Respiratory syncytial virus (RSV) is the major cause of lower respiratory tract infections in infants and an important respiratory pathogen in the elderly and the immunocompromised patients. It is estimated that 23-65 thousand deaths in infants are caused by RSV in 2013. RSV is divided into two subgroups (RSV-A and RSV-B) based on its genetic and antigenic differences. The two subgroups are further subdivided into genotypes by sequencing of the second Highly Variable Region (HRV2) at the C-terminal end of the G gene. To date,36 genotypes have been described worldwide since the discovery of RSV in 1955. Besides the point mutations, large fragment insertion mutation was also detected in the HRV2 of the G gene, which lead to two new genotypes, BA in RSV-B and ON1 in RSV-A. Early evidences showed that the RSV_G protein, one of an important immunomodulatory protein, acquired accumulative mutations of translated amino acid over years, suggesting that the G protein hotspots mutation played a key role in persistent RSV epidemics. However, most previous studies on the genetic variability of RSV focused on the HRV2 of the G gene and only for limited time. Morover,16 new genotypes of RSV were detected since 2005, and a recent phenomenon was observed that the old genotypes are replacing by more new genotypes. However, the evolutionary characteristics of RSV were still not elucidated. How about the molecular epidemiological characteristics of RSV in china, and whether the epidemic patents are similar with that in other countries? What is the role of genetic mutation in RSV evolution? Is there synergistic effect of amino acid mutation in the first Highly Variable Region (HRV1) of the G gene during RSV evolution? To elucidate the molecular epidemic features of RSV, we studied the genetic variations of RSV between 2005 to 2015 in china. Based on the RSV gene data acquired from our study and the deposit data in GenBank since 1976, we calculated and analyzed the evolutionary characteristics of RSV by using phylodynamics method in this study.Total 1,601 RSV-positive respiratory specimens were collected previously in China from 2005 to 2015, and we obtained 514 RSVG full-length gene sequences, of which 289 were RSV-A and 225 were RSV-B; and 1,238 HRV2 gene sequences, of which 691 were RSV-A and 547 were RSV-B, by using nested RT-PCR methods. The phylogenetic tree was constructed by maximum likelihood method to cluster sequences into genotypes. In our study, the acquired sequences were cluster into 14 genotypes, including six genotypes in RSV-A (NA1, NA2, NA3, NA4, GA5, ONI) and eight genotypes in RSV-B (BA4, BA7, BA9, BA10, BAC, GB3, CB1, SAB4). Time distribution graph showed that RSV-A and RSV-B can co-detect in the same epidemic season and both have a biennial seasonal pattern. For example, RSV-ANA1 predominated in epidemic year of 2007/2008,2010/2011 and 2011/2012, while RSV-BBA9 predominated in 2008/2009, 2009/2010 and 2012/2013. RSV-A_ON1, which was characterized by a 72-nt duplication in the G gene, has been detected firstly in 2013. The detection rate of ON1 increased to 1.85% and 8.33% at 2013/2014 season and 2014/2015 season, while that of NA1 were kept at 1.44% and 1.30% during the same period. This observation suggested that RSV-A_ON1 was replacing the A_NA1, become the predominant genotype in China. Besides, BA9 was also one predominant genotype in subgroup B. We also found that all genotypes except A_NA1, B_BA9 and B_BA10 before 2011 were replaced. Interestingly, the ON1 positive case had higher pneumonia rates, while BA10-positve case had higher congenital heart disease rates than any other genotype case. There’s no significant difference in positive detection rate of genotypes between male and female.To analysis the evolution variability of RSV, we analyze the evolution dynamics of RSV-A and RSV-B based on full length of G gene and HVR2, with published data extracted from GenBank in 36 years. Bsyesian MCMC and selection pressure analysis showed that RSV-B have higher nucleotide substitution rate than RSV-A. The nucleotide substitution rate of G full-length gene and HRV2 over the period of RSV-B and B were estimated to be 2.83× 10-3,1.82×10-3, and 4.34χ10-3 and 3.41×10-3 site/year, respectively. It was found that the genotype substitution was more frequency in RSV-A, while BA9 and BA10 maintaining the pandemic in RSV-B. The results indicated that there might be a possible relationship between the different evolution characteristics of RSV-A and RSV-B and the different frequency of genotype substitution, or the epidemic time of predominant genotypesPositive selective press analysis based on G gene and HRV2 region showed four (225V,250P,274P, and 314P) and ten (215P,226L,233K,247L,256P,258H,262E, 274L,280Y, and 314P) positive selection sites, respectively (P>95%, Genbank accession number of reference strain:AB754590). Three genotypes showed positive selection on G and HRV2 regions, including 126L,262E and 321Q amino acid (aa) sites of RSV-A_GA5; 273Y aa site of A_ON1 and 237D, and 274P in NA1 genotype. In RSV-B, we found five (4H,219L,223P,252V, and 288V) and three (221P,236K, and 249V) positive selection site based on G gene and HRV2 region, respectively (P>95%, Genbank accession number of reference strain:JX198152). Five genotypes showed positive selections on G and HRVs region, including 216P in BA3,267F and 285Y in BA11,152P in GB1,4N,219L,2681,2751,276A,285Y and 292Y in BA9,249V in GB3, and 256G, and 3121 in GB4. All of the positive selection sites were located on the antigenic epitope, suggesting that it will have functions in imune regulation during RSV evolution, but need further investigations. Furthermore, the positive selection sites were not all the same between G full-length gene and HRV2, it may due to the selective sequences. Less aa positive selection sites were found in HRV1 than that in HRV2 gene regions, indicating the stronger selection pressure in HRV2. In A and B subgroup, the different aa positive selection sites results analyzed based on all genotypes and specific genotypes indicated the genotype-distinguishing codon sites and positively selected sites play roles in the ongoing genetic evolution. Most Recent Common Ancestor (MRCA) analysis results showed that the RSV-A group MRCA probably arised in 1950s and RSV-B in 1960s.In summary, this study reveals 6 epidemic genotypes in RSV-A and 8 epidemic genotypes in RSV-B in our country, and also shown that shift of predominant subgroup between RSV-A and RSV-B, shift of the genotypes and multiple genotypes co-circulation were the main epidemic characteristics. Though there are some difference on the genotypes epidemic in different regions in worldwide, RSV-ANA1 and RSV-ABA were the predominant genotypes, and also the replacement of NA1 by ON1 in recent two years. Besides, we found that RSV-B have higher nucleotide substitution rate than RSV-A. All of the aa positive selection sites in RSV-A were related to gene mutation generating genotypes, but not related to time. Two aa sites 221 and 268 were reverse-mutations. Our study provide an important data basis for explaining the evolutionary characteristics of RSV, and through analysis of the characteristics and potential functions of amino acid hotspots mutation, it will help us to intensively understand the immune escape mechanism of RSV. |