Share this article on:

Molecular Epidemiological Study of HIV-1 CRF01_AE Transmission in Hong Kong

Chen, J H K PhD*; Wong, K H MD; Li, P MD; Chan, K C MD; Lee, M P MD; Lam, H Y BSc*; Cheng, V C C MD*; Yuen, K Y MD*; Yam, W C PhD*

JAIDS Journal of Acquired Immune Deficiency Syndromes: August 2009 - Volume 51 - Issue 5 - p 530-535
doi: 10.1097/QAI.0b013e3181aac516
Basic Science

Objectives: The objective of this study was to investigate the transmission history of the HIV-1 CRF01_AE epidemics in Hong Kong between 1994 and 2007.

Methods: A total of 465 HIV-1 CRF01_AE pol sequences were derived from an in-house or a commercial HIV-1 genotyping system. Phylogenies of CRF01_AE sequences were analyzed by the Bayesian coalescent method.

Results: CRF01_AE patient population included 363 males (78.1%) and 102 females (21.9%), whereas 65% (314 of 465) were local Chinese. Major transmission routes were heterosexual contact (63%), followed by intravenous drug use (IDU) (19%) and men having sex with men (MSM) (17%). From phylogenetic analysis, local CRF01_AE strains were from multiple origins with 3 separate transmission clusters identified. Cluster 1 consisted mainly of Chinese male IDUs and heterosexuals. Clusters 2 and 3 included mainly local Chinese MSM and non-Chinese Asian IDUs, respectively. Chinese reference isolates available from China (Fujian, Guangxi, or Liaoning) were clonally related to our transmission clusters, demonstrating the epidemiological linkage of CRF01_AE infections between Hong Kong and China. The 3 individual local transmission clusters were estimated to have initiated since late 1980s and late 1990s, causing subsequent epidemics in the early 2000s.

Conclusions: This is the first comprehensive molecular epidemiological study of HIV-1 CRF01_AE in Hong Kong. It revealed that MSM contact is becoming a major route of local CRF01_AE transmission in Hong Kong. Epidemiological linkage of CRF01_AE between Hong Kong and China observed in this study indicates the importance of regular molecular epidemiological surveillance for the HIV-1 epidemic in our region.

From the *Department of Microbiology, Queen Mary Hospital, University of Hong Kong, Hong Kong, China; †Integrated Treatment Centre, Department of Health, Special Preventive Programme, Centre of Health Protection, Hong Kong, China; and ‡Department of Medicine, Queen Elizabeth Hospital, Hong Kong, China.

Received for publication December 14, 2008; accepted April 17, 2009.

Financial support: AIDS Trust Fund, Hong Kong Special Administrative Region. Supported by the AIDS Trust Fund of Hong Kong Special Administrative Region Government (Grant MSS-152R, 155R, and 164R).

The opinions and assertions contained in this study are personal views of the authors and do not necessarily reflect those of the Centre for Health Protection, Hong Kong Department of Health.

Correspondence to: Dr. W. C. Yam, PhD, Department of Microbiology, Queen Mary Hospital, University of Hong Kong, Hong Kong, China (e-mail:

Back to Top | Article Outline


HIV-1 infections spread rapidly throughout the world in the past 20 years. According to the UNAIDS/World Health Organization report in 2007, an estimated 5 million people in Asia are living with HIV.1 CRF01_AE is one of the major non-B genotypes that is circulating among the southeast Asian countries. This HIV-1 recombinant form was first identified in Thailand2,3 in 1990 and later spread to the neighboring countries including China, Vietnam, Cambodia, Myanmar, and India, which caused different non-B HIV-1 epidemics.1,4-7

The CRF01_AE pandemic in China was first detected among the intravenous drug user (IDU) population in Yunnan Province in 1989.8 Explosive numbers of infections were later reported in provinces of southern China including Fujian, Guangdong, Guangxi, Henan, and Sichuan.9 In recent years, the number of HIV-1 CRF01_AE infections increased rapidly among individuals with sexually transmitted diseases, especially those living in southern China.4,10,11 The low level of knowledge on HIV and reproductive health, reluctance of clients of commercial sex workers to use condoms, and the illegal status of sex work in some provinces in southern China fueled the rapid spread of HIV-1 infection.1

The Hong Kong Special Administrative Region is located in the southern part of China. HIV infection was first detected in Hong Kong in 1984, and since then, over 3600 people have been infected with HIV-1.12 From the 2007 surveillance report prepared by the Department of Health, Hong Kong Government (, the cumulative number of HIV infection in 2007 was 4 times of that in 1997 and is on a rise every year. In previous studies, CRF01_AE was found to be one of the predominant genotypes, which accounted for nearly 50% of the HIV-1 infections in Hong Kong.13-15 HIV-1 infections in Hong Kong were reported to be of multiple origin with no significant secondary spread in the locality before 2002.15 The exponential increase in HIV-1 infections among local Chinese male adults started from the early 1990s, and heterosexual contact was the major risk group during that period.15 At the turn of the century, the increase in infections among local men having sex with men (MSM) complicated the HIV-1 transmission pattern.16

In the 10 years after sovereignty transfer of Hong Kong from the United Kingdom to the People's of Republic China, there has been an increased population movement between Mainland China and Hong Kong, and thus, the epidemiology situation of HIV-1 infection in Hong Kong may have changed. Molecular epidemiological investigation on the spread and transmission of HIV-1 subtype B in Hong Kong was reported in our recent investigation.17 Similar study also revealed the demographic history and divergence date of the HIV-1 subtype B causing wide epidemic in Asia.18 As CRF01_AE is another predominant HIV-1 genotype circulating in Asia, we aim to extend our study to assess the impact of CRF01_AE virus on the HIV epidemic in Hong Kong and reconstruct the transmission history.

Back to Top | Article Outline


Study Population

A total of 465 HIV-1 CRF01_AE pol gene sequences were generated from plasma samples collected from the Integrated Treatment Centre of the Department of Health and the Queen Elizabeth Hospital of Hong Kong. These 2 clinics serve over 95% of patients under HIV care in Hong Kong. The samples were submitted for routine genotyping resistance testing between 1996 and 2007, which included samples from treatment-naive, treatment-responsive, and treatment failure patients.

Back to Top | Article Outline

Protease and Reverse Transcriptase Nucleotide Sequence Determination

The pol gene sequences including the entire protease and the first 289 codons of the reverse transcriptase (total 1166 nucleotides in length) were generated by the ViroSeq HIV-1 Genotyping System Version 2.0 (Celera Diagnostics, Alameda, CA) or a previously published in-house genotyping method.13 The sequences were then analyzed by the ABI PRISM 3700 or a newly installed PRISM 3130 Genetic Analyzer (Applied Biosystems, Foster City, CA). The working protocol for PRISM 3130 was slightly modified by loading only 1/4 of the cycle sequencing products for detection by comparison with the PRISM 3700. Sequences generated by the ABI Prism 3700 and Prism 3130 were evaluated to give perfectly matched results.

Back to Top | Article Outline

Phylogenetic Reconstruction

The genotype of the sequences was further confirmed by the REGA HIV-1 Genotyping Tool, version 2.19 The sequences were then aligned with another 122 CRF01_AE pol sequences isolated from different geographical regions (78 from Thailand, 28 from China, 8 from Japan, 5 from United States, 2 from Central African Republic, and 1 from Hong Kong) and the HXB2 subtype B pol sequences that were available in the Los Alamos HIV database ( by using Clustal X, version 1.83.20 The aligned sequences were later manually edited to exclude the alignment gaps. A neighbor-joining (NJ) phylogenetic tree with 1000 bootstrap replicates rooted with subtype B HXB2 (K03455) was constructed using PAUP*, version 4.0b10, under the GTR model of nucleotide substitution,21 with gamma-distributed rate variation (Γ) and a proportion of invariable sites (I), as selected by MrModeltest, version 2.2.22 From the bootstrapped NJ tree, the bootstrap replicate of nodes was used as phylogenetic support for each transmission cluster group. Transmission history of clusters with size over 20 sequences and bootstrap value >70 was further analyzed with the Bayesian coalescent method.

Back to Top | Article Outline

Estimation of the Transmission History of CRF01_AE in Hong Kong

After identifying the CRF01_AE transmission clusters in Hong Kong, the dated phylogenies were estimated using the Bayesian Markov chain Monte Carlo (MCMC) inference in BEAST version The date used for each sequence in the analysis was the number of months since the earliest sequence isolation date. Analyses were performed using the GTR + I + Γ substitution model under the relaxed lognormal molecular clock. For each cluster, 2 separate MCMC chains were run for 20-50 million generations for each cluster, with a 2-5 million burn-in. The posterior distribution, previously estimated from an independent data set of 106 subtype B pol sequences sampled between 1983 and 2000 in Amsterdam,24 was subsequently used as an empirical prior distribution in the coalescent analyses. All parameters were estimated from an effective sampling size >200. The Bayesian MCMC results were analyzed by the program TRACER, version 1.4 (

The epidemiological profile of these transmission clusters was traced back from the Integrated Treatment Centre, Department of Health, and Queen Elizabeth Hospital. The information retrieved included sex, age, ethnic group, transmission route, viral load, and CD4+ cell count. By combining the epidemiological information and the dated phylogenies, the CRF01_AE virus transmission history in Hong Kong could be estimated.

Back to Top | Article Outline


Demographic Characteristics of CRF01_AE-Infected Individuals

The 465 CRF01_AE-infected individuals recruited locally between 1994 and 2007 had median age of 36 years, and the median viral load and CD4+ cell count were 43 × 103 copies per milliliter and 209 cells per cubic millimeter, respectively. Among them, 78.1% (363 of 465) were males and 21.9% (102 of 465) were females. Chinese accounted for 67.5% (314 of 465) of the individuals and was the dominant ethnic group among the HIV CRF01_AE-infected population in Hong Kong. The non-Chinese Asian and non-Chinese non-Asian accounted for 27.3% (127 of 465) and 5.2% (24 of 465) of CRF01_AE-infected population, respectively. As many as 63.0% (293 of 465) of the HIV-1-infected cases acquired the virus through heterosexual transmission, whereas 19% (88 of 465) of cases were IDUs, 17% (79 of 465) were MSM, 0.6% (3 of 465) were bisexuals, 0.2% (1 of 465) had mother-to-child transmission, and 0.2% (1 of 465) of cases had undetermined cause of transmission. The epidemiological characteristics of the 465 HIV-1-infected individuals are summarized in Table 1.



Back to Top | Article Outline

Molecular Epidemiology of the HIV-1 Infection

Phylogenetic analysis using NJ method constructed from 465 Hong Kong and 122 worldwide CRF01_AE pol sequences showed that local CRF01_AE sequences were mainly polyphyletic with multiple origins. In the phylogenetic tree, distinct phylogenetic relationships could only be demonstrated among 144 Hong Kong sequences, which were identified into 3 individual local CRF01_AE transmission clusters (clusters 1, 2, and 3), each with >20 members and >70 bootstrap value (Fig. 1). The other 321 Hong Kong sequences demonstrated a weak phylogenetic association (<70 bootstrap value and <20 members) in the phylogenetic analysis.



Among the 3 transmission clusters, transmission cluster 1 had 28 members, including 36 local sequences isolated from 2000 to 2007 and 2 reference sequences (EF036529 and EF122535) isolated from 2 IDUs residing in Fujian and Liaoning, China, in 2004 and 2005 (Fig. 2A).25,26 All the 36 local sequences were isolated from local Chinese with median age 42 years, of which 69.4% (25 of 36) were males and 30.6% (11 of 36) were females. Heterosexual (63.9%, 23 of 36) and IDU (36.1%, 13 of 36) accounted for the major transmission routes for this cluster. For transmission cluster 2, 23 sequences were isolated from 22 local Chinese males (95.7%) and 1 non-Chinese Asian female (4.3%) with median age of 30 years and were found to cluster with 5 reference sequences (EF122532, EF122534, EF122536, EF122537, and EF122541) isolated from patients with HIV-1 in Liaoning, China, in 2003 and 2004 (Fig. 2B).25,27 Among the 23 patients, 19 local Chinese males belonged to the MSM risk group (82.7%), whereas the other 4 Chinese males were infected through heterosexual (8.7%, 2 of 23) and bisexual (4.3%, 1 of 23) contacts. Available information indicated that the non-Chinese Asian female was a Thai woman infected through heterosexual contact (4.3%, 1 of 23). Transmission cluster 3 included 85 sequences collected between 2001 and 2007 and 2 reference sequences from HIV-1-infected IDUs in Pingxiang and Nanning of Guangxi, China, in 1997 (AY008714 and AY008718)28 (Fig. 2C). Among the 85 local isolates, 76.5% (65 of 85) were collected from Vietnamese male IDUs. Another 21.1% (18 of 85) were collected from 9 Vietnamese males, 5 Vietnamese females, 1 Filippino female, and 3 Chinese males who were infected through heterosexual contacts. The other 2.4% (2 of 85) included 1 Chinese and 1 white MSM.



Back to Top | Article Outline

Evolutionary Rates and Divergence Dates of the Local Spreading Clusters

Estimates of the substitution rate and the time of the most recent common ancestor (T mrca) were obtained separately for the 3 local spreading clusters using a Bayesian inference as implemented in BEAST version 1.4.7. The mean estimated HIV-1 evolutionary rates for different clusters were 1.2 × 10-4 (cluster 1), 1.4 × 10-4 (cluster 2), and 1.6 × 10-4 (cluster 3) substitutions per site per month. The evolution rate estimates for all the 3 clusters gave similar mean values and overlap high posterior density intervals with the prior interval (1.25 × 10-4 to 2.9 × 10-4 substitutions·site-1·mo-1) defined by using the subtype B pol sequences.24

With these nucleic acid substitution rates, the median estimated T mrca for different clusters were November 1997 (cluster 1), July 1996 (cluster 2), and August 1987 (cluster 3). Because the T mrca estimates for the 3 clusters were almost totally dependent on the prior value used for the substitution rate, very large high posterior density intervals were observed around these estimates (Table 2).



Back to Top | Article Outline


This study demonstrated that local Chinese males represented the major risk population for CRF01_AE transmission in Hong Kong throughout the past 12 years, although the major transmission route was heterosexual contact. In comparison to the epidemiological study performed in 2004, the percentage of CRF01_AE transmission occurring among MSM and IDUs had increased gradually.15 This observation correlated well with the annual report of the local health authority ( Our findings revealed that MSM contact is becoming a major route of HIV-1 transmission in Hong Kong other than heterosexual contact.

Through this molecular epidemiology study, HIV-1 pol region, which was primarily targeted for genotyping resistance test, was also applicable to monitor the transmission pattern among HIV-1 CRF01_AE-infected patients in Hong Kong.13 Although the HIV-1 env gene has been widely used as the conventional genetic region for HIV-1 phylogenetic analysis,29,30 our findings concord with previous studies,30,31 demonstrating the feasibility of using single pol gene for HIV-1 molecular epidemiological studies.

Similar to a previous study, the polyphyletic branching pattern in the phylogenetic tree of this study demonstrated that most of the CRF01_AE cases in Hong Kong caused no large-scale secondary transmission as the subtype B did in previous years.15 These CRF01_AE infections were highly diverse in sources of infection within the past 12 years. They were identified among different Chinese heterosexual couples with weak social interrelationships.

Among the 465 CRF01_AE isolates collected in this study, phylogenetic analysis revealed at least 3 separate HIV-1 CRF01_AE epidemics that were introduced into the IDUs and sexually active local populations in the early 2000s. Results suggest the presence of secondary spread of CRF01_AE in the locality after 2000. The major spreading route of cluster 1 included unprotected heterosexual contact and intravenous drug usage, indicating that local IDUs and heterosexual individuals have close epidemiological link in our locality. Among the 3 clusters, each of them included some clinical reference strains isolated in the provinces of southern China (Fujian and Guangxi) and in Liaoning, which were available in the GenBank recently.25-28 The clustering of local sequences and the Chinese isolated sequences revealed the high clonal relationships between HIV-1 strains circulating in Hong Kong and Mainland China. Despite the absence of evidence for HIV-1 epidemic link between Hong Kong and China before 2002,15 our findings indicate recent HIV-1 transmissions between these 2 regions. This is probably explained by the increase of population flow between 2 regions, a larger sample size included in this study, and more recent Chinese HIV sequences available in the GenBank after 2004.

According to the Bayesian coalescent analysis, the estimated substitution rates of the pol gene of the 3 local CRF01_AE clusters in this study were lower than those determined in previous studies on subtype B and F pol sequences.24,32 This may be accounted by variation of viral genetic evolutionary rate among different subtypes. However, further study will be necessary to clarify on this claim.

For cluster 3, it was probably introduced into a group of Vietnamese IDUs 10 years earlier (late 1980s) than the other 2 transmission clusters (mid-1990s). Due to the large influx of illegal Vietnamese immigrants to Hong Kong in late 1980s, this group of patients could have acquired infections before reaching Hong Kong as HIV surveillance of IDUs did not reveal significant local spread. Our findings correlate well to the demographic history of Hong Kong. The CRF01_AE epidemics among local Chinese MSM and heterosexual population were estimated to be introduced in the 1990s.

In conclusion, our study revealed the history of the CRF01_AE epidemics in Hong Kong within the past 12 years through molecular epidemiological investigation. It also revealed that MSM contact is becoming a major route of local CRF01_AE transmission in Hong Kong other than heterosexual contact. Epidemiological linkage of the HIV-1 strains detected in Hong Kong and Mainland China populations initiated since late 1980s, indicating the importance of regular, molecular, epidemiological surveillance for the monitoring of HIV-1 epidemic in our region.

Back to Top | Article Outline


1. UNAIDS, WHO. AIDS epidemic update 2007. Available at: Accessed November 19, 2007.
2. McCutchan FE, Hegerich PA, Brennan TP, et al. Genetic variants of HIV-1 in Thailand. AIDS Res Hum Retroviruses. 1992;8:1887-1895.
3. Ou CY, Takebe Y, Luo CC, et al. Wide distribution of two subtypes of HIV-1 in Thailand. AIDS Res Hum Retroviruses. 1992;8:1471-1472.
4. Bao L, Vidal N, Fang H, et al. Molecular tracing of sexual HIV type 1 transmission in the southwest border of China. AIDS Res Hum Retroviruses. 2008;24:733-742.
5. Motomura K, Kusagawa S, Lwin HH, et al. Different subtype distributions in two cities in Myanmar: evidence for independent clusters of HIV-1 transmission. AIDS. 2003;17:633-636.
6. Ruxrungtham K, Brown T, Phanuphak P. HIV/AIDS in Asia. Lancet. 2004;364:69-82.
7. Tran TT, Maljkovic I, Swartling S, et al. HIV-1 CRF01_AE in intravenous drug users in Hanoi, Vietnam. AIDS Res Hum Retroviruses. 2004;20:341-345.
8. Zheng X, Tian C, Choi KH, et al. Injecting drug use and HIV infection in southwest China. AIDS. 1994;8:1141-1147.
9. China MoH. 2005 Update on the HIV/AIDS Epidemic and Response in China. Beijing, China: Ministry of Health China, UNAIDS, WHO; 2006.
10. Zhang Y, Lu L, Ba L, et al. Dominance of HIV-1 subtype CRF01_AE in sexually acquired cases leads to a new epidemic in Yunnan province of China. PLoS Med. 2006;3:e443.
11. Laeyendecker O, Zhang GW, Quinn TC, et al. Molecular epidemiology of HIV-1 subtypes in southern China. J Acquir Immune Defic Syndr. 2005;38:356-362.
12. Wong KH, Lee SS, Chan KC. Twenty years of clinical human immunodeficiency virus (HIV) and acquired immunodeficiency syndrome (AIDS) in Hong Kong. Hong Kong Med J. 2006;12:133-140.
13. Chen JH, Wong KH, Chan K, et al. Evaluation of an in-house genotyping resistance test for HIV-1 drug resistance interpretation and genotyping. J Clin Virol. 2007;39:125-131.
14. Yam WC, Chen JH, Wong KH, et al. Clinical utility of genotyping resistance test on determining the mutation patterns in HIV-1 CRF01_AE and subtype B patients receiving antiretroviral therapy in Hong Kong. J Clin Virol. 2006;35:454-457.
15. Lim WL, Xing H, Wong KH, et al. The lack of epidemiological link between the HIV type 1 infections in Hong Kong and Mainland China. AIDS Res Hum Retroviruses. 2004;20:259-262.
16. Lee KCK, Wong KH. et al. Epidemiology of HIV infection in Hong Kong and beyond. In: Lee SS, Wu JCY, Wong KH, eds. HIV Manual 2007. Stanley Ho Centre for Emerging Infectious Diseases, The Chinese University of Hong Kong and Centre for Health Protection, Department of Health, Hong Kong SAR, China; 2007:17-27.
17. Chen JH, Wong KH, Chan KC, et al. Molecular epidemiology and divergence of HIV type 1 protease codon 35 inserted strains among treatment-naive patients in Hong Kong. AIDS Res Hum Retroviruses. 2008;24:537-542.
18. Deng X, Liu H, Shao Y, et al. The epidemic origin and molecular properties of B': a founder strain of the HIV-1 transmission in Asia. AIDS. 2008;22:1851-1858.
19. de Oliveira T, Deforche K, Cassol S, et al. An automated genotyping system for analysis of HIV-1 and other microbial sequences. Bioinformatics. 2005;21:3797-3800.
20. Thompson JD, Gibson TJ, Plewniak F, et al. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 1997;25:4876-4882.
21. Rodriguez F, Oliver JL, Marin A, et al. The general stochastic model of nucleotide substitution. J Theor Biol. 1990;142:485-501.
22. Nylander JA, Ronquist F, Huelsenbeck JP, et al. Bayesian phylogenetic analysis of combined data. Syst Biol. 2004;53:47-67.
23. Drummond AJ, Rambaut A. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evol Biol. 2007;7:214.
24. Hue S, Pillay D, Clewley JP, et al. Genetic analysis reveals the complex structure of HIV-1 transmission within defined risk groups. Proc Natl Acad Sci U S A. 2005;102:4425-4429.
25. Han X, Zhang M, Dai D, et al. Genotypic resistance mutations to antiretroviral drugs in treatment-naive HIV/AIDS patients living in Liaoning Province, China: baseline prevalence and subtype-specific difference. AIDS Res Hum Retroviruses. 2007;23:357-364.
26. Huang HL, Jian Z, Yan PP, et al. Genetic characterization of CRF01_AE full-length human immunodeficiency virus type 1 sequences from Fujian, China. AIDS Res Hum Retroviruses. 2007;23:569-574.
27. Han XX, Zhang M, Dai D, et al. [Background study of HIV-1 drug resistant mutations in treatment-naive patients in Liaoning province]. Zhongguo Yi Xue Ke Xue Yuan Xue Bao. 2006;28:632-636.
28. Piyasirisilp S, McCutchan FE, Carr JK, et al. A recent outbreak of human immunodeficiency virus type 1 infection in southern China was initiated by two highly homogeneous, geographically separated strains, circulating recombinant form AE and a novel BC recombinant. J Virol. 2000;74:11286-11295.
29. Sturmer M, Preiser W, Gute P, et al. Phylogenetic analysis of HIV-1 transmission: pol gene sequences are insufficient to clarify true relationships between patient isolates. AIDS. 2004;18:2109-2113.
30. Hue S, Clewley JP, Cane PA, et al. HIV-1 pol gene variation is sufficient for reconstruction of transmissions in the era of antiretroviral therapy. AIDS. 2004;18:719-728.
31. Kaye M, Chibo D, Birch C. Phylogenetic investigation of transmission pathways of drug-resistant HIV-1 utilizing pol sequences derived from resistance genotyping. J Acquir Immune Defic Syndr. 2008;49:9-16.
32. Bello G, Eyer-Silva WA, Couto-Fernandez JC, et al. Demographic history of HIV-1 subtypes B and F in Brazil. Infect Genet Evol. 2007;7:263-270.

China; CRF01_AE; Hong Kong; molecular epidemiology; transmission

© 2009 Lippincott Williams & Wilkins, Inc.