Share this article on:

Characterization of HIV-1 subtypes and viral antiretroviral drug resistance in men who have sex with men in Beijing, China

Zhang, Xiaoyana; Li, Shenweia; Li, Xinpinga; Li, Xinxua; Xu, Jianqinga; Li, Dongliangb; Ruan, Yuhuaa; Xing, Huia; Zhang, Xiaoxib; Shao, Yiminga

doi: 10.1097/01.aids.0000304698.47261.b1
Original papers

Objectives: To characterize the HIV subtypes prevalent among men who have sex with men (MSM) in Beijing and to perform baseline genotypic analysis of anti-HIV drug resistance in this population.

Design: In 2005, half of new HIV-1 infections occurred through unprotected sex in China. MSM have become the second most vulnerable group to HIV infection. HIV-1-prevalent subtypes among this population as well as their genetic and biological characteristics have not been well defined.

Methods: A cohort consisting of 54 HIV-seropositive MSM were recruited with written informed consent. Samples of plasma and whole blood were collected to characterize prevalent HIV-1 subtypes with overlapped polymerase chain reaction followed by sequencing and phylogenic analysis. The genotypes of anti-HIV drug resistance were analysed.

Results: Among the amplified gag sequences, HIV-1 subtype B accounted for 71.1% (32/45), followed by CRF01_AE for 24.4% (11/45) and CRF07_BC for 4.4% (2/45). A similar trend was observed among the amplified env sequences. Six antiretroviral therapy (ART)-naive participants (15%) carried drug-resistant mutations, with intermediate to high-level resistance both to drugs used in China including zidovudine, didanosine, nevirapine, stavudine, and lamivudine, and drugs not used in China such as delavirdine, efavirenz, tenofovir, emtricitabine and abacavir. We also have concerns over nelfinavir and atazanavir regarding their future use in China because low-level resistance was also seen against those drugs.

Conclusion: The HIV-1 strains prevalent among Beijing MSM include complex subtypes derived from recombination. High rates of HIV drug-resistant mutations in ART-naive patients represent a serious challenge for HIV prevention and treatment programmes in China.

From the aState Key Laboratory for Infectious Disease Prevention and Control, National Center for AIDS/STD Control and Prevention, China CDC, Beijing 100050, China

bChaoyang District Center for AIDS/STD Control and Prevention, Beijing 100021, China.

Correspondence to Yiming Shao, National Center for AIDS/STD Control and Prevention, China CDC, Beijing 100050, China. Tel: +86 10 87789398; fax: +86 10 87789386; e-mail:

Back to Top | Article Outline


HIV-1 prevalence is currently on the increase in China through multiple transmission routes. It was estimated in 2005 that 19.6% of HIV cases were infected through commercial heterosexual contacts, and 7.3% were through homosexual contacts among men in 2005 [1]. Men who have sex with men (MSM) may play a bridging role in spreading HIV and other sexually transmitted diseases (STD). Engagement in high-risk behaviours, such as unprotected sex with multiple male or female partners, has rendered this group the second most vulnerable to HIV infection in China, after injection drug users [2]. Studies on HIV prevalence among MSM populations have been carried out in China, and preliminary data have been generated on social behaviours and HIV seroprevalence among MSM [2,3]. Recent studies have revealed a high rate of MSM engaging in unprotected sex with their male partners, a significant proportion of whom also have sex with women (17–41%) or sell sex (17%) [4,5]. According to a survey conducted in metropolitan areas of China, the estimated HIV-1 infection rate among MSM was over 1% in 2004 [1]. In Beijing, 3.0–4.6% of MSM were found to be HIV positive [5,6]. Data from the latter study in 2005 also indicated that one in five men was ignorant of his risk of HIV infection and more than two-thirds of respondents reported engaging in unprotected sex during the previous 6 months [5].

Limited information is available on HIV-1 viral characteristics in the MSM population in China, as previous studies have focused on HIV prevalence and risk behaviours. Only one study of the molecular profile of HIV-1 in MSM has reported on the subtypes through sequence analyses of the C2–V3 region of the env gene from 1998 to 2001 in Beijing [7]. HIV-1 exhibited extreme genetic diversity and at least 16 recombinant subtypes were identified through evolutional change that can result in more pathogenic and virulent HIV-1 strains. Diverse subtypes, altered viral fitness, tropism, and multiple drug resistance may hamper the development of subtype-based vaccines [8]. Full characterization of HIV-1 and continuous surveillance of high-risk populations are needed in China, including MSM in Beijing.

Antiretroviral treatment (ART) has been made available free of charge to AIDS patients since 2003. This has had the effect of greatly reducing patients' viral load, facilitating the reconstitution of their immune system and improving the quality of life in individuals who are identified and get access to ART [9]. As the free ART programme has progressed, an increase in drug resistance has been observed among AIDS patients in China. Some Chinese patients have been confirmed to have been infected by drug-resistant HIV-1 [10,11]. In western countries including the United States, the average level of primary HIV drug resistance has been reported at approximately 10–26% [12,13]. Another study on 195 MSM (all infected with HIV subtype B) from six cities in the United States showed that 15.9% of participants showed primary HIV drug resistance during 1999–2003 [14]. In 2005, our group initiated a comprehensive study of HIV-1 characteristics among MSM in Beijing, China.

Back to Top | Article Outline


Study subjects

An initial cross-sectional study on MSM had been conducted in 2005 and identified 17 from 526 MSM (3.2%) as HIV-1 seropositive [15]. An additional 37 HIV-1-seropositive MSM whose HIV infection status was identified by other studies conducted in Beijing during 2002–2006 were recruited in January/February 2006 from the HIV-seropositive MSM management programme in Chaoyang CDC, Beijing. These 54 HIV-infected MSM were enrolled to permit a panel of laboratory assays on their HIV-1 properties to be examined. The study was approved by the Committee for Human Research in the National Center for AIDS/STD Control and Prevention, China CDC.

All 57 study subjects were self-reported to be treatment-naive. Potential participants were referred through the internet or informal social networks, or were approached by peer recruiters in MSM-identified bars, clubs, parks and bathhouses. They were screened for eligibility (age 18 years or older and Beijing residents). Eligible men provided written informed consent and completed a 30-min questionnaire.

Back to Top | Article Outline

Sample collection and quantification of CD4 T cells

Samples of whole blood were collected in sterile ethylenediammine tetraacetic acid-treated tubes and transferred to the laboratory no more than 6 h after collection. Fresh whole blood samples were used for CD4/CD8 T-cell counting; and sera were stored at −20°C for HIV confirmation testing. The plasma samples were stored at −80°C for viral RNA extraction and viral load estimation. HIV infection was confirmed in all subjects by Western blot (HIV blot 2.2; Genelab Technologies, Inc., Singapore), according to the manufacturer's instructions. CD4 T lymphocytes in whole blood were quantitated by flow cytometry using reagents and equipment provided by Becton Dickinson Biosciences (San Jose, California, USA).

Back to Top | Article Outline

Viral RNA purification and quantitation

Viral RNA was isolated from 200 μl ethylenediammine tetraacetic acid anticoagulated plasma using a QIAamp viral RNA kit (Qiagen Inc., Germany) and treated with DNase I on a spin column (DNase 1 set; Qiagen), according to the manufacturer's instructions. RNA was eluted in 60 μl AVE buffer and analysed immediately.

Real-time reverse transcriptase polymerase chain reaction (RT–PCR) was used to determine plasma viral load by the amplification of a 102 base pair (bp) segment of the HIV-1 gag gene. The reaction composition and cycling conditions followed the manufacturer's instructions (real-time RT–PCR kit; Shenzhen PG Biotech Co., China). Reactions were performed in triplicate using a Cobas Amplicor analyser (Roche Diagnostics, Basel, Switzerland). The lower limit of detection was 500 copies/ml.

Back to Top | Article Outline

HIV-1 complimentary DNA purification and subtype determination

HIV-1 cDNA was obtained by RT–PCR using the TaKaRa One Step RNA PCR kit (TaKaRa Biotechnology Co. Ltd., Dalian, China), and then subjected to nested multiple polymerase chain reaction (PCR) for the amplification of env, gag and pol genes with corresponding outer primer pairs and inner primer pairs as listed in Table 1. Reaction conditions are described elsewhere [16,17]. PCR products were separated by agarose gel electrophoresis, and purified using a Qiagen gel extraction kit (Qiagen). The primers used for PCR direct sequencing included a forward primer (Env7) and a reverse primer (Env8) for C2∼V5 of the env region, and primers 306 and c-gag for a portion of the p24 and p17 gag region. Overlapping primers were used for the pol region (1315 bp, encoding the protease gene and part of the reverse transcriptase gene). DNA sequencing was performed using fluorescent dye terminators (Prism BigDye terminator cycle sequencing ready reaction kit; Applied Biosystems, Foster City, California, USA) and an automated DNA sequencer (model 377; Applied Biosystems) [16,17]. The sequences were then aligned with reference sequences of HIV-1 strains of various subtypes from the Los Alamos HIV-1 database ( Multiple alignments were done automatically by CLUSTAL X12 with minor manual adjustments. Phylogenetic analysis of the aligned sequences was performed by the neighbour-joining method. The statistical robustness of the neighbour-joining tree and reliability of the branching patterns were confirmed by bootstrapping (1000 replicates). The trees were midpoint rooted.

Table 1

Table 1

Back to Top | Article Outline

Genotypic baseline of HIV-1 drug resistance

Sequences for the pol gene were compared with the consensus sequence using HIV db software (Stanford HIV Drug Resistance Database,, version 4.2.5) to detect drug-resistance mutations including PI major and minor resistance mutations, nucleoside reverse transcriptase inhibitor (NRTI) and non-nucleoside reverse transcriptase inhibitor (NNRTI) resistance mutations. Susceptibility to drugs used in China (zidovudine, didanosine, nevirapine, stavudine, and lamivudine), as well as other drugs available worldwide (Table 2), was estimated. Genotypic susceptibility was classified into two groups: (i) the resistant group, including high-level resistance, intermediate-level resistance and low-level resistance; and (ii) the sensitive group, including sensitive and potential resistance.

Table 2

Table 2

Back to Top | Article Outline


Cohort of HIV-seropositive men who have sex with men

The average age of HIV-infected MSM was 31 years (range 19–54). The median number of lifetime sexual partners was 27 (range 1–500), and the median was two partners in the past 6 months. Seventeen (31.5%) had engaged in sex with both men and women. Three individuals were ART experienced (M50, M53, and M54). Their CD4 T-cell counts and viral loads varied widely, from 78 to 1619 cells/μl for CD4 T cells, and from 2.58 × 103 to 7.01 × 106 copies/ml for viral loads.

Back to Top | Article Outline

Prevalent HIV-1 subtypes among men who have sex with men

Forty-five gag sequences (83.3%), 29 env sequences (53.7%), and 40 pol sequences (74.1%) were successfully amplified. Four samples failed to amplify any portion of the three genes. Phylogenetic analysis of 45 gag (671 bp, encoding a portion of p24 and p17), 29 env (668 bp of the C2∼V5 region), and 40 pol (1315 bp, encoding the protease gene and part of the reverse transcriptase gene region) genes were performed. The representative data for env are shown in Fig. 1. A significant proportion (18/54) in this cohort with undetectable viral loads accounted for the low rate of successfully amplified fragments.

Fig. 1

Fig. 1

Among the amplified gag sequences, HIV-1 subtype B accounted for 71.1% (32/45), CRF01_AE for 24.4% (11/45), and CRF07_BC for 4.4% (2/45). A similar trend was observed among the amplified env sequences. The majority of the amplified env sequences were also HIV-1 subtype B (25/29, 86.2%). Two sequences of CRF07_BC (6.9%) were found in non-B samples, one CRF01_AE and one Thailand B were found (3.4%). For pol, only two subtypes were identified among the amplified sequences; 92.5% was HIV-1 subtype B (37/40), and 7.5% was CRF01_AE (3/40). At least over 70% of samples were sorted into subtype B, and two were subtype CRF01_AE (3.7%). One new subtype, CRF15_01B (MSM07), was identified. The remaining sequences were too complicated to be determined and additional sequences will be required for defining subtypes.

Back to Top | Article Outline

Genotypic analysis of HIV-1 drug resistance

Pol genes were successfully amplified from 40 HIV-positive samples and were subjected to drug-resistance analysis including PI resistance mutations, NRTI and NNRTI resistance mutations through the Stanford HIV Drug Resistance Database. Fifteen out of 40 pol sequences showed no drug resistance-related mutations. The remaining 25 sequences contained one or more mutations (Table 2). Overall, 33 out of 40 sequences (82.5%) displayed sensitive genotypes to all classes of antiretroviral drugs. Viruses of six individuals (15%), including MSM20, MSM23, MSM25, M43, MSM44, and MSM49, showed intermediate to high-level resistance to zidovudine (1), didanosine (2), nevirapine (5), stavudine (1), and lamivudine (4). Five of these six subjects showed mutations resistant to both NRTI and NNRTI reverse transcriptase inhibitors, the exception being the virus of MSM20 that developed resistance to NRTI only (Table 2). Among those mutations, K103R and Y181C (that confer resistance to NNRTI) and M184V and K65R (that cause resistance to NRTI) were observed as key mutations. The V118I mutation, which does not confer any drug resistance to NRTI by itself, appeared in 19 sequences; the biological significance of this mutation remains to be determined.

We also identified several minor or low level resistant mutations among HIV-positive MSM. One subject (MSM07) with a M46L mutation showed low-level resistance to both atazanavir and nelfinavir, and potential low-level resistance to fosamprenavir, indinavir, and lopinavir. MSM43 developed two mutations (V118I and K238N), which may confer intermediate resistance to delavirdine and nevirapine, and potential low-level resistance to efavirenz.

Back to Top | Article Outline


It has been estimated that there are 17.82 million MSM in China [18]. As a result of strong cultural taboos surrounding MSM, many have married women, presumably to avoid social pressure. The majority of spouses are, however, not aware of their husbands' sexual preference. In our study, 31.5% of HIV-positive MSM engaged in sex with both men and women within and out of marriage. This behaviour facilitates the transmission of HIV and other STD from high-risk populations to the general population, and may be contributing to the growing number of HIV-positive women infected through unprotected sex.

HIV-1 molecular epidemic surveillance carried out during 2001–2002 revealed that there were at least seven major HIV-1 subtypes circulating in China [19]. The most prevalent subtypes were B/C recombinant (∼50%; mainly found in injecting drug users), followed by Thailand B (29.1%; former plasma/blood donors), and CRF01_AE (15.5%; heterosexual transmission). Only 2.6% of subtype B was identified and was concentrated in MSM.

Similar to a previous report from 2002 [7], we observed that the majority (57.4%) of the HIV isolates among MSM belonged to subtype B. In addition, CRF01_AE (3.7%) was also identified, which is usually transmitted through heterosexual activities. Furthermore, one new subtype CRF15_01B (MSM07) was identified in this cohort. From the remaining HIV-1-infected samples, recombinants of subtype B and CRF01_AE or CRF07_BC, as well as other minor subtypes were observed in env, gag and pol gene sequences amplified from one patient.

The HIV-infected MSM with subtype CRF01_AE identified in this study were confirmed as homosexual but not bisexual. There was no statistically significant difference observed in HIV subtypes between homosexual and bisexual MSM in this study, although our numbers would not be adequate to assess any other than large differences. An intensive ‘case investigation on HIV-positive MSM’ including their sexual partner network needs to be performed to clarify further the influence of sexual preference on HIV-1 subtypes in the future.

The phenomenon of HIV-1 diversity observed in Beijing MSM was similar to other studies. Although the majority of HIV-1 subtypes among MSM worldwide were subtype B [20–22], other subtypes such as subtype D in South Africa [20], F in Colombia [23], and CRF12_BF and a recombinant of B, C, and F in Argentina [24] were also identified. These mixtures of HIV-1 subtypes suggest the presence of either co-infection or superinfection of multiple HIV-1 strains among MSM. They may also be indicating that new recombinant subtypes are prevalent among this MSM cohort. In either scenario, the presence of these subtypes strongly indicates continued interaction between MSM and other high-risk groups.

In addition to the complicated subtypes prevalent in MSM, there was a high prevalence of HIV drug-resistant mutations, which represents a challenging situation for HIV prevention and control in China. In a study conducted in the United States with 1099 AIDS patients, HIV drug resistance to at least one drug was 76%, to NRTI, PI, and NNRTI resistance was observed in 71, 41 and 25% of samples, respectively. HIV drug resistance to two and three classes of drug was 48 and 13%, respectively [12]. The overall drug resistance was 60%, and primary HIV drug resistance was at least 10%. In 2005, high HIV drug-resistant mutations (20.8%) were reported in patients (n = 124) treated with ART in Henan province, central China [25]. In addition, the primary HIV drug resistance was reported as 0.6% to PI, 5.8% to NRTI, and 1.5% to NNRTI in 2002 in a nationwide investigation [10], and 4.4% in Liaoning province among ART-naive AIDS patients [11]. In this study, six self-reported ART-naive participants (15%) carried drug-resistant mutations, a much higher proportion than noted from earlier studies on non-MSM populations conducted in China. It is worth noting that all six patients were infected by subtype B HIV-1. They carried intermediate to high-level resistance both to the drugs used in China including zidovudine, didanosine, nevirapine, stavudine, and lamivudine, and the drugs not used in China such as delavirdine, efavirena, tenofovir, emtricitabine and abacavir. We also have concerns about nelfinavir and atazanavir regarding their future use in China because low-level resistance was also seen for these drugs. In China, the B subtype HIV-1 strains are mainly found in the MSM population, especially those men who live in big cities where social and sexual contacts with westerners are more frequent than other regions [19]. In addition, our MSM population was better educated than other at-risk populations in China, and has many opportunities to interact with the international MSM community. Therefore, it is possible they have been primarily infected with drug-resistant subtype B viruses derived from western countries with more diverse and prolonged ART experience. It was also possible that some participants misreported their ART status as naive rather than ‘not currently taking’, because ART use was self-reported. If this were true, it could lead to an overestimation of HIV primary drug resistance. In light of this, our data may only represent primary drug resistance in the MSM population and therefore may not be generalizable to other high-risk groups. Nevertheless, these data warn us that HIV drug-resistant isolates may be prevalent among MSM, which will no doubt compromise the success of future ART in these patients and the women to whom they pass these drug-resistant viruses.

In conclusion, the HIV-1 strains prevalent among this MSM cohort have evolved to complicated HIV-1 subtypes, partly because of sexual HIV transmission within the MSM population and across to heterosexual individuals. High rates of HIV drug-resistant mutations in ART-naive patients in the study represent a serious challenge for the future HIV treatment programme in China.

Back to Top | Article Outline


The authors would like to thank Dr Feng Gao from Duke University for his critical comments on this manuscript.

The first two authors contributed equally to this paper.

Sponsorship: This study was supported by grants from the Ministry of Science and Technology of China (2004BA719A01), and the National Institute of Health (CIPRA, U19AI51915).

Conflicts of interest: None.

Back to Top | Article Outline


1. The Ministry of Health of the People's Republic of China, Joint United Nations Programme on HIV/AIDS and World Health Organization. 2004 and 2005 Update on the HIV/AIDS epidemic and response in China [in Chinese]. Available at: Accessed: November 2007.
2. Zhang BC, Chu QS. MSM and HIV/AIDS in China. Cell Res 2005; 15:858–864.
3. Choi KH, Liu H, Guo Y, Han L, Mandel JS, Rutherford GW, et al. Emerging HIV-1 epidemic in China in men who have sex with men. Lancet 2003; 361:2125–2126.
4. Jiang J, Cao N, Zhang J, Xia Q, Gong X, Xue H, et al. High prevalence of sexually transmitted diseases among men who have sex with men in Jiangsu Province, China. SexTransm Dis 2006; 33:118–123.
5. UNAIDS/WHO. AIDS epidemic update: December 2006. Available at: Accessed: November 2007.
6. Choi KH, Lui H, Guo Y, Han L, Mandel JS. Lack of HIV testing and awareness of HIV infection among men who have sex with men, Beijing, China. AIDS Educ Prev 2006; 18:33–43.
7. Yao J, Zhang FJ, He ZP, Zhao HX, Li XW, Feng X, et al. Subtype and sequence analysis of the C2-V3 region of env gene among HIV-1 infected homosexual men in Beijing. Chin J STD/AIDS Prev Cont 2002; 8:131–133.
8. Steain MC, Wang B, Dwyer DE, Saksena NK. HIV-1 co-infection, superinfection and recombination. Sex Health 2004; 1:239–250.
9. Ye S, Feng YL, Ke YF, Yu FC. Analysis of results in antiviral treatment of 30 AIDS patients. Chin Trop Med 2006; 6:1143–1144.
10. Si XF, Huang HL, Wei M, Guan Q, Song YH, Ma P, et al. Prevalence of drug resistance mutations among antiretroviral drug-naive HIV-1-infected patients in China. Chin J Exp Clin Virol 2004; 4:308–311.
11. Han XX, Zhang M, Dai T, Lu CM, Zhang ZN, Liu J, et al. Background study of HIV-1 drug resistant mutations in treatment-naïve patients in Liaoning province. Acta Acad Med Sin 2006; 5:632–635.
12. Richman DD, Morton SC, Wrin T, Hellmann N, Berry S, Shapiro MF, et al. The prevalence of antiretroviral drug resistance in the United States. AIDS 2004; 18:1393–1401.
13. Smith AJ, Wang H, Bennett D, Teshale E, Buskin S, Morse A, et al. Prevalence of mutations associated with antiretroviral drug resistance in a cohort of treated individuals in four US cities [Abstract 101]. Antivir Ther 2004; 9:S112.
14. Eshleman SH, Husnik M, Hudelson S, Donnell D, Huang Y, Huang W, et al. Antiretroviral drug resistance, HIV-1 tropism, and HIV-1 subtype among men who have sex with men with recent HIV-1 infection. AIDS 2007; 21(9):1165–1174.
15. Zhang XY, Wang C, Li XX, Zhang XX, Song YH, Li DL, et al. Study of HIV infection among High Risk populations of Men who have sex with men in Beijing. Chin J AIDS STD 2006; 12:294–296.
16. Wei M, Guan Q, Liang H, Chen J, Chen Z, Hei F, et al. Simple subtyping assay for human immunodeficiency virus type 1 subtypes B, C,CRF01-AE, CRF07-BC, and CRF08-BC. J Clin Microbiol 2004; 42:4261–4267.
17. Guan Q, Wei M, Huang HL, Xing H, Hong KX, Ma PF, et al. Sequence analysis of gag-pol gene of HIV-1 B/C recombinant viruses in China. Zhonghua Yi Xue Za Zhi 2004; 84:387–391.
18. Zhang BC, Li XF, Shi TX, Cao N, Hu T. Survey on the high risk behaviors and other AIDS/STI related factors among men who have sex with men (MSM) in mainland China (2001). Chin J Dermatol 2002; 35:214–216.
19. Shao Y: Report on the 2nd National HIV Molecular Epidemiology Survey. Basic science update by Chinese Ministry of Science and Technology [in Chinese]. 2004. Available at: Accessed: November 2007.
20. Van Harmelen J, Wood R, Lambrick M, Rybicki EP, Williamson AL, Williamson C. An association between HIV-1 subtypes and mode of transmission in Cape Town, South Africa. AIDS 1997; 11:81–87.
21. Babi c DZ, Poljak M, Seme K, Tomazic J, Vidmar L. Molecular epidemiology of HIV-1 subtypes based on analysis of pol sequences in Slovenia, 1996–2005. J Med Virol 2006; 78:997–1002.
22. Alaeus A. Significance of HIV-1 genetic subtypes. Scand J Infect Dis 2000; 32:455–463.
23. Eyzaguirre L, Bautista CT, Ayala C, Acosta J, Negrete M, Sateren WB, et al. First case of HIV type 1 subtype F among men who have sex with men in Colombia. AIDS Res Hum Retroviruses 2006; 22:808–811.
24. Pando MA, Eyzaguirre LM, Segura M, Bautista CT, Marone R, Ceballos A, et al. First report of an HIV-1 triple recombinant of subtypes B, C and F in Buenos Aires, Argentina. Retrovirology 2006; 3:59.
25. Wang Z, Liang Y, Xin T. Investigation on curative effect and drug resistance of HIV/AIDS patients receiving HAART. J Med Forum 2005; 26:1–3.

drug resistance; genotype; HIV-1; homosexual; men who have sex with men; subtype

© 2007 Lippincott Williams & Wilkins, Inc.