-0.8 in LB medium containing
100 mg/L ampicillin at 30°C, induced overnight by the addition of 1 mM isopropyl
β-D-1-thiogalactopyranoside and then harvested. The culture cells were centrifugated, resuspended in 20 mL (per gram cell weight) buffer A (50 mM Tris-HCl,
pH 8.0, 10 mM imidazole, 500 mM NaCl) and lysed at 25 MPa using a French
press. After centrifugation for 30 min at 28,000 3 g at 4°C, the supernatant was
collected, loaded onto Ni21-affinity column (HiTrap Affinity resin, Pharmacia)
and eluted with linear gradient from 10% to 60% of buffer B (50 mM Tris-HCl,
pH 8.0, 500 mM imidazole, 500 mM NaCl). Fractions containing the purified PGP
I protein were pooled and dialyzed against buffer C (25 mM Tris-HCl, pH 8.0,
50 mM NaCl) and concentrated by Amicon (10 kDa cut-off, Millipore). The concentrated proteins were loaded onto SP100-gel filtration column (HiTrap Affinity
resin, Pharmacia) and eluted by buffer C. Fractions containing the purified PGP
I protein were pooled and concentrated against 25 mM Tris-HCl, pH 8.0, 50 mM
NaCl, 0.25 mM TCEP (tris(2-carboxyethyl)phosphine) concentrated by Amicon
(10 kDa cut-off, Millipore). The stock concentration of PGP I is about 220 mg/mL
and frozen at 280°C. The molecular weight of PGP I was determined by MALDI
mass spectrometry.
Activity Assay of PGP I
To measure the inhibitory activity of platinum(II) complexes on PGP I, the experiment was modified from the published method (29). The experiments were performed in 2.5 mL reaction mixture 50 mM Tris-HCl buffer, pH 8.0, containing 2.7
nM PGP I and various concentration of platinum(II) complexes. The mixtures were
incubated for 10 min at 25°C and 150 μM pyroGln-β-naphthylamide was added to
start the reaction. The reaction was monitored continuously for 5 min at 410 nm
by using Perkin-Elmer LS-55 fluorescence spectrometer. Upon drug treatment, the
PGP I activity is expressed as a percentage of that determined in the untreated
control sample. The IC50 values were calculated based on the results of inhibitory
assay toward PGP I activity.
Crystallographic Analysis
Crystallization condition for PGP I was screened using sitting drop methods and
SaltRx kit (Hampton Research). The condition for PGP I crystallization was optimized by mixing 2 μL PGP I (concentration of 55 mg/mL) and 2 μL [1.8 M sodium
phosphate monobasic monohydrate, potassium phosphate dibasic/pH 6.9], from
which large, single crystals were obtained after 15 days. The TP-Pt(II)-PGP I
complex crystals were obtained by using the same condition expect by adding
1 μL 10 mM TP-Pt(II)-Cl. Large, single crystals with pink color, typical of TPPt(II)-proteins, were obtained after 17 days. The cryoprotectant contained 1.1 M
sodium phosphate monobasic monohydrate/potassium phosphate dibasic, 25%
glycerol, 5% ethylene glycol, and 7% sucrose. The TP-Pt(II)-SsPTP complex
crystals were obtained by using 0.1 M sodium citrate (pH 5.5) and 20% PEG3000
as in the previously optimized condition (27), and the molar ratio of TP-Pt(II)Cl:SsPTP was 1:1.
X-ray data of PGP I crystals were recorded using a Rigaku R-Aixs IV11 detector
at a temperature of 100 K and a wavelength of 1.54 Å by in-house Rigaku rotating
anode X-ray FR-E generator and processed with the HKL suite (30). Additional
X-ray data at 1.66 Å resolution of TP-Pt(II)-PGP I complex crystals were collected
on an ADSC Q215 detector at a temperature of 100 K and a wavelength of 1.0 Å
at the synchrotron beamline 12B2 (NSRRC, Japan) and processed with the HKL
suite. These crystals belong to the monoclinic space group C2 with unit-cell dimensions a 5 290 Å, b 5 46 Å, c 5 68 Å, b 5 91.3 and contain four molecules in an
asymmetric unit.
The monomer structure of PGP I (PDB code: 1AUG, space group: P2) was used as
a search model for molecular replacement (28). All data of PGP I (25-2.01 Å) and
TP-Pt(II)-PGP I (30-1.66 Å) were used in the calculations by MOLREP of CCP4
(31, 32), from which an unambiguous solution was obtained, yielding an initial R
of 23.7% (Rfree 5 23.9%) after rigid body refinement by the program CNS (33). The
structure of TP-Pt(II)-SsPTP was solved by using SsPTP (PDB code: 2I6I, space
group: P41) as a model for molecular replacement (27). The refinement parameters
are summarized in Table I. Manual rebuilding of the model was carried out by
using the program XFIT and COOT based on σA-weighted 2Fo−Fc and Fo−Fc electron density maps (34, 35). Further refinement led to Rwork 5 0.19 and Rfree 5 0.21.
The atomic coordinates of PGP I, TP-Pt(II)-PGP I and TP-Pt(II)-SsPTP have been
deposited in the Protein Data Bank, with accession codes 3RNZ, 3RO0 and 3RO1.
Figures were made with the program Pymol (36).
Detection of the TP-Pt(II)-CVB3 3Cpro and TP-Pt(II)-GspA Adduct by MALDI
TOF-TOF Mass Spectrometry
The TP-Pt(II)-CVB3 3Cpro protein adduct was prepared by molar ratio 1:1.1 ([protein] : [TP-Pt(II)]). The TP-Pt(II)-GspA protein adduct was prepared by molar ratio
1:2.2 ([protein] : [TP-Pt(II) complexes]). The reaction mixture of all three protein
adducts were pre-incubated in 50 mM Tris-HCl buffer at 37°C for 3 h and subsequently incubated at 4°C for 48 h. The reaction mixture was removed by a PD10
(Sephadex G25) desalting column. After desalting process, the protein adducts
were digested by trypsin at a trypsin/protein ratio of 1:50 at 37°C for overnight
in 50 mM (NH4)2CO3, 10% acetonitrile. The mixture peptides were eluted twice
with 60% acetonitrile, 0.1% TFA from a ZipTip C18 column (Millipore). The eluate was then premixed 1:1 with a 2,5-DHB matrix (2,5-Dihydroxybenzoic acid,
Bruker) and spotted onto the sample plate. A Bruker Daltonics AutoFlex III smartbeam TOF-TOF 200 mass spectrometer was used to detect the peptide mass.
Results
Inhibition of PGP I Activity by the Platinum(II) Amine Complexes
To examine the inhibitory activities of TP-Pt(II) complexes toward PGP I, the IC50
values were determined by including different concentrations of the compounds
in the activity measurements. Among the platinum(II) complexes, BP-Pt(II) had
marked inhibitory activity (Figure 2 and Table II). Here, the weakest inhibitor is
carboplatin, which is a prodrug and thus has lower inhibitory activity than cisplatin.
Interestingly, the inhibitory activities of BP-Pt(II) and HTP-TP-Pt(II) are vary similar, and they are slight stronger than those of TP-Pt(II) and NAP-TP-Pt(II). Probably the HTP, NAP and chloride were replaced by water molecules before reacting
with enzyme. In sum, the order of inhibitory effects of several platinum(II) complexes (Figure 1) on PGP I is polypyridine platinum(II) complexes  cisplatin 
transplatin and carboplatin (Table II). Increasing the size of aromatic ring of a
platinum(II) complex enhanced the inhibitory activity toward PGP I, likely due to
the hydrophobic interactions between the aromatic ring of the platinum(II) complexes and the hydrophobic side chains of the enzyme.
Crystallographic Features of the TP-Pt(II)-PGP I Complex
The structural basis of the inhibition of PGP I by TP-Pt(II) was then investigated
by X-ray crystallography. The TP-Pt(II)-PGP I complex crystals belong to the
271
TP-Pt(II) Complexes Inhibit
Cysteine Proteases
272
Lo et al.
monoclinic space group C2 and contain four independent protein molecules (with
chain names A/B/C/D) in an asymmetric unit (Table I). In addition to the four TPPt(II) complexes, each bound to the active-site cysteine of a PGP I chain, a fifth
TP-Pt(II) was coordinated to a surface His161 residue in the D chain. The residues
Figure 2: Inhibitory effects of the platinum complexes on PGP I. (A) cisplatin; (B) transplatin; (C) carboplatin; (D) bipyridine platinum; (E) terpyridine platinum;
(F) HTP-TP-Pt(II); (G) NAP-TP-Pt(II). Experiments were done in triplicate.
273
His161 of A and C chains are buried in the interface; consequently, the TP-Pt(II)
complex is difficult to interact with the His161 residues. No such density was
observed in the native crystal; nor was it seen in the B chain. The interaction between
the fifth TP-Pt(II) and PGP I seems to be non-specific and may be caused by the
slightly excess amount of TP-Pt(II) (the molar ratio of TP-Pt(II) to PGP I was 1.2:1).
The r.m.s. deviations between four TP-Pt(II)-PGP I monomers are 0.16-0.20 Å
TP-Pt(II) Complexes Inhibit
Cysteine Proteases
Table II
The PGP I activity was inhibited by various platinum(II) complexes.
Compounds
IC50 (µM)
1
2
76.28 6 4.24
169.70 6 7.69
3
4
5
200 18.34 6 3.91 31.89 6 0.64
6
7
21.33 6 0.57
31.77 6 1.54
The data shown represent the mean ± standard deviation from three independent experiments.
for 206 Cα-atoms. The TP-Pt(II)-PGP I monomer structure is also similar to free
PGP I (see Table I) with r.m.s.d. of 0.15 Å for 206 Cα-atoms. Figure 3A shows that
TP-Pt(II) is directly ligated to the active-site Cys144 of PGP I. The well resolved
electron density clearly delineates the conformation of the TP-Pt(II)-Cys144 of the
PGP I complex (Figure 3B). The structure of PGP I in complex with the inhibitor
pyroglutaminal (PGL) has been solved (37). In Figure 3C, superimposition of the
active sites of TP-Pt(II)-PGP I and PGL-PGP I structures shows that the interaction
Figure 3: The active sites of TP-Pt(II)-PGP I. (A) Ribbon presentation of TP-Pt(II)-PGP I binding site (top view). TPT is the residue name of TP-Pt(II) complex,
shown in orange. (B) 2Fo2Fc electron density map at 1.66 Å contoured at 1.0 σ level (blue) for the TP-Pt(II) and Cys144 of PGP I, and at 12 σ level for the Pt atom
(red). (C) The superposition of TP-Pt(II)-PGP I and PGL-PGP I structures.
274
Lo et al.
networks are different in which Phe10, Phe13, Ile92, and Phe142 form a hydrophobic pocket to enhance the TP-Pt(II) binding. Besides, the main-chain carbonyl group of Gln71 (Figure 3C) is pointed toward to the platinum(II) atom (3.6 Å)
similar to the O6 atoms of the DNA guanine rings stacking over the platinum atom
(3.4 Å) for stabilization (38). In the PGL-PGP I structure, PGL was sandwiched
between Phe10 and Phe13 and formed hydrogen bonds with the main-chain atoms
of Gln71 (37).
This crystal structure of TP-Pt(II)-PGP I correlates well with our PGP I activity
assay (Table II). The hydrophobic environment of PGP I enables BP-Pt(II) and
TP-Pt(II) to exhibit higher inhibitory potential than cisplatin. Superimposing of the
TP-Pt(II)-PGP I and PGL-PGP I structures in Figure 3C suggests that the active site
of PGP I is large enough to accommodate either TP-Pt(II) complex or PGL.
MALDI-TOF-TOF Analysis of TP-Pt(II) Chloride Binding to 3Cpro
To further investigate whether platinum(II) complexes could also bind to cysteine
proteases with different structure folds, mass spectrometry was used. TP-Pt(II)-3Cpro
(human coxsackievrius) and TP-Pt(II)-GspA (E. coli) were used as models to examine
the reaction between platinum(II) complexes and cysteine proteases. Compared with
the untreated proteins, the TP-Pt(II)-treated 3Cpro and GspA gained extra peaks with
an increase of 428 Da, suggesting that both proteins could be tagged by the TP-Pt(II)
complex (in Figure S2). To identify the TP-Pt(II) complex modification site in 3Cpro
and GspA, the untreated and TP-Pt(II) treated proteins were digested with trypsin and
then analyzed by MALDI-TOF-TOF. The detailed peak assignments of 3Cpro and
TP-Pt(II)-3Cpro are shown in Figure 4A and Table III. Most notably, in the TP-Pt(II)3Cpro mass spectrum, the active-site peptide (144AGQCGGVLMSTGK156, MW 5
1208.553) plus one TP-Pt(II) (MW 5 428.35) is represented by a significant peak
with an M/Z ratio of 1634.677 (Figure 4B). There are three isotopes of platinum: 194
(32.9%), 195 (33.8%) and 196 (25.2%). To confirm that the TP-Pt(II) complex was
bound to the active-site cysteine residue, the modified fragment 144AGQC-[TP-Pt(II)]
GGVLMSTGK156 was further analyzed by MS/MS (shown in Figure 4C). The singly charged C-terminal product ions (y2-y6 and y9-y12) and H2O-loss fragment ions
(y5 and y10) were observed. The fragment ions (y10-H2O and y9) had their masses
increased by 512 Da [TP-Pt(II)-Cys147], suggesting that the TP-Pt(II) was bound
directly to the sulfur in the 3Cpro active-site motif 144AGQCGGVLMSTGK156.
TP-Pt(II) Complexes Binding to GspA
As to the analysis of GspA, the detailed peaks assignments of TP-Pt(II)-treated
and untreated samples were shown in Figure 5A and Table IV. The peak assigned
to 57WQC-[TP-Pt(II)]VEFAR64, was observed at 1464.685 M/Z in the TP-Pt(II)GspA mass spectrum (in Figure 5B). In the TOF-TOF analysis, the singly charged
C-terminal product ions (y1-y7) were observed and the fragment ions (y5 and y6)
had their masses increased by 529 Da [TP-Pt(II)-Cys59], suggesting that the
TP-Pt(II) is bound directly to the sulfur in the active-site 57WQCVEFAR64 motif
of GspA (in Figure 5C).
Table III
Peptides identified by MALDI-TOF from 3Cpro and TP-Pt(II)-3Cpro.
Position
Peptide sequence
20-33
34-39
93-108
135-143
144-156
144-156
157-176
TEYGEFTMLGIYDR
WAVLPR
EEVEVNEAVLAINTSK
MLMYNFPTR
AGQCGGVLMSTGK
AGQC-(TPT)GGVLMSTGK
VLGIHVGGNGHQGFSAALLK
Theoretical Mass
Observed Mass
1694.77
741.44
1744.90
1172.56
1208.58
1635.92
1975.09
1694.86
741.39
1744.76
1172.62
1208.55
1634.69
1975.14
In addition, the TP-Pt(II)-modified TP-Pt(II)-3Cpro and TP-Pt(II)-GspA displayed
three fragments, 428.165 (TP-Pt(II)), 460.144 (TP-Pt(II)-S), and 474.150 (TPPt(II)-S-CH2) in their TOF-TOF spectra (see Figure 6 for the spectra of 3Cpro), confirming that the TP-Pt(II) complex covalently modifies the active-site cysteine of
3Cpro and GspA. In conclusion, the TP-Pt(II) complexes can bind to the active-site
cysteine of different cysteine proteases, regardless of their structural folds.
275
TP-Pt(II) Complexes Inhibit
Cysteine Proteases
Figure 4: Trypsin digestion result of 3Cpro. The protein was pre-incubated with (lower panel) or without (upper panel) TP-Pt(II) complex. (A) The mass spectrum
shows whether the cysteine-containing peptide in the active site was modified with TP-Pt(II) complexes (MW 428.06) or not. (B) The spectrum shows the TP-Pt(II)peptide adduct. (C) The mass spectrum of the [M 1 H]1 ion at m/z from the TP-Pt(II)-modified fragment with the sequence AGQC-(TPT)GGVLMSTGK.
276
TP-Pt(II)of HTP-TP-Pt(II) and NAP-TP-Pt(II) Complexes Transferring to GspA
Lo et al.
In a previous study, 4′-Cl-TP-Pt(II) of pyridine-4-thione-4′-Cl-TP-Pt(II) was
shown to be transfered and attached to Cys34 in human serum albumin (39). The
pKa value of Cys34 of human serum albumin is about 5. We found that HTP-TPPt(II) and NAP-TP-Pt(II) could also interact with GspA. In Figure 7, the peaks at
1464.685 M/Z are assigned to 57WQC-[TP-Pt(II)]VEFAR64, and observed in all
Figure 5: Trypsin digestion of GspA. (A) The GspA was pre-incubated with or without TP-Pt(II) complexes before trypsin digestion and the following MS
analysis. The spectrum shows whether cysteine-containing peptide of GspA active site interacted with TP-Pt(II) complexes or not. (B) The spectrum shows that
TP-Pt(II)-peptide adduct. (C) The TOF/TOF spectrum of the [M 1 H]1 ion at m/z from the TP-Pt(II)-modified GspA active site sequence WQC-(TPT)VEFAR.
Table IV
The observed and theoretical masses of peptide sequences of GspA and TP-Pt(II)-GspA after
tryptic-digested.
Position
Peptide sequence
65-75
76-83
76-83
85-107
111-129
130-141
147-161
164-182
183-196
183-196
SYIDDEYMGHK
WQCVEFAR
WQC-(TPT)VEFAR
FLFLNYGVVFTDVGMAWEIFSLR
EVVNDNILPLQAFPNGSPR
APVAGALLIWDK
DTGHVAIITQLHGNK
IAEQNVIHSPLPQGQQWTR
ELEMVVENGCYTLK
ELEMVVENGC-(TPT)YTLK
Theoretical Mass
Observed Mass
1357.57
1038.48
1465.82
2724.39
2080.08
1253.73
1603.86
2202.14
1627.77
2055.11
1357.60
1038.55
1464.66
2724.62
2081.24
1253.75
1603.57
2202.32
1625.98
2054.75
277
TP-Pt(II) Complexes Inhibit
Cysteine Proteases
of HTP-TP-Pt(II)-GspA, NAP-TP-Pt(II)-GspA and TP-Pt(II)-GspA mass spectra.
These implied that HTP-TP-Pt(II) and NAP-TP-Pt(II) can be transfered to the
active-site cysteine of cysteine proteases. Normally, the pKa of a cysteine residue
is 8.5, but a reduced value occurs in the active site cysteine of cysteine proteases.
It is noted that the active-site cysteine of GspA has low pKa values of about 3.05
(40). Here, the sulfhydryl group of the low-pKa cysteine residue of proteases
will form a reactive nucelophilc thiolate. The reactive nucelophilic thiolate of
active-site cysteine residue of cysteine proteases attaches to the platinum atom
of TP-Pt(II) complexes and replaces the chloride atom, HTP and NAP as leaving
groups. This transfer phenomenon to a low-pKa cysteine residue of proteins is
elucidated by our results. Not only TP-Pt(II) complexes have the ligand exchange
ability, but other metal complexes also have this ability (like Ru and Au) (2). The
low-pKa cysteine residue in the active site of proteins is more reactive for metal
complexes.
Figure 6: The MS/MS spectrum of TP-Pt(II)-labeled active site cysteine residue of 3Cpro. TPT denotes the TP-Pt(II) complex.
278
Discussion
Lo et al.
A recent study showed that the TP-Pt(II) complexes could irreversibly inhibit
hTrxR1 at nanomolar concentration (21), in contrast to the micromolar concentration level inhibition of PGP I in this study. The selenocysteine residue in the
C-terminal active site of TrxR is a highly reactive nucleophile (pKa 5 5.3) under
physiological conditions. The different effect on hTrxR1 and PGP I may be caused
Figure 7: The MS/MS spectrum of TP-Pt(II)-labeled GspA. (A) HTP-TP-Pt(II); (B) NAP-TP-Pt(II); (C) TP-Pt(II)-Cl. The GspA-TP-Pt(II) adducts were trypsindigested and subjected to MS analysis. The spectra show that the cysteine-containing peptide of GspA active site was labeled by TP-Pt(II).
by the different binding environment. In Figure 3A, the TP-Pt(II) is perpendicularly
positioned in the active site of PGP I, bound to Cys144 and in contact with hydrophobic residues through CH/π interactions. In comparison, TP-Pt(II) complexes
bind to hTrxR1 via stacking with Trp114 and covalent bonding to the active-site
selenocysteine. The active sites of TryR and hTrxR1 have hydrophobic and stacking environments for providing TP-Pt(II) complexes a niche to bind to and thus
cause the prominent inhibitory effect. In contrast, the PGP I only has a hydrophobic
environment for interacting with TP-Pt(II) complexes.
The studies of platinum(II)-amine complexes labeling to papain-like cysteine proteases, including papain and cathepsin B, have been reported with different results
(19, 20). The obvious difference between platinum(II) alkyl-amine complex and
TP-Pt(II) complex is that TP-Pt(II) complex has aromatic polypyridine rings which
increase their interaction with papain-like cysteine proteases. Although PGP I,
GspA, and 3Cpro have the same catalytic triad, their structure folds are quite different (Figure S1). The active sites of PGP I, GspA, and 3Cpro contain hydrophobic
phenylalanine, tryptophan and tyrosine residues, and have large enough cavities for
binding with the TP-Pt(II) complex.
It is worthy to note that not all cysteine enzymes can bind TP-Pt(II) complexes specifically at the active site. We have tested the binding of TP-Pt(II) to protein tyrosine
phosphatase from Sulfolobus sofataricus (SsPTP) by co-crystallizing TP-Pt(II) with
SsPTP (27). The catalytic center (cysteine, arginine and aspartate) of PTPs is not
like that of other cysteine proteases which have triad or diad (cysteine, histidine
and a glutamate or asparagine). The crystal structure determined at 1.9 Å resolution
revealed that TP-Pt(II) binds at the entrance of the active site, stacking on the tryptophan side chain of Trp39 residue, but not coordinating to the Cys96 in the active
site (Figure 8, Table I). We also examined the effects of platinum(II) complexes
against SsPTP activity. The results indicated that the TP-Pt(II) complexes did not
have significant inhibitory activity toward SsPTP at 10 µM while some platinum(II)
complexes had slight inhibitory activity at 10 µM. Further inhibitory characteristics
of the platinum(II) complexes toward PTPs remain to be investigated.
279
TP-Pt(II) Complexes Inhibit
Cysteine Proteases
In trypanosome, the trypanothione reductase (TryR)-trypanothione synthetase/amidase (TSA) system is the major redox regulator (41). TP-Pt(II) complexes have
been known to inhibit TryR (42) activity and we found they can tag on GspA,
which contains a papain-like cysteine protease domain highly analogous to the N
terminus of TSA. Moreover, in trypanosome, several chemotherapeutic targets,
such as Trypanosoma brucei cathepsin B-like enzyme (TbCat) and Rhodesain, are
papain-like cysteine proteases (43). Accordingly, our study will be useful for developing treatment of trypanosome parasitic disease.
TP-Pt(II) chloride complex has been shown to label to the active-site hsitidine residue of chymotrypsin (serine protease) (44). The chloride of TP-Pt(II) complex is
replaced by a water to form TP-Pt(II) hydrate and then the TP-Pt(II) complex forms
a coordination bond with the histidine residue of chymotrypsin (Figure 9A). From
Figure 8: The structure of TP-Pt(II)-SsPTP. (A) 1.9 Å 2Fo2Fc electron density map (blue) of TP-Pt(II) and E40 of SsPTP is contoured at 1 σ level and 2Fo2Fc
electron density map (red) of Pt atom is contoured at 9 σ level. TPT is colored orange. (B) Structure of TP-Pt(II)-SsPTP presented as surface. (red: oxygen atom,
blue: nitrogen atom, gray: carbon atom, yellow: sulfur atom and magenta: water) (C) Structure of TP-Pt(II)-SsPTP presented as stereoscopic ribbon diagrams. There
is a TP-Pt(II) stacking on the W39 and the chloride of TP-Pt(II) complex is replaced by water to form TP-Pt(II) hydrate.
280
+
(A)
[TP-Pt(II)Cl] + H2O
[TP-Pt(II)(H 2O)]
2+
2+
Lo et al.
Pt
N
N
N
OH2
N
–
O
NH
OH
Pt
N
N
NH
OH
O
α -chymotrypsin
(B)
–
2+
N
N
+ Cl
H 2O
–
O
O
α -chymotrypsin
2+
+
N
N
Pt
N
N
OH2
N
SH
–
O
NH
O
Cysteine Proteases
N
+
Pt
H 3O
N
S
N
–
O
NH
O
Cysteine Proteases
Figure 9: The mode of interaction between platinum(II) complexes and cysteine protease. (A) The
water hydrates TP-Pt(II) chloride complex to form TP-Pt(II) hydrate. The residue of histidine in the
active site of α-chymotrypsin is tagged with TP-Pt(II). (B) The TP-Pt(II) complex forms a covalent
bond with the active-site cysteine residue of cysteine proteases.
our data, the TP-Pt(II) complexes tag on the active-site cysteine residue of 3Cpro,
which is a chymotrypsin-like cysteine proteases (Figure 9B), implying that the
active-site cysteine residue has higher reactivity toward the TP-Pt(II) complexes
than the histidine residue in cysteine protease. A sulfhydryl group is a stronger
nucleophile than an imidazolium cation, and thiolate is stronger than imidazole
in papain binding site (19). Cisplatin was shown to inactivate human caspase 3
activity (18) and BP-Pt(II)-Cl2 was shown to inactive PGP I activity. Our mass
analysis further indicated that BP-Pt(II)-Cl2 complex tagged to GspA (Figure S3).
We cannot exclude the possibility, however, that the active site architecture could
favor the interaction of Cys144 of PGP I with the TP-Pt(II) complex to form the
covalent bond, and the higher reactivity of Cys144 may or may not be the driving factor. In fact, the bidentate platinum(II) complexes (like cisplatin, bipyridine
platinum(II)) may interact with both the active-site cysteine and histidine residues
(like zinc complexes binding to 3Cpro) (45).
Besides, terpyridine is a chelating compound and can coordinate to a variety of
metal ions (like Au, Pd, Zn and Ru) (46). These metal complexes are in a high
structural diversity as caused by the different coordination numbers and geometries
(47). Therefore, terpyridine can form complexes with platinum(II) or other metal
ions and can be used as pharmacophores for designing anti-cancer, anti-parasite
and anit-virus drugs. Although TP-Pt(II) complex is an irreversible covalent inhibitor of cysteine proteases, it still has benefit for drug design. For example, in HDACi
(histone deacetylase inhibitors) resistant cancer cells, hTrxR1 was upregulated
(48). TP-Pt(II) can be used in conjunction with the HDACi inhibitors to form dual
function anticancer drug. Not only hTrxR1 expression was upregulated in cancer
cells but cathepsin B expression was also increased in various tumors (like bladder,
colon and prostate carcinomas) (49). In our opinion, the increased protein level,
for example hTrxR1, of cancer cell will result in the increased total hTrxR1 in our
body. The irreversible inhibitor will lead to the loss of hTrxR1 function (equivalent
to the decrease of hTrxR1 level) that can render the tumor growth controllable.
While the irreversible nature of the inhibitors needs more extensive toxicological
studies to cope with, the complex crystal structures can provide us with detailed
information for increasing binding affinity and decreasing toxicity.
Supplementary material giving details on the classification of cysteine proteases
structure folds and whole molecular weights of TP-Pt(II)-GspA, TP-Pt(II)3Cpro and BP-Pt(II)-GspA are posted at the JBSD web site where the article
appears.
Acknowledgements
We gratefully acknowledge the National Synchrotron Radiation Research Center, Taiwan, and Core Facility for Protein X-ray Crystallography in Academia
Sinica. We thank Dr. Cheng-Chung Lee, Dr. Hsing-Mao Chu, Dr. Chien-Hua
Pai, and Mr. Tsung-Lin Chou (all from Institute of Biological Chemistry Academia Sinica, Taiwan) for their proteins in the mass experiments, Miss Hsilin
Cheng from Institute of Molecule Biology Mass Spectrometry Facility for mass
analysis, Professor Kiyoshi Ito for providing the coordinates of the PGL-PGP
I structure, Dr. Shiuan-Woei LinWu and Miss Andrea Tseng Lai for editing the
manuscript.
References
1. L. Lu and M. Zhu. Anticancer Agents Med Chem 11, 164-171 (2011).
2. S. P. Fricker. Metallomics 2, 366-377 (2010).
3. H. Mansouri-Torshizi, M. I. Moghaddam, A. Divsalar, and A. A. Saboury. J Biomol Struct
Dyn 26, 575-586 (2009).
4. A. Divsalar, A. A. Saboury, H. Mansoori-Torshizi, M. I. Moghaddam, F. Ahmad, and
G. H. Hakimelahi. J Biomol Struct Dyn 26, 587-597 (2009).
5. P. J. Sadler and Z. Guo. Pure and Appl Chem 70, 863-871 (1998).
6. A. A. Moosavi-Movahedi, S. J. Mousavy, A. Divsalar, A. Babaahmadi, K. Karimian,
A. Shafiee, M. Kamarie, N. Poursasan, B. Farzami, G. H. Riazi, G. H. Hakimelahi,
F.-Y. Tsai, F. Ahmad, M. Amani, and A. A. Saboury. J Biomol Struct Dyn 27, 319-329 (2009).
7. M. Selim and K. K. Mukherjea. J Biomol Struct Dyn 26, 561-566 (2009).
8. Z. R. Lü, Y. J. Wang, D. Y. Lee, Y. D. Park, H. C. Zou, and F. Zou. J Biomol Struct Dyn
26, 567-574 (2009).
9. Q. Sheng, Z. R. Lu, H. Mu, H. C. Zou, F. Zou, and S. J. Yao. J Biomol Struct Dyn 27, 59-64
(2009).
10. P. Sharma, S. Sharma, A. Mitra, and H. Singh. J Biomol Struct Dyn 27, 65-81 (2009).
11. T. C. Ramalho, M. V. J. Rocha, E. F. F. Da Cunha, L. C. A. Oliveira, and K. T. G. Carvalho.
J Biomol Struct Dyn 28, 227-238 (2010).
12. T. W. Hambley. J Chem Soc Dalton Trans, 369-372 (2001).
13. G. Lowe, J. A. McCloskey, J. S. Ni, and T. Vilaivan. Bioorg Med Chem 4, 1007-1013
(1996).
14. V. Calderone, A. Casini, S. Mangani, L. Messori, and P. L. Oriol. Angew Chem Int Ed Engl
45, 1267-1269 (2006).
15. A. Casini, G. Mastrobuoni, C. Temperini, C. Gabbiani, S. Francese, G. Moneti,
C. T. Supuran, A. Scozzafan, and L. Messori. Chem Commun, 156-158 (2007).
16. A. K. Boal and A. C. Rosenzweig. J Am Chem Soc 131, 14196-14197 (2009).
17. E. Sabini, H. Schubert, G. Murshudov, K. S. Wilson, M. Siika-Aho, and M. Penttilä. Acta
Cryst D 56, 3-13 (2000).
18. J.-N. Shin, Y.-W. Seo, M. Kim, S. Y. Park, M.-J. Lee, B.-R. Lee, J.-W. Oh, D.-W. Seol, and
T.-H. Kim. J Biol Chem 280, 10509-10515 (2005).
19. S. L. Pinnow, H. M. Brothers II, and N. M. Kostić. Croat Chim Acta 64, 519-528 (1991).
20. S. van Zutphen, M. Kraus, C. Driessen, G. A. van der Marel, H. S. Overkleeft, and J. Reedijk,
J Inorg Biochem 99, 1384-1389 (2005).
21. Y.-C. Lo, T.-P. Ko, W.-C. Su, T.-L. Su, and A. H.-J. Wang. J Inorg Biochem 103,
1082-1092 (2009).
22. L. Tong. Chem Rev 102, 4609-4626 (2002).
23. C.-C. Lee, C.-J. Kuo, T.-P. Ko, M.-F. Hsu, Y.-C. Tsui, S.-C. Chang, S. Yang, S.-J. Chen,
H.-C. Chen, M.-C. Hsu, S.-R. Shih, P.-H. Liang, and A. H.-J. Wang. J Biol Chem 284,
7646-7655 (2009).
281
TP-Pt(II) Complexes Inhibit
Cysteine Proteases
282
Lo et al.
24. B.-Y. Chiang, T.-C. Chen, C.-H. Pai, C.-C. Chou, H.-H. Chen, T.-P. Ko, W.-H. Hsu,
C.-Y. Chang, W.-F. Wu, A. H.-J. Wang, and C. H. Lin. J Biol Chem 285, 25345-25353
(2010).
25. A. C. Awade, P. Cleuziat, T. Gonzales, and J. Robert-Baudouy. Proteins: Struct Funct
Genet 20, 34-51 (1994).
26. P. M. Cummins and B. O’Connor. Biochim Biophys Acta 1429, 1-17 (1998).
27. H.-M. Chu and A. H.-J. Wang. Proteins: Struct Funct Genet 66, 996-1003 (2007).
28. Y. Odagaki, A. Hayashi, K. Okada, K. Hirotsu, T. Kabashima, K. Ito, T. Yoshimoto,
D. Tsuru, M. Sato, and J. Clardy. Structure 7, 399-411 (1999).
29. M. Schaeffer, A. Miranda, J. C Mottram, and G. H. Coombs. Molecul Biochem Parasitol
150, 318-329 (2006).
30. Z. Otwinowski and W. Minor. Methods Enzymol 276, 307-326 (1997).
31. A. Vagin and A. Teplyakov. Acta Cryst D 56, 1622-1624 (2000).
32. Collaborative Computational Project, Number 4 Collaborative Computational Project. Acta
Cryst D 50, 760-763 (1994).
33. A. T. Brunger, P. D. Adams, G. M. Clore, W. L. DeLano, P. Gros, R. W. Grosse-Kunstleve,
J.-S. Jiang, J. Kuszewski, M. Nilges, N. S. Pannu, R. J. Read, L. M. Rice, T. Simonson, and
G. L. Warren. Acta Cryst D 54, 905-921 (1998).
34. T. A. Jones, J. Y. Zou, S. W. Cowan, and M. Kjeldgaard. Acta Cryst A 47, 110-119 (1991).
35. P. Emsley, B. Lohkamp, W. G. Scott, and K. Cowtan. Acta Cryst D 66, 486-501 (2010).
36. W. L. DeLano. The PyMOL molecular graphics system. DeLano Scientific, San Carlos,
Calif (2002).
37. K. Ito, T. Inoue, T. Takahashi, H. S. Huang, T. Esumi, S. Hatakeyama, N. Tanaka,
K. T. Nakamura, and T. Yoshimoto. J Biol Chem 276, 18557-18562 (2001).
38. A. H.-J. Wang, J. Nathans, G. van der Marel, J. H. van Boom, and A. Rich. Nature 276,
471-474 (1978).
39. S. A. Ross, C. A. Carr, J.-W. Brïet, and G. Lowe. Anti-Cancer Drug Design 15, 431-439
(2000).
40. C.-H. Pai, H.-J. Wu, C.-H. Lin, and A. H.-J. Wang. Proteins Sci 20, 557-566 (2011).
41. S. Krieger, W. Schwarz, M. R. Ariyanayagam, A. H. Fairlamb, R. L. Krauth-Siegel, and
C. Clayton. Mol Microbiol 35, 542-552 (2000).
42. S. Bonse, J. M. Richards, S. A. Ross, G. Lowe, and R. L. Krauth-Siegel. J Med Chem 43,
4812-4821 (2000).
43. I. D. Kerr, P. Wu, R. Marion-Tsukmaki, Z. B. Mackey, and L. S. Brinen. PLoS Negl Trop
Dis 4, e701 (2010).
44. H. M. Brothers II, and N. M. Kostić. Biochemistry 29, 7468-7474 (1990).
45. C.-C. Lee, C.-J. Kuo, M.-F. Hsu, P.-H. Liang, J. M. Fang, J. J. Shie, and A. H.-J. Wang.
FEBS Letters 581, 5454-5458 (2007).
46. A. Wild, A. Winter, F. Schlütter, and U. S. Schubert. Chem Soc Rev 40, 1459-1511 (2011).
47. E. Meggers. Chem Comm, 1001-1010 (2009).
48. G. Chen, A. Li, M. Zhao, Y. Gao, T. Zhou, Y. Xu, Z. Du, X. Zhang, and X. Yu. J. Proteome
Res 7, 2733-2742 (2008).
49. F. Lecaille, J. Kaleta, and D. Bromme. Chem Rev 102, 4459-4488 (2002).
Date Received: May 28, 2011
Communicated by the Editor Ramaswamy H. Sarma