Secondary Logo

Journal Logo

Epidemiology and Social

Demographic but not geographic insularity in HIV transmission among young black MSM

Oster, Alexandra M.a,b; Pieniazek, Danutab; Zhang, Xinjianb; Switzer, William M.b; Ziebell, Rebecca A.c; Mena, Leandro A.d,e; Wei, Xierongb; Johnson, Kendra L.d; Singh, Sonita K.d; Thomas, Peter E.b; Elmore, Kimberlee A.b; Heffelfinger, James D.b

Author Information
doi: 10.1097/QAD.0b013e32834bfde9
  • Free

Abstract

Introduction

In the United States, black MSM are disproportionately affected by the HIV epidemic [1]. Particularly, young black MSM are at substantially increased risk compared with young MSM of other races/ethnicities [2]. National data also show that diagnoses of HIV infection among black MSM aged 13–24 years have increased substantially in recent years [3].

It is not fully understood to what extent HIV transmission occurs between black MSM and other groups. Media and public health reports have focused attention on MSM and women; this has contributed to a perception that transmission of HIV from MSM and women to their female partners is a primary cause of high HIV prevalence in the black community [4–6]. The National HIV Behavioral Surveillance System found that 28% of black MSM surveyed during 2003–2005 reported sex with female partners; this proportion was substantially higher than that for MSM in other racial/ethnic categories [7]. Behavioral surveys have described the proportion of MSM who have sex with female partners, but it is unclear the extent to which this results in HIV transmission to women.

Phylogenetic analysis can be used to describe the transmission dynamics in different populations and geographic areas [8–13]. Transmission that is primarily confined to a specific demographic group may require a prevention approach targeted to that group, whereas transmission distributed more broadly among various demographic and risk groups may require a different approach. Moreover, understanding whether HIV infections distributed throughout a jurisdiction represent one interconnected epidemic (with transmission between persons in different geographic areas) or multiple geographically isolated epidemics (without substantial transmission between geographic areas) may impact delivery of prevention interventions. Therefore, understanding transmission dynamics is important for decreasing HIV transmission. In epidemics in which transmission of HIV strains with drug resistance-associated mutations (DRMs) is common, these concerns are of even higher priority.

During 2007, the Mississippi State Department of Health (MSDH) noted an increase in HIV diagnoses among black MSM aged 16–25 years in Jackson, Mississippi. The Centers for Disease Control and Prevention (CDC) and MSDH conducted a multifaceted investigation to understand HIV transmission in this population and enhance HIV prevention efforts [14,15]. Formative interviews with community members and stakeholders indicated that transmission between black MSM and women was of concern in the community and suggested that there might be differences between older and younger black MSM with respect to openness about sexual identity. These interviews also raised the possibility that persons who travel to have sex might facilitate the spread of HIV across the state.

As part of this investigation, we analyzed HIV sequences from newly diagnosed HIV-infected persons in Mississippi to evaluate phylogenetic clustering and determined demographic, behavioral, and geographic characteristics of persons in clusters that included young black MSM, with the goal of understanding the HIV transmission dynamics of young black MSM and others in Mississippi.

Methods

Specimen collection and phylogenetic analyses

All sequences used in this analysis were from antiretroviral-naive persons newly diagnosed with HIV infection in Mississippi from January 2005 to April 2008. All specimens were taken from remnant diagnostic serum or plasma drawn at participating public facilities within 3 months of HIV diagnosis. Sequences from persons later determined to have been previously diagnosed with HIV infection were excluded.

Sequences were obtained from two sources: the Variant, Atypical, and Resistant HIV Surveillance system (VARHS) and additional local surveillance among young (aged 16–25 years) black men. VARHS is a supplemental surveillance system funded by CDC in a number of locations [16]. MSDH received funding to conduct VARHS during 2005–2007 and continued to collect remnant specimens during early 2008 from persons newly diagnosed with HIV infection. Specimens that had been collected from young black men during January–April 2008 were included in this analysis. Following PCR amplification, specimens underwent sequencing of the polymerase (pol) region at Stanford University Clinical Virology Laboratory, Palo Alto, California, USA (VARHS specimens, n = 767) or at CDC, Atlanta, Georgia, USA (additional 2008 specimens, n = 32).

Phylogenetic analysis was used to infer the HIV-1 subtype. Pol nucleotide sequences of specimens from Mississippi and reference strains of groups M, N, and O extracted from the Los Alamos Database (http://hiv.lanl.gov) were aligned using Clustal W v1.83 included in the GeneStudio package (http://www.genestudio.com) [17]. Neighbor joining phylogenetic analysis was performed on a 1 011-nucleotide pol sequence spanning protease and the first 714 nucleotides of reverse transcriptase. Nucleotide distances were calculated by Kimura's two-parameter method in PHYLIP version 3.5c, with 1 000 bootstrap replicates [18]. The SIVcpz sequence (GenBank accession number X52154) was used as the outgroup.

Phylogenetic relationships among viral sequences were initially assessed using the neighbor-joining method, as described above. Close relationships among two or more sequences (referred to as clusters) were inferred if the following two stringent conditions were fulfilled: a bootstrap value of at least 99% and an average genetic distance (i.e., branch length) less than 0.015 nucleotide substitutions per site. The presence of such clusters was subsequently independently verified with both the maximum likelihood method using fastDNAml included in the GeneStudio package [19] and Bayesian Markov Chain Monte Carlo (MCMC) inference using BEAST (http://beast.bio.ed.ac.uk) [20]. Two independent BEAST runs were performed consisting of 400 million MCMC generations for the pol alignment with a sampling every 1000 generations, an uncorrelated exponential growth relaxed molecular clock, a general time-reversible nucleotide substitution model, and a burn-in of 4 million generations. The constant coalescent was used as a tree prior to inferring the pol tree topologies. Convergence of the MCMC was assessed by calculating the effective sampling size (ESS) of the runs using the program Tracer v1.5 (http://beast.bio.ed.ac.uk/Tracer). All parameter estimates showed significant ESSs (> 400). The tree with the maximum product of the posterior clade probabilities (maximum clade credibility tree) was chosen from the posterior distribution of 9001 sampled trees after burning in the first 1000 sampled trees with the program TreeAnnotator v1.5.4 included in the BEAST software package. Trees were viewed and edited using FigTree v1.3.1 (http://tree.bio.ed.ac.uk/software/figtree).

Neighbor joining, maximum likelihood, and Bayesian methods all inferred identical sets of clusters for this analysis. The neighbor joining phylogenetic tree presented is limited to the 21 clusters that included at least one young black MSM and includes bootstrap and posterior probability support for each cluster.

For the analysis of transmitted DRMs (TDRMs), sequence processing and mutation identification were completed by the Stanford laboratory (VARHS specimens) or electronically using Sierra web service program (additional 2008 specimens, http://hivdb.stanford.edu/pages/webservices/). Using SAS v9.2 (SAS Institute, Cary, North Carolina, USA), sequences were analyzed to identify mutations on the ‘list of drug resistance-associated mutations for surveillance in the United States,’ which was generated to maximize surveillance for TDRM for subtype B HIV strains that are predominant in the United States [16].

Analysis of demographic characteristics

We used surveillance data from the HIV/AIDS Reporting System (HARS) to examine demographic attributes and transmission categories of the entire sample and of persons in clusters that included at least one MSM. Additionally, we examined the characteristics of members of clusters that included at least one young black MSM. We considered persons to be young black MSM if their HARS records indicated that they were black or African–American, aged 16–25 years, and their transmission category was MSM or MSM/IDU. We also examined the characteristics of persons in clusters that did not include young black MSM, but did include at least one older black MSM (defined as black or African–American, aged > 25 years, transmission category MSM or MSM/IDU). We compared the characteristics of these clusters to those with young black MSM using the Mantel–Haenszel χ2-test performed in SAS v9.2. Additionally, we assessed whether clusters with at least one young black MSM were composed of persons from one geographic region in Mississippi or more than one region. To define regions for this analysis, we used the nine public health districts in Mississippi, although we combined some districts with small numbers of cases into larger regions. A small number of persons included in this analysis were diagnosed with HIV infection in Mississippi, but later determined to have been residents of other states at the time of diagnosis; they have been included in this analysis and were considered to have out-of-state residency.

Behavioral survey

As part of the investigation of increased HIV diagnoses among young black MSM in Jackson, we interviewed black MSM aged 16–25 years who were diagnosed with HIV infection from January 2006 to April 2008 and living in or diagnosed in the Jackson area (interviewed cases), as previously described [14,15]. Briefly, interviewed cases were identified using surveillance records and recruited for participation by phone, by mail, or in person. Interviewed cases completed a self-administered survey containing questions about demographic characteristics, substance use, and risk behaviors on a laptop or handheld computer. All questions assessed behaviors during the 12 months before HIV diagnosis. For the 26 interviewed cases for whom HIV-1 sequences were available, we compared behavioral characteristics of those who were and were not in clusters.

Protection of the identity of individuals and the security of data

VARHS operates under HIV/AIDS surveillance authority and is not considered research. The only additional data collected for this analysis were interview data; the interview portion of the investigation was conducted in the context of a public health epidemiologic investigation, and it was determined by the National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention at the CDC that this investigation did not require approval from CDC or local institutional review boards. No data that could identify individuals are presented. Phylogenetic data were not disclosed to participants. Surveillance data are protected by an assurance of confidentiality and cannot be released; therefore, sequences were not deposited in GenBank.

Results

Study population description

Of 1848 persons reported with HIV infection in Mississippi from January 2005 to April 2008, 799 (43%) had sequences available and met eligibility criteria. Of these, 767 were obtained through VARHS and 32 from additional 2008 surveillance among young black men. Those for whom sequences were available were not significantly different from those without sequences with respect to sex or residence in the Jackson area (vs. elsewhere in the state). However, persons with sequences available were significantly more likely than those without available sequences to be black (80% with sequences vs. 72% without) and younger than 30 years (47 vs. 32%).

The sample of persons with available sequences was primarily male (71%) and black (79%) (Table 1). Median age was 31 years. Transmission category was MSM for 39%, heterosexual contact for 9%, unknown for 48%, and IDU or MSM/IDU for 4%. Of 799 persons, 130 (16%) were young black MSM and 109 (14%) were older (aged > 25 years) black MSM; 192 (24%) were black women.

Table 1
Table 1:
Characteristics of persons in sample and, specifically, in phylogenetic clusters containing young black MSM or older black MSM, Mississippi, 2005–2008.

Cluster analysis

Of the 799 sequences analyzed, 794 (99%) were subtype B, three (0.4%) were subtype A, and two (0.3%) were distinct circulating recombinant forms (CRF02-AG and CRF10-CD). In all, 228 (29%) sequences clustered with at least one other sequence; likelihood of clustering was not significantly associated with sex, race/ethnicity, or transmission category. All clustered sequences were subtype B. These formed 82 clusters with two to 10 sequences per cluster.

Of the 82 phylogenetic clusters identified, 10 (12%) included strains from white or Latino MSM and 34 (41%) included strains from black MSM. Characteristics of persons in these clusters are depicted in Fig. 1(a and b, respectively) with each person represented by a circle (man) or triangle (woman). Of the 36 persons in clusters with at least one white or Latino MSM (Fig. 1a), only one was female, and all had a transmission category of MSM (69%) or unknown (31%). These clusters included substantial numbers of persons from each racial/ethnic group (black, white, or Latino).

Fig. 1
Fig. 1:
Depiction of members of phylogenetic clusters of HIV-infected persons in Mississippi by sex, race/ethnicity, and transmission category.Panel (a) depicts all members of clusters that include at least one white or Latino MSM. Almost all cluster members are men, there are no members with heterosexual transmission mode, and cluster members are diverse with respect to race/ethnicity. Panel (b) depicts all members of clusters that include at least one black MSM. These clusters include a number of women and persons with heterosexual transmission mode and are less diverse with respect to race/ethnicity. Panel (c) depicts clusters including at least one black MSM aged 25 years or less. These clusters are nearly homogeneous with respect to sex, transmission mode, and race/ethnicity. Panel (d) depicts clusters containing older black MSM but no young black MSM. The members of these clusters are diverse with respect to sex, transmission mode, and race/ethnicity. Persons reported to the HIV/AIDS Reporting System with no identified transmission category are classified as ‘unknown’. No persons with transmission category of IDU or MSM/IDU were in clusters with MSM.

In contrast with the clusters including strains from white and Latino MSM, clusters including strains from black MSM (Fig. 1b) had a higher proportion of women (9%) and included four persons with heterosexual contact as their transmission category. To determine whether this pattern varied by age, we assessed clusters with black MSM that did and did not include young black MSM.

Clusters with black MSM that included at least one young black MSM (21 clusters including strains from 69 persons) were nearly homogeneous with respect to race/ethnicity, sex, and transmission category, including no women or persons with heterosexual contact as their transmission category, and very few persons who were not black (Fig. 1c). Moreover, 80% of persons in these clusters were young (Table 1). In contrast, clusters with black MSM that did not include young black MSM (13 clusters including strains from 40 persons) included a number of women and persons whose transmission category was heterosexual contact (Fig. 1d). Persons in clusters containing young black MSM were significantly different from persons in clusters containing only older black MSM with respect to sex, race/ethnicity, age, and transmission category (Table 1, all P < 0.01). No MSM clusters contained persons with IDU or MSM/IDU as their mode of transmission.

Geographic analysis

Figure 2 depicts 21 clusters that included at least one young black MSM; each cluster is delineated by a bracket, and font color is used to represent region of residency at diagnosis. Of these clusters, 10 contained residents of a single geographic region in Mississippi; however, all five regions were represented in these clusters (Fig. 2). The remaining 11 clusters contained sequences from residents of two or more regions in Mississippi, and three of these contained persons from other states.

Fig. 2
Fig. 2:
Inference of HIV-1 clusters involving black MSM and others in Mississippi.Neighbor joining (NJ) tree showing phylogenetic relationships between HIV-1 subtype B polymerase sequences from 16 to 25-year-old (young) black MSM newly diagnosed with HIV infection in Mississippi during 2005–2008. Other closely related sequences from persons diagnosed with HIV infection during 2005–2008 in Mississippi are included. Sequences from young black MSM are underlined. Font color represents region of residence at HIV diagnosis (see inset map). Brown font indicates residence in another state; black font indicates missing data for residence. Each cluster is delineated by the NJ bootstrap value (displayed when ≥99%), Bayesian posterior probability (in parentheses) and a bracket to the right. All clusters shown had maximum likelihood P < 0.01. Drug resistance-associated mutations (drug class and specific mutation) are shown. The prefix of the sequence ID (05, 06, 07) indicates the year of specimen collection for Variant, Atypical, and Resistant HIV Surveillance (VARHS) specimens. Specimens from additional surveillance in 2008 begin with ‘Miss’. The scale bar indicates an evolutionary distance of 0.01; vertical distances are for clarity only.

Transmitted drug resistance

Of 799 sequences, 133 (17%) had TDRMs (Table 1). Of the 228 sequences that clustered with at least one other sequence, 35 (15%) had TDRM; this proportion was higher (29%) for sequences in clusters containing at least one young black MSM. Among sequences in clusters containing at least one young black MSM, the most common mutations conferring drug resistance were K103N and L90M, which confer resistance to the nonnucleoside reverse transcriptase inhibitors and protease inhibitors, respectively; sequences with TDRM tended to cluster together (Fig. 2). None of the sequences in clusters containing older black MSM, but no young black MSM, had TDRM.

Behavioral characteristics of interviewed cases

Among the 26 interviewed young black MSM, HIV strains from 13 (50%) were included in 10 clusters. The 13 interviewed cases who were in clusters were more likely to report marijuana use (62 vs. 23%, P = 0.05) or exchange sex (31 vs. 0%, P = 0.03) than the 13 interviewed cases who were not in clusters (Table 2).

Table 2
Table 2:
Behavioral characteristics of clustered vs. nonclustered HIV-infected black MSM aged 16–25 years during the 12 months before HIV diagnosis, Mississippi, 2005–2008.

Discussion

This analysis identified multiple clusters involving HIV-infected young black MSM in Mississippi. Moreover, we determined that these clusters were fairly homogeneous with respect to sex, race/ethnicity, age, and transmission category, suggesting insularity of young black MSM with respect to HIV transmission. Reducing HIV transmission among young black MSM in Mississippi may require prevention strategies that are tailored to young black MSM.

On the contrary, clusters that did not contain young black MSM but did contain older black MSM were much less homogeneous and were similar to the overall population included in this analysis. This suggests that there is mixing between older black MSM and other populations with respect to sex, race, and transmission category. The fact that many of the persons in these clusters were reported to have unknown transmission category in the HIV/AIDS Reporting System suggests that they may be less open about their sexual behavior with public health staff (and others). Likewise, the inclusion of women in these clusters suggests that there may be transmission between some older black MSM and women.

Black MSM are less likely than white MSM to identify as gay or to disclose their sexual identity and/or behavior to others [21]. Our analysis suggests that there may be a generational difference in openness about sexuality; young black MSM may feel less pressured than older black MSM to engage in normative heterosexual relationships and, thus, may be less likely to transmit HIV infection to women. This may be due to changes in comfort with sexuality or the fact that these young black MSM have not yet reached an age at which relationships with women are expected.

Nevertheless, young black MSM continue to be at unacceptably high risk for HIV infection [22]. Our analysis suggests that young black MSM in Mississippi are likely to transmit HIV to others like themselves. The insularity of their networks suggests that HIV prevalence in this population may continue to rise unless successful interventions are put into place for young black MSM.

Our geographic analysis demonstrated that, although the clusters that included young black MSM were homogeneous with respect to demographic and risk characteristics, they were heterogeneous with respect to area of residence. More than half of clusters included persons from more than one region of Mississippi, and many of these included persons from multiple regions of Mississippi or other states. Additionally, given that we only had sequences for persons diagnosed with HIV infection in Mississippi, it is likely that we are underestimating the geographic heterogeneity. Data from the behavioral survey described in this article indicate that 20% of interviewed cases reported traveling to another region of Mississippi or out of state to have sex with a man during the 12 months before HIV diagnosis. This analysis suggests that, to reduce HIV transmission among young black MSM in Mississippi, HIV prevention interventions should be delivered in a way that is accessible to MSM, particularly young black MSM, from all regions of Mississippi.

Sequences from persons in clusters containing young black MSM had high prevalence of TDRM compared with those from the entire sample, and some large clusters consisted exclusively of persons with TDRM. This suggests that interrupting HIV transmission among young black MSM is particularly important, because acquisition of strains with TDRM can limit first-line treatment options and decrease the effectiveness of other treatment regimens. Providers should consider implementing enhanced prevention interventions with HIV-infected persons with TDRM to decrease transmission of strains with TDRM.

This analysis demonstrates that phylogenetic analysis can be a useful tool for assessing HIV transmission dynamics within a population. HIV-1 viral sequencing is routinely performed to monitor drug resistance for clinical purposes. Moreover, a number of sites in the United States participate in surveillance for variant, atypical, and resistant HIV strains through the VARHS program. The uniform approach of VARHS has expanded the ability to perform analyses that employ a population-based phylogenetic approach to understand HIV transmission. As HIV-1 sequence data have become available to surveillance programs, an opportunity has emerged for jurisdictions (e.g., cities, states, or countries) to better understand the transmission dynamics within their areas. These analyses may be most informative when performed at the state or country level. The information gained from these analyses can, in turn, be used to tailor prevention efforts.

The representativeness of our phylogenetic analysis is limited by the completeness of sequences available for analysis. There were some demographic differences between those for whom we did and did not have sequences; this likely resulted from the fact that certain groups (blacks, younger persons) were more likely to receive HIV tests at public facilities. Nonetheless, we had sequences for a high proportion (43%) of those diagnosed with HIV infection in Mississippi during our period. However, we were likely missing sequences for those who were HIV-infected but not yet diagnosed (or reported), as well as those who were diagnosed outside of Mississippi or outside of our time frame. This could potentially lead to an underestimate of the number of clusters or the size of clusters.

Additionally, only the major HIV strain was sequenced for each sample; because mutations can revert to wild type over time in the absence of selective pressure, more sensitive assays that detect additional minor strains in those with multiple sources of HIV infection might have resulted in a higher prevalence of TDRM, as has been previously shown [23]; identification of these strains might have provided information to support additional clusters. We identified black MSM using data from the HIV/AIDS Reporting System; some of those with unknown transmission category were likely MSM, which may have led to underestimation of the number of black MSM. Moreover, it is important to note that these clusters do not establish an epidemiologic link in HIV transmission between any two persons; the similarity of their viruses could have resulted from intermediate steps in transmission or another person serving as a common source of infection. Finally, our findings cannot be considered to be representative of transmission dynamics in other groups or in areas outside of Mississippi.

Our findings suggest that transmission of HIV between young black MSM and other groups in Mississippi is limited. Therefore, prevention efforts that are focused on young and black MSM [24–26] or network approaches may be effective methods to promote HIV testing and prevention in this area. Interrupting HIV transmission in this group is particularly important given the high proportion with transmitted TDRM. Phylogenetic analysis is a tool that can be used to understand transmission dynamics in communities and to appropriately tailor prevention efforts.

Acknowledgments

A.M.O., D.P., W.M.S., L.A.M., P.E.T., and J.D.H conceived of and designed the analysis. R.A.Z., K.L.J., S.K.S., A.M.O., and L.A.M acquired data from VARHS (R.A.Z., S.K.S), the HIV/AIDS Reporting System (R.A.Z., K.L.J.), and the behavioral survey (A.M.O., L.A.M.). X.W. performed sequencing. D.P., X.Z, and W.M.S performed the phylogenetic analysis. R.A.Z. performed the drug resistance analysis. A.M.O. performed the demographic, behavioral, and geographic analyses. K.A.E. created the maps. A.M.O. and D.P. drafted the manuscript. All authors critically revised the manuscript.

The authors thank Christina Dorell, Carlos Toledo, Craig Thompson, Walid Heneine, and Jeff Johnson for their assistance in conducting this investigation.

Conflicts of interest

The Centers for Disease Control and Prevention (CDC) funds HIV/AIDS surveillance and Variant, Atypical, and Resistant HIV Surveillance (VARHS). CDC and the Mississippi State Department of Health funded the behavioral survey.

The findings and conclusion in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

There are no conflicts of interest.

References

1. Centers for Disease Control and Prevention. HIV prevalence, unrecognized infection, and HIV testing among men who have sex with men: five U.S. cities, June 2004-April 2005. MMWR Morb Mortal Wkly Rep 2005; 54:597–601.
2. MacKellar DA, Valleroy LA, Secura GM, Behel S, Bingham T, Celentano DD, et al. Unrecognized HIV infection, risk behaviors, and perceptions of risk among young men who have sex with men: opportunities for advancing HIV prevention in the third decade of HIV/AIDS. J Acquir Immune Defic Syndr 2005; 38:603–614.
3. Centers for Disease Control and Prevention. Trends in HIV/AIDS diagnoses among men who have sex with men: 33 States 2001–2006. MMWR Morb Mortal Wkly Rep 2008; 57:681–686.
4. Boykin K. Beyond the down low: sex, lies and denial in black America. New York:Carroll & Graf; 2005.
5. Denizet-Lewis B. Double lives on the down low.New York Times Magazine 2003; section 6, 28–33.
6. Malebranche DJ. Bisexually active black men in the United States and HIV: acknowledging more than the ‘down low’. Arch Sex Behav 2008; 37:810–816.
7. Sanchez T, Finlayson T, Drake A, Behel S, Cribbin M, Dinenno E, et al. Human immunodeficiency virus (HIV) risk, prevention, and testing behaviors: United States, National HIV Behavioral Surveillance System – men who have sex with men, November 2003–April 2005. MMWR Surveill Summ 2006; 55:1–16.
8. Holmes EC, Zhang LQ, Robertson P, Cleland A, Harvey E, Simmonds P, Leigh Brown AJ. The molecular epidemiology of human immunodeficiency virus type 1 in Edinburgh. J Infect Dis 1995; 171:45–53.
9. Kouyos RD, von Wyl V, Yerly S, Boni J, Taffe P, Shah C, et al. Molecular epidemiology reveals long-term changes in HIV type 1 subtype B transmission in Switzerland. J Infect Dis 2010; 201:1488–1497.
10. Brenner BG, Roger M, Routy JP, Moisi D, Ntemgwa M, Matte C, et al. High rates of forward transmission events after acute/early HIV-1 infection. J Infect Dis 2007; 195:951–959.
11. Smith DM, May SJ, Tweeten S, Drumright L, Pacold ME, Kosakovsky Pond SL, et al. A public health model for the molecular surveillance of HIV transmission in San Diego, California. AIDS 2009; 23:225–232.
12. Yerly S, Junier T, Gayet-Ageron A, Amari EB, von Wyl V, Gunthard HF, et al. The impact of transmission clusters on primary drug resistance in newly diagnosed HIV-1 infection. AIDS 2009; 23:1415–1423.
13. Yerly S, Vora S, Rizzardi P, Chave JP, Vernazza PL, Flepp M, et al. Acute HIV infection: impact on the spread of HIV and transmission of drug resistance. AIDS 2001; 15:2287–2292.
14. Oster AM, Dorell CG, Mena LA, Thomas PE, Toledo CA, Heffelfinger JD. HIV risk among young African American men who have sex with men: a case-control study in Mississippi. Am J Public Health 2011; 101:137–143.
15. Centers for Disease Control and Prevention. HIV infection among young black men who have sex with men: Jackson, Mississippi 2006-2008. MMWR Morb Mortal Wkly Rep 2009; 58:77–81.
16. Wheeler WH, Ziebell RA, Zabina H, Pieniazek D, Prejean J, Bodnar UR, et al. Prevalence of transmitted drug resistance associated mutations and HIV-1 subtypes in new HIV-1 diagnoses, U.S.: 2006. AIDS 2010; 24:1203–1212.
17. Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 1994; 22:4673–4680.
18. Felsenstein J. PHYLIP-phylogeny inference package (version 3.2). Cladistics 1989; 5:164–166.
19. Olsen GJ, Matsuda H, Hagstrom R, Overbeek R. fastDNAmL: a tool for construction of phylogenetic trees of DNA sequences using maximum likelihood. Comput Appl Biosci 1994; 10:41–48.
20. Drummond AJ, Rambaut A. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evol Biol 2007; 7:214.
21. Millett GA, Peterson JL, Wolitski RJ, Stall R. Greater risk for HIV infection of black men who have sex with men: a critical literature review. Am J Public Health 2006; 96:1007–1019.
22. Centers for Disease Control and Prevention. Prevalence and awareness of HIV infection among men who have sex with men: 21 cities, United States 2008. MMWR Morb Mortal Wkly Rep 2010; 59:1201–1207.
23. Johnson JA, Li JF, Wei X, Lipscomb J, Irlbeck D, Craig C, et al. Minority HIV-1 drug resistance mutations are present in antiretroviral treatment-naive populations and associate with reduced treatment efficacy. PLoS Med 2008; 5:e158.
24. Jones KT, Gray P, Whiteside YO, Wang T, Bost D, Dunbar E, et al. Evaluation of an HIV prevention intervention adapted for black men who have sex with men. Am J Public Health 2008; 98:1043–1050.
25. Wilton L, Herbst JH, Coury-Doniger P, Painter TM, English G, Alvarez ME, et al. Efficacy of an HIV/STI prevention intervention for black men who have sex with men: findings from the Many Men, Many Voices (3MV) project. AIDS Behav 2009; 13:532–544.
26. Kegeles SM, Hays RB, Coates TJ. The Mpowerment Project: a community-level HIV prevention intervention for young gay men. Am J Public Health 1996; 86:1129–1136.
Keywords:

African–Americans; HIV infections/epidemiology/transmission; homosexuality; male; molecular epidemiology; phylogeny; risk factors

© 2011 Lippincott Williams & Wilkins, Inc.