Modelling of partial structure of human cytomegalovirus PUL97 protein and drugresistant analysis of new non-synonymous mutations

FANG Feng-qin,XIE Qiong,ZHANG Yue,MAO Ke-zi,ZHANG Li,LU Yi-de,HUA Li,CAO Guo-jun,JI Yu-hua
2010-01-01
Abstract:Objective To preliminarily explore the 3-dimensional modelling of partial structure(329-572aa)of cytomegalovirus PUL97 protein and elementarily analyse the possibility of drug-resistance of new non-synonymous mutations.Methods Nested PCR was employed to amplify 776 bp fragment of UL97 gene.And some non-synonymous mutations in it were found by sequencing.After that,the online server of SWISS-MODEL and other related softwares were used for modelling and optimizing the partial structure of wildtype PUL97 protein based on the alignment mode.Then,the modelled structures were evaluated and screened by various classical indicators or online servers.Finally,the best structure was obtained and employed for preliminarily analysis of potential drug-resistant possibility of non-synonymous mutations.Results The finally-screened model was Model A based upon the template of 1YDTE and the alignment file of L1.FASTA.The whole energy of the structure was the lowest,while the prediction result of ProQ was on the top compared with the rest.Using the method of molecular docking,the potential binding sites of ATP and GCV were found in it.And most important,the mutation analysis based upon the modelled structure and positive/negative drug-resistant mutations confirmed that this structure was reasonable.Finally,the newly-found mutations(namely,point mutation C518Y and combined mutation E517G/C518Y)showed some evidences of having the possibility of drug-resistance under the benefit of the aboveestablished system.Such evidences were the significant differences in structural information before and after mutagenesis. Conclusion The above-described system of PUL97.structure modelling and mutation analysis is the valuable and alternative method for preliminarily predicting the possibility of drug-resistance of new mutations.It is more convenient and quick,and can be regarded as the screener before experimental verification.However,the accuracy of modelled structure itself is the big prerequisite for successful application of this system.
What problem does this paper attempt to address?