Retraction
Following the publication of this article [1], concerns were raised regarding western blot results in Fig 5 and the use of a potentially contaminated cell line. Specifically,
- The p38α panel in Fig 5C and phosphorylated ATF-2 panel in Fig 5D appear to have vertical discontinuities between lanes 1/2 and 2/3 suggested of splicing.
- The p38α panel in Fig 5C of this article [1] appears similar to the p38α panel in Fig 4C of [2]. Additionally, lanes 1 and 3 of this panel appear similar to lanes 1 and 2 of the P38β panel of Fig 4B in [3].
- Lanes 1 and 2 of the phosphorylated ATF-2 panel in Fig 5D of this article [1] appear similar to lanes 1 and 2 of the phosphorylated ATF-2 panel in Fig 4D of [2] and lanes 1 and 3 of the phosphorylated ATF-2 panel in Fig 4B of [3].
- The KB cell line described as an oral cancer cell lines has been shown to be potentially contaminated or a likely derivative of HeLa cervical cancer cell line [4].
The corresponding author stated that all primary data underlying this article remain available. They provided files described as the original raw images underlying the p38α and phosphorylated ATF-2 panels in Fig 5. These images also appeared to have discontinuities between the lanes, and the author’s explanation did not resolve the concerns.
The corresponding author agreed that the p38α and phosphorylated ATF-2 panels in Fig 5 of this article [1] appear similar to the indicated panels in the other articles [2, 3], and they claimed that the similarity was due to the comparable experimental conditions. The editors remain concerned about the level of similarity between images representing different experimental conditions.
The corresponding author acknowledged the concerns regarding potential contamination of the KB cell line; however, they stated that STR profile analysis is not available to confirm the identity of the cells used in this study.
In light of the concerns that question the reliability and integrity of these data, the PLOS ONE Editors retract this article.
SD did not agree with the retraction. All other authors either did not respond directly or could not be reached.
Figs 5C and D appear to report previously published material from [2, 3], published in 2012 by Elsevier and John Wiley & Sons, which are not offered under a CC BY license and are therefore excluded from this article’s [1] license. At the time of retraction, the article [1] was republished to note this exclusion in the Fig 5 legend and the article’s copyright statement.
8 Dec 2023: The PLOS ONE Editors (2023) Retraction: The Rational Design of Specific Peptide Inhibitor against p38α MAPK at Allosteric-Site: A Therapeutic Modality for HNSCC. PLOS ONE 18(12): e0295637. https://doi.org/10.1371/journal.pone.0295637 View retraction
Figures
Abstract
p38α is a significant target for drug designing against cancer. The overproduction of p38α MAPK promotes tumorigenesis in head and neck squamous cell carcinoma (HNSCC). The ATP binding and an allosteric site referred as DFG are the key sites of the p38α mitogen activated protein kinase (MAPK) exploited for the design of inhibitors. This study demonstrated design of peptide inhibitor on the basis of allosteric site using Glide molecular docking software and the biochemical analysis of the best modeled peptide. The best fitted tetrapeptide (FWCS) in the allosteric site inhibited the pure recombinant and serum p38α of HNSCC patients by 74 and 72%, respectively. The potency of the peptide was demonstrated by its IC50 (4.6 nM) and KD (3.41×10−10 M) values, determined by ELISA and by surface plasmon resonance (SPR) technology, respectively. The cell viability of oral cancer i.e. KB cell line was reduced in dose dependent manner by 60 and 97% by the treatment of peptide and the IC50 was 600 and 210 µM after 24 and 72 h incubation, respectively. Our result provides an insight for the development of a proficient small peptide as a promising anticancer agent targeting DFG site of p38α kinase.
Citation: Gill K, Nigam L, Singh R, Kumar S, Subbarao N, Chauhan SS, et al. (2014) The Rational Design of Specific Peptide Inhibitor against p38α MAPK at Allosteric-Site: A Therapeutic Modality for HNSCC. PLoS ONE 9(7): e101525. https://doi.org/10.1371/journal.pone.0101525
Editor: Ahmad Waseem, Institute of Dentistry, Barts & The London School of Medicine and Dentistry, Queen Mary University of London, United Kingdom
Received: January 14, 2014; Accepted: June 7, 2014; Published: July 1, 2014
Copyright: © 2014 Gill et al. Except for the p38α and phosphorylated ATF-2 panels in Figure 5, this is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Author acknowledge Research Section of All India Institute of Medical Sciences (AIIMS), New Delhi, India, for providing funds for the consumable items; Indian Council Medical Research, Government of India, for the fellowship of Kamaldeep Gill and are thankful to Prof. Jiahuai Han, The Scripps Research Institute, United States of America, for providing p38α cDNA. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Cancer drug discovery is a great challenge in recent years. Scientists have learnt a great deal about how faulty genes and proteins contribute to cancer development. This has opened up a new approach for screening the anticancer molecules to enhance the affinity, selectivity (to reduce the potential side effects), efficacy/potency, metabolic stability and oral bioavailability. This work focused on the development of anti oral-cancer inhibitor targeting p38α mitogen activated protein kinase (MAPK). p38α has emerged as an attractive target for chemotherapeutic intervention for the treatment of cancer. p38α MAPK is a broadly expressed signaling molecule that participates in the regulation of cellular responses to stress as well as in the control of proliferation, apoptosis and differentiation in a manner that is dependent on the cellular contents. It is known to be vital in regulating the expression of inflammatory cytokoines such as TNFα, IL6 and IL12 in response to proinflammatory signals [1]. Cytokines developed by activating immune cells during chronic inflammation are the major promoters for cancer growth and progression [2], [3]. The over production of theses cytokines causes tumor growth or cancer as well as has a critical role in the development and progression of cancer [4].
p38α is evident to be over-expressed in many cancers like oral [5] breast [6], gastric [7] and non small lung cancer [8]. The role of p38 MAPK in inflammation and cancer makes it as an attractive drug target. Generally, kinases share a similar conserved secondary structure, ATP binding site and catalyze analogous reaction of protein phosphorylation but also possess unique structural properties viz. protein-protein interaction sites and allosteric site [9]–[14]. Now a days the two important sites of kinase enzyme that are being focussed for inhibitor designs are the ATP binding site and the adjacent DFG-site. The majority of p38α MAPK inhibitors developed to date are competitive inhibitors targeting the ATP binding site. Our previous study also reported a specific competitive peptide inhibitor, VWCS for p38α MAPK designed on the basis of ATP binding site [15]. However, the crystal structure of p38α has revealed, an adjacent secondary site called DFG- site (Asp-Phe-Gly), also addressed as an allosteric binding site. The binding of inhibitor to the allosteric site involves strong conformational changes, as during the activity of the enzyme aromatic ring of phenylalanine of DFG-site plays a major role. The inhibitors like Gleevac, Nexavar and BIRB-796 are reported for the DFG-site for interaction [16].
Head and Neck Squamous cell Carcinoma (HNSCC) is associated with high recurrence, metastatic rate as well as poor prognosis. It has already been reported that p38α is overexpressed in HNSCC and declined after therapy [17]. Moreover, p38α kinase is an important parameter in promoting the tumor micro-environment in HNSCC [5]. This study attempted to establish a novel peptide inhibitor based on DFG-site of p38α as an anti-cancer agent.
Methods
Ethics
The Ethics Committee of All India Institute of Medical Sciences (AIIMS) approved the study protocol (A-39/4.08.2008) and informed consent was obtained. The study was performed compliant to the rules and regulations of the Ethics Committee, all subjects gave written informed consent.
Expression and purification of p38α
The pET14b expression vector containing human p38α-cDNA was transformed in bacterial E. coli BL21 (DE3) competent cells (Novagen, USA) and was grown in Luria-Bertani broth at 37°C containing 100 µg/ml ampicilin. The cells were grown for 16 h. The expression was induced with 1 mM IPTG and the cell pellet was resuspended in 10 ml of lysis buffer (25 mM Tris (pH7.9), 300 mM NaCl, 0.8 mM PMSF and 10 mM imidazole). The cells lysate was centrifuged at 10,000 g for 20 mins and was loaded onto Ni2+-NTA-Agarose column (1 ml, QIAGEN, Netherlands) pre-equilibrated with the lysis buffer. Eventually, the protein was eluted with the same buffer containing 100 mM imidazole.
The elutant containing p38α was further purified by anion-exchange chromatography on DEAE-Sephadex-A50 column pre-equilibrated with buffer (25 mM Tris (pH7.9), 50 mM NaCl, 10 mM MgCl2, 5% glycerol and 1 mM DTT) at 4°C and was eluted with a concentration gradient of 0.05–0.2 M NaCl.
Characterization of recombinant p38α
The eluted p38α (9.0 mg/ml) protein was assessed for its molecular-weight and purity by SDS-PAGE using Laemmli-buffers system [18]. Protein spots were excised from the gel and subjected to in-gel reduction, alkylation and tryptic digestion [19]. The mass and sequence of the protein was determined by MS/MS (Bruker Daltonics, USA). Each of the resultant peptide was used to BLAST search the protein and was identified by Mascot search programme (matrixscience) [20].
In-vitro p38α kinase assay
The activity of p38α was measured by an ELISA based Forrer et al. method [21] using activated transcription factor-2 (ATF-2) as a substrate. The experiment was performed in triplicates. Microtiter-plates were coated with 10 µl ATF-2 solution (Sigma-Aldrich, USA) (10 µg/ml) at 4°C. The plates were washed and purified p38α protein (100 µM) in kinase-buffer (50 mM Tris (pH7.5), 10 mM MgCl2, 10 mM β-glycerophosphate, 100 µg/ml BSA, 1 mM DTT, 0.1 mM Na3VO4, 100 µM ATP) was incubated in the wells for 1 h at 37°C. The kinase mixture without the p38α-protein was used as a blank. The mixture was incubated for 1 h at 37°C with anti-phosphoATF-2 primary antibody (1∶400) (Biovision, USA) that interacted with the phosphorylated form of ATF-2. The subsequent incubation was done with alkaline phosphatase conjugated goat anti-rabbit secondary antibody (1∶4000) (Chemicon, USA). The chromogenic substrate 4-nitrophenyl phosphate (4-NPP) (Cayman Chemical Company, USA) (5 mg/ml) in 0.1 M Tris-HCl (pH 8.0) containing 0.01% MgCl2 was added for 1.5 h at 37°C. Finally, the reaction was stopped with 3N NaOH and the formation of Nitrophenolate was measured at 405 nm using ELISA reader (Quanta Biotech, UK).
Design of peptide inhibitor using molecular docking
A library of 1.6 lac tetra-peptides [15] was built and subjected to Ligprep preparation wizard of GLIDE tool [22] generating a maximum of 32 tautomers per tetra-peptide in the pH range of 7.0±2.0. Similarly, BIRB-796 an inhibitor for DFG-out confirmation was also prepared.
The four isoforms of p38 kinase namely α, β, γ and δ exhibit high sequence identity specifically in the active site residues. The crystal structure sequences in FASTA format for all isoforms were downloaded from NCBI database (Accession ID's Q16539, Q15759, P53778, and O15264) to determine the similarities and then multiple sequence alignment (MSA) was performed using Clustal Omega (1.1.0) [23]. MSA resonates more than 95% similarities in isoform sequences, indicating that the DFG site is highly identical having conserved amino acids. Further, the sequence similarity was confirmed by performing structural alignment using MUSTANG v3.2.1 [24] of the four PDB structures for p38 isoforms viz. 1KV2, 3GDP, 1CM8, 3COI.
Crystal structures of p38α (PDB id 1KV2) co-crystallized with known inhibitor BIRB 796, human mitogen activated protein kinase 11 (p38β) in complex with nilotinib (PDB id 3GDP), p38γ (PDB id 1CM8), and p38δ kinase (PDB id 3COI) were downloaded from PDB database (www.rcsb.org).The heteroatoms and water molecules co-crystallized with the proteins were extracted out. Then prime module of GLIDE was used to build missing residues in the crystal structures of p38 isoforms. The protein structures were prepared using protein preparation wizard of GLIDE v 9.1 docking tool by adding hydrogen atoms and assigning bond orders. The structures were minimized using impref minimization wizard keeping OPLS_2005 molecular mechanics force field as default settings.
BIRB-796 co-crystallized with p38α was used for determination of active site residues of crystal structure 1KV2 including amino acids Val(50), Lys(53), Ser(61), Arg(67), Glu(71), Leu(74), Val(83), Thr(106), Met(109), Ile(141), Ile(146), Arg(149), Ile(166), Asp(168), Phe(169) and Gly(170) [25]. A gird of 17 Å was prepared using GLIDE grid preparation wizard and in silico docking was performed by keeping the grid as rigid. All tautomeric confirmations of tetra-peptides treated as flexible were screened against p38α. The peptide library was initially screened using High Throughput Virtual Screening (HTVS) and Standard Precision (SP) docking modes. The resultant top 16000 tetra-peptides were subjected to Extra Precision (XP) mode and finally five docked tetra-peptides conformers with minimum energy were selected. They were also docked against p38β, p38γ and p38δ using the same parameters and residues. Further to investigate the effects and specificity of the screened tetra-peptides against DFG site, we performed docking experiments on the ATP binding site. Since all the four isoforms share the same active site with some variations, individual grids for the crystal structures (PDB id 1KV2, 3GDP, 1CM8, 3COI) of the four isoforms were prepared using active site residues of p38α [15]. Grids for p38γ (Met109, Pro110, Phe111) and δ (Met107, Pro108, Phe109) were also included [26] as they are the key residues rendering variations amongst the isoforms. Post-docking analysis was performed to cross validate the docking performance by using X-score v1.2.1 [27] which calculates negative logarithm of ligand's dissociation constant with respect to the target protein. The docked poses were also analyzed using Pymol [28] and Chimera [29] visualization tools.
DFG-site is present in many other proteins and in order to further clarify the specificity of our results we also performed in silico docking for c-ABL and b-RAF proteins. For this purpose we used crystal structures of c-ABL and b-RAF with PDB id 3OXZ (resolution 2.20 Å) [30] and 4KSP (resolution 2.90 Å) [31] and prepared the proteins as discussed above. Keeping DFG motif as our target the active site for both the proteins were defined with residues Glu (500), Thr (507), Trp (530), Cys (531), Gly (533), Asp (593), Phe (594) for b-RAF and residues Tyr (253), Glu (286), Thr (315), Phe (317), Met (318), Ile (360), His (361), Asp (381), Phe (382) for c-ABL protein. For both the proteins Extra presicion mode of docking was performed to dock the screened tetra-peptides and BIRB-796.
Synthesis of peptides
The peptides were synthesized by solid phase peptide synthesis in PS3 synthesizer (Protein technology, USA) using Fmoc and Wang resin chemistry [32]. Dimethylformamide (DMF) was used as a solvent. The uronium salt 2-(1H-Benzotriazole-1-yl)-1, 1, 3, 3-tetramethyluronium-hexafluorophosphate (HBTU) and N-methylmorpholine (NMM) were used as activators of Fmoc amino-acids (Chem Impex, USA).
The Wang resin used was attached to the C-terminal amino-acid and N-terminal amino-acid was protected by Fmoc for each desired peptide amino-acid. In the first step, the Fmoc protected amino-acid attached to Wang resin (500 mg: 0.5 mmol) was deprotected by 20% piperidine in DMF. The next amino-acid protected by Fmoc at N-terminal was activated by HBTU (455 mg; 1 mmol) in the presence of base NMM (0.4 M) to form the acetic ester (551 mg; 0.5 mmol) and was coupled to the amino-acid attached to resin forming a dipeptide. Similarly, the remaining amino-acids were coupled forming the complete sequence. Finally, the Fmoc was removed by using 20% pipyridine solution in DMF.
Peptide-resin was washed with DMF, dichloromethane (DCM) and ethanol/methanol and dried in a dessicator. For 500 mg resin; 30 ml trifluoroacetic acid (TFA), 1.5 ml anisole, 1.5 ml thioanisole and 1.5 ml water were added and stirred for 2 h (in peptides containing Met, Arg, Cys, or Trp, thioanosole was added). TFA was removed using a rotary evaporator and anhydrous ether was added to get precipitate of peptide. The precipitate was then washed with ether till the scavengers were completely removed. Precipitate was dissolved in 10% acetic acid solution and lyophilized to obtain as powder.
Analytical RP-HPLC of peptides
The purity of peptides was established by analytical RP-HPLC using C18 reversed phase column (RPC) (1.6×10 cm, Amersham Bioscience, USA). 1 mg/ml of peptide was loaded onto RPC. The linear gradients were formed by passing two different solvents, where solvent-A was 0.05% aqueous TFA (pH2) and solvent-B was 0.05% TFA in acetonitrile at a flow rate of 0.25 ml/min at room temperature.
Binding studies of peptides with p38α using Surface Plasmon Resonance (SPR)
The recombinant-p38α containing His-tag was immobilized on the NTA sensor chip via Ni2+/NTA chelation at 25°C using the SPR technology in BIAcore-2000 apparatus (Pharmacia-Biosensor, Sweden) that monitors the real-time bimolecular interaction. The surface of the chip was activated with 2 mM NiCl2 at a flow rate of 5 µl/min where, Ni+2 forms a chelating complex with NTA that further binds with His-tagged recombinant-p38α. One flow-cell having Ni+2 was used as reference cell and on the other, the protein was immobilized by injecting 20 µl of p38α (9 mg/ml) at a flow rate of 5 µl/min.
The four different concentrations (1.31, 2.62, 6.57 and 13.13 µM) of each peptide were passed over the immobilized-p38α and the response units (RUs) were obtained to determine the binding constants. Eventually, the sensorgram obtained was subjected to BIAevaluation-3.0 software to calculate the dissociation-constant (KD).
Kinetic analysis of peptide inhibitors by ELISA
The assay was performed in ATF-2 coated 96-well plate at 37°C by following the in vitro kinase assay. Ten different concentrations (0–100 µM) of peptide were pre-incubated thrice with 100 µM of p38α; first for 30 min, 2 h and 3 h. 100 µM of ATP was added to each reaction mixture. Each concentration was used in triplicates. The extent of phosphorylation of ATF-2 was measured at 405 nm and the mean±SD values were used to calculate the % inhibition of peptides from which IC50 value of FWCS was determined.
HNSCC serum Inhibitory assay of peptide against p38α
The peptides were also tested against the serum p38 µ of HNSCC patients by the aforementioned in vitro kinase assay using coated ATF-2 and ATP as mentioned above. The protein content of the serum was estimated using BCA-protein assay Kit (Thermo Scientific, USA). 50 µg of serum protein was incubated with 2, 4, 6, 8, 10, 12, 14, 16, 18 and 20 nM peptide concentrations separately for 3 h. The %inhibition of serum p38α by peptides was calculated.
Measurement of Minimal hemolytic concentration (MHC)
Human blood was collected under sterile conditions from a healthy volunteer. The human red blood cells were separated by centrifugation at 3000 rpm for 10 mins, washed and resuspended to a concentration of 4% (v/v) with sterile Phosphate Buffer Saline (PBS). 100 µl each of 920 µM FWCS and gentamycin (positive control) was added to the polypropylene 96-well plate in triplicates and was twofold serially diluted in PBS (920–1.79 µM). To each well 100 µl of RBC suspension was added and incubated at 37°C for 1 h. The PBS and detergent Triton-X100 were used for 0% and 100% haemolysis, respectively. All the concentrations were analyzed in triplicates. The plate was centrifuged at 1000 g for 10 mins and the quantity of hemoglobin released was measured in the form of absorbance at 541 nm.
Protease assay
The peptide FWCS was tested for its protease activity by using casein as a substrate. The casein solution was prepared by dissolving 1 g casein in 5 ml of 0.2 M NaOH and 35 ml of distilled-water with pH adjusted to 7.5. It was heated for 15 mins, cooled and diluted with 50 ml of 10 mM Tris-HCl (pH8.0) buffer containing 40 mM CaCl2. Eventually, the solution was centrifuged at 13,000 rpm for 15 mins. 350 µl of casein solution was incubated separately with FWCS and trypsin (positive control). 1 ml of 4% (w/v) trichloroacetic-acid (TCA) was added, incubated at room temperature for 30 mins and centrifuged at 15,000 rpm for 15 mins. The proteolytic action was determined spectrophotometrically at 280 nm against water as a blank.
Cytotoxic effect of peptide on oral-cancer cell-line
The cytotoxic analysis of inhibitory peptide was done on KB oral-cancer cell-line as well as with PBMC; normal human cells using MTT dye reduction assay. The cell-line, was obtained from National Centre for Cell Sciences (Pune, India) and was maintained in minimum essential medium (MEM) in CO2 incubator to synchronize the culture. The cells (4×103 cells/100 µl media) were seeded in 96-well plates and were grown for 24 h. The experiment was performed in triplicates. The cells were incubated with different concentrations of the FWCS (250, 500, 750, 1000, 1250 and 1500 µM) for 24, 48 and 72 h. The plates were then incubated with 100 µl/well MTT-solution (1 mg/ml in MEM for KB cells and RPMI-1640 for PBMCs) for 4 h at 37°C. The formazan crystals formed were dissolved by the addition of DMSO (100 µl/well) and the absorbance was measured at 570 nm. The blank solution used to correct the background was 100 µl media with 100 µl MTT stock-solution and 100 µl DMSO. The data obtained was presented as percentage viability in the best-fit (linear) dose-response curves. The IC50 was calculated from mean±SD values.
Western-blot analysis of the peptide treated KB cells
KB cells were grown, treated with 600 µM FWCS peptide for 24 h and were lysed in Passive Lysis Buffer. The pure protein and the HepG2 cell-line expressing ATF-2 were taken as control. The proteins in the lysate were fractionated on 10% SDS-PAGE gel, transferred to PVDF, washed, and blocked in TBS with 3% BSA. The membranes were washed and incubated overnight with primary antibodies (1∶500) viz. anti-p38α and anti-Phospho-ATF2, respectively. The species specific Horse Radish Peroxidase (HRP)-conjugated secondary antibodies were used for 1 h at 37°C. The bands were detected by the Enhanced-Chemiluminescent system. β-actin was used for loading control. The PVDF membranes were stripped and assessed again with the primary (1∶500) β-actin antibody. Eventually, they were incubated with HRP-conjugated secondary antibody and developed using ECL-kit.
Results
Characterization of purified protein
The final p38α protein eluted at 0.1 M NaCl concentration from the DEAE-sephadex column was found to be >95% pure (single band of ∼40 kDa) on SDS-PAGE gel. The bands were excised from the gel. The trypsinized fragments obtained were identified and analyzed by MS-MS as well as with mascot search program that confirmed the protein to be p38α with a score of 117 based on probability analysis.
Screening of peptides
In-silico docking studies were used to predict probable candidates exhibiting inhibitory function against target protein. MSA (Figure 1A) and structural sequences (Figure 1B) alignments with emphasis on the DFG-site were performed indicating high degree of sequence similarities. Five tetra-peptides viz. FWCS, WFCS, YWFS, VFCS and FWVS against p38α were filtered from a library of 1.6 lac tetra-peptides on the basis of higher glide score and number of interacting hydrogen bonds with respect to BIRB-796. These tetra-peptides were further docked against other p38 isoforms viz. β, γ and δ as well as with c-ABL and b-RAF proteins under same parameters (Table 1). The FWCS tetra-peptide demonstrated highest affinity for p38α (Glide-score) and formed more hydrogen bonds than other isoforms in comparison to BIRB-796. Eventually, FWCS exhibited highest affinity for p38α among all the isoforms. It forms six hydrogen bonds with Asp (168) and Glu (71) at DFG-site (Figure 2). There was no hydrogen bond interaction with the hinge region. FWCS was showing hydrophobic interaction with Leu(75), Asp(168), Met(78), Arg(149) at allosteric site of p38α (Table S1). The dissociation constant of FWCS calculated by X-score v1.2.1 was 1.41×10−10 M, which was the higher strength of binding capability with p38α among the other peptides (Table S2).
(A) Multiple sequence alignment using Clustal Omega (1.1.0) indicating the conserved DFG domain in isoforms. (B) Structural alignment of p38 isoforms viz. 1KV2(α), 3GDP(β), 1CM8(γ), 3COI(δ). The active site residues are blue. Residue 169 codes for Phe in 1KV2 (red), 3GDP (yellow) and 3COI (cyan) whereas in 1CM8 (p38γ) (brown) residue 172 codes for Phe.
(A) Surface structure of p38α with inhibitory tetra-peptide where DFG active site is in yellow and FWCS is in magenta color. (B) Interactions of FWCS (purple) with the DFG site (green) of p38α (cyan) in DFG-out confirmation of Phe-169 (red).
The Glide score of FWCS for ATP binding site is comparatively showed lower affinity than DFG site of p38α isoform. Further the binding affinity of the top screened tetra-peptides was also found to be less in all other isoforms suggesting their lower specificity with other p38 isoforms (Table S3). The post-docking analysis of FWCS with all the four isoforms showed less number of hydrogen bonds formation and hydrophobic interactions suggesting their lower specificity for the ATP-binding site within all p38 isoforms (Table S4) On the basis of our computational studies we suggest that FWCS tetra-peptide can be a potent inhibitor of p38α at DFG site.
All the five peptides were synthesised, purified by HPLC (Figure S1) and analysed for binding efficiency using biochemical assay.
Determination of binding constant (KD) by SPR
A high quality biosensor data depends on the homogeneity of the target protein that is immobilized on the sensor surface. The standard immobilization procedure resulted in greater than 95% purity of the protein and was shown by the analysis of the p38α biosensor surface. The SPR signal for p38α immobilization was 6357.6 RUs. The sensitive and specific peptide interactions was determined in the form of binding capacity on to immobilized protein as the RU changed with different concentrations of peptides denoting the change in bound mass on the sensor chip with time giving the KD value of the peptides (Table 2). The KD value of FWCS (3.41×10−10M) was comparable with commercial inhibitor, BIRB-796 (KD = 0.1×10−9M) [25]. The KD value of FWCS illustrated a higher strength of interaction with the DFG-site of p38α than that of other peptides and BIRB-796. The sensorgram in the Figure 3A shows the binding of varying concentrations of FWCS over the p38α.
(A) The sensorgram shows concentrations of FWCS passed over the immobilized p38α on the Ni-NTA chip for the calculation of binding constants. (B) The line diagram of IC50 obtained by in vitro kinase assay after incubating the peptide with p38α for 30 mins, 2 h and 3 h with ten different concentrations of FWCS respectively using ATF-2 as a substrate. Error bar shows mean±SD.
Evaluation of IC50 by kinase assay
The inhibitory activity of peptides was tested against both pure and serum p38α. The peptides were preincubated for 30 min and also for 3 h with p38α as it has been reported that the increase in the preincubation time of inhibitor decreases the apparent IC50 value [32]. The reaction mixture was then incubated with ATP. During incubation the inhibitor interacted with the protein at the allosteric site resulting in a conformation change in p38α thereby making the protein unfit to interact with ATP. Thus, the phosphorylation of the ATF-2 was prevented.
The peptide FWCS was found to exhibit a % inhibition of 74.0±0.2 and 72.0±0.1% against the pure p38α and the serum-p38α, respectively after 3 h preincubation which was highest among the designed peptides (Table 3). The IC50 value of FWCS against the pure p38α was 42.4 nM, 24.2 and 4.6 nM after 30 min, 2 h and 3 h incubation, respectively and was comparable with the BIRB-796 (control) (8 nm) [33]. This figure display that the IC50 values improved linearly as the incubation period increased (Figure 3B).
Hemolytic assay
The peptide FWCS was assessed on 4% RBC solution. The peptide exhibited 6.38% hemolysis at a high concentration (920 µM) (Figure 4). Thus, the peptide was demonstrated to be nontoxic to human cells.
The peptide assessed against 4% human RBC was found to be non-toxic at the IC50 values obtained after treatment on KB cells and was only 6.38% hemolytic at a high concentration of 920 µM. Error bar shows mean±SD.
Protease assay
The protease activity of FWCS was tested on casein. The peptide FWCS did not exhibit protease activity. This was demonstrated by the absorbance values of the resultant casein fragments as the absorbance of FWCS treated casein was very less as compared to those produced by trypsin (Data not shown).
Cytotoxicity assay
The cytotoxicity of FWCS was assessed using MTT assay on KB oral-cancer cell line and normal PBMCs. The cells were treated with 0–1500 µM of FWCS peptide for 24, 48, and 72 h. After the treatment of different peptide doses, the IC50 values of the peptide were observed to be 600, 350 and 210 µM at 24, 48 and 72 h, respectively. At these concentrations only 50% cells remained viable as seen in a dose and time response curve and no cell death was observed in case of normal PBMCs (Figure 5A, B).
(A) The dose and time dependent response of peptide on KB cells (B) PBMCs assessed by MTT assay. The western blot analysis of treated and untreated KB cells after 24 h (C) using p38α antibody and (D) using anti-phosphoATF-2 antibody demonstrating that inhibition of p38α with FWCS results in a declined expression of downstream ATF-2 protein. β-actin was taken as a loading control. Error bar shows mean±SD. The p38α and phosphorylated ATF-2 panels in Figure 5 are excluded from this article's CC BY license. See the accompanying retraction notice for more information.
Western-blot analysis of peptide treated KB cells
The western-blot analysis was performed for FWCS treated and untreated cell lysate. The lysate was resolved on SDS-PAGE gel and the bands were observed at the appropriate molecular weight. The western-blot was first performed using p38α primary antibody. In Figure 5C, anti-p38α antibody interacted with the pure-protein, untreated and treated cell lysate as shown in lane 1, 2 and 3, respectively. To further validate the selectivity of FWCS, analysis of cell lysates was also done by using anti-phospho-ATF-2 antibody that interacts only with phosphorylated-ATF-2. ATF-2 is a downstream signaling substrate for p38α kinase and the inhibition of p38α prevented the phosphorylation of ATF-2. Therefore, no bands were observed on the blot in treated cell lysate. In Figure 5D, the lane 1 and 2 showed the interaction of anti-phosphoATF-2 with ATF-2 protein in control cell lysate and in untreated cells, respectively whereas no protein band was detected in treated cell lysate (lane 3).
Discussion
The MAPK cascade presents a new approach for the development of novel cancer therapies intended to be less toxic and provides a valid alternative to conventional chemotherapeutic drugs. Inhibition of any of these MAPK target has the potential for translational pharmacodynamic assessment of target suppression. The discovery of small molecule kinase inhibitors has attracted interest for novel drug development and for better understanding of biological role of these proteins. The four isoforms of p38 MAPK are activated by dual phosphorylation on Thr and Tyr within a conserved Thr-X-Tyr tri-peptide motif in an activation loop proximal to the ATP and substrate binding sites [34]. MAPK has an N-terminal and a C-terminal domain joined by a short hinge region about which the two domains can rotate with respect to each other. The activation and the catalytic loop in kinases are responsible for the orientation of ATP and the positioning of the substrate. In almost all kinases, sequence motif “Asp-Phe-Gly” (DFG-motif) at the beginning of the activation loop is conserved and involved in the correct orientation of ATP in the binding site as well as enables the entry of the substrates [35]. The phosphorylation of protein in the activation loop opens the allosteric DFG-site where the position of the phenyl ring of Phe determines active or inactive conformation of kinase. Many kinase inhibitors developed to date target the ATP binding site and suffer from widespread cross-reactivity among the other target kinase [36]. Only a few inhibitors were designed complementary to the conserved DFG residues. The binding of inhibitors to this site create a large conformational change in highly conserved DFG-motif within the active site of the kinase. One of the most potent compounds belonging to this series of inhibitors is BIRB-796 [37].
Although BIRB-796 impairs the phosphorylation of p38 but it was found to be more potent and selective for p38α than p38β followed by p38γ and p38δ which could be due to its interaction with Thr(106), a residue unique to p38α and p38β in the hydrophobic pocket adjacent to the DFG-site [38]. This study focused on the allosteric DFG-site where five best peptides designed on the basis of the structural information of DFG-site were analyzed biochemically. Here, it was observed on the basis of molecular-docking studies that FWCS peptide best fitted the DFG-site of p38α followed by that of isoforms viz. p38β, p38γ and p38δ, as well as with other proteins having DFG site like c-ABL and b-RAF. It formed six hydrogen bonds with p38α DFG-site, where Phe of peptide forms 2 and 1 hydrogen bonds with Asp (168) and Glu (71), respectively, Trp formed bond with Glu (71), Cys bonded with Asp (168) and Ser bonded with Asp (168) but there was no hydrogen bond with the hinge region. Similar hydrogen bonds were also reported in the central urea of BIRB-796 where the carbonyl and NH of urea forms hydrogen-bonds with the NH of Asp (168) and side-chain of Glu (71), respectively [25]. The calculated dissociation constant of FWCS revealed to be a potent inhibitor of p38α.
The low binding affinity with ATP binding site with less number of hydrogen bonds and low hydrophobic interaction also suggested the specificity of FWCS with DFG binding site of p38α isoform.
The peptide also showed a strong biochemical interaction with p38α. The KD value of FWCS obtained for allosteric p38α-site was comparable and superior to the known inhibitor BIRB-796. FWCS indirectly competed with ATP in the inhibition assay by pre-incubating with pure p38α for three different time intervals viz. 30 mins, 2 h and 3 h. The resultant rapid gradual linear decline in IC50 from 42.4 nM to 4.6 nM obtained with the increased pre-incubation time confirmed the slow binding kinetics of the peptide. A similar decline in apparent IC50 was observed incase of BIRB-796 with the increase in pre-incubation period which confirmed the slow binding behavior [33], [34]. This was probably the conformational change in the DFG site due to the movement of Phe residues as mentioned earlier [38].
Moreover, the peptide was found to be nearly non-toxic to human erythrocytes as well as to human PBMCs. Generally blood serum proteases perform their activity in specific condition and digestive protease needs certain site and size for cleavage. The present peptide does not have the site and the size to get recognized by any proteases. Hence, peptide can be delivered intravenously. In HNSCC, the overproduction of various cytokines like IL-6 and IL-8 takes place which are responsible in oncogenic processes such as angiogenensis and metastasis. In addition, p38MAPK was shown to be activated by phorbol-12-myristate-13-acetate in HNSCC escalating the secretion of cytokines IL-6 and IL-8 [5]. An important and accepted therapeutic approach for potential drug intervention in this disease is the reduction of proinflammatory cytokines. The structure based drug designing is the rational and very efficient way of development of therapeutic molecule. It was reported that BIRB-796 in presence of compound Bortezomib inhibits p38α MAPK and blocked the phosphorylation leading to enhanced cytotoxicity in multiple myeloma cell-lines and paracrine tumors but alone did not induce significant growth inhibition [39]. The reduction in phophorylation of ATF-2, a substrate for p38α was observed during kinetic analysis after the incubation of FWCS with pure-p38α. The peptide FWCS efficiently reduced the cell viability of oral-cancer KB cell-line in MTT assay which was probably due to the inhibition of p38α but it did not effected on normal PBMCs. The time and dose dependent cytotoxicity of FWCS was evident by the IC50 values of peptide for 24, 48 and 72 h. The specificity of peptide on the KB cells was further assessed by performing western-blot analysis of the cell lysates. The band was not observed in the peptide treated cells with primary anti-phosphorylated ATF-2 antibody indicating that FWCS inhibited the activity of p38α MAPK and prevented the phosphorylation of ATF-2. In contrast, the phosphorylated ATF-2 band was obtained in untreated cells. So the peptide FWCS can be considered as an effective inhibitor of p38α. Hence, FWCS can function as an anticancer agent as it potentially inhibits cell viability in a dose and time dependent manner in oral-cancer cell line. Besides, it also reduced the proficiency of serum-p38α in HNSCC patients.
In conclusion, the peptide inhibitor may prove to be a potent and new generation of therapeutic agent against human oral-cancer targeting DFG-site of p38α. Modern drug discovery involves the identification by screening using molecular modeling, biochemical assay and optimizing it to increase the affinity, selectivity and potency. This peptide was screened by modeling for selectivity and performed for affinity as well as potency by biochemical assay. This work may be employed as a new lead generation of inhibitor for p38α MAPK in cancer.
Supporting Information
Figure S1.
Purification profile of FWCS peptide obtained by HPLC.
https://doi.org/10.1371/journal.pone.0101525.s001
(TIF)
Table S1.
Hydrophobic interactions of peptides and BIRB-796 with p38 isoforms.
https://doi.org/10.1371/journal.pone.0101525.s002
(DOC)
Table S2.
The predicted dissociation constant of peptides and BIRB-796 with p38 isoforms as per Xscore.
https://doi.org/10.1371/journal.pone.0101525.s003
(DOC)
Table S3.
Glide Docking Score for the screened tetrapeptides against ATP binding site for the hits that were screened against DFG-out site for p38α.
https://doi.org/10.1371/journal.pone.0101525.s004
(DOC)
Table S4.
Post-docking analysis of FWCS against ATP-binding site of all the four isoforms of p38. Hydrogen bonds were determined using Chimera and hydrophobic interactions were determined by GETNEARES program.
https://doi.org/10.1371/journal.pone.0101525.s005
(DOC)
Acknowledgments
Authors acknowledge Indian Council Medical Research, Government of India for the fellowship of Kamaldeep Gill and thankful to Prof. Jiahuai Han, the Scripps Research Institute, USA for providing p38α cDNA.
Author Contributions
Conceived and designed the experiments: SD. Performed the experiments: KG RS LN SK. Analyzed the data: SD KG SSC NS SD. Contributed reagents/materials/analysis tools: SD. Wrote the paper: SD KG.
References
- 1. Kumar S, Boehm J, Lee JC (2003) p38 MAP kinases: key signalling molecules as therapeutic targets for inflammatory diseases. Nat Rev Drug Discov 2(9): 717–726.
- 2. Karin M, Lawrence T, Nizet V (2006) Innate immunity gone awry: linking microbial infections to chronic inflammation and cancer. Cell 124(4): 823–835.
- 3. Naugler WE, Sakurai T, Kim S, Maeda S, Kim K, et al. (2007) Gender disparity in liver cancer due to sex differences in MyD88-dependent IL-6 production. Science 317(5834): 121–124.
- 4. Dhillon AS, Hagan S, Rath O, Kolch W (2007) MAP kinase signalling pathways in cancer. Oncogene 26(22): 3279–3290.
- 5. Riebe C, Pries R, Kemkers A, Wollenberg B (2007) Increased cytokine secretion in head and neck cancer upon p38 mitogen-activated protein kinase activation. Int J Mol Med 20(6): 883–887.
- 6. Davidson B, Konstantinovsky S, Kleinberg L, Nguyen MT, Bassarova A, et al. (2006) The mitogen-activated protein kinases (MAPK) p38 and JNK are markers of tumor progression in breast carcinoma. Gynecol Oncol 102(3): 453–461.
- 7. Liang B, Wang S, Zhu XG, Yu YX, Cui ZR, et al. (2005) Increased expression of mitogen-activated protein kinase and its upstream regulating signal in human gastric cancer. World J Gastroenterol 11(5): 623–628.
- 8. Greenberg AK, Basu S, Hu J, Yie TA, Tchou-Wong KM, et al. (2002) Selective p38 activation in human non-small cell lung cancer. Am J Respir Cell Mol Biol 26(5): 558–564.
- 9. Kulkarni RG, Achaiah G, Sastry GN (2006) Novel targets for antiinflammatory and antiarthritic agents. Curr Pharm Des 12(19): 2437–2454.
- 10. Noble ME, Endicott JA, Johnson LN (2004) Protein kinase inhibitors: insights into drug design from structure. Science 303(5665): 1800–1805.
- 11. Morphy R (2010) selectively nonselective kinase inhibition: striking the right balance. J Med Chem 53(4): 1413–1437.
- 12. Simard JR, Grutter C, Pawar V, Aust B, Wolf A, et al. (2009) High-throughput screening to identify inhibitors which stabilize inactive kinase conformations in p38alpha. J Am Chem Soc 131(51): 18478–18488.
- 13. Vulpetti A, Bosotti R (2004) Sequence and structural analysis of kinase ATP pocket residues. Farmaco 59(10): 759–765.
- 14. Huse M, Kuriyan J (2002) The conformational plasticity of protein kinases. Cell 109(3): 275–282.
- 15. Gill K, Singh AK, Kapoor V, Nigam L, Kumar R, et al. (2013) Development of peptide inhibitor as a therapeutic agent against head and neck squamous cell carcinoma (HNSCC) targeting p38alpha MAP kinase. Biochim Biophys Acta 1830(3): 2763–2769.
- 16. Dietrich J, Hulme C, Hurley LH (2010) The design, synthesis, and evaluation of 8 hybrid DFG-out allosteric kinase inhibitors: a structural analysis of the binding interactions of Gleevec, Nexavar, and BIRB-796. Bioorg Med Chem 18(15): 5738–5748.
- 17. Gill K, Mohanti BK, Ashraf MS, Singh AK, Dey S (2012) Quantification of p38alphaMAP kinase: a prognostic marker in HNSCC with respect to radiation therapy. Clin Chim Acta 413(1–2): 219–225.
- 18. Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227(5259): 680–685.
- 19. Boussac M, Garin J (2000) Calcium-dependent secretion in human neutrophils: a proteomic approach. Electrophoresis 21(3): 665–672.
- 20.
Kinter M, Sherman NE (2000) Protein Sequencing and Identification Using Tandem Mass Spectrometry. Wiley-Interscience: New York.
- 21. Forrer P, Tamaskovic R, Jaussi R (1998) Enzyme-linked immunosorbent assay for measurement of JNK, ERK, and p38 kinase activities. Biol Chem 379(8–9): 1101–1111.
- 22.
Glide, version 9.1 (2007) Schrödinger, L. L. C.; New York, NY.
- 23. Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, et al. (2011) Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol 7: 539.
- 24. Konagurthu AS, Whisstock JC, Stuckey PJ, Lesk AM (2006) MUSTANG: a multiple structural alignment algorithm. Proteins 64(3): 559–574.
- 25. Pargellis C, Tong L, Churchill L, Cirillo PF, Gilmore T, et al. (2002) Inhibition of p38 MAP kinase by utilizing a novel allosteric binding site. Nat Struct Biol 9(4): 268–272.
- 26. John CL, Kassis S, Kumar S, Badger A, Jerry LA (1999) p38 Mitogen-Activated Protein Kinase Inhibitors -Mechanisms and Therapeutic Potentials. Pharmacol Ther 82: 389–397.
- 27. Wang R, Lai L, Wang S (2002) Further development and validation of empirical scoring functions for structure-based binding affinity prediction. J Comput Aided Mol Des 16(1): 11–26.
- 28.
DeLano WL (2002) The PyMOL Molecular Graphics System DeLano Scientific, San Carlos, CA, USA. http://www.pymol.org.
- 29. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, et al. (2004) UCSF Chimera–a visualization system for exploratory research and analysis. J Comput Chem 25(13): 1605–1612.
- 30. Zhou T, Commodore L, Huang WS, Wang Y, Thomas M, et al. (2011) Structural mechanism of the Pan-BCR-ABL inhibitor ponatinib (AP24534): lessons for overcoming kinase inhibitor resistance. Chem Biol Drug Des 77(1): 1–11.
- 31. Okaniwa M, Hirose M, Arita T, Yabuki M, Nakamura A, et al. Discovery of a selective kinase inhibitor (TAK-632) targeting pan-RAF inhibition: design, synthesis, and biological evaluation of C-7-substituted 1,3-benzothiazole derivatives. J Med Chem 56(16): 6478–6494.
- 32.
Merrifield RB (1963) Solid Phase Peptide Synthesis. I. The Synthesis of a Tetrapeptide. J Am Chem Soc 85 (14): : 2149–2154.
- 33. Kroe RR, Regan J, Proto A, Peet GW, Roy T, et al. (2003) Thermal denaturation: a method to rank slow binding, high-affinity P38alpha MAP kinase inhibitors. J Med Chem 46(22): 4669–4675.
- 34.
Klebl B, Müller G, Hamacher M (2003) Methods and Principles in Medicinal Chemistry, Protein Kinases as Drug Targets, Wiley International.
- 35. Young PR, McLaughlin MM, Kumar S, Kassis S, Doyle ML, et al. (1997) Pyridinyl imidazole inhibitors of p38 mitogen-activated protein kinase bind in the ATP site. J Biol Chem 272(18): 12116–12121.
- 36. Schäfer M, Egner U (2006) Structural Aspects of Drugability and Selectivity of Protein Kinases in Inflammation. Anti-Inflamm Anti-Allergy Agents Med Chem 6: 5–17.
- 37. Zuccotto F, Ardini E, Casale E, Angiolini M (2010) Through the “gatekeeper door”: exploiting the active kinase conformation. J Med Chem 53(7): 2681–2694.
- 38. Kuma Y, Sabio G, Bain J, Shpiro N, Marquez R, et al. (2005) BIRB796 inhibits all p38 MAPK isoforms in vitro and in vivo. J Biol Chem 280(20): 19472–19479.
- 39. Yasui H, Hideshima T, Ikeda H, Jin J, Ocio EM, et al. (2007) BIRB 796 enhances cytotoxicity triggered by bortezomib, heat shock protein (Hsp) 90 inhibitor, and dexamethasone via inhibition of p38 mitogen-activated protein kinase/Hsp27 pathway in multiple myeloma cell lines and inhibits paracrine tumour growth. Br J Haematol 136(3): 414–423.