With the development of next-generate sequencing (NGS) technology and bioinformation analysis methods, whole-genome sequencing of HCV has until now been a convenient, easy and accurate procedure that has now been adopted in scientific research and clinical applications16, 17. The NGS methodology we used in this study was able to generate HCV whole genomes from blood donor’s samples, furthermore, it could accurately and reliably define the HCV subtype as the high average depths and strict access standards.
The molecular evolution of HCV 6a has previously only been investigated using single genome region or two concatenated partial genome regions18. Utilizing Bayesian MCMC phylogenetic methods, we analyzed 107 whole genome sequences and found that HCV 6a worldwide evolved from a most recent common ancestor (tMRCA) in 1905 (95%CI:1031-1943), which was younger than any other subtypes of HCV gt619. The topology of the MCMC tree showed that Vietnam and other Asian countries could be the origin of 6a all over the world, which was consistent with hypothesis of previous studies 3,12. Even though there is no spatio-temporal phylogenetic analysis, they provided considerable evidence to support their hypothesis20. In addition, we found that HCV 6a was introduced into Guangdong around 1960-1970 by two important transmission events. Originated from Vietnam, HCV 6a was transmitted to Hongkong. There were some evidences for the association between Hong Kong and Vietnamese Immigration21, our study also showed that Hongkong 6a was originated from Vietnam. One of the important transmission events was that HCV 6a in Guangdong were originated from Hongkong in 1966 (1941-1975), and then formed a local endemic in Guangdong province. We suspected that was because of the exchange between Guangdong Province and Hongkong in the early founding of the country22. Another transmission event showed that HCV6a was directly disseminate to Guangdong and formed a regional epidemic. From the topology of tree, we found that majority of sequences in group C was from Guangdong province not only in this study but also in reference sequences. Therefore, we conclude that this clade is clustered together, because the basic characteristics or lifestyles of the population in this area are similar, which resulting in a certain concentration of infected virus sequences.
Our study showed that E2 has the fastest evolutionary rate, roughly ten times than the majority partitioned genes of HCV 6a. It is well known that E2 contains three hypervariable regions (HVR), known as HVR1-323, which are under continuous selection pressure exerted by the host immune system. Therefore, it is reasonable that E2 evolve fastest among all the partitional genes. The evolutionary rate of E1, NS3, NS5A and NS4A was the same as a rate typical of RNA viruses 1.00×10-3s/s/y, almost equal to whole genome evolutionary rate. E1, an HCV gene, usually represent whole-genome for HCV genotyping, as the variability of E1 can approximately replace HCV whole-genome22. Our result also supported this insight. NS3 is the HCV protease and NTPase/helicase, NS4A acts as a cofactor for the NS3 protease23, which can explain why the NS3 and NS4A have the same evolutionary rate. The evolutionary rate of NS2 was 9.94×10-4s/s/y, NS5B and NS4A was slightly slower than NS2 gene. Core and 5’UTR had the slowest evolutionary rate among all the HCV 6a partitional genes. Study also showed that HCV core protein was the lowest variability and have most pan-genotypic consensus positions24. Combination with these two results, we speculated that was because HCV core protein forms the viral capsid and interacts with multiple cellular proteins, which may require a high level of conserved residues. 5’UTR was so a conversed gene that do not distinguish some of the HCV-6 subtypes prevalent in Southeast Asia, and that these subtypes should instead be classified as genotype 1 or 1b25 when based on primers from the 5’UTR. Using core sequencing data as well as 5’ UTR data, it shown a remarkable improvement as it in genotyping accuracy and differentiation between HCV-1 and HCV-6 variants 26.
Most published studies detecting positively selected sites in this virus have analyzed for different major subtypes such as HCV 1a and HCV 1b, no considering the region of the sequence source27, 28. In this study, we found that the proportion of positively selected sites along HCV genomes was difference between Guangdong and non-Guangdong sequences, which indicating that geographical location is related viral adaptation between HCV and the cellular immune response. Analyses at the genomic level also revealed that Guangdong and non-Guangdong HCV 6a evolve under difference forces and constraints (CD8 T cell epitopes favoring selection in non-Guangdong HCV 6a). We speculate that the virus transmission pattern together with host genetic background in different regions leading to the different positive sites. The highest proportion of positive selection sites was found in E1 and E2 in both geographical locations, which is consistent with previous published studies24, 29, 30. These regions were reported to have the high genetic variability along the HCV genomes31, which reflect the generation of escape mutants whereby the interaction between positive selection and host immune response. As we know, E1 and E2 glycoproteins determines binding and fusion properties of HCV during cell entry32. Consequently, the positive site was the consequence of the human immune response and virus mutation when infected a new host. On the contrary, the proteins with the lowest proportion of sites under positive selection were NS3 which is a potential target for direct-acting antivirals that function to inhibit virus replication33. In addition, NS3 was also one of the most frequently HCV protein targets of the CD4 and CD8 T cell response34, 35. The lowest positive selection sites in this gene would make DAA treatment more effective and facilitate viral escape from the immune system whereby CD4 and CD8 T mediated cell immune response.
The frequency of RASs varied by both HCV subtype and geographic regions 33, therefore, we analyzed the frequency and distribution of naturally occurring RASs of HCV 6a in Guangdong regions using NGS sequence data. The RASs of HCV 6a has previously only been investigated using sanger sequencing, which can only detect more than 15%-25% of HCV variations, while deep sequencing can detect as low as 1% of HCV variation 36, 37. In this study, mutations V36G in NS3 region and K24R in NS5A region were not found when the cutoff >95%, but after adjusting the cutoff to 5%, the prevalence was as high as 28.57% (4/14) and 14.3% (2/14) of the patients. Other RASs reported in this study were also found in other region’s 6a sequences38. Even HCV 6a RASs could compromise treatment in robust infectious cell culture models39, but 100% SVR rates have been reported in a few reports on the outcomes of DAA therapy in gt6-infected patients40, 41. More functional and clinical trials are needed to validate the RASs found in this study.
In conclusion, this study firstly analyzed the propagation mode and evolution rate of HCV 6a by whole genome sequence, revealed the transmission mode of HCV 6a in Guangdong. We have shown that, Guangdong and non-Guangdong HCV 6a share different selective pressures along their genomes, which give information about the effect of some of the interactions between HCV and its host on HCV variability. It may be useful for antiviral research against this virus. Overall, the results from this study provided data support for the formulation of HCV 6a prevention strategy in Guangdong, China.