Secondary Logo

Journal Logo


HIV-1 and hepatitis C virus selection bottleneck in Chinese people who inject drugs

Li, Fana; Ma, Liyinga; Feng, Yia; Ruan, Yuhuaa; Hu, Jinga; Song, Hongshuoa; Liu, Pengtaoa; Ma, Junb; Rui, Baolinc; Kerpen, Kated; Scheinfeld, Benjamind; Srivastava, Tuhinad; Metzger, Davidd,e; Li, Huid; Bar, Katharine J.d,*; Shao, Yiminga,f,g,*

Author Information
doi: 10.1097/QAD.0000000000001702



For both HIV-1 and hepatitis C virus (HCV), it is believed that preventive methods, including vaccines, will be most effective in blocking the virus or virus-infected cells at or near the moment of transmission or in early infection preceding systemic dissemination [1,2]. Thus, understanding the earliest events of transmission and early infection are key to curbing these pandemics. To this end, the enumeration of transmitted/founder (TF) viruses has allowed quantification of the stringency of the selection bottleneck and highlighted differences between modes of transmission [3–17].

Both HIV-1 and HCV can be transmitted via sexual and parenteral transmission. For HIV-1, the more than 60 million infections comprising the global pandemic have been primarily sexual transmissions, but injection drug use accounts for nearly 10% globally [18]. An estimated 71 million individuals are living with HCV infection globally [19], largely driven by parenteral exposure via IDU, contaminated blood supplies or medical equipment; sexual transmission has played a more minor role [20].

Recent studies have consistently shown that mucosal transmission of HIV-1 is defined by a highly stringent bottleneck, with analyses of heterosexual transmission demonstrating the vast majority (>80%) of infections resulting from a single TF virus [3–5]. HIV-1 transmission in MSM is biologically and epidemiologically distinct from heterosexual transmission [21,22] and studies of HIV-1 transmission in MSM show more frequent multivariant transmission (MVT) and a greater range in the number of TF viruses than heterosexuals [6–8].

HIV-1 transmission in people who inject drugs (PWID) is expected to involve different biology than mucosal transmission. The specific target cells, immune pressures and inoculum size of parenteral transmission are unclear, but the physical barrier and immune functionalities comprising the mucosal transmission process are absent. A study of high-risk PWID in Montreal demonstrated a high rate of MVT, with examples of very high numbers of TF viruses in one individual [9]. Similarly, a study using single genome sequencing (SGS) to elucidate the failure of postexposure prophylaxis after a high-inoculum needlestick demonstrated the transmission of multiple (>15) TF viruses [23]. In contrast, studies of lower risk cohorts (e.g. PWID enrolled in drug treatment or clinical trials) have reported rates of MVT that are only marginally higher than in heterosexuals [10,11]. Thus, the frequency of MVT in PWID is unclear.

For HCV infection, recent studies in acutely HCV-infected plasma donors have employed molecular sequencing and modelling to characterize HCV transmission and early virus diversification [12–16]. These studies revealed a wide range frequency of MVT establishing HCV infection. Notably, the individuals studied either lacked reliable behavioural data, so the causative modes of transmission –including injection and sexual contact among MSM – remain undefined. Molecular characterization of transmission processes for parenteral acquisition of these pathogens is, therefore, a scientific priority, especially in cohorts with characterized behaviours.

Our study aims to quantify the HIV-1 and HCV transmission bottleneck in a well characterized cohort of PWID in China. We specifically ask what is the rate of MVT in PWID, how do HIV-1 and HCV transmission rates compare in the same cohort, and can we link any specific behaviours to MVT? We studied PWID from Urumqi, the capitol city of Xinjiang Province, which has a high prevalence of both infections [24,25]. All study individuals were participants of HIV Prevention Trials Network (HPTN) studies in Urumqi between 2003 and 2012 [26,27], who were HIV-1 seronegative upon entry, provided detailed demographic and behavioural data, and then were followed longitudinally for HIV-1 and HCV seroconversion. Using SGS, validated models of early virus diversification and a Bayesian model of virus evolution, we provide the first description of the stringency of both the HIV-1 and HCV selection bottlenecks in a large, well described cohort of PWID.

Materials and methods

Ethics statement

This study was approved by the institutional review boards of the Chinese National Center for AIDS/STD Control and Prevention, the Chinese Center for Disease Control and Prevention, and the University of Pennsylvania. All study participants provided written informed consent prior to sample collection through HPTN033 and HPTN058, which were reviewed and approved by the China CDC and the Johns Hopkins IRBs [25,27].

Hepatitis C virus and HIV-1 testing

HCV and HIV-1 infected individuals were identified by prospective serological testing during the HPTN trials; acute infection was determined by retrospective testing for vRNA in stored preseroconversion plasma. The HIV-1 samples were classified by Fiebig staging [28]. The window period of HCV seroconversion was estimated at 56 days [29].

Viral RNA extraction, cDNA synthesis and SGS

Viral RNA was extracted from plasma using the QIAamp Viral RNA Mini kit (Qiagen, Valencia, California, USA) and reverse transcribed by SuperScript III (Invitrogen, Carlsbad, California, USA). Full-length HIV-1 gp160 or gp41 and 5’ half HCV genome sequences ranging from the complete half genome to a more than 1000 nucleotide amplicon were amplified by nested PCR from cDNA (primers in Table S1, SGS was carried out as previously described [5,9]; PCR products were sequenced by using BigDye Terminator chemistry (Applied Biosystems, Foster City, California, USA) and edited using Sequencher, version 4.7 (Gene Codes, Ann Arbor, Michigan, USA). All sequences were aligned using Gene Cutter ( and adjusted manually with Bioedit software [30]. All sequences were deposited in GenBank under accession numbers KY345421-KY346492, KY405041-KY405825.

Sequence analysis

A clustering approach also used for estimating numbers of linage variants. Within-individual nucleotide sequence diversity was analysed using DIVEIN [31] ( Sequences were visualized in Highlighter plots [5]. Neighbour-joining (NJ) phylogenetic trees were generated by Mega 6.0 [32]. Maximum likelihood (ML) phylogenies of all sequences for HIV-1 and HCV were generated by RaxML [33]. Sequences were assessed for fit to a model of random virus evolution by Poisson–Fitter ( [34].

Mutation rate estimation and Bayesian analysis

Mutation rates for HIV-1 and HCV were calculated using Treerate ( with longitudinal sampling from our cohort and other acute infection cohorts [5,35,36]. Average mutation rates based on published estimates [37,38], of 1.7e−5 (HIV-1 gp160), 1.2 e−5 (HIV-1 gp41) and 2.2e−5 (HCV) [39] substitutions/site/day, were used in Bayesian analyses. The time since the most recent common ancestor (tMRCA) was estimated using a Bayesian Markov Chain Monte Carlo (MCMC) approach, implemented in BEAST v1.8.2 [40,41]. All analyses were carried out using the HKY85 substitution model with gamma-distributed rate heterogeneity (four gamma categories), selected jModelTest 2.1.7 [42,43]. We assumed exponential population growth and a strict molecular clock [34]. The MCMC algorithm was run for at least 108 generations (logging every 1000 generations; burn in was set to 10% of the original chain). The results were visualized in TRACER V1.6 [44].


Study population and clinical staging of samples

The present study analysed plasma samples collected in Urumqi, Xinjiang, through the HPTN033 and HPTN058 clinical studies [25,27]. HPTN033 was an observational cohort study conducted in 2003–2004; HPTN058 was a randomized trial testing buprenorphine/naloxone as a HIV-1 prevention strategy conducted in 2009–2012. Both trials enrolled individuals reporting active injection practices who were HIV-1 antibody negative and HCV-positive or negative. After enrolment, samples were collected at 6-month intervals. The present analysis included all incident HIV-1 and HCV infections from both trials with detectable viremia, totaling 14 acute (Fiebig II-V) and 19 early (Febig VI) HIV-1 infections and 20 acute (preseroconversion) and seven early (postseroconversion) HCV infections (Tables 1 and 2, S2-3, [28].

Table 1:
Diversity of SGS-derived env sequences in participants with primary HIV-1 infections.
Table 2:
Diversity of SGS-derived 5’half sequences in participants with primary HCV infections.

Clinical and behavioural characterization

In both HPTN033 and HPTN058, the HCV prevalence at enrolment was high (79 and 81% seropositivity, respectively) [25,27]. The annual incidence of HCV was 52% during HPTN033 and 32% during HPTN058. Incidence also fell for HIV-1; it was 8.8% in HPTN033 and less than 2% in HPTN058 [18,36]. We studied 33 participants with incident HIV-1 and 27 participants with incident HCV. Baseline demographic and behavioural data of these two subgroups are summarized in Table 3. Briefly, participants were an average of 28 years, predominantly men and of Uyghur ethnicity, more likely to be single and without secondary or postsecondary education. Behaviourally, all reported active injection of heroin with the modal frequency of at least daily; a minority (28%) reported sharing injection equipment. The majority were sexually active with a minority, 22%, reporting multiple partners. We have no data on the sex of individual's partners, nor self-reports of MSM activity.

Table 3:
Baseline characteristics and behaviours of newly infected participants.

Sequence analyses of HIV-1 primary infections

We generated 1070 SGS-derived HIV-1 env sequences (median 30 per individual, range 14–71). For 22 individuals, full env gp160 regions were sequenced; for 11 individuals, gp41 sequences were generated (Table 1). When the gp41 sequences of the 33 HIV infections were analysed together in a maximum likelihood phylogeny, each individual's sequences clustered independently, with the exception of three groups of linked infections (Fig. S1, The limited genetic distance between nodes reflects the homogeneity of the CRF07_BC epidemic and closely related networks of PWID in Urumqi [45,46].

For the 33 HIV-1 seroconversions, we enumerated TF viruses through several methodologies, including visualization of highlighter plots and phylogenies, Poisson-fitter modelling [34] and comparison with Bayesian estimations of time to tMRCA (Table 1). Four representative examples are shown in Fig. 1. Individual 33.12 (Fig. 1a), sampled in Fiebig stage II, had a homogeneous sequence set with a maximum diversity of 0.002 that formed a single low-diversity lineage in both the Highlighter plot and the NJ tree. Poisson–Fitter modelling suggested a single virus transmission (SVT) (Table 1). The Bayesian estimate of tMRCA of 18 days [95% confidence interval (95% CI) 10–26] was within the range of the time since infection suggested by clinical staging, as shown in the maximum clade credibility (MCC) tree in Fig. 1a. These corroborating lines of evidence all suggest SVT. Similarly, Individual 33.12, sampled in Fiebig stage V with a maximum diversity of 0.0047, also suggested SVT by Highlighter plot, NJ phylogeny and Bayesian estimates of tMRCA. Regions of shared mutations that are visible in the Highlighter plot and NJ tree that violated Poisson-fitter model conditions are consistent with virus escape from early immune pressure [11,47,48]. In contrast, individuals 58.10 and 33.03, sampled in Fiebig stage II and VI, respectively, demonstrated multiple genetically distinct virus populations, grossly violated Poisson-fitter modelling and had estimated tMRCA that far exceed the maximum time since infection determined by clinical staging and timing of samples (Table 1). As shown in Fig. 1c and d, at the maximum time since infection, the time-weighted MCC phylogeny demonstrated multiple distinct branches, corroborating the inference of the minimum number of virus lineages establishing productive infection by NJ tree and Highlighter in these participants with MVT. In individuals 58.10 and 33.03, therefore, we inferred MVT with six and two TF viruses, respectively. These approaches were applied to the 33 HIV-1 seroconversions and revealed 20 individuals with SVT. When analysed by Poisson-fitter [34], 16 of these 20 SVT individuals conformed to a model of random virus replication, demonstrating Star-like phylogeny and a Poisson distribution of mutations (Table 1). Individuals with SVT who violated model conditions generally did so with a small number of shared mutations, which when removed allowed for sequences to fit model conditions. In total, we found that 39.4% (13 of the 33) HIV-1 infections were founded by multiple viruses (range 2–6). Estimations of single versus MVT and exact enumerations between phylogenetic, Poisson-Fitter and Bayesian strategies were highly congruent, even in individuals sampled postseroconversion (Fig. 1, Table 1).

Fig. 1:
Highlighter plot, NJ phylogenetic tree and MCC tree depictions of HIV-1 env sequences.The left column displays Highlighter plots, in which nucleotide polymorphisms are displayed as coloured tics (T: red, A: green, C: blue, G: yellow) compared with the master sequence [denoted with (m) on top line of alignment]. The centre column depicts NJ phylogenies, with genetic distance shown by scale bar. The right column depicts MCC trees, based on Bayesian modelling, with the root of the tree representing the tMRCA, and the blue line representing the maximum time to infection as determined by sampling dates and clinical staging. Participants 33.12 (a) and 33.18 (b),demonstrate a single low-diversity lineage and MCC estimated tMRCAs after maximum infection times, consistent with infection by a single virus. Participant 58.10 (c) and 33.03 (d), demonstrated multiple low-diversity lineages, with maximum infection time far after the MCC estimate of tMRCA.

Sequence analyses of hepatitis C virus seroconversions

A total of 773 HCV 5’ half genome sequences (median 27 per individual, range 11–57) were generated by SGS. When sufficient numbers of complete half genomes were not obtained or when plasma samples were limited, we amplified shorter regions within the 5’ half genome (between 700 and 4879 nucleotides) (Table 2). Virus subtype was determined for each sample; we detected the predominant genotypes reported to be circulating in Xinjiang (3a, 3b and 1b) [42]. When analysed within a maximum likelihood phylogeny with reference sequences, the sequences clustered by genotype with strong bootstrap support and individual participants’ sequences formed distinct, but closely related lineages, reflecting a narrow genetic range of the Xinjiang HCV epidemic in PWID [24].

Individual participant lineages demonstrated a substantial range in maximum within-patient diversity, from 0.0005 to 0.093 (Table 2). To determine the multiplicity of HCV infection, we visualized sequences by NJ phylogeny and Highlighter plot and analysed the sequences with Poisson-Fitter, a model of random virus evolution that has been validated in early HCV infection [13,35]. Four representative individuals’ sequences are shown in Figs. 1 and 2. Participants 33.28 and 58.15, sampled in acute HCV infection, demonstrated a single low-diversity lineage and MCC estimated tMRCAs that were after maximum infection times, consistent with infection by a single virus (Fig. 2a and b). Thirteen individuals had similar low diversity sequences (0.0005–0.0071), which clustered into single low-diversity lineages by NJ phylogeny and Highlighter analysis and generally demonstrated a Poisson distribution of mutations [5,34]. Fourteen individuals, illustrated by participants 58.19 and 58.16 (Fig. 2c and d), had greater sequence diversity (0.0068–0.093) that clustered into multiple lineages by NJ phylogeny and Highlighter plot. In these individuals, tMRCA preceded the maximum time since infection, suggesting MVT (Table 2). Enumeration of low-diversity lineages in NJ and MCC phylogenies demonstrated between two and six genetically distinct virus populations in these individuals with MVT. In total, we found 52% (14 of 27) MVT of HCV, with a range of two to six virus lineages per participant.

Fig. 2:
Highlighter plot, NJ phylogenetic tree and MCC tree depictions of hepatitis C virus 5’ half sequences.The annotation is the same as given for Fig. 1. Participants 33.28 (a) and 58.15 (b), both sampled in acute infection, demonstrate a single low-diversity lineage and MCC estimated tMRCAs after maximum infection times, consistent with infection by a single virus. Participants 58.19 (c) and 58.16 (d), sampled post and before seroconversion, respectively, demonstrate multiple low-diversity lineages by Highlighter plot and NJ tree, and maximum times to infection that substantially precede estimated tMRCAs.

To mitigate concerns about the effects of immune selection in HCV infection after nearly 80–100 days [14–16], we performed analyses excluding all individuals sampled postseroconversion. Restricting the analysis to only HCV transmissions sampled in acute infection (preseroconversion), the frequency of MVT is similar, with 47% (9/19) MVT and a range of two to six TF variants.

In summary, we found between 47 and 52% (14/27 or 9/19) MVT of HCV, with a range of two to six virus lineages per individual.

Correlations between clinical and behavioural data and multivariant transmission

We performed multivariate analysis for risk factors associated with MVT in both HCV and HIV-1 seroconversions. We assessed demographic characteristics, including age, sex, education level and marital status, and risk behaviours, including shared needles and equipment, practices such as backloading and drug type. In multivariate analyses of both HIV-1 and HCV, no demographic characteristics or drug or sexual behaviours were significantly correlated with MVT (Table S4,


The present study provides new insights into the transmission of HIV-1 and HCV in PWID. Using multiple complementary sequence analysis strategies, we determined the stringency of the transmission process in well described cohorts of Chinese PWID for these two critically important viral pathogens.

For HIV-1, we found a rate of 39% MVT, which suggests a substantially more permissive transmission process in parenteral transmission than in sexual transmission (Table S5, [3–8]. For HCV, results provide the first estimate of the multiplicity of infection specifically in PWID. We found 47% MVT in preseroconversion HCV infections and 52% in both acute and early HCV infections in the Xinjiang PWID.

The increased frequency of MVT in PWID compared with mucosal transmission cohorts is important for several reasons. First, it suggests a distinct biology of transmission in parenteral exposures that is congruent with the higher epidemiologic risk of infections in parenteral exposure. Similar rates of MVT have been seen in analyses of other cohorts, and our findings support and strengthen this body of evidence. Including our study, analysis of 75 total PWID using similar methodologies yields a rate of 43% MVT in HIV-1(Table S5, That the frequency of MVT is similar in studies performed in different geographical regions, with diverse virus subtypes [9–11], and in cohorts with different risk behaviours provides additional confidence to the idea that parenteral transmission is a significantly more permissive process than is sexual transmission in both heterosexual cohorts (18–9% MVT, P = 0.0003, Fisher's exact test) and MSM (25% MVT, P = 0.0123) Table S5, Notably, we did not find examples of very high numbers of TF viruses in our cohort, as has been reported in high-risk PWID [9], large inocula parenteral exposures [23] or other studies of HCV transmission [13]. This may be due to later sampling and limited sampling depth, but we note the high numbers of TF viruses seen in the cited examples persisted through early infection when tested [9,13,23]. Our quantitative measure of the HCV selection bottleneck in PWID is, to our knowledge, the first study to link a sensitive molecular estimate of the multiplicity of HCV infection with a mode of transmission in larger cohort of individuals. Unique to our study, the virologic measures are linked to behavioural data. All individuals reported ongoing injection practices with minimal sexual risk factors (just 22% reported multiple partners). Although we found no clear associations between any specific behaviours and MVT, the behavioural data allow us to more clearly establish the link between injection and increased multiplicity of infection.

Our results also have clinical implications. MVT of HIV-1 has been shown to be associated with worse virologic and clinical outcomes in large cohorts [49,50]. In conjunction with the detrimental effects of injection drug use unrelated to HIV-1 and HCV [51–53], these data inform our understanding of the health challenges of PWID.

Finally, a less stringent transmission process has important theoretical consequences for prevention strategies, wherein interdiction of MVT may be more challenging than SVT. Recent analyses of acute and early infection suggest that MVT may allow the virus greater flexibility to rapidly escape host immune pressures [54]. In theory, therefore, the efficacy of preventive strategies, for example vaccines or preexposure prophylaxis with antiretroviral medications, may need to be independently evaluated in individuals with predominantly parenteral transmission risks.

This study also provided the unique opportunity to compare the multiplicity of infection of both HIV-1 and HCV in the same cohorts. We had just one individual, 33.03, with both incident HIV-1 and HCV; he or she was infected with at least two HIV-1 variants and a single HCV variant. 33.03 was sampled pre-HCV antibody seroconversion (within ∼56 days of HCV infection) and at Fiebig stage VI of HIV infection (>100 days of HIV-1 infection), suggesting that he or she may have acquired viral infections via the different events and possibly from different donors. Thus, the direct comparison of HIV-1 and HCV transmission is challenging in this individual. In the cohort as a whole, we note that the rate of HIV-1 acquisition decreased substantially between HPTN033 and HPTN058 (from 8.8% to <2%), while the rate of HCV infection only decreased modestly (52 to 32%). Similarly, we saw a statistically nonsignificant decrease in the rate of MVT HIV-1 between trials (43 to 33%), but not in HCV (47 to 55%) (P = 0.33, by chi-square test) [24,25,27,55]. While many factors likely contributed to the differential decrease in HIV-1 compared with HCV, it is reasonable to infer that the prevention strategies in place, including treatment with buprenorphine/naloxone combined with behavioural drug and risk counselling [27], while associated with a reduction in incident HIV-1, were insufficient to greatly impact the local HCV epidemic. Further study into coinfected individuals to investigate these issues is warranted.

This study had several important limitations inherent to the use of this cohort and our chosen methods. First, some individuals had samples obtained after antibody seroconversion. Although it is challenging to identify the exact sequence of the TF viruses, we argue it is feasible to infer a minimal estimate of the diversity at transmission using the novel Bayesian model of virus evolution as both HIV and HCV, which utilizes exponential viral growth with sequence diversity increased linearly within ∼150 days after infection [56–58]. Using this strategy, we can confidently determine a threshold diversity in which SVT was not possible. In addition, later sampling prevented the opportunity to sequence some HCV-infected individuals who spontaneously cleared infection in the interval before sampling. To mitigate effects of later sampling our sequence analyses, we assessed virus evolution by multiple methods. Importantly, the Bayesian, Poisson-Fitter and phylogenetic estimates of the minimum number of virus lineage were highly congruous (Tables 1 and 2) and comparisons of HCV infections studied before and after seroconversion led to similar estimates of the frequency of MVT and the range of TF viruses. An additional limitation of the study is that our enumeration of virus lineages are minimum estimates, as our methods may have underestimated minor variants and closely related variants [5,6]. The sensitivity to detect minor variants is directly related to sequencing depth; given the median numbers of sequences for HIV-1 and HCV (30 and 27, respectively), we have 95% confidence of detect variants circulating at least nearly 10% frequency, and may miss more minor constituents [5]. Further, the later sampling limited discrimination of multiple closely related variants, which can only be distinguished when acutely infected individuals are sampled extremely early in infection [5,12,13,34]. Another limitation was sample degradation, which precluded amplification of full env genomes (HIV-1) or complete 5’ half genomes (HCV) in several participants. We performed sensitivity analyses to determine the ability to estimate tMRCA with larger (between 3 and 5 kB) and smaller (∼1 kB) sequence spans and found only minor differences between sequence lengths (Table S6,

In summary, we found a relatively permissive transmission bottleneck to both HIV-1 and HCV in PWID in Xinjiang, China. The estimations of the multiplicity of infection in PWID provide a quantitative measure of the HIV-1 and HCV transmission process in this critically important population. Results provide an estimate of the challenge to be overcome by preventive strategies needed to end these pandemics.


The authors are grateful to colleagues in the Center for Disease Control and Prevention of Xinjiang and Urumqi for their assistance during the HPTN058 study and to Nina for her help in HCV serological testing and vRNA testing during the pilot test of this study.

Y.S., K.J.B. and L.M. designed the study; F.L., H.L., K.K., B.S., T.S. performed the SGA experiments; J.H. and F.L. performed screening experiments; F.L., Y.F., H.S. and K.J.B. analysed the sequencing data; Y.R., J.M. and B.R. contributed to cohort design and sample collection; D.M. and P.L performed population study and behaviour analysis; F.L., K.J.B., D.M. and Y.S. contributed to drafting and editing the article; all authors reviewed the manuscript.

This work was supported by the National Natural Science Foundation of China [81361120407], State Key Laboratory for Infectious Disease Prevention and Control China, SKLID, National Institutes of Health [R01 DA037244] and the Penn Center for AIDS Research (P30 AI045008); This study is approved by the Committee for Research in Human Subjects National Center for AIDS/STD Control and Prevention, Chinese.

Conflicts of interest

All authors declare that they have no competing conflicts of interests.


1. Houghton M, Abrignani S. Prospects for a vaccine against the hepatitis C virus. Nature 2005; 436:961–966.
2. Pope M, Haase AT. Transmission, acute HIV-1 infection and the quest for strategies to prevent infection. Nat Med 2003; 9:847–852.
3. Abrahams MR, Anderson JA, Giorgi EE, Seoighe C, Mlisana K, Ping LH, et al. Quantitating the multiplicity of infection with human immunodeficiency virus type 1 subtype C reveals a nonpoisson distribution of transmitted variants. J Virol 2009; 83:3556–3567.
4. Haaland RE, Hawkins PA, Salazar-Gonzalez J, Johnson A, Tichacek A, Karita E, et al. Inflammatory genital infections mitigate a severe genetic bottleneck in heterosexual transmission of subtype A and C HIV-1. PLoS Pathog 2009; 5:e1000274.
5. Keele BF, Giorgi EE, Salazar-Gonzalez JF, Decker JM, Pham KT, Salazar MG, et al. Identification and characterization of transmitted and early founder virus envelopes in primary HIV-1 infection. Proc Natl Acad Sci U S A 2008; 105:7552–7557.
6. Li H, Bar KJ, Wang S, Decker JM, Chen Y, Sun C, et al. High multiplicity infection by HIV-1 in men who have sex with men. PLoS Pathog 2010; 6:e1000890.
7. Rolland M, Edlefsen PT, Larsen BB, Tovanabutra S, Sanders-Buell E, Hertz T, et al. Increased HIV-1 vaccine efficacy against viruses with genetic signatures in Env V2. Nature 2012; 490:417–420.
8. Tully DC, Ogilvie CB, Batorsky RE, Bean DJ, Power KA, Ghebremichael M, et al. Differences in the selection bottleneck between modes of sexual transmission influence the genetic composition of the HIV-1 founder virus. PLoS Pathog 2016; 12:e1005619.
9. Bar KJ, Li H, Chamberland A, Tremblay C, Routy JP, Grayson T, et al. Wide variation in the multiplicity of HIV-1 infection among injection drug users. J Virol 2010; 84:6241–6247.
10. Masharsky AE, Dukhovlinova EN, Verevochkin SV, Toussova OV, Skochilov RV, Anderson JA, et al. A substantial transmission bottleneck among newly and recently HIV-1-infected injection drug users in St Petersburg, Russia. J Infect Dis 2010; 201:1697–1702.
11. Sterrett S, Learn GH, Edlefsen PT, Haynes BF, Hahn BH, Shaw GM, et al. Low multiplicity of HIV-1 infection and no vaccine enhancement in VAX003 injection drug users. Open Forum Infect Dis 2014; 1:ofu056.
12. Yu W, Ge M, Lu S, Shi J, Li X, Zhang J, et al. Anti-inflammatory effects of interleukin-35 in acquired aplastic anemia. Cytokine 2015; 76:409–416.
13. Li H, Stoddard MB, Wang S, Blair LM, Giorgi EE, Parrish EH, et al. Elucidation of hepatitis C virus transmission and early diversification by single genome sequencing. PLoS Pathog 2012; 8:e1002880.
14. Bull RA, Luciani F, McElroy K, Gaudieri S, Pham ST, Chopra A, et al. Sequential bottlenecks drive viral evolution in early acute hepatitis C virus infection. PLoS Pathog 2011; 7:e1002243.
15. Wang GP, Sherrill-Mix SA, Chang KM, Quince C, Bushman FD. Hepatitis C virus transmission bottlenecks analyzed by deep sequencing. J Virol 2010; 84:6218–6228.
16. Shen C, Gupta P, Xu X, Sanyal A, Rinaldo C, Seaberg E, et al. Transmission and evolution of hepatitis C virus in HCV seroconverters in HIV infected subjects. Virology 2014; 449:339–349.
17. Herbeck JT, Rolland M, Liu Y, McLaughlin S, McNevin J, Zhao H, et al. Demographic processes affect HIV-1 evolution in primary infection before the onset of selective processes. J Virol 2011; 85:7523–7534.
18. HIV/AIDS JUNPo. Global AIDS update 2016. Geneva: Joint United Nations Programme on HIV/AIDS; 2016.
19. Organization GWH. Global hepatitis report, 2017. Geneva: WHO; 2017.
20. Wiktor SZ, Hutin YJ. The global burden of viral hepatitis: better estimates to guide hepatitis elimination efforts. Lancet 2016; 388:1030–1031.
21. Poles MA, Elliott J, Taing P, Anton PA, Chen IS. A preponderance of CCR5(+) CXCR4(+) mononuclear cells enhances gastrointestinal mucosal susceptibility to human immunodeficiency virus type 1 infection. J Virol 2001; 75:8390–8399.
22. Patel P, Borkowf CB, Brooks JT, Lasry A, Lansky A, Mermin J. Estimating per-act HIV transmission risk: a systematic review. AIDS 2014; 28:1509–1519.
23. Li H, Blair L, Chen Y, Learn G, Pfafferott K, John M, et al. Molecular mechanisms of HIV type 1 prophylaxis failure revealed by single-genome sequencing. J Infect Dis 2013; 208:1598–1603.
24. Jackson JB, Wei L, Liping F, Aramrattana A, Celentano DD, Walshe L, et al. Prevalence and seroincidence of hepatitis B and hepatitis C infection in high risk people who inject drugs in china and Thailand. Hepat Res Treat 2014; 2014:296958.
25. Zhang Y, Shan H, Trizzino J, Ruan Y, Beauchamp G, Masse B, et al. HIV incidence, retention rate, and baseline predictors of HIV incidence and retention in a prospective cohort study of injection drug users in Xinjiang, China. Int J Infect Dis 2007; 11:318–323.
26. Zhang Y, Shan H, Trizzino J, Ruan Y, Beauchamp G, Masse B, et al. Demographic characteristics and risk behaviors associated with HIV positive injecting drug users in Xinjiang, China. J Infect 2007; 54:285–290.
27. Metzger DS, Donnell D, Celentano DD, Jackson JB, Shao Y, Aramrattana A, et al. Expanding substance use treatment options for HIV prevention with buprenorphine-naloxone: HIV Prevention Trials Network 058. J Acquir Immune Defic Syndr 2015; 68:554–561.
28. Fiebig EW, Wright DJ, Rawal BD, Garrett PE, Schumacher RT, Peddada L, et al. Dynamics of HIV viremia and antibody seroconversion in plasma donors: implications for diagnosis and staging of primary HIV infection. AIDS 2003; 17:1871–1879.
29. Kucirka LM, Sarathy H, Govindan P, Wolf JH, Ellison TA, Hart LJ, et al. Risk of window period hepatitis-C infection in high infectious risk donors: systematic review and meta-analysis. Am J Transplant 2011; 11:1188–1200.
30. Hall TA. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl Acids Symp Series 1999; 41:95–98.
31. Deng W, Maust BS, Nickle DC, Learn GH, Liu Y, Heath L, et al. DIVEIN: a web server to analyze phylogenies, sequence divergence, diversity, and informative sites. Biotechniques 2010; 48:405–408.
32. Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. MEGA6: molecular evolutionary genetics analysis version 6.0. Mol Biol Evol 2013; 30:2725–2729.
33. Stamatakis A. RAxML version 8: a tool for phylogenetic analysis and postanalysis of large phylogenies. Bioinformatics 2014; 30:1312–1313.
34. Giorgi EE, Funkhouser B, Athreya G, Perelson AS, Korber BT, Bhattacharya T. Estimating time since infection in early homogeneous HIV-1 samples using a poisson model. BMC Bioinformatics 2010; 11:532.
35. Maljkovic Berry I, Ribeiro R, Kothari M, Athreya G, Daniels M, Lee HY, et al. Unequal evolutionary rates in the human immunodeficiency virus type 1 (HIV-1) pandemic: the evolutionary rate of HIV-1 slows down when the epidemic rate increases. J Virol 2007; 81:10625–10635.
36. Maljkovic Berry I, Athreya G, Kothari M, Daniels M, Bruno WJ, Korber B, et al. The evolutionary rate dynamically tracks changes in HIV-1 epidemics: application of a simple method for optimizing the evolutionary rate in phylogenetic trees with longitudinal data. Epidemics 2009; 1:230–239.
37. Mansky LM, Temin HM. Lower in vivo mutation rate of human immunodeficiency virus type 1 than that predicted from the fidelity of purified reverse transcriptase. J Virol 1995; 69:5087–5094.
38. Ribeiro RM, Li H, Wang S, Stoddard MB, Learn GH, Korber BT, et al. Quantifying the diversification of hepatitis C virus (HCV) during primary infection: estimates of the in vivo mutation rate. PLoS Pathog 2012; 8:e1002881.
39. Salazar-Gonzalez JF, Bailes E, Pham KT, Salazar MG, Guffey MB, Keele BF, et al. Deciphering human immunodeficiency virus type 1 transmission and early envelope diversification by single-genome amplification and sequencing. J Virol 2008; 82:3952–3970.
40. Drummond AJ, Suchard MA, Xie D, Rambaut A. Bayesian phylogenetics with BEAUti and the BEAST 1.7. Mol Biol Evol 2012; 29:1969–1973.
41. Drummond AJ, Ho SY, Phillips MJ, Rambaut A. Relaxed phylogenetics and dating with confidence. PLoS Biol 2006; 4:e88.
42. Posada D. Selection of models of DNA evolution with jModelTest. Methods Mol Biol 2009; 537:93–112.
43. Hasegawa M, Kishino H, Yano T. Dating of the human-ape splitting by a molecular clock of mitochondrial DNA. J Mol Evol 1985; 22:160–174.
44. Drummond AJ, Rambaut A, Shapiro B, Pybus OG. Bayesian coalescent inference of past population dynamics from molecular sequences. Mol Biol Evol 2005; 22:1185–1192.
45. Ma L, Guo Y, Yuan L, Huang Y, Sun J, Qu S, et al. Phenotypic and genotypic characterization of human immunodeficiency virus type 1 CRF07_BC strains circulating in the Xinjiang Province of China. Retrovirology 2009; 6:45.
46. Tee KK, Pybus OG, Li XJ, Han X, Shang H, Kamarulzaman A, et al. Temporal and spatial dynamics of human immunodeficiency virus type 1 circulating recombinant forms 08_BC and 07_BC in Asia. J Virol 2008; 82:9206–9215.
47. Bar KJ, Tsao CY, Iyer SS, Decker JM, Yang Y, Bonsignori M, et al. Early low-titer neutralizing antibodies impede HIV-1 replication and select for virus escape. PLoS Pathog 2012; 8:e1002721.
48. Salazar-Gonzalez JF, Salazar MG, Keele BF, Learn GH, Giorgi EE, Li H, et al. Genetic identity, biological phenotype, and evolutionary pathways of transmitted/founder viruses in acute and early HIV-1 infection. J Exp Med 2009; 206:1273–1289.
49. Sagar M, Lavreys L, Baeten JM, Richardson BA, Mandaliya K, Chohan BH, et al. Infection with multiple human immunodeficiency virus type 1 variants is associated with faster disease progression. J Virol 2003; 77:12921–12926.
50. Janes H, Herbeck JT, Tovanabutra S, Thomas R, Frahm N, Duerr A, et al. HIV-1 infections with multiple founders are associated with higher viral loads than infections with single founders. Nat Med 2015; 21:1139–1141.
51. Muga R, Sanvisens A, Bolao F, Tor J, Santesmases J, Pujol R, et al. Significant reductions of HIV prevalence but not of hepatitis C virus infections in injection drug users from metropolitan Barcelona. Drug Alcohol Depend 2006; 82 (Suppl 1):S29–S33.
52. Abadie R, Welch-Lazoritz M, Gelpi-Acosta C, Reyes JC, Dombrowski K. Understanding differences in HIV/HCV prevalence according to differentiated risk behaviors in a sample of PWID in rural Puerto Rico. Harm Reduct J 2016; 13:10.
53. De Vos AS, Kretzschmar ME. The efficiency of targeted intervention in limiting the spread of HIV and Hepatitis C virus among injecting drug users. J Theor Biol 2013; 333:126–134.
54. Kijak GH, Sanders-Buell E, Chenine AL, Eller MA, Goonetilleke N, Thomas R, et al. Rare HIV-1 transmitted/founder lineages identified by deep viral sequencing contribute to rapid shifts in dominant quasispecies during acute and early infection. PLoS Pathog 2017; 13:e1006510.
55. Lucas GM, Young A, Donnell D, Richardson P, Aramrattana A, Shao Y, et al. Hepatotoxicity in a 52-week randomized trial of short-term versus long-term treatment with buprenorphine/naloxone in HIV-negative injection opioid users in China and Thailand. Drug Alcohol Depend 2014; 142:139–145.
56. Henn MR, Boutwell CL, Charlebois P, Lennon NJ, Power KA, Macalalad AR, et al. Whole genome deep sequencing of HIV-1 reveals the impact of early minor variants upon immune recognition during acute infection. PLoS Pathog 2012; 8:e1002529.
57. Bull RA, Leung P, Gaudieri S, Deshpande P, Cameron B, Walker M, et al. Transmitted/founder viruses rapidly escape from CD8+ T cell responses in acute hepatitis C virus infection. J Virol 2015; 89:5478–5490.
58. Park SY, Love TM, Perelson AS, Mack WJ, Lee HY. Molecular clock of HIV-1 envelope genes under early immune selection. Retrovirology 2016; 13:38.

* Katharine J. Bar and Yiming Shao contributed equally to the writing of this article.


acute infection; hepatitis C virus; HIV-1; people who inject drugs; transmitted/founder virus

Supplemental Digital Content

Copyright © 2018 Wolters Kluwer Health, Inc.