AFFINITY PEPTIDE LIGAND OF MOUSE POLYOMAVIRUS CAPSOMER AND DESIGNED SCREENING METHOD THEREOF

Abstract

Affinity peptide ligands of mouse polyomavirus capsomers and a designed screening method thereof. The affinity peptide ligands can be used for separation and purification of the capsomers.

Claims

1. Novel affinity peptide ligands of mouse polyomavirus capsomers, comprising DWDLRLLY, DWDLRLIY, DWNLRLIY, DWFLNLFY, DWSLKLVY, DWSLRLKY and DWNLHLPY.

2. A designed screening method of the novel affinity peptide ligands of the mouse polyomavirus capsomers of claim 1, wherein a novel peptide ligand library of the capsomers is constructed on the basis of the crystal structure of a capsomer and VP2-C complex existing in the natural world, with the feature sequence being DWXLXLXY, wherein the X denotes 19 amino acids except cysteine.

3. The designed screening method of claim 2, wherein the molecular interaction mechanism of the capsomer and VP2-C complex is examined using a molecular mechanics-Poisson-Boltzmann surface area (MM/PBSA) method, where hydrophobic interaction is identified as the major driving force while V283, P285, D286, W287, L289, L293, and Y296 are identified as key residues, making significant contribution to binding, in VP2-C.

4. The designed screening method of claim 2, wherein the peptide library is constructed on the basis of five key residues in VP2-C, i.e., D286, W287, L289, L293, and Y296; and within the range of the peptide library, a candidate peptide modular library is constructed using an amino acid locating method.

5. The designed screening method of claim 2, wherein the high-affinity peptide ligands of the mouse polyomavirus capsomers are screened by molecular docking screening, root mean square deviation (RMSD) comparison, and molecular dynamics (MD) simulation coupled with free energy calculation.

6. The designed screening method of claim 5, wherein the peptide ligands in the peptide ligand library are sequentially docked to the mouse polyomavirus capsomers by molecular docking software VINA, and a total of 1158 peptide ligands with binding free energy less than 6.5 kcal/mol are selected.

7. The designed screening method of claim 5, wherein the RMSD values between the 1158 peptide ligands obtained from docking by VINA and the corresponding key residues in VP2-C are calculated by a program g_rms provided by molecular simulation software GROMACS, and 334 peptide ligands are selected for researching.

8. The designed screening method of claim 5, wherein the 334 peptide ligands are then rescreened on the basis of the C-terminus orientation, and 227 peptide ligands having the similar orientation with VP2-C are selected and rescreened by a docking experiment ROSETTA FlexPepDock, and then ten optimal peptide ligands are selected for MD simulations.

9. The designed screening method of claim 5, wherein the ten peptide ligands obtained by screening are subjected to MD simulations with the complex of the mouse polyomavirus capsomers, and free energy calculation is conducted using the MM/PBSA method to further evaluate the affinity and specificity of the peptide ligands, thereby obtaining seven peptide ligands with relatively high affinity, which are DWDLRLLY, DWDLRLIY, DWNLRLIY, DWFLNLFY, DWSLKLVY, DWSLRLKY and DWNLHLPY.

10. Applications of the designed screening method of claim 2, wherein a modification strategy for the peptide library includes: adding one or more amino acid residues to the N-terminus; introducing one or more amino acid residues to the C-terminus; inserting one or more amino acid residues between adjacent residues of the peptides; replacing certain one or more resides in the peptides with other amino acid residues.

11. Applications of the designed screening method of claim 5, wherein a modification strategy for the peptide library includes: adding one or more amino acid residues to the N-terminus; introducing one or more amino acid residues to the C-terminus; inserting one or more amino acid residues between adjacent residues of the peptides; replacing certain one or more resides in the peptides with other amino acid residues.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0028] FIG. 1 is a chart of a design and screening strategy of novel affinity peptide ligands of MPV Cap.

DETAILED DESCRIPTION

[0029] Combined with the specific embodiments, the followings are the further description of the invention in detail.

Embodiment 1

Affinity Interaction Analysis Between VP2-C and Caps and Construction of a VP2-C Simplified Affinity Binding Model

[0030] The crystal structure of a Cap and VP2-C complex, adopted in the invention, is taken from the Protein Data Bank (PDB ID: 1CN3, http://www.rcsb.org/pdb/). Cap comprises five VP1 (residues from 34-316) while VP2-C contains 19 residues (residues from 279 to 297). In this work, MD simulations are performed using a GROMACS 4.5.3 package with an all-atom CHARMM27 force field. The Cap and VP2-C complex is first solvated in a cubic box (12.16312.16312.163 nm) using TIP3P as the water model. 225 Na.sup.+ and 190 Cl.sup. are added. After energy minimization, the system is equilibrated for 200 ps under NVT ensemble and further 200 ps under NPT ensemble. The temperature of the system is controlled at 298.15 K by the velocity-rescale (V-rescale) method. Periodic boundary conditions are adopted for all the simulations. The calculation of non-binding interactions adopts 1.2 nm cut-off. The long-range electrostatic interactions are treated with the particle mesh Ewald (PME) method. All hydrogen atoms are constrained with the Lincs algorithm with a 2 fs time step. The simulation is carried out for 20 ns.

[0031] The RMSD and the potential energy of the complex reach equilibrium at a time scale of 5 ns according to the simulation result. Therefore, 75 snapshots of conformations extracted from the 17-20 ns trajectory of the complex at an interval of 40 ps are collected for the calculation and analysis of free energy. The hydrophobic interaction contribution of the Cap and VP2-C complex is 79 kcal/mol while the electrostatic interaction contribution is 3 kcal/mol, which indicates that the major driving force for affinity between Cap and VP2-C is hydrophobic interaction, which is in line with experimental facts in terms of the X-ray crystal diffraction structure. For electrostatic interaction, most inter-molecular electrostatic interaction contribution (171 kcal/mol) favorable for molecular binding is compensated by the unfavorable electrostatic solvation interaction energy (177 kcal/mol). So the total electrostatic interaction energy is very small relatively.

[0032] The free energy decomposition of VP2-C in the Cap and VP2-C complex is utilized to identify the hot spot residues. In this invention, the hot spot residues are identified as the residues that have large contribution to the binding free energy and that are involved in the important intermolecular interaction formation to compensate the unfavorable solvation interaction. The residues contributing a lot to the free energy are identified on the basis of the criterion of 2.5 kcal/mol. For VP2-C, six hot spot residues (V283, P285, D286, W287, L289 and Y296) are found in calculation. It should be noted that the contribution of L293 to the free energy is not satisfied with the criterion of the hot spot residues. It is reported that the mutation L181E on simian virus 40 (SV40) minor coat protein VP3 causes reduction of VP3 and Cap binding by 33%, where L181 in SV40 VP3 is just the counterpart of L293 in MPV VP2 herein. Therefore, in consideration of the high conserved region in polyomaviruses, L293 is considered to be important in the binding of VP2-C and Cap and thus included in the process of construction of the affinity binding model. Finally, the VP2-C simplified affinity binding model is construction according to the affinity mechanism of VP2-C and Cap and the distribution of the hot spot residues of VP2-C, and comprises V283, P285, D286, W287, L289, L293 and Y296.

Embodiment 2

Construction of Polypeptide Library

[0033] The key residues, i.e., D286, W287, L289, L293 and Y296, located nearest to the base of Cap, were selected as the starting point of polypeptide construction, thereby maximally avoiding the steric-hindrance effect existing in the actual operation. W287, L289, L293 and Y296 are almost aligned in a straight line, which is favorable for the design of short peptide ligands without needing consideration of the spatial conformations of VP2. It is known that the lengths of a peptide bond and an amino acid backbone are about 1.33 and 2.78 , respectively. The insertion of one amino acid reside requires the lengths of two peptide bonds and the length of one amino acid backbone, i.e., about 21.33+2.78=5.44 . The VMD calculation obtains the results that W287L289=5.42 , L289L293=5.72 , and L293Y296=5.61 . Accordingly, one amino acid can be inserted between every two adjacent hot spot residues. Thus, an octapeptide library of DWXLXLXY, where X represents residues (except Cys), containing 6859 sequences, each having the aforementioned five hot spot residues, is rationally designed and constructed finally by CHARMM invoked by a perl script.

Embodiment 3

Docking of Peptides to Caps

1. VINA Docking

[0034] 6859 peptides are first docked into the binding region located on the inner cavity surface of Caps using VINA and ranked from 4 to 8 kcal/mol, which is in line with the requirement that affinity ligands should have medium affinity (the binding constant is in the range of 10.sup.4-10.sup.8 M.sup.1). Then, in order to avoid missing the promising ligands, 1158 peptide candidates with the binding free energy being less than 6.5 kcal/mol are selected according to the empirical value and distribution result.

2. RMSD Calculation

[0035] The program g_rms provided by the GROMACS 4.5.3 simulation package is used to calculate the RMSD values between the hot spot residues in the 1158 peptide sequences obtained by VINA docking and those corresponding in Cap. Smaller RMSD value indicates higher structure similarity of the docked conformation of the hot spot residues contained in the peptides to the conformation of those corresponding in VP2-C. The results show that the distribution of RMSD is in the range of 0.3-0.6 nm, and 334 peptide sequences with RMSD values less than 0.4 nm are selected for the next step of analysis.

3. C-Terminus Orientation Comparison

[0036] In the Cap and VP2-C complex, the VP2-C runs from the bottom to the top of the inner cavity of Cap and the C-terminus is thus located at the bottom of Cap. Then the peptides without similar orientation with VP2-C, e.g., with N-terminus toward the bottom of Cap, are not considered in the following screening. Then 227 peptides are selected for next screening.

4. Rescreening Peptides Using FlexPepDock

[0037] The selected peptides are then rescreened by using a ROSETTA FlexPepDock web server. FlexPepDock mainly consists of two modules that optimize the peptide backbone and rigid body orientation, respectively. The starting structure is refined in 200 independent FlexPepDock simulations. 100 of the simulations are carried out strictly in a high-resolution mode, while 100 of the simulations include a low-resolution pre-optimization step, followed by the high-resolution refinement. A total of 200 models are thus created and then ranked based on their generic full-atom energy scores. 10 optimal conformations for each peptide are created. The top ten peptides with high scoring function scores are finally screened out, i.e., DWDLRLLY, DWDLRLIY, DWGLRLKY, DWSLKLVY, DWFLNLFY, DWSLDLWY, DWGLKLIY, DWNLRLIY, DWSLRLKY and DWNLHLPY.

Embodiment 4

MD Simulations and MM/PBSA Free Energy Calculation

[0038] To further predict the affinity interaction between Caps and the peptide ligand complex more precisely by using computer-aided ligand design, MD simulations coupled with free energy calculation with more calculation but also more accuracy are utilized to further evaluate the magnitude of affinity of the complex. The MD simulation parameters are in accordance with the embodiment 1. The final and original conformations of the peptides are compared using VMD. It is shown that except DWGLKLIY, DWGLRLKY and DWSLDLWY, the other seven peptides (DWDLRLLY, DWDLRLIY, DWSLKLVY, DWFLNLFY, DWNLRLIY, DWSLRLKY, and DWNLHLPY) keep a stable binding conformation with Caps during the MD simulations and thereby are predicated to be effective affinity ligands of Caps.

[0039] The binding free energies of these peptide ligands on the inner surface of Caps are calculated by MM/PBSA and the results are listed according to the binding free energies from high to low: DWDLRLLY, DWDLRLIY, DWNLRLIY, DWFLNLFY, DWSLKLVY, DWSLRLKY, DWNLHLPY, DWSLDLWY, DWGLRLKY, DWGLKLIY. The last three peptides present three ligands with unstable binding conformations with Caps during the MD simulations. For binding of all the ten peptides with Caps, the free energy calculation results indicate that the hydrophobic interaction contribution is dominant, which is in agreement with the initial design thinking, demonstrating the accuracy of rational design. The optimal peptide is DWDLRLLY according to the free energy calculation, as it has a binding free energy of 61 kcal/mol with Caps, close to the binding free energy (76 kcal/mol) of VP2-C with Caps. Thus, DWDLRLLY is considered as the high-specificity affinity ligand of Caps.

Embodiment 5

Affinity Chromatography Experiment Validation

[0040] 1. Preparation of Affinity Medium DWDLRLLY-6B and E.coli Lysate

[0041] A Thiopropyl Sepharose 6B (GE Healthcare) medium is washed with transmembrane water and then pre-equilibrated in linking buffer (0.5 M NaCl, 1 mM EDTA, 0.1 M PBS, pH 6.5) for 12 h. After being drained, 1 g of the wet medium is transferred into an Erlenmeyer flask containing 2.57 mg of peptides and 10 mL of linking buffer. After fully mixed, the mixture reacts in a shaking bath at 25 C. and 170 rpm for 2 h. The affinity peptide medium with the ligand density being 2 mol/(g drained wet medium) (thereafter termed DWDLRLLY-6B) is obtained.

[0042] A plasmid pGEX-Ssp DnaB-Ser-VP1 is constructed and transformed into E.coli BL21(DE3). The recombinant E.coli BL21(DE3) is inoculated in 25 mL of TB medium [100 mg/L ampicillin, 12 g/L tryptone, 24 g/L yeast extract, 0.4%(v/v) glycerol, 2.31 g/L KH.sub.2PO.sub.4, 12.54 g/L K.sub.2HPO.sub.4] overnight at 37 C., 170 rpm. Then the culture is diluted 1:1000 in 250 mL TB medium and grown to OD.sub.6000.5-0.6 (37 C., 170 rpm), after which protein expression is induced for 24 h by the addition of 0.3 mM IPTG at 26 C., 170 rpm. The cell pellets from 250 mL of the fermentation broth are harvested, re-suspended and lysed by sonication on ice. The supernatant of the E.coli lysate is collected by centrifugation for further use.

2. Static Adsorption Experiment of Caps on Affinity Medium

[0043] The DWDLRLLY-6B medium is equilibrated with equilibrating buffer (50 mM PBS, 200 mM NaCl, pH 6.0) for 12 h and then drained. 0.1 g of the medium is weighed out accurately, added to a 25 mL Erlenmeyer flask containing 5 mL of the supernatant of the lysate, and then incubated at 25 C., 170 rpm for 24 h in a shaking bath. The mixture is centrifuged at 10,000 rpm for 1 mM and the supernatant is taken for SDS-PAGE analyses. It is shown that almost all Caps in the supernatant of the lysate bind to the affinity peptide medium. In contrast, Caps cannot be adsorbed to a blank medium. At the same time, most impure proteins are retained in the supernatant, demonstrating that the affinity peptide medium can selectively recognize Caps. It should be noted that adsorption at relatively high salinity (200 mM) suggests the domination role of hydrophobic interaction between DWDLRLLY and Caps, which is in accordance with the simulation results.

3. Chromatographic Separation and Purification Experiment

[0044] The DWDLRLLY-6B affinity medium is loaded into a 1 mL glass column (Tricorn 55) by gravity sedimentation. The equilibrating buffer (50 mM PBS, 200 mM NaCl, pH 7.0) is used for elution until the baseline is level. After loading 200 L of the supernatant of the lysate, the equilibrating buffer is used again for elution until the baseline is level. The bound protein is eluted with eluting buffer (50 mM citrate buffer, pH 3.0) of 5-10 column volumes and the regeneration of the affinity medium is carried out by 2-5 column volumes of regenerating buffer (100 mM Gly-HCl, pH 2.4) after the elution peak is separated completely. Equilibrating, washing, elution and regeneration are all performed at the flow rate of 0.5 mL/min while that of 0.2 mL/min is applied for loading. Components of the flow-through peak and the elution peak of the separated supernatant of the E.coli lysate are subjected to SDS-PAGE. The electrophoresis image is analyzed by the software Gel-Pro Analyzer 3.1 to determine the purity of VP1. The results indicate that the affinity medium could dramatically improve the purity of VP1 from the supernatant of the E.coli lysate from 15.6% to 70.1% by one-step chromatographic purification.

[0045] Therefore, the experiments indicate that the affinity peptide ligand DWDLRLLY could effectively purify target proteins and thus is a high-affinity peptide ligand for MPV Caps. It should be noted that, all the 6859 sequences of the peptide ligand library are able to be affinity peptide ligands of Caps in theory as they all have identical design thinking (key residues of VP2-C: D286, W287, L289, 1293, and Y296) and structural features. According to the available experimental data, the affinity peptide ligands in the peptide library are predicated to be effective affinity ligands of MPV Caps.

[0046] The invention provides novel affinity peptide ligands of mouse polyomavirus capsomers and a designed screening method thereof. The top one affinity peptide ligand DWDLRLLY is experimentally validated to be an effective affinity ligand of Caps and is capable of further separating and purifying Caps from the supernatant of the E.coli lysate, thereby having a broad application prospect in preparation of virus-like particles. However, it should be noted that there also exist some problems in DWDLRLLY, such as too strong hydrophobicity. Thereafter, to obtain higher-affinity and -specificity ligands, modifications of the DWDLRLLY and other peptides in the library are required in the following experiments. A modification strategy includes: adding one or more amino acids to the N/C-terminus of the peptides; inserting one or more amino acids between adjacent residues in the peptides; and replacing certain one or more residues in the peptides with other amino acids.