The classification of HIV-1 into distinct genetic subtypes has enabled the study of its global molecular epidemiology. 1 These subtypes can indicate the origin of an infection, reveal outbreaks within population subgroups, and provide a means of monitoring the spread of infection between exposure groups. 2–4 HIV-1 genetic diversity has been shown to affect the accuracy of diagnostic assays 5 and may affect quantification of plasma HIV RNA concentration. 6,7 Differences in the susceptibility to antiretroviral drugs have been reported for subtypes F and G, 8,9 and the diverse group O viruses possess natural resistance to the nonnucleoside reverse transcriptase inhibitor class of drugs. 10 Given the potential public health impact of HIV-1 genetic diversity, the monitoring of subtype trends is an essential component of national HIV surveillance in England and Wales. 11
The subtypes of HIV-1 segregate broadly by geographical region, with all currently identified viral types having been detected in sub-Saharan Africa. 12–14 Historically, subtype B infections have been prevalent in Western countries, predominantly among men who have sex with men and injecting drug users (IDUs). 15–17 The subtype distribution among the heterosexual population is genetically diverse, with non-B subtypes and recombinant strains having been documented in several European countries. 16–22 Small numbers of non-B viruses have also been detected in the United States. 23 In Western countries, heterosexually acquired non-B subtype infections are more commonly found in individuals with links to areas of the world where multiple subtypes co-circulate, such as sub-Saharan Africa. Since 1999, among newly diagnosed HIV infections in the United Kingdom, heterosexual transmission has become the most common route of HIV-1 acquisition. 24–26 Molecular epidemiology provides a sensitive tool for monitoring trends in HIV infections and for tracing the likely origin of these infections, thereby providing a better understanding for the development of public health interventions. 17,20
We have previously generated baseline estimates of the subtype distribution in England and Wales for 1997, based on assigning a subtype in the env gene region. 16 In that study we concluded that approximately a quarter of HIV-1-positive patients were infected with a non-B subtype. 16 Furthermore, in a separate study we estimated that one-fifth of non-B infections in the UK were due to recombinant viruses. 19 In the present study, an improved subtyping algorithm has been applied to specimens from heterosexual sexually transmitted infection (STI) clinic attendees between 1997 and 2000. Using this new algorithm, we present results on the genetic diversity of the heterosexual HIV epidemic in England and Wales during the period 1997–2000.
MATERIALS AND METHODS
In England and Wales, the Unlinked Anonymous HIV Prevalence Monitoring Program (UAPMP) has surveyed HIV infection in individuals presenting at selected clinics since the early 1990s. 27 Residual serum specimens (50–1000 μL) following routine syphilis testing were obtained from 15 UAPMP STI clinics (7 in London and 8 outside London) in England and Wales. Patient identifiers were irreversibly unlinked from each specimen with limited demographic, epidemiologic, and clinical data being retained, 28 i.e., the STI clinic attended, calendar year and yearly quarter of attendance, gender, sexual orientation, IDU, age group, presence of an acute STI, anti-HIV-positive diagnosis at time of leaving the clinic, and world region of birth (WRB). Anti-HIV screening was performed either at the laboratory serving the participating clinic or at the Central Public Health Laboratory (CPHL), with confirmatory testing of reactive samples at CPHL. All anti-HIV-1 positive specimens obtained from individuals who indicated their sexual orientation as heterosexual and attended STI clinics during the calendar years 1997–2000 were included in the study (n = 1055). Seventy-nine anti-HIV-1-positive specimens (7.5%) were unavailable for RNA extraction due to insufficient residual serum.
Genotyping of Specimens
Nucleic acid was extracted from the residual serum specimens using the Boom method. 29 The protocol was modified during the year 2000 to allow virus to be pelleted from serum at 28,110 g at 4°C for 1 hour, prior to nucleic acid extraction. Specimens received during 1997 were genotyped as previously described. 16,30,31 The protocol was revised for subsequent years to include a mixed specific-primer reverse transcription (RT) reaction using primers DT7 (gag), Pout4 (pol), and ED12 (env), and a multiplexed first-round polymerase chain reaction (PCR) with primer pairs DT1/DT7 (gag), Pout3/Pout4 (pol), and ED5/ED12 (env). 31–33 Second-round gag and env PCRs were performed as previously described, and second-round pol amplification was performed using the primer pair Pin3/Pin4. 16,31,33
Heteroduplex mobility assays (HMAs) of the env amplicons were performed as previously described. 30 The gag HMAs were essentially as previously described but were resolved on 5% polyacrylamide, 6M urea Criterion (BioRad Laboratories, Hercules, CA) precast gels. 31 Screening HMAs were performed using amplicons generated from single reference plasmids representing subtypes A–C for env and A–D for gag, with H2O as a control. 30,31 Any ambiguous HMA results, or results discordant between gene regions, were resolved by sequencing. Second-round PCR amplicons were gel-purified using either QIAquick Gel Extraction (Qiagen, Crawley, UK) kits or Ultrafree-DA columns (Millipore, Watford, UK). Sequencing of specimens was as previously described 16 or, more recently, using a CEQ 2000XL DNA Analysis System capillary sequencer (Beckman Coulter, High Wycombe, UK), according to the manufacturer’s instructions.
Reference sequences of group M subtypes (A–K), circulating recombinant forms (CRFs), and groups O and N were obtained from the Los Alamos HIV Sequence Database (2001 version). Alignments of the study and reference sequences were generated using ClustalW within Bioedit v4.8.5. 34 The parameters of the optimal model of evolution were estimated using Modeltest v3.0 35 and Paup*, 36 and neighbor-joining trees (bootstrapped ×1000) were generated.
The subtype proportions and analyses presented refer to specimens for which both gag and env subtypes were available, unless otherwise stated. This provides a consistent sampling frame over the study period, and allows the detection of recombinant genomes. Statistical analyses were also performed using each individual gene region to confirm any associations. Where the subtypes obtained from different gene regions were concordant, those viral genomes were defined as belonging to a single subtype. Similarly, where phylogenetic analysis indicated a sequence similar to a CRF in ≥1 gene region, and the genome composition was consistent with that type, the virus was defined as a CRF. Viruses discordant between gene regions but not resembling known CRFs were defined as unique recombinant forms (URFs). Where the phylogenetic analyses showed evidence of nearly identical viral sequences, the epidemiologic data were analyzed to determine whether the sequences might have been obtained from the same individual. There was only a single instance where both the sequence and epidemiologic data suggested that this may be the case; however, these were not excluded from the analysis.
HIV-1 molecular data were linked to the demographic information and analyzed using subtype as the dependent variable, and molecular and demographic data as explanatory variables. Demographic data and single-variable analysis of subtype distributions were analyzed using Fisher’s Exact tests, χ2, and χ2 tests-for-trend. Multivariable analyses were performed using logistic regression where the dependent variable was binomial, and polytomous regression where the dependent variable contained >2 categories. The variables included in the logistic regression analyses are described in the “Study Population” section and in Table 1. Statistical analyses were performed using the software packages Intercooled Stata v7.0 for Windows (Stata Corp., College Station, TX) and EpiInfo v6.04 (Centers for Disease Control and Prevention, Atlanta, GA).
Amplification Success from Viral RNA
Amplification success differed among the 3 gene regions. The inclusion of a multiplex PCR protocol into the subtyping algorithm from 1998 specimens onwards (see “Methods”) resulted in unequal denominators in the various datasets. Amplification was attempted using env primers from all samples received during the study (n = 976) and in all 3 gene regions (gag, pol, and env) for samples received from 1998 onwards (n = 714). Amplification using gag primers was performed on a limited number of samples received during 1997 (n = 150). Amplification using pol primers was not attempted on samples received during 1997. PCR amplicons were obtained from 76.9% of specimens in pol, 67.6% in gag, and 63.0% in env. Amplicons were obtained in both gag and env from 67.7% (499/737) of specimens for which amplification was attempted in both gene regions. The amplification success increased significantly over the study period for all 3 gene regions (P < 0.05). Logistic regression analyses showed the amplification success to be significantly lower in specimens from women in both gag (P = 0.01) and pol (P = 0.04). When subtype B specimens were removed from the analysis there was no significant difference in the amplification success between the sexes in any of the gene regions.
HIV-1 Subtype Distribution
A subtype was assigned to 99.7% of PCR amplicons in gag, 96.3% in env, and 96.0% in pol (pol amplicons, n = 149, from 1998 were not subtyped). The main reason for not determining a subtype was a failure to amplify from a specimen. There was complete concordance between the subtype result obtained where both HMA and sequence-derived subtypes were available. A subtype was assigned in all 3 gene regions to 306/743 (41.2%) of specimens subtyped, with a further 35.1% (261/743) subtyped in only 2 gene regions, and 23.7% (176/743) in a single gene region. A subtype was determined for all specimens where amplification was possible in both gag and env (n = 499). To provide a consistent sampling frame, all subsequent subtype distributions refer to samples for which both a gag and env subtype was determined, unless otherwise stated.
Overall, the most common type was HIV-1 subtype C, comprising 162 (32.5%) of the specimens for which a subtype was determined in both gag and env. HIV-1 subtype B accounted for 144 specimens (28.9%), followed by subtype A (72; 12.4%). There were also 38 infections (7.6%) with other subtypes (25 subtype D, 4 F, 8 G, and 1 H). CRFs accounted for 45 (9.0%) of the specimens subtyped, the most prevalent being CRF_02AG (33/45; 73.3% of all CRFs). Six genomes resembled CRF_01AE, two were similar to CRF_14BG, and one each of the CRFs _03AB, _05DF, _06cpx, and _11cpx were identified. In addition, URFs accounted for 38 (7.6%) of all specimens subtyped. There was no significant difference in the subtype distribution between individual STI clinics or between clinics located either inside or outside London.
HIV-1 Subtype Trends and Association With Demographic Variables
Logistic regression analyses were performed taking the proportion of subtype B infections as the baseline comparison group in all analyses. Likelihood-ratio tests indicated the subtype distribution to be significantly associated with individuals who gave an African WRB (P < 0.01), the gender of the individual (P = 0.03), and the HIV-1 diagnostic status of individuals upon leaving the STI clinic (P < 0.01). No other demographic variables were significantly associated with any HIV-1 subtype.
Logistic regression analyses showed there was no significant change in the overall subtype distribution in heterosexual STI clinic attendees over the period 1997–2000 (Fig. 1). However, several significant associations were observed between subtype and individual demographic variables (Table 1). Firstly, the proportions of non-B subtypes, CRFs, and URFs were significantly higher among African-born individuals when compared with those born in the UK. Similarly, the proportions of these subtypes were also significantly higher among women. Finally, the proportions of subtype A, C, CRFs, and URFs were significantly higher among individuals in whom HIV infection was newly diagnosed at the time of their clinic attendance compared with those in whom HIV infection remained undiagnosed.
HIV-1 Subtypes in Male and Female Heterosexual STI Clinic Attendees
Of 737 anti-HIV-1-positive specimens from heterosexuals for which amplification was performed in both gag and env, 359 (48.7%) were from men and 378 (51.3%) from women. A subtype was assigned in both gag and env to specimens from 260 men (72.4%) and 239 women (63.2%). There was no significant change in the subtype distribution for either gender during the study period. However, there was a significant difference in the subtype distribution between the genders, with women 2.4 times more likely to be infected with a non-B subtype (Table 1). HIV-1 subtype B was the most common subtype in men, accounting for 102 (39.2%), followed by subtypes C (72; 27.7%) and A (31; 11.9%). There were 17 (6.5%) single-subtype infections with subtypes other than A, B, and C. Twenty-three (8.9%) were identified as CRFs and 15 (5.8%) as URFs. In contrast, HIV-1 subtype C was most common among women, accounting for 90 (37.7%) of subtyped infections, followed by subtypes B (42; 17.6%) and A (41; 17.2%). Twenty-one infections (8.8%) with other subtypes (D–H) were detected, 22 (9.2%) CRFs, and 23 (9.6%) URFs. Logistic regression analyses of the subtype distribution in each gender indicated similar associations with the demographic data to those observed when analyzing the complete data set of heterosexual STI attendees.
HIV-1 Subtypes According to World Region of Birth
Subtype B was most common among infections in both European (62%) and UK-born (51.5%) individuals (Fig. 2). In contrast, only 10 of 232 infections (4%) in African-born individuals were with a subtype B virus. Subtype C was most common in both African- (46%) and Asian- (40%) born individuals, with the proportions of subtype A and CRFs being similar in each of the 4 most common WRBs in this study (Fig. 2).
HIV-1 Subtypes Identified as a Result of pol Amplification in 1999 and 2000
A total of 384 specimens collected during 1999 and 2000 were assigned a subtype in pol. The pol sequence data enabled subtyping of an additional 70 specimens (29 in 1999; 41 in 2000) for which a subtype had not been determined for both gag and env. Of these, 22 (31.4%) were subtype C, 11 (15.7%) subtype B, 8 (11.4%) subtype A, 14 (20.0%) CRFs, 8 (11.4%) other subtypes (D and G), and 7 (10.0%) URFs. There was no significant difference between the additional subtypes detected due to pol sequencing and those assigned by gag and env alone (P = 0.5). Overall, no significant difference was observed between the subtype distribution when assigning a subtype by both gag and env, or when pol sequence data were also included (P = 0.1).
Since the early 1990s, the unlinked anonymous testing of routine clinical specimens has provided minimally biased estimates of HIV prevalence in England and Wales among population groups at varying risk of infection. An algorithm has been developed that combines HMA and sequencing of 2 gene regions (gag and env) and sequencing of a third (pol) to subtype specimens collected as part of the UAPMP. The rates of recovery of viral RNA were acceptable for the samples in this study, i.e., residual serum specimens that had been frozen and thawed and may not have been stored optimally prior to arrival at our laboratory. In addition, the administration of antiretroviral therapy to individuals in whom HIV infection had been previously diagnosed would be expected to lower plasma viral RNA to undetectable levels. Use of the same subtyping methods with fresh clinical samples submitted for diagnosis gives close to 100% recovery. Where amplification was possible, a subtype could be assigned to >96% of specimens. This multigenic approach maximizes amplification success and enables the detection of intergenic recombinant viruses. 19,31 Application of the subtyping algorithm to specimens obtained via the UAPMP has enabled the detection of group M subtypes A–K, in addition to group O viruses. HIV subtyping provides a tool to aid understanding of differing transmission patterns, to detect the emergence of novel viral forms, and to target and monitor appropriate public health interventions. 11
Previous studies have shown subtype B to be the predominant type circulating in the UK. 16,18,20,37 Historically, HIV-1 subtypes segregated broadly by exposure category and geographic region, with subtype B predominating among men who have sex with men and IDUs in Western countries, and non-B subtypes, CRFs, and URFs in heterosexually acquired infections worldwide. 16,17,20,23,38 The historical links between the UK and Commonwealth countries in both sub-Saharan Africa and the Indian subcontinent are potentially an important route for the importation of diverse viral genomes. This, coupled with increases in sexual risk behaviors over the past decade, necessitates the continued monitoring of HIV diversity within the UK. 39
Previous findings from UAPMP specimens obtained during 1996/1997 showed subtype B infections to be the most common single type detected among heterosexuals in the UK. 16 In our present study, although there is no statistically significant change in the overall subtype distribution over the study period, subtype C is now found to be the most common single type, followed by subtypes B and A (Fig. 1). This is consistent with what we would predict within the UK, where the majority of heterosexually acquired infections diagnosed were probably acquired in Africa, and of those acquired within the UK the majority are likely to have originated from a partner infected outside of Europe. 26 The prominence of subtype C among HIV-infected heterosexuals in England and Wales reflects its predominance worldwide and is consistent with a shift from heterosexual infections diagnosed in the UK originating in eastern Africa (dominated by subtype A) to those from southeastern Africa (particularly Zimbabwe, where subtype C predominates) among those originating from this continent. 26
CRFs are recognized to play an increasingly significant role in the global epidemic. 40,41 In this study, CRFs accounted for 9% of subtyped infections, with nearly three-quarters being CRF_02AG. Of these, 60% were associated with a western African WRB, where this type is relatively common. 40 More than half of all infections with URFs and CRFs were identified in African-born individuals. URFs have previously been reported in the UK, although no single combination has been detected at high prevalence in this, or previous, studies. 19,20,42 The relatively high proportion of recombinant viruses among UK-born individuals suggests that mixing between migrants and UK-born individuals, coupled with the primary transmission of recombinant viruses, contributes to increased levels of genetic diversity. The incorporation of pol sequencing for specimens received during 1999 and 2000 has enhanced the detection of URFs and CRFs and enabled the prediction of antiretroviral drug resistance. However, inclusion of the pol sequence data did not significantly alter the overall subtype distribution derived from gag and env alone.
Previous studies of HIV genetic diversity among IDUs in England and Wales, Scotland, and France have shown this exposure group to contain almost exclusively subtype B infections. 16,17,20,43 A small proportion of heterosexuals for whom a subtype was obtained also indicated IDU as an HIV risk factor. While a large proportion of these infections in IDUs were with subtype B (76%), infections with subtypes A, C, and URFs were also observed (data not shown). This may indicate an increase in the genetic diversity within this risk group due to transmission through sharing of injecting equipment. However, injecting drug use among STI clinic attendees is recorded if they report ever having injected nonprescription drugs, and no information is sought about recent or current injecting behavior. In view of this, and the stable low prevalence of HIV among IDUs in England and Wales, HIV-1 acquisition via sexual exposure or other contact cannot be excluded. 44
There was a significant difference in the distribution of HIV-1 subtypes between the sexes, with women more than twice as likely as men to be infected with a non-B subtype. In addition, 60% of HIV-1-infected female clinic attendees were born in an African country (18% were UK-born), compared with 41% of males (34% were UK-born). The difference in the subtype distribution between the genders may, therefore, have arisen as a result of the acquisition of infection in different geographic regions. As we have suggested previously, some UK-born male STI attendees who declared their sexuality to be heterosexual are suspected to be homo-/bisexual, a population group in which subtype B predominates. 16 In an attempt to account for this potential bias, separate analyses were performed having removed all UK- and European-born, predominantly subtype B–infected, individuals from the analyses. Despite this, the association between non-B, other subtypes, and URFs in women remained.
HIV prevalence in younger age groups may be taken as a marker of more recent transmission events. The proportions of subtype C and CRFs were significantly higher among clinic attendees ≤24 years of age compared with older attendees. Recent evidence suggests younger people are adopting less-safe sexual practices. 39 This fact, together with our finding that subtype C has recently become the most common subtype among heterosexuals in England and Wales, suggests that the higher levels of subtype C in the younger age group may reflect recent HIV-1 transmission. Moreover, heterosexually acquired infections tend to be diagnosed later during the course of infection than those acquired through homosexual sex or injecting drug use, as evidenced by the significantly lower median CD4 count at first diagnosis in the former group. 45 Therefore, the higher proportion of subtype B infections in older age groups may indicate detection of longer-standing HIV infections, reflecting a historic transmission pattern when subtype B was the predominant subtype in the UK. A higher frequency of undisclosed homosexual risk behavior, or injecting drug use in both sexes, among older people may also contribute to the number of subtype B infections in this population group.
The burgeoning epidemic in south and southeast Asia, in which subtype C infections are most common, appears, as yet, to have had little impact in the UK. 46 Clinic attendees born in Asia accounted for only 3% of all typed specimens and were infected with subtypes typical of this area of the world. 41,47,48 It is possible that people born in south and southeast Asia are less likely to attend STI clinics; should this be the case, estimates of the impact of HIV infection in this population group on the UK epidemic would need to be revised. 46
HIV infection in individuals unaware of their anti-HIV-positive status represents a reservoir for ongoing HIV transmission. 49 Although historical records of individual HIV diagnoses are not directly available through the UAPMP, their HIV infection status at the time of clinic attendance is recorded. Individuals whose HIV infection was newly diagnosed were >3 times as likely to be infected with a non-B subtype than those in whom infection remained undiagnosed. HIV infections probably acquired abroad accounted for 90% (1380/1525) of the new diagnoses of heterosexually acquired infections in 2000. 44 This is consistent with the high proportion of non-B subtypes found in this study among newly diagnosed infections, and that infection with a non-B subtype is associated with probable acquisition of infection in areas of the world where multiple subtypes co-circulate. 16,17,20,42
In conclusion, these findings are the basis for national estimates of HIV-1 subtype trends and demographic associations in England and Wales. Our findings indicate a high level of HIV diversity among the heterosexual population in England and Wales and show that non-B subtypes are also prevalent among UK-born individuals. Given the present global pattern of population migration, the level of diversity in this population may be expected to further increase. The genetic variability of HIV-1 has the potential to affect the accuracy of diagnostic tests, antiretroviral treatment, and vaccine strategies, and monitoring the global distribution and subtype dynamics in the epidemic remains essential.
The authors thank Philip Mortimer, Pauline Rogers (PHLS Statistics Unit), and Gary Murphy for their critical appraisal of this article. We also wish to thank Laura Jordan, Christine McGarrigle, Katy Sinka, and Barry Evans at the HIV/STI Division, Communicable Disease Surveillance Centre (CDSC) for provision of demographic data and helpful comments. The pol primer sequences were provided by Pat Cane and Deenan Pillay, PHLS Antiviral Susceptibility Reference Unit, Birmingham. We also thank the many contributors to the UAPMP throughout England and Wales, the HIV Surveillance Group at the Sexually Transmitted and Blood Borne Virus Laboratory (CPHL), and the HIV/STI Division, CDSC.
2. Kunanusont C, Foy HM, Kreiss JK, et al. HIV-1 subtypes and male-to-female transmission in Thailand. Lancet
3. Yirrell DL, Robertson P, Goldberg DJ, et al. Molecular investigation into outbreak of HIV in a Scottish prison. BMJ
4. Trask SA, Derdeyn CA, Fideli U, et al. Molecular epidemiology of human immunodeficiency virus type 1 transmission in a heterosexual cohort of discordant couples in Zambia. J Virol
5. Barlow KL, Tosswill JH, Parry JV, et al. Performance of the Amplicor human immunodeficiency virus type 1 PCR and analysis of specimens with false-negative results. J Clin Microbiol
6. Amendola A, Bordi L, Angeletti C, et al. Underevaluation of HIV-1 plasma viral load by a commercially available assay in a cluster of patients infected with HIV-1 A/G circulating recombinant form (CRF02). J Acquir Immune Defic Syndr
7. Swanson P, Soriano V, Devare SG, et al. Comparative performance of three viral load assays on human immunodeficiency virus type 1 (HIV-1) isolates representing group M (subtypes A to G) and group O: LCx HIV RNA quantitative, AMPLICOR HIV-1 MONITOR version 1.5, and Quantiplex HIV-1 RNA version 3.0. J Clin Microbiol
8. Apetrei C, Descamps D, Collin G, et al. Human immunodeficiency virus type 1 subtype F reverse transcriptase sequence and drug susceptibility. J Virol
9. Descamps D, Apetrei C, Collin G, et al. Naturally occurring decreased susceptibility of HIV-1 subtype G to protease inhibitors. AIDS
10. Descamps D, Collin G, Letourneur F, et al. Susceptibility of human immunodeficiency virus type 1 group O isolates to antiretroviral agents:in vitro
phenotypic and genotypic analyses. J Virol
11. Tatt ID, Barlow KL, Nicoll A, et al. The public health significance of HIV-1 subtypes. AIDS
12. Simon F, Mauclere P, Roques P, et al. Identification of a new human immunodeficiency virus type 1 distinct from group M and group O. Nat Med
13. Vidal N, Peeters M, Mulanga-Kabeya C, et al. Unprecedented degree of human immunodeficiency virus type 1 (HIV-1) group M genetic diversity in the Democratic Republic of Congo suggests that the HIV-1 pandemic originated in Central Africa. J Virol
14. Gurtler LG, Hauser PH, Eberle J, et al. A new subtype of human immunodeficiency virus type 1 (MVP-5180) from Cameroon. J Virol
15. Expert Group of the Joint United Nations Programme on HIV/AIDS. Implications of HIV variability for transmission: scientific and policy issues. AIDS
16. Parry JV, Murphy G, Barlow KL, et al. National surveillance of HIV-1 subtypes for England and Wales: design, methods, and initial findings. J Acquir Immune Defic Syndr
17. Couturier E, Damond F, Roques P, et al. HIV-1 diversity in France, 1996–1998: the AC 11 laboratory network. AIDS
18. Clewley JP, Arnold C, Barlow KL, et al. Diverse HIV-1 genetic subtypes in UK. Lancet
19. Barlow KL, Tatt ID, Cane PA, et al. Recombinant strains of HIV type 1 in the United Kingdom. AIDS Res Hum Retroviruses
20. Yirrell DL, Goldberg DJ, Whitelaw J, et al. Viral subtype and heterosexual acquisition of HIV infections diagnosed in Scotland. Sex Transm Infect
21. Perez-Alvarez L, Cuevas MT, Villahermosa ML, et al. Prevalence of drug resistance mutations in B, non-B subtypes, and recombinant forms of human immunodeficiency virus type 1 in infected individuals in Spain (Galicia). J Hum Virol
22. Alaeus A, Leitner T, Lidman K, et al. Most HIV-1 genetic subtypes have entered Sweden. AIDS
23. Sullivan PS, Do AN, Ellenberger D, et al. Human immunodeficiency virus (HIV) subtype surveillance of African-born persons at risk for group O and group N HIV infections in the United States. J Infect Dis
24. Kent G. HIV infections hit a record high in United Kingdom. BMJ
25. Communicable Disease Reports. AIDS and HIV infection in the United Kingdom: monthly report. Commun Dis Rep Wkly
26. Health Protection Agency, SCIEH, ISD, National Public Health Service for Wales, and CDSC Northern Ireland and the UASSG. Renewing the Focus. HIV and Other Sexually Transmitted Infections in the United Kingdom in 2002
. London: Health Protection Agency; 2003.
27. Gill ON, Adler MW, Day NE. Monitoring the prevalence of HIV. BMJ
28. Nicoll A, Gill ON, Peckham CS, et al. The public health applications of unlinked anonymous seroprevalence monitoring for HIV in the United Kingdom. Int J Epidemiol
29. Boom R, Sol CJ, Salimans MM, et al. Rapid and simple method for purification of nucleic acids. J Clin Microbiol
30. Belda FJ, Barlow KL, Clewley JP. Subtyping HIV-1 by improved resolution of heteroduplexes on agarose gels. J AIDS Hum Retrovirol
31. Tatt ID, Barlow KL, Clewley JP. A gag
gene heteroduplex mobility assay for subtyping HIV-1. J Virol Methods
32. Delwart EL, Herring B, Rodrigo AG, et al. Genetic subtyping of human immunodeficiency virus using a heteroduplex mobility assay. PCR Methods Appl
33. Cane PA, de Ruiter A, Rice P, et al. Resistance-associated mutations in the human immunodeficiency virus type 1 subtype C protease gene from treated and untreated patients in the United Kingdom. J Clin Microbiol
34. Hall TA. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp Ser
35. Posada D, Crandall KA. MODELTEST: testing the model of DNA substitution. Bioinformatics
36. Paup*. Phylogenetic Analysis Using Parsimony (*and Other Methods) [computer program]. Sinauer Association; 1999.
37. Devereux H, Loveday C, Burke A, et al. The prevalence of non-B subtype HIV-1 in a London HIV/AIDS outpatient clinic. AIDS
38. Holguin A, Aracil B, Alvarez A, et al. Prevalence of human immunodeficiency virus type 1 (HIV-1) non-B subtypes in foreigners living in Madrid, Spain, and comparison of the performances of the AMPLICOR HIV-1 MONITOR version 1.0 and the new automated version 1.5. J Clin Microbiol
39. Johnson AM, Mercer CH, Erens B, et al. Sexual behaviour in Britain: partnerships, practices, and HIV risk behaviours. Lancet
40. Montavon C, Toure-Kane C, Liegeois F, et al. Most env
subtype A HIV-1 viruses circulating in West and West Central Africa are similar to the prototype AG recombinant virus IBNG. J Acquir Immune Defic Syndr
41. 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
42. Arnold C, Barlow KL, Parry JV, et al. At least five HIV-1 sequence subtypes (A, B, C, D, A/E) occur in England. AIDS Res Hum Retroviruses
43. Brown AJ, Lobidel D, Wade CM, et al. The molecular epidemiology of human immunodeficiency virus type 1 in six cities in Britain and Ireland. Virology
44. Unlinked Anonymous Surveys Steering Group. Prevalence of HIV and Hepatitis Infections in the United Kingdom 2000
. London: Department of Health; 2001.
45. Gupta SB, Gilbert RL, Brady AR, et al. CD4 cell counts in adults with newly diagnosed HIV infection: results of surveillance in England and Wales, 1990–1998. CD4 Surveillance Scheme Advisory Group. AIDS
46. Cliffe S, Mortimer J, McGarrigle C, et al. Surveillance for the impact in the UK of HIV epidemics in South Asia. Ethn Health
47. Su L, Graf M, Zhang Y, et al. Characterization of a virtually full-length human immunodeficiency virus type 1 genome of a prevalent intersubtype (C/B′) recombinant strain in China. J Virol
48. Lole KS, Bollinger RC, Paranjape RS, et al. Full-length human immunodeficiency virus type 1 genomes from subtype C-infected seroconverters in India, with evidence of intersubtype recombination. J Virol
49. Catchpole MA, Mercey DE, Nicoll A, et al. Continuing transmission of sexually transmitted diseases among patients infected with HIV-1 attending genitourinary medicine clinics in England and Wales. BMJ