| Objective:To systematically predict the conformational epitopes of coxsackievirus A10(CVA10)using the bioinformatics algorithms of human enterovirus conformational epitopes developed by our laboratory.And then,we extended these algorithms to picornavirus.To systematically predict the conformational epitopes of Human parechovirus 1(HPe V1)and Human parechovirus 3(HPe V3),and explore the distribution of the conformational epitopes of these three picornaviruses.Methods:1.Three strains of CVA10(FY01 / AH / CHN / 2013,CVA10-FJ-01 and A10 / S0148b)were select,the three-dimensional structure files of these three strains were downloaded from RCSB PDB database and the corresponding amino acid sequences of structural proteins were downloaded from the NCBI Nucleotide database.The amino acid sequences and secondary structures of CVA10 structural protein were aligned by the MEGA 7 and ESPript,and the three-dimensional structures and surface characteristics of the viral capsid of CVA10 were generated using Py MOL.Use the bioinformatics algorithms of human enterovirus conformational epitopes developed by our laboratory to predict the conformation epitopes of CVA10.The experimental epitopes of CVA10 were obtained by literature search.The "footprint" of CVA10 was generated using RIVEM and we evaluated the accuracy of the prediction algorithm by comparing experimental epitopes with predicted epitopes and analyzed the relationship between antibodies and conformational epitopes.2.The representative strains of HPe V1(strain Harris)and HPe V3(strain A308/99)were choosed and the three-dimensional structure files of HPe V1 and HPe V3 were downloaded from the RCSB PDB database and the corresponding amino acid sequences of structural proteins were retrieved from the NCBI Nucleotide database.In addition,we randomly selected 16 HPe V1-6 strains(HPe V1: 4 strains;HPe V2: 2 strains;HPe V3: 4 strains;HPe V4: 2 strains)from the Nucleotide database which had the complete amino acid sequences of structural proteins to analyze the sequence conservation.The amino acid sequences,secondary structures,three-dimensional structures and capsid surface structure of HPe V1,HPe V3 and CVA10 were compared using MEGA,ESPript,Py MOL and UCSF Chimera.The bioinformatics algorithms of human enterovirus conformational epitopes developed by our laboratory were extended to picornavirus to predict the conformation epitopes of HPe V1 and HPe V3.The experimental epitopes of HPe V1 and HPe V3 were obtained by literature search.The "footprint" of HPe V1 and HPe V3 were generated using RIVEM and we evaluated the accuracy of the prediction algorithm by comparing experimental epitopes with predicted epitopes and analyzed the relationship between antibodies and conformational epitopes.Results:1.The 45 amino-acid residues were clustered into three sites: site 1,site 2,and site 3.Site 1was located in the "northern rim" near the five-fold axis and consisted of the BC loop and HI loop of VP1.Site 2 was located in the "puff" of "southern rim" and mainly composed of the GH loop and C-terminus of VP1 and the EF loop of VP2.Site 3 was divided into two parts.One was located in the "knob" of "southern rim" and composed of C-terminus of VP1 and N-terminus of VP3.It was closer to site 1 and site 2 and were distributed at three sites as the Chinese character “品”.The other part was close to the threefold vertex which was mainly consisted of the HI loop of VP2 and the HI loop of VP3.Site 2 and site 3 overlapped highly with two epitopes 2G8 and VP2-P28.2.The results showed that the distribution patterns of conformational epitopes in HPe V1 and HPe V3 are highly consistent and both the conformational epitopes of HPe V1 and HPe V3 clustered into three sites(site 1,site 2 and site 3).Site 1 was located in the "northern rim" near the fivefold vertex and consisted of the BC loop and the HI loop of VP1;Site 2 was in the "puff" of "southern rim" and was mainly composed of the C-terminus of VP1;Site 3 was divided into two parts,one was located in the "knob" and mainly composed of the N-terminus of VP3;the other was close to the threefold vertex which was mainly consisted of the HI loop of VP0/VP2 and the HI loop of VP3.Compared with CVA10,there were fewer residues in the EF loop of VP0(VP2)and GH loop of VP1 of HPe V1 and HPe V3,resulting in a flattened puff region and then making it difficult to form conformational epitopes.So,there were differences between the conformational epitopes in site 2.At the same time,due to the longer C-terminus of VP1 of CVA10,there was also a difference between the conformational epitopes that located in the "knob" of site 3.The two residues VP1-85N-87 D on site 1 may be the key residues of AT12-015 binding to HPe V3.The three amino acids VP3-91N-92 H and VP0-257 S of HPe V1 were not only the key residues for the monoclonal antibody AM28 binding,but also had vital effects on the serological distinguished between HPe V1 and HPe V3.Conclusions:1.HPe V1 and HPe V3 both had the same distribution patterns of conformational epitopes and they both clustered into three sites(site 1,site 2 and site 3).2.The conformation epitopes in site 2 and site 3 of HPe V1 and HPe V3 are significantly different from CVA10.The three key amino acid residues in site 3 may determine the serological differences between HPe V1 and HPe V3. |
