Basic Science: Original Papers
Emerging recombinant human immunodeficiency viruses: uneven representation of the envelope V3 region
Renjifo, Borisa; Gilbert, Peterb; Chaplin, Betha; Vannberg, Fredrika; Mwakagile, Davisc; Msamanga, Gernardd; Hunter, Davide; Fawzi, Wafaief; Essex, Maxa
From the aHarvard AIDS Institute, the bDepartment of Biostatistics, the eDepartment of Epidemiology and the fDepartment of Nutrition, Harvard School of Public Health, Boston, Massachusetts, USA, the cDepartment of Microbiology and Immunology and the dDepartment of Community Health, Muhimbili Medical Centre, Dar es Salaam, Tanzania.
Correspondence to Max Essex, Harvard AIDS Institute, 651 Huntington Avenue, Boston, MA 02115-6017, USA. Tel: +1 617 432 2334; fax: +1 617 739 8348; e-mail: email@example.com
Received: 9 February 1999; revised: 12 May 1999; accepted: 9 June 1999.
Objective: To determine whether the envelope V3 region from HIV-1 subtypes A, C or D had the same probability of being present in intersubtype recombinant genomes.
Materials and methods: The envelope C2-C5 and the gag p24-p7 regions from one hundred infants infected perinatally in Tanzania were compared using phylogenetic and recombination analysis. Exact binomial and Fisher‚s exact tests were used to assess if various genomic regions were more likely to be overrepresented in intersubtype recombinants.
Results: Of one hundred HIV-1 positive infants analyzed, twenty-two (22%) showed exclusively subtype A sequence in gag and env. Subtype C accounted for twenty-two infants (22%) whereas nineteen infants (19%) were infected by HIV-1 subtype D. Intersubtype recombinant genomes accounted for thirty-seven infections (37%). The V3 region from subtype A was found in all fifteen A-D recombinants (P=0.00003) and the V3 region from subtype C was found in all twelve C-D recombinants (P=0.0002). Conversely, subtype D gag sequences were preferentially represented in the gag of A-D recombinants (P=0.0003) as well as C-D recombinants (P=0.002). In A-D recombinants, the V3 region of subtype A was generally surrounded by subtype A C3-C5 sequences. In contrast, the V3 region from subtype C was surrounded by subtype D C3-C5 sequences in C-D recombinants. Significant differences were not found in the number of subtype A or subtype C sequences in A-C recombinants.
Conclusion: We have shown that several recombinant HIV-1 viruses have been generated and efficiently transmitted to infants in Tanzania. The recombination patterns showed that the V3 region of subtypes A or C was always selected in A-D and C-D recombinants. This selection suggests that the fitness of subtype D-V3 in perinatal transmission may be reduced with respect to V3 from subtype A and/or subtype C. The elevated number of recombinants transmitted perinatally suggests that co-infection or super-infection by two HIV-1 subtypes is not uncommon in this population.
The human immunodeficiency virus type 1 (HIV-1) pandemic is caused by at least 10 subtypes designated A to J within the M group[1-3]. The global HIV-1 pandemic is made up of distinct regional epidemics with unique subtypes and risk factors. Countries in sub-Saharan Africa account for more than half the HIV-1 infections in the world. In Africa, heterosexual and perinatal transmissions are the most common routes of infection. HIV-1 subtypes A and D predominate in the epidemic of East Africa, and HIV-1 subtype C has dominated the expanding epidemic in Southern Africa and Ethiopia.
Perinatal transmission of HIV-1 is the only major source of infection in children. Perinatal transmission rates of 15 to 43% have been described in different geographical locations. HIV-1 transmission may occur during pregnancy, at delivery, or later through infected milk. Several maternal factors, including seroconversion during pregnancy, high viral load, low CD4 cell counts, and progression to AIDS have been associated with higher rates of perinatal transmission[7-14]. Virological determinants and their role in perinatal transmission have not been well established. HIV-1 in the infant may represent a minor viral quasispecies from the mother, and macrophage-tropic non-syncytium-inducing viruses are the most frequently transmitted viral phenotype[15-22]. It is still unknown which and how many genetic determinants drive this selection. In addition, it is still unknown if HIV-1 subtypes differ in their perinatal transmission rates.
Due to the ability of HIV-1 to recombine and the presence of multiple subtypes in the population, several recombinant genomes have been described, some of which showed different HIV-1 subtypes between the gag and env genes[23-25]. The biological relevance of recombinant genomes in the HIV-1 epidemic has been well recognized in Thailand with the spread of HIV-1 subtype E in the heterosexual population. We recently reported that HIV-1 subtype C is expanding in Tanzania where HIV-1 subtype A and D once dominated the epidemic. By studying recombinant genomes in a population such as this, where multiple subtypes are represented at significant levels, it may be possible to map viral determinants associated with the preferential expansion of the new viruses. Since variation in the envelope gene, particularly the V3 region, has been shown to correlate with co-receptor affinity and cell tropism as well as immune evasion, we evaluated characteristics of HIV-1 env from 100 infants neonatally exposed to different subtypes of HIV-1.
The samples analyzed were from infants participating in a randomized double-blind trial to determine whether vitamin supplements could reduce the rate of perinatal transmission of HIV-1 in Dar es Salaam, Tanzania. Information regarding the treatment received by the infant-mother pairs is still under code and not available at this time. The samples in this report represented the first 100 HIV-1-positive infants for which the gag and env genes were examined. Whole blood was collected from infants ranging from 3 to 50 weeks of age.
PCR amplification, cloning and DNA sequencing
Peripheral blood mononuclear cell separation, lysate preparation, and determination of HIV-1 infection by polymerase chain reaction (PCR) have been previously described (27). A nested PCR amplification protocol was performed to amplify the C2-C5 envelope regions and the 3′-p24/5′-p7 region of gag. For envelope amplification, twenty microliters of crude cell extract were used for the first PCR amplification with primers vpu157 (position 6209-6228 according to HxB2 numbering (3)), vpu232 (6283-6303), p130 (7945-7924) and p131 (7970-7948). Four microliters of the first PCR amplification was subjected to nested PCR with primers Bstq2+ (6858-6880) and env1556 (7815-7783). Cycles for both envelope amplifications consisted of 39 cycles of 1 min at 94°C, 1 min at 52°C, 3 min at 72°C and a final extension cycle for 10 min at 72°C. Two individual reactions with HIV-1-negative cell extracts and a reaction without DNA were included in every set of PCR reactions. To amplify gag sequences, 20μl of cell extract was used for the first PCR amplification with primers p108 (764-782) and p109 (2281-2264). The nested gag PCR amplification was done with primers p91 (1550-1572) and p92 (2043-2024). PCR conditions were similar to those for envelope amplification but annealing was done for 1 min. Amplicons were gel-purified using a Gel DNA purification kit (Boehringer Mannheim, Indianapolis, IN, USA) and cloned into a TopoVector (Invitrogen, Carlsbad, CA, USA). Individual bacterial colonies were screened for the presence of the insert with restriction enzyme analysis of plasmid DNA. Env and gag clones were sequenced in both directions by cycle sequencing with dye terminators using an ABI 373 system (Applied Biosystems, Foster City, CA, USA). In some samples, a new sequence was obtained directly from PCR products to corroborate the sequences obtained from cloned fragments. Nucleotide sequences have been submitted to GenBank under accession numbers AF038052- AF038064, AF038067-AF038075, AF038077- AF038080, AF038082-AF038091, AF038093- AF038094, AF038098, AF038100-AF038106, AF038108-AF038110, AF038112-AF038121, AF106332-AF106472.
Individual gag and env sequences were classified into HIV-1 subtypes by comparison with control HIV-1 subtype sequences from the Los Alamos HIV-1 Sequence Database. Nucleotide sequences from samples and controls were aligned using the Clustal W software package. Alignments were subjected to phylogenetic analysis using the neighbor-joining method and the Kimura two-parameter model for calculation of distances. Bootstrap resampling (1000 replicas) was used to assess the robustness of the phylogenetic trees. In all cases, clones from one infant clustered together and separately from any clones from other infants.
Identification of intragenic recombinants
The Recombination Identification Program (RIP) and neighbor-joining trees were used to identify gag and env intragenic recombinants. Nucleotide sequences from one representative clone of each infant was compared to a master gag and env nucleotide consensus alignment of HIV-1 subtypes. A four-sequence alignment including the consensus sequence of the two putative subtype parents and an outgroup sequence (SIVcpz GAB) was used to localize breakpoints. To further establish the presence of two subtypes, nucleotide sequences on both sides of the breakpoint were analyzed separately by constructing phylogenetic trees. The nomenclature of recombinants was based on the subtypes involved and the location in the genome (5′ end to 3′ end) of each subtype. To determine whether the samples with similar recombinant patterns originated independently, the Viral Epidemiology Signature Program Analysis (VESPA) and Monte Carlo randomization were used to assess the statistical significance of differences in amino acids frequencies. Background genome sequences of HIV-1 subtypes A, C, or D corresponded to all the samples in this study classified as non-recombinant subtype A (22), subtype C (22), or subtype D (19).
Subtype contribution in recombinants
For sequences identified as a recombinant of two subtypes, an exact binomial test was used to assess if various genomic regions were more likely to consist of one subtype than the other. The null hypothesis for this test consisted of the V3 region being equally likely to belong to either subtype involved in the recombinant. Fisher‚s exact tests (two-sided) were used to assess whether the proportional representation of subtypes in V3 differed from that in the C3-C5 region of env. A similar analysis was carried out to compare V3 to gag.
Classification of HIV-1 subtypes transmitted perinatally
A summary of the results of phylogenetic analysis and RIP analysis of env and gag sequences is shown in Fig. 1. Of 100 HIV-1 positive infants analyzed, 22 samples (22%) showed exclusively HIV-1 subtype A sequences in both gag and env. HIV-1 subtype C accounted for 22 infections (22%) whereas 19 samples (19%) showed HIV-1 subtype D sequences in gag and env. Thirty-six infants (36%) showed intersubtype recombinant genomes from two HIV-1 subtypes and one sample (1%) showed evidence for recombination between three HIV-1 subtypes.
Not all possible HIV-1 gag and env combinations where found in infants infected with recombinant viruses. Based on the localization of the breakpoints and the subtypes involved in the recombinant, 11 recombination patterns were found in 37 infants infected with recombinants viruses (Fig. 1). In 18 of 37 (49%) recombinants, the breakpoint could not located inside the regions sequenced in gag or env. In the remaining 19 recombinants (51%), it was possible to detect a breakpoint between two subtypes in gag and/or env. Seven recombinants showed breakpoints in gag (three C/A, three D/A, and one C/D) whereas 12 recombinants showed breakpoints in env (nine D/C/D and three D/A/D) (Fig. 1). We did not detect subtype A envelopes with V3 regions from subtypes C or D, nor did we find subtype C envelopes with V3 regions from subtypes A or D. Strikingly, none of the 13 recombinant subtype D envelopes had a subtype D-V3 region.
Recombinants composed of subtype D(gag)-A(env) and D(gag)-D/C/D(env) accounted for 18 of 36 (50%) intersubtype recombinants involving two HIV-1 subtypes. To determine if these two recombination patterns originated from independent recombination events or if they originated from a single recombination event, which expanded into the population, we analyzed them by phylogenetic trees and VESPA. We hypothesized that if recombinants with the same subtype conformation originated from a single recombination event, they would form a distinct cluster under phylogenetic analysis and signature sequences would be found throughout the entire length of recombinant sequences. We compared each region of the recombinants against the corresponding region from non-recombinants genomes belonging to the same subtype. These controls were used to avoid a bias due to geographical origin of the controls. Phylogenetic trees of gag (Fig. 2a), env V3 (Fig. 2b) or env C3-C5 (Fig. 2c) from D(gag)-A(env) recombinants showed that recombinants samples do not cluster when we compared them against 19 non-recombinant subtype D gag sequences or 22 non-recombinant V3 or C3-C5 subtype A env sequences. A total of eight samples were D(gag)-D/C/D(env) and the envelope V3 and C3-C5 regions were analyzed separately since they belong to two different subtypes. Figures 2d, e, and f show that the gag or C3-C5 sequences from the D(gag)-D/C/D(env) recombinants did not cluster together when we compared them to the background subtype D sequences. Although there was a tendency for the V3 region from recombinant samples to cluster together, the bootstrap value for this sub-cluster was too low to support an epidemiological linkage between these samples (Fig. 2). Amino acid analysis of D(gag)-D/C/D(env) sequences using VESPA and Monte Carlo showed that amino acid position 375 (T instead of S) in C3 was preferentially found in the recombinant set (adjusted P value < 0.005) (Table 1).
Significant differences in the V3 region were not found between recombinants and background sequences.
Together, these analyses suggest that either the recombinants originated from independent recombination events or they had been present in the population for a prolonged period. This is supported by the lack of phylogenetic clustering and by the variability of the RIP patterns showing >90% certainty in sequences of the same subtype or recombinant structure (Figs. 1 and 2).
To further address whether a founder effect may have been involved in the distribution of transmitted recombinants, we also amplified and sequenced the long terminal repeat (LTR) from several samples with D(gag)-A(env), D(gag)-D/C/D(env), C(gag)-A(env) or D(gag)-D/A/D(env). The lack of identity between LTR, gag, and env in with D(gag)-A(env), C(gag)-A(env) and D(gag)-D/A/D(env) supports our previous findings that many of these recombinants could have originated via independent events.
Genetic contribution of subtype A, C and D to the recombinants
If HIV-1 subtypes A, C, and D have the same intrinsic ability to be transmitted perinatally as well as to be represented in a recombinant virus, the genetic composition of a recombinant should not have any preferential representation of any particular HIV-1 subtype. We compared three groups of recombinants to determine whether different subtypes had the same representation in recombinants between subtypes A and D, A and C, or C and D. Despite the amplified envelope including C2 to C5 regions, we did not include in the comparison the C2, V4, or V5 regions. Their small size (C2) and frequent insertion/deletions (V4, V5) made it difficult to assign a particular subtype when subjected to phylogenetic analysis as individual nucleotide segments. The contribution of each subtype in V3, C3, C4, C5, and gag for A-D, A-C, and C-D recombinants are shown in Table 2. Subtype A was preferentially represented in the V3 region from A-D recombinants, with all 15 samples possessing a V3 from subtype A (exact binomial test P = 0.00003). Conversely, subtype D was preferentially represented in gag in A-D recombinants (exact binomial test P = 0.0003). A Fisher‚s exact test comparing the proportion A V3 to the proportion of A gag was highly significant (P = 0.00001). Similar results was obtained when we compared the proportion of subtype A C3-C5 to subtype D gag. This suggests that A-D recombinants with A V3, C3-C5 and D gag may hold an advantage for perinatal transmission. On the other hand, comparison the proportion of subtype A in V3 to the proportion of subtype A in C3-C5 was not significant (P= 0.22). Based on this result, it did not appear that a V3 region of subtype A was a determinant separate from the C3-C5 region of env. All (12 of 12) C-D recombinants had a V3 region from subtype C (P = 0.0002) whereas 0.5 had a gag from subtype C (P=0.002). The Fisher‚s exact test comparing proportions was highly significant (P = 0.0001). Unlike A-D recombinants, the proportion of C-D recombinants with V3 region of subtype C was larger than the proportion of C-D samples with C3-C5 region of subtype C (P = 0.0013), suggesting that the presence of subtype C in the V3 region could be a determinant separate from C3-C5. The proportion of subtype A or subtype C sequences in A-C recombinants did not reach statistical significance even though the number of subtype A sequences was higher than the number of subtype C sequences.
Mother-to-child-transmission of HIV-1 is a multifaceted phenomenon involving host resistance factors as well as the properties of the transmitted virus. It has been shown that viral samples taken from young infants are substantially homogeneous, suggesting that one or few infectious units infected the infant[15-19]. Phenotypic studies of transmitted viruses suggested that most early viral isolates from perinatally infected children are tropic for monocyte-derived macrophages (MDM). These results raised the possibility that sequences in the V3 region could be responsible for escaping maternal immune responses while preserving determinants specific for tropism of cells exposed around or during birth.
At the present time, the viral genetic determinants responsible for perinatal infection as well as whether genetic differences detected between HIV-1 subtypes have biological consequences are largely unknown. As such, intersubtype recombinants could represent useful molecular tools to map viral determinants associated with perinatal transmission in the context of HIV-1 subtypes. In this study, we found 22% of infants infected with HIV-1 subtype A, 22% infected with HIV-1 subtype C, 19% infected with HIV-1 subtype D, and 37% percent infected with recombinant viruses. These estimates do not indicate the subtype-specific transmission rates. It is necessary to identify the prevalence of HIV-1 subtypes in the mothers to determine whether HIV-1 subtypes A, C, D, and recombinants have different rates of transmission.
Analysis of genomic patterns showed that viruses with similar recombination patterns infected several infants. Despite the fact that these recombinant patterns clearly originated from independent events involving different subtypes, they all supported a selection bias for the V3 region. These recombinants could be grouped into two main categories _ D/A/D(env) and D/C/D(env). Alike D/A/D envelopes occurred with A(gag) or D/A(gag) sequences. Similarly, resembling D/C/D(env) occurred with D(gag) or A(gag) (Fig. 1). The fact that V3 regions from subtype D were never found in A-D or in C-D recombinants, suggests that the fitness of D-V3 for perinatal transmission is reduced with respect to V3 from subtype A and/or subtype C.
Viruses isolated from adult individuals shortly after seroconversion and during the asymptomatic phase of HIV-1 infections are mainly MDM-tropic and non-syncytium-inducing (NSI). Viral isolates from perinatally infected children have likewise been described as MDM-tropic and NSI [34,36]. The env V3 region has been shown to be a major determinant influencing viral tropism for macrophages and the ability of a virus isolate to cause syncytia. More recently, the β-chemokine receptor CCR5 has been shown to be the major coreceptor for primary non-T cell line-adapted viruses, and the agr;-chemokine receptor CXCR4 to act as the co-receptor for primary T cell line-adapted viruses[37,38]. Differences in coreceptor usage between different HIV-1 subtypes have recently been described. All (15 of 15) HIV-1 subtype C isolates used the CCR5 co-receptor exclusively. Subtype A isolates preferentially used CCR5, but some isolates used CXCR4 or CCR5 and CXCR4. Since viral determinants for macrophage and T-cell tropism are located in the V3 region, several reports have shown that sequences in the V3 region also determine coreceptor usage of HIV-1 isolates[40-42]. It is plausible that recombinant D/A/D and D/C/D envelopes might have acquired an enhanced ability to infect MDM cells, providing them a selective advantage for transmission over non-recombinant subtype D viruses. It has been shown that HIV-1 subtypes A and C have the most similar V3 amino acid sequences among phylogenetically distinct subtypes, while HIV-1 subtype D viruses have the most divergent set of V3 regions. This selection of conserved V3 regions in subtypes A and C has been proposed as the result of selection for transmission[44,45]. The uneven representation of V3 sequences from subtypes A, C, and D in recombinant samples is compatible with the natural history of subtypes in Tanzania. HIV-1 subtype C appears to have been introduced after HIV-1 subtypes A and D were already established and has since quickly become one of the major subtypes, while HIV-1 subtype D infections appears to have declined in Dar es Salaam.
How and where these recombinants were generated and transmitted is still unknown. Contrary to the subtype E and subtype G subtypes for which parent sequences have not been found, parent genomes are currently circulating in the Dar es Salaam population[24,27]. Comparisons between recombinant sequences and sequences from non-recombinant viruses using phylogenetic and VESPA analysis showed that recombinant regions were indistinguishable from non-recombinant parent sequences. Since we found the same subtype in all clones from each infant, we could probably rule out that recombinant viruses were generated and preferentially expanded in the infant (Renjifo and Essex, unpublished results). Each of the major recombinant patterns could have originated as a single recombination event in one individual and expanded in the general population in Tanzania. Our analysis of samples with D(gag)-A(env) and D(gag)-D/C/D(env) did not support this possibility. However, it is still plausible that these recombinants originated a long time ago, allowing for saturation of mutations that made the recombinants indistinguishable from non-recombinant viruses. If this is the case, these recombinants appear to have gained an advantage for transmission in the general population, and the V3 region might represent an important region for this selective expansion.
In summary, we have shown that a variety of recombinant HIV-1 viruses have been generated and efficiently transmitted to infants in Tanzania. The recombination patterns in these samples suggest that the V3 region could be a key determinant for transmission. The elevated number of recombinants transmitted perinatally suggests that co-infection or super-infection by two HIV-1 subtypes is not uncommon in this population.
We thank members of the Tanzania Vitamin Group. We are grateful to F. McCutchan and P. Kanki for providing comments on this manuscript and to R. Rawat for editorial assistance. This study was supported in part by NIH grants R35 CA39805 and RO1 HD32257. D.M. was supported by grant D43 TW00004 from the Fogarty International Center.
1. Louwagie J, McCutchan FE, Peeters M, et al. Phylogenetic analysis of gag genes from 70 international HIV-1 isolates provides evidence for multiple genotypes. AIDS 1993, 7:769-780.
2. Louwagie J, Janssens W, Mascola J, et al. Genetic diversity of the envelope glycoprotein from human immunodeficiency virus type 1 isolates of African origin. J Virol 1995, 69:263-271.
3. Myers, G, Korber B, Foley B, Jeang K-T, Mellors JW, Wain-Hobson S. Human Retroviruses and AIDS, 1997: a Compilation and Analysis of Nucleic Acid and Amino Acid Sequences. Los Alamos, New Mexico: Los Alamos National Laboratory; 1997.
4. Stanecki KA, Way PO. The dynamics of the HIV/AIDS pandemic. In AIDS in the World II. Edited by Mann J, Tarantola D. London: Oxford University Press; 1996:41-56.
5. Janssens W, Buve A, Nkengasong JN. The puzzle of HIV-1 subtypes in Africa. AIDS 1997, 11:705-712.
6. The Working Group on Mother-To-Child Transmission of HIV. Rates of mother-to-child transmission of HIV-1 in Africa, America, and Europe: results from 13 perinatal studies. J Acquir Immune Defic Syndr Hum Retrovirol 1995, 8:506-510.
7. Albert J, Gaines H, Sonnerborg A, et al. Isolation of human immunodeficiency virus (HIV) from plasma during primary HIV infection. J Med Virol 1987, 23:67-73.
8. Anderson RM, Medley, GF: Epidemiology of HIV infection and AIDS: incubation and infectious periods, survival and vertical transmission. AIDS 1988, 2(suppl 1): S57-S63.
9. Blanche S, Rouzioux C, Moscato ML, et al. A prospective study of infants born to women seropositive for human immunodeficiency virus type 1. HIV Infection in Newborns French Collaborative Study Group. N Engl J Med 1989, 320:1643-1648.
10. The European Colaborative Study. Mother-to-child transmission of HIV infection. Lancet 1988, ii:1039-1043.
11. Goedert JJ, Duliege AM, Amos CI, Felton S, Biggar RJ. High risk of HIV-1 infection for first-born twins. The International Registry of HIV-exposed Twins. Lancet 1991, 338:1471-1475.
12. Hira SK, Kamanga J, Bhat GJ, et al. Perinatal transmission of HIV-I in Zambia. BMJ 1989, 299:1250-1252.
13. Lallemant M, Baillou A, Lallemant-Le Coeur S, et al. Maternal antibody response at delivery and perinatal transmission of human immunodeficiency virus type 1 in African women. Lancet 1994, 343:1001-1005.
14. Rossi P, Moschese V, Broliden PA, et al. Presence of maternal antibodies to human immunodeficiency virus 1 envelope glycoprotein gp120 epitopes correlates with the uninfected status of children born to seropositive mothers. Proc Natl Acad Sci USA 1989, 86:8055-8058.
15. Wolinsky SM, Wike CM, Korber BT, et al. Selective transmission of human immunodeficiency virus type-1 variants from mothers to infants. Science 1992, 255:1134-1137.
16. Mulder-Kampinga GA, Kuiken C, Dekker J, Scherpbier HJ, Boer K, Goudsmit J. Genomic human immunodeficiency virus type 1 RNA variation in mother and child following intra-uterine virus transmission. J Gen Virol 1993, 74:1747-1756.
17. Ahmad N, Baroudy BM, Baker RC, Chappey C. Genetic analysis of human immunodeficiency virus type 1 envelope V3 region isolates from mothers and infants after perinatal transmission. J Virol 1995, 69:1001-1012.
18. Mulder-Kampinga GA, Simonon A, Kuiken CL, et al. Similarity in env and gag genes between genomic RNAs of human immunodeficiency virus type 1 (HIV-1) from mother and infant is unrelated to time of HIV-1 RNA positivity in the child. J Virol 1995, 69:2285-2296.
19. Scarlatti G, Leitner T, Halapi E, et al. Comparison of variable region 3 sequences of human immunodeficiency virus type 1 from infected children with the RNA and DNA sequences of the virus populations of their mothers. Proc Natl Acad Sci USA 1993, 90:1721-1725.
20. Munoz-Fernandez MA, Navarro J, Obregon E, et al. Immunological and virological markers of disease progression in HIV-infected children. Acta Paediatr, Suppl 1997 421:46-51.
21. Spencer LT, Ogino MT, Dankner WM, Spector SA. Clinical significance of human immunodeficiency virus type 1 phenotypes in infected children. J Infect Dis 1994, 169:491-495.
22. De Rossi A, Pasti M, Mammano F, Ometto L, Giaquinto C, Chieco-Bianchi L. Perinatal infection by human immunodeficiency virus type 1 (HIV-1): relationship between proviral copy number in vivo, viral properties in vitro, and clinical outcome. J Med Virol 1991, 35:283-289.
23. Robertson DL, Hahn BH, Sharp PM. Recombination in AIDS viruses. J Mol Evol 1995, 40:249-259.
24. McCutchan FE, Salminen MO, Carr JK, Burke DS. HIV-1 genetic diversity. AIDS 1996, 10(suppl 3):S13-S20.
25. Robertson DL, Sharp PM, McCutchan FE, Hahn BH. Recombination in HIV-1. Nature 1995, 374:124-126.
26. Weniger BG, Takebe Y, Ou CY, Yamazaki S. The molecular epidemiology of HIV in Asia. AIDS 1994, 8(suppl 2):S13-S28.
27. Renjifo B, Chaplin B, Mwakagile D, et al. Epidemic expansion of HIV type 1 subtype C and recombinant genotypes in Tanzania. AIDS Res Hum Retroviruses 1998, 14:635-638.
28. Fawzi WW, Msamanga GI, Spiegelman D, et al. Randomised trial of effects of vitamin supplements on pregnancy outcomes and T cell counts in HIV-1-infected women in Tanzania. Lancet 1998, 351:1477-1482.
29. 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. Nucl Acids Res 1994, 22:4673-4680.
30. Saitou N, Nei M. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 1987, 4:406-425.
31. Siepel AC, Halpern AL, Macken C, Korber BT. A computer program designed to screen rapidly for HIV type 1 intersubtype recombinant sequences. AIDS Res Hum Retroviruses 1995, 11:1413-1416.
32. Korber BT, Kunstman KJ, Patterson BK, et al. Genetic differences between blood- and brain-derived viral sequences from human immunodeficiency virus type 1-infected patients: evidence of conserved elements in the V3 region of the envelope protein of brain-derived sequences. J Virol 1994, 68:7467-7481.
33. Blackard J, Renjifo B, Mwakagile D, Montano M, Fawzi W, Essex M. Transmission of HIV-1 viruses with intersubtype recombinant LTR sequences. Virology 1999, 254:220-225
34. De Rossi A, Ometto L, Masiero S, Zanchetta M, Chieco-Bianchi L. Viral phenotype in mother-to-child HIV-1 transmission and disease progression of vertically acquired HIV-1 infection. Acta Paediatr, Suppl 1997 421:22-28.
35. Zhu T, Mo H, Wang N, Nam DS, Cao Y, Koup RA, Ho DD. Genotypic and phenotypic characterization of HIV-1 patients with primary infection. Science 1993, 261:1179-1181.
36. Ganeshan S, Dickover RE, Korber, BT, Bryson YJ, Wolinsky SM. Human immunodeficiency virus type 1 genetic evolution in children with different rates of development of disease. J Virol 1997, 71:663-677.
37. Deng H, Liu R, Ellmeier W, et al. Identification of a major co-receptor for primary isolates of HIV-1. Nature 1996, 381:661-666.
38. Feng Y, Broder CC, Kennedy PE, Berger EA. HIV-1 entry cofactor: functional cDNA cloning of a seven-transmembrane, G protein-coupled receptor. Science 1996, 272:872-877.
39. Tscherning C, Alaeus A, Fredriksson R, et al. Differences in chemokine coreceptor usage between genetic subtypes of HIV-1. Virology 1998, 24:181-188.
40. Smyth RJ, Yi Y, Singh A, Collman RG. Determinants of entry cofactor utilization and tropism in a dualtropic human immunodeficiency virus type 1 primary isolate. J Virol 1998, 72:4478-4484.
41. Wang WK, Dudek T, Zhao YJ, Brumblay HG, Essex M, Lee TH. CCR5 coreceptor utilization involves a highly conserved arginine residue of HIV type 1 gp120. Proc Natl Acad Sci USA 1998, 95:5740-5745.
42. Speck RF, Wehrly K, Platt EJ, et al. Selective employment of chemokine receptors as human immunodeficiency virus type 1 coreceptors determined by individual amino acids within the envelope V3 loop. J Virol 1998, 71:7136-7139.
43. Korber BT, MacInnes K, Smith RF, Myers G. Mutational trends in V3 loop protein sequences observed in different genetic lineages of human immunodeficiency virus type 1. J Virol 1994, 68:6730-6744.
44. Zhang LQ, MacKenzie P, Cleland A, Holmes EC, Brown AJ, Simmonds P. Selection for specific sequences in the external envelope protein of human immunodeficiency virus type 1 upon primary infection. J Virol 1993, 67:3345-3356.
45. Chesebro B, Wehrly K, Nishio J, et al. Macrophage-tropic human immunodeficiency virus isolates from different patients exhibit unusual V3 envelope sequence homogeneity in comparison with T-cell-tropic isolates: definition of critical amino acids involved in cell tropism. J Virol 1992, 66:6547-6554.
HIV-1; genetic subtypes; env; gag; V3 region; perinatal; transmission; recombination
© 1999 Lippincott Williams & Wilkins, Inc.
Highlight selected keywords in the article text.