HCV 6a prevalence in Guangdong province had the origin from Vietnam and recent dissemination to other regions of China: phylogeographic analyses

Abstract

Background: Recently in China, HCV 6a infection has shown a fast increase among patients and blood donors, possibly due to IDU linked transmission.

Methodology/findings: We recruited 210 drug users in Shanwei city, Guangdong province. Among them, HCV RNA was detected in 150 (71.4%), both E1 and NS5B genes were sequenced in 136, and 6a genotyped in 70. Of the 6a sequences, most were grouped into three clusters while 23% represent emerging strains. For coalescent analysis, additional 6a sequences were determined among 21 blood donors from Vietnam, 22 donors from 12 provinces of China, and 36 IDUs from Liuzhou City in Guangxi Province. Phylogeographic analyses indicated that Vietnam could be the origin of 6a in China. The Guangxi Province, which borders Vietnam, could be the first region to accept 6a for circulation. Migration from Yunnan, which also borders Vietnam, might be equally important, but it was only detected among IDUs in limited regions. From Guangxi, 6a could have further spread to Guangdong, Yunnan, Hainan, and Hubei provinces. However, evidence showed that only in Guangdong has 6a become a local epidemic, making Guangdong the second source region to disseminate 6a to the other 12 provinces. With a rate of 2.737×10⁻³ (95% CI: 1.792×10⁻³ to 3.745×10⁻³), a Bayesian Skyline Plot was portrayed. It revealed an exponential 6a growth during 1994-1998, while before and after 1994-1998 slow 6a growths were maintained. Concurrently, 1994-1998 corresponded to a period when contaminated blood transfusion was common, which caused many people being infected with HIV and HCV, until the Chinese government outlawed the use of paid blood donations in 1998.

Conclusions/significance: With an origin from Vietnam, 6a has become a local epidemic in Guangdong Province, where an increasing prevalence has subsequently led to 6a spread to many other regions of China.

Figure 1. Subtype 6a phylogenies estimated from (A) E1 and (B) NS5B region sequences, corresponding to H77 nucleotide positions of 869–1289 and 8276–8615, respectively.

Subtype 6b sequence D84262 was used as an outgroup. Green pies label sequences from our previous studies , . Red and yellow pies label sequences from this study, in which yellow pies mark isolates from IDUs with multiple HCV infections. Sequences without pies were retrieved from Genbank. In each tree, five rectangles highlight the further classification of 6a isolates into I, II, III, VI, and VII clusters. Scale bar represents 0.02 nucleotide substitutions per site.

Figure 2. MCC tree estimated under the model of BSP + Uncorrelated Exponential (see Table 4 ).

In addition to the isolates presented in Figure 1A, sequences determined among 21 blood donors from Vietnam, 22 blood donors from the 12 provinces of China, and 36 IDUs from Liuzhou City in the Guangxi Province were also included. Branches were colored according to their respective geographic origins. Posterior probabilities greater than 0.9 were shown at the respective nodes, while the estimated time points of the origin for the nodes were parenthesized. Below the tree is a time scale from 1960–2010, which measures 6a origin and evolution.

Figure 3. Map of HCV 6a migration in China.

With origin from Vietnam, 6a was spread to the neighboring Guangxi and Yunnan Provinces for circulation (shown in pink arrows). From Guangxi, 6a was further spread to the Guangdong, Hainan, Yunnan, and Hubei provinces (shown in green arrows), while from Yunnan, 6a was brought by IDUs to the central part of China (shown in a blue arrow). For unknown reasons, 6a has become a local epidemic in Guangdong and caused infections not only among IDUs, but also among blood donors and patients. As the prevalence is increasing, Guangdong has become the second source region to radiate 6a to other regions (shown in red arrows).

Figure 4. The overlapped BSPs.

Three indicated solid lines represent the estimated effective population sizes through time under three combination models: BSP + Uncorrected Exponential, BSP + Uncorrected Lognormal, and BSP + Strict Clock, respectively. The colored areas (pink = Exponential, blue = Lognormal, and yellow = Strict) around the solid lines represent the 95% highest posterior density confidence intervals for these estimates. The vertical ruler on the left scales the effective population size while the horizontal ruler on the bottom measures time.