recently shown to support entry of both M- and T-tropic isolates free survival in HIV-1 infected adults [18, 20, 25 27].

Transcription

1 1019 Impact of Heterozygosity for the Chemokine Receptor CCR5 32-bp Deleted Allele on Plasma Virus Load and CD4 T Lymphocytes in Perinatally Human Immunodeficiency Virus Infected Children at 8 Years of Age F. Buseyne, G. Janvier, J. P. Teglas, S. Ivanoff, Unité de Virologie et Immunologie Cellulaire, Institut Pasteur and M. Burgard, E. Bui, M.-J. Mayaux, S. Blanche, Laboratoire de Virologie and Unité d Immunologie et d Hématologie, Hôpital Necker-Enfants Malades, Paris; INSERM U292, Hôpital du C. Rouzioux, and Y. Rivière Kremlin-Bicêtre, Le Kremlin-Bicêtre, France The CCR5 gene encodes one of the major human immunodeficiency virus type 1 (HIV-1) coreceptors. A 32-bp deletion in this gene (Dccr5) is associated with relative resistance to disease progression in heterozygous HIV-1 infected persons. The effect of this mutation on virologic and immunologic parameters was determined in a cohort of 45 perinatally HIV-1 infected children prospectively followed after 5 years of age. At a median age of 8.3 years, heterozygous children had significantly lower virus load than homozygous children (median, 3.3 vs. 4.1 log copies/ml, P õ.009) and higher percentages of CD4 T cells (median, 26% vs. 17%, P õ.07). However, there was no discernible influence of the CCR5 genotype on the percentages of CD8 T cells (P õ.27) or on HIV-specific cytotoxic T lymphocyte activities (P õ.65). There was a trend for lower rates of progression to AIDS (CDC stage C) in heterozygous children. These data confirm a major role for the CCR5 coreceptor in HIV-1 pathogenesis in children. Human immunodeficiency virus type 1 (HIV-1) can broadly amounts by the relatively HIV-1 resistant CD4 T cells from be divided into macrophage-tropic (M-tropic) and T cell line some uninfected but multiply exposed individuals [4, 14]. A tropic (T-tropic) isolates. M-tropic isolates infect primary mac- 32-bp deletion in the second extracellular loop of CCR5 results rophages but fail to infect transformed T cell lines, while in the synthesis of a severely truncated form of CCR5 (Dccr5) T-tropic isolates have the reciprocal tropism. Both classes of that fails to reach the cell surface [15, 16]. As a result, the HIV-1 efficiently infect CD4 T cells isolated from peripheral cells are relatively resistant to infection by M-tropic viruses blood. The molecular basis of HIV-1 tropism appears to lie in [15 17]. Genetic analyses of the CCR5 genotype showed that the ability of the HIV envelope protein to interact with different homozygosity for the mutant Dccr5 allele seems to confer coreceptors. T-tropic viruses tend to use CXCR4 [1 3], resistance to infection with HIV-1 [16, 18 20]. However 4 whereas M-tropic viruses use primarily CCR5 [4 6]. In addition, cases of HIV-1 infection of homozygous Dccr5/Dccr5 individcoreceptors a minority of M-tropic isolates uses CCR2b or CCR3 as uals have been reported [21 24]. Most studies showed that [7, 8], and three putative chemokine receptors were heterozygosity for the 32-bp deletion leads to longer AIDS- recently shown to support entry of both M- and T-tropic isolates free survival in HIV-1 infected adults [18, 20, 25 27]. [9 12]. Some virus isolates can use several coreceptors [7, 8]. The protective effect of the Dccr5 allele was limited to Several observations show that one of these viral corecep- patients that harbor non syncytium-inducing viruses [27]. tors, CCR5, plays an important role in HIV-1 pathogenesis. Most viruses isolated from perinatally HIV-1 infected new- CCR5 is the receptor for the b-chemokines RANTES, macro- borns during the first year of life have the non syncytiuminducing phage inflammatory protein (MIP)-1a, and MIP-1b. These phenotype [28]. This viral phenotype is closely linked molecules inhibit replication of M-tropic HIV-1 isolates, and to CCR5 use as a coreceptor [29 32]. Most primary isolates they are some of the major HIV-suppressive factors produced obtained early in infection from 9 perinatally HIV-1 infected by CD8 T lymphocytes [13]. They are released in higher children are sensitive to RANTES, MIP-1a, and MIP-1b [33]. These data led us to postulate that the CCR5 genotype may be relevant in pediatric HIV pathogenesis, as was previously shown for HIV-1 infected adults. A study of 512 children Received 18 March 1998; revised 5 June born to HIV-infected mothers in the Paris area showed that Informed consent was obtained from the legal guardians of the children. The project received relevant authorization from the Ethical Committee (Comité heterozygosity for the Dccr5 allele did not prevent HIV trans- d éthique). mission but strongly reduced the progression to AIDS and to Financial support: Institut Pasteur and Agence Nationale de la Recherche severe immunodeficiency in infected children [34]. However, sur le SIDA. Y.R. is an Elisabeth Glaser scientist. Reprints or correspondence: Dr. Y. Rivière, Unité de Virologie et Immunologie most children were õ7 years of age, and the persistence of the Cellulaire, Institut Pasteur, 28 rue du Dr Roux, Paris, France (rivi- protective effect of the mutated allele has not been studied. ere@pasteur.fr). We present a group of children included in the longitudinal The Journal of Infectious Diseases 1998;178: follow-up of cytotoxic T lymphocyte (CTL) activities in the 1998 by the Infectious Diseases Society of America. All rights reserved /98/ $02.00 Paris area who have been followed for a longer period [35].

2 1020 Buseyne et al. JID 1998;178 (October) Materials and Methods HIV-1 infected children followed at Institut Pasteur for CTL Patient population. We studied children previously included in activity [35]. None of the children tested was homozygous for the longitudinal follow-up of the HIV-specific CTL activities program the Dccr5 allele. None of the 22 children of sub-saharan Africa at the Institut Pasteur, Paris [35]. These infants are followed at the carried the Dccr5 allele, and they were excluded from the study Département de Pédiatrie, Hôpital Necker, Paris. Some of these chil- group. Seven (18%) of 38 children of caucasian origin were dren were enrolled in the French Pediatric Cohort and have been CCR5/Dccr5 heterozygous. Two of 7 non-caucasian HIV-1 included in the previous analysis of the CCR5 genotype in mother- infected children were CCR5/Dccr5 heterozygous. to-child HIV transmission [34]. Children of sub-saharan origin were Relationship between CCR5 genotype and evolution to CDC excluded from the study. We studied between the CCR5 genotype stage C. Forty-five perinatally HIV-1 infected children were and clinical progression in 45 HIV-1 infected children born to HIVanalyzed for evolution toward disease. At last clinical examina- 1 positive mothers. Biologic data from 30 of these children, who have been followed since 5 years of age, were studied. tion (median age, 9.5 years), 13 (36%) of 36 homozygous Clinical analysis. The diagnosis of HIV infection was based CCR5/CCR5 children had progressed to CDC stage C at a on HIV isolation from peripheral blood mononuclear cells (PBMC) median age of 3.7 years (range, ; table 1). One child in the first months of life or persistence of HIV-specific antibodies (11%) of 9 heterozygous for the Dccr5 allele progressed to after 18 months of age [36]. The clinical status of the children is AIDS and died at 3 months of age. Median ages at last clinical recorded according to the 1994 revised pediatric HIV classification examination were higher for heterozygous children than for system (CDC) [37]. The CDC stage at the last clinical examination homozygous children (table 1), so we may underestimate the recorded in October 1997 was taken to analyze the influence of rates of progression to CDC stage C for CCR5/CCR5 children. the CCR5 genotype on disease progression. Numbers of children receiving no or various combinations of Cytotoxic assays. Detection of HIV-specific cytotoxic activiantiretroviral treatments were not different between the 2 ties in the peripheral blood of children has been described pregroups (Fisher s exact test, P õ.18). In conclusion, heterozyviously [35]. CTL activities from phytohemagglutinin-stimulated PBMC directed against HIV-1 Env, Gag, Pol, and Nef proteins gosity for the Dccr5 allele seems to be associated with a lower were evaluated. Results from CTL assays were expressed as the rate of progression to AIDS in this group of perinatally HIV-1 number of HIV proteins recognized (0, 1, 2, 3, or 4). infected children, although the results did not achieve statistical Viral quantitation. HIV RNA in plasma (stored at 080 C) was significance (Fisher s exact test, P õ.24). quantified by the Amplicor HIV Monitor test (Roche, Neuilly, Relationship between CCR5 genotypes and virus load. France). This technique is based on reverse transcription of the Nine children carried the CCR5/Dccr5 genotype. One of them RNA corresponding to the HIV gag gene, followed by amplificadied at 3 months of age, whereas the 8 other children were tion of cdna by polymerase chain reaction (PCR), in the presence alive at last clinical examination (ages years). Mean of a quantitation standard [38]. The detection limit of the technique was 2.3 log age at biologic examination of children carrying the CCR5/ 10 HIV RNA copies/ml. CCR5 genotyping. PCR amplification of CCR5 sequences was Dccr5 genotype was 8.3 years, so for each child, biologic performed using genomic DNA extracted from cryopreserved Epformed examination closest to 8.3 years was selected. Evaluations per- stein-barr virus transformed B lymphocytes or phytohemagglutinin-stimulated before 5 or after 11 years of age were excluded, as PBMC, as described by Liu et al. [15], with some were data obtained when the children were under three-drug modifications. Primers 5 -CCAGATCTCAAAAAGGTCT-3 and therapy. The wild-type homozygous and CCR5/Dccr5 hetero- 5 -GGTCTTCTCTTTTATTTGTTAGT-3, which flank the 32-bp zygous children had similar ages at examination (CCR5/CCR5: deletion, were used to generate wild-type and deleted fragments median, 8.3 years; range, ; CCR5/Dccr5: median, 8.3 of 244 and 212 bp, respectively. The PCR reaction mixture conyears; range, ). tained 10 pmol of each primer, 5 mm dntps, 1.5 mm MgCl 2, and We investigated the effect of CCR5 genotype on plasma 0.2 U of Taq polymerase in 11 reaction buffer (United States Biochemicals, Cleveland). Each PCR amplification consisted of virus load. Plasma HIV RNA levels from the 2 groups were 40 cycles; the first 5 cycles were 94 C for 30 s, 62 C for 45 s, and 72 C for 90 s, followed by 35 cycles of 94 C for 30 s, 55 C Table 1. CCR5/CCR5 or CCR5/Dccr5 genotype and CDC stage at for 45 s, and 72 C for 90 s. The PCR products were applied to last clinical evaluation of 45 HIV-1-infected children. 2% agarose gel, and the bands were visualized under UV light after ethidium bromide staining. CDC stage CCR5/CCR5 CCR5/Dccr5 Statistical methods. Rates of progression to CDC stage C in children who carried the CCR5/CCR5 or CCR5/Dccr5 genotypes N, A, or B were analyzed by Fisher s exact test. Virus load and lymphocyte No. of children 23 8 subsets from both groups were compared by use of the Mannyears Median length of follow-up, Whitney test. Numbers of HIV proteins recognized by CTL were (range)* 9.9 ( ) 10.4 ( ) C compared by Fisher s exact test. No. of children 13 1 Results Median length of follow-up, CCR5 genotyping. We used a PCR-based assay to detect years (range)* 7.1 ( ) 0.25 the presence or absence of the Dccr5 allele in 67 perinatally * Age at last clinical examination.

3 JID 1998;178 (October) CCR5 Mutant Allele in HIV-1 Infected Children 1021 Figure 1. Virus load and lymphocyte populations of children with CCR5/CCR5 (n Å 22) and CCR5/Dccr5 (n Å 8) genotypes at 8.3 years of age. For each child, data collected at age closest to 8.3 years were selected. A, HIV RNA copies/ml of plasma; broken line indicates cutoff value. B, % CD4 T cells. C, % CD8 T cells. significantly different (Mann-Whitney test, P õ.009). CCR5/ in heterozygous children, but the differences did not achieve CCR5 children had ,000 HIV RNA copies/ml (median, statistical significance (median, 360 and 533/mL, respectively, 24,000; figure 1A). Virus load from CCR5/Dccr5 children P õ.13). Percentages and absolute CD8 T cell numbers were varied between 1700 and copies/ml (median, 5100). not statistically different between the 2 groups (median, 49% Since the numbers of children receiving no, one, or two antiret- and 43% for CCR5/CCR5 and CCR5/Dccr5 children, respectively, roviral treatments are not significantly different between groups P õ.26; figure 1C and data not shown). (Fisher s exact test, P õ.77), these results demonstrate that CTL activities of mitogen-stimulated PBMC from these children heterozygosity for the Dccr5 allele is significantly related to were tested against HIV-1 Env, Gag, Pol, and Nef proheterozygosity lower virus load after long-term pediatric infection. teins. Five CTL classes were defined as the numbers of viral Relationship between CCR5 genotypes and immunologic parameters. antigens recognized by in vitro activated CTL. No difference We next studied the T lymphocyte subpopulations of the distribution in CTL classes was found between the 2 in the same 2 groups of children. Children carrying the deletion groups (P õ.52, table 2). However, 7 of the 20 wild-type had higher percentages of CD4 T cells (median, 17% and homozygous children had no CTL response, but only 1 of 26% for CCR5/CCR5 and CCR5/Dccr5 children, respectively, the 8 heterozygous children was a nonresponder. Numbers of P õ.07; figure 1B). CD4 T cells counts were also higher patients recognizing each protein were not different between

4 1022 Buseyne et al. JID 1998;178 (October) Table 2. Number of HIV-1 antigens recognized by in vitro activated agree with a protective role of the Dccr5 allele against disease cytotoxic T lymphocyte (CTL) lines. progression. No. of children/% In vitro, the efficiency of HIV-1 entry and replication is decreased in CD4 T cells from heterozygous patients [15]. This No. of antigens recognized CCR5/CCR5 CCR5/Dccr5 could lead to lower virus burden and slower CD4 T cell decline in vivo. Early in infection, virus load was decreased in hetero- 0 7 (35) 1 (13) zygous infected adults [19, 26]. For each 1-year period from 1 1 (5) 2 (25) 0 to 6 years of age, we tested 1 3 CCR5/Dccr5 heterozygous 2 6 (30) 3 (38) 3 5 (25) 2 (25) children. Except for the first year of age, most of these children 4 1 (5) 0 had virus loads below the mean of those from wild-type homo- Total 20 8 zygous children (data not shown). At a median age of 8.3 years, we found a significantly lower virus load in heterozygous children (figure 1). Our data show that the effect of heterozygosity persists several years after perinatal infection. This result the 2 groups (data not shown). When results of CTL assays supports the in vivo effect of reduced expression of viral corewere scored as the percent specific lysis at the highest effector ceptor CCR5 on viral replication in children. cell to target cell ratio or as lytic units, no difference was CD4 T lymphocyte decline was found to be delayed in both observed between wild-type homozygous and heterozygous CCR5/Dccr5 heterozygous adults and children [19, 25, 34]. children (data not shown). So, our study shows no impact of Another study did not find a difference in CD4 T lymphocyte CCR5 genotype on the functional antiviral response. counts between homozygous and heterozygous individuals, but only data from the first 2 years of infection were analyzed [26]. We now show that, late in infection, heterozygous children had Discussion higher numbers of CD4 T cells (figure 1). This correlation was weak (P õ.07) but is probably relevant, as exclusion of children Our present work extends the previous observations of a with AIDS did not modify this observation (data not beneficial role of the Dccr5 allele in heterozygous HIV-1 shown). infected children, by the study of children older than 5 years. The three b-chemokines that bind to the CCR5 receptor are Heterozygosity is significantly associated with lower virus load. produced by T cells and are involved in the chemotaxis and It is also linked to a lower rate of progression to AIDS and activation of these same cells [39]. We found no difference in higher percentages of CD4 T cells. However, no link between the HIV-specific CTL response of children heterozygous for the CCR5 genotype and CD8 T cell counts or their anti-hiv the Dccr5 allele compared with homozygous CCR5/CCR5 CTL activities was found. children. This antiviral response might be independent of b- In the present study, 18% of HIV-1 infected children of chemokines, or redundancy of chemokine function and receptor caucasian origin were heterozygous for the Dccr5 allele. This may compensate the lower expression of the CCR5 receptor proportion of heterozygous children in our cohort is significantly on T cells. We did not find any effect of the Dccr5 allele on higher than the previously reported frequency of 9.8% this antiviral response, and this result suggests that inactivation among children from the same area (x 2 test, P õ.05) [34]. of the CCR5 gene has a direct effect on viral entry rather than This difference might be due to a random effect or to the an indirect effect on the immune system. selection in the present study of older children that had survived In conclusion, heterozygosity for the mutation that abrogates HIV infection. Independent determination of CCR5 genotype the expression of a major HIV-1 coreceptor has a strong beneficial from 40 samples in two laboratories with slightly different effect in vivo. This effect is probably mediated by the techniques gave the same results. The frequency of the deleted reduced capacity of cells to support viral replication. allele in this group is in the same range as those observed in adult caucasians [16, 19, 20, 25 27]. As with other investigators, we did not find subjects from sub-saharan Africa origin Acknowledgments to carry the mutated Dccr5 allele [16, 19, 20]. We thank A. Blanchard, C. Callebaut, and D. Rebouillat for Most studies of cohorts of HIV-1 infected adults from Euhelp in designing conditions for the PCR-based assay and C. Khan rope or North America showed slower progression to AIDS in for the HIV quantification. heterozygous than in wild-type homozygous individuals [18 20, 25 27]. A study of 276 infected children showed that the heterozygous genotype delays progression to disease [34]. References Despite the small numbers of children in our group, we found a 1. Oberlin E, Amara A, Bachelerie F, et al. The CXC chemokine SDF-1 is lower incidence of CDC stage C symptoms in the heterozygous the ligand for LESTR/fusin and prevents infection by T-cell-line children at a median age of follow-up of 9.5 years. These data adapted HIV-1. Nature 1996;382:833 5.

5 JID 1998;178 (October) CCR5 Mutant Allele in HIV-1 Infected Children Feng Y, Broder CC, Kennedy PE, Berger EA. HIV-1 entry cofactor: tor 5: studies in populations with contrasting clinical phenotypes, defined functional cdna cloning of a seven-transmembrane, G protein coupled racial background, and quantified risk. Mol Med 1997;3: receptor. Science 1996;272: Balotta C, Bagnarelli P, Violin M, et al. Homozygous D-32 deletion of 3. Bleul CC, Farzan M, Choe H, et al. The lymphocyte chemoattractant SDF- the CCR-5 chemokine receptor gene in an HIV-1 infected patient. 1 is a ligand for LESTR/fusin and blocks HIV-1 entry. Nature 1996; AIDS 1997;11: : Biti R, French R, Young J, Bennetts B, Stewart G, Liang T. HIV-1 infec- 4. Dragic T, Litwin V, Allaway GP, et al. HIV-1 entry into CD4 / cells is tion in an individual homozygous for the CCR5 deletion allele. Nature mediated by the chemokine receptor CC-CKR-5. Nature 1996;381: Med 1997;3: Obrien TR, Winkler C, Dean M, et al. HIV-1 infection in a man homozygous 5. Deng H, Liu R, Ellmeier W, et al. Identification of a major co-receptor for CCR5-D-32. Lancet 1997;349:1219. for primary isolates of HIV-1. Nature 1996;381: Theodorou I, Meyer L, Magierowska M, Katlama C, Rouzioux C. HIV-1 6. Alkhatib G, Combadiere C, Broder CC, et al. CC CKR5: a RANTES, infection in an individual homozygous for CCR5-D-32. Lancet 1997; MIP-1a, MIP-1b receptor as a fusion cofactor for macrophage-tropic 349: HIV-1. Science 1996;272: Eugen-Olsen J, Iversen AK, Garred P, et al. Heterozygosity for a deletion 7. Doranz BJ, Rucker J, Yi Y, et al. A dual-tropic primary HIV-1 isolate in the CKR-5 gene leads to prolonged AIDS-free survival and slower that uses fusin and the b-chemokine receptors CKR-5, CKR-3, and CD4 T-cell decline in a cohort of HIV-seropositive individuals. AIDS CKR-2b as fusion cofactors. Cell 1996;85: ;11: Choe H, Farzan M, Sun Y, et al. The b-chemokine receptors CCR3 and 26. Meyer L, Magierowska M, Hubert JB, et al. Early protective effect of CCR5 facilitate infection by primary HIV-1 isolates. Cell 1996;85: CCR-5 D-32 heterozygosity on HIV-1 disease progression relationship with viral load. AIDS 1997;11:F Loetscher M, Amara A, Oberlin E, et al. TYMSTR, a putative chemokine 27. Michael NL, Chang G, Louie LG, et al. The role of viral phenotype and receptor selectively expressed in activated T cells, exhibits HIV-1 coreceptor CCR-5 gene defects in HIV-1 transmission and disease progression. function. Curr Biol 1997;7: Nature Med 1997;3: Liao F, Alkhatib G, Peden KW, Sharma G, Berger EA, Farber JM. 28. Spencer LT, Ogino MT, Dankner WM, Spector SA. Clinical significance of STRL33, a novel chemokine receptor like protein, functions as a fusion human immunodeficiency virus type 1 phenotypes in infected children. J cofactor for both macrophage-tropic and T cell line tropic HIV-1. J Infect Dis 1994;169: Exp Med 1997;185: Zhang L, Carruthers CD, He T, et al. HIV type 1 subtypes, coreceptor 11. Deng HK, Unutmaz D, KewalRamani VN, Littman DR. Expression cloning usage, and CCR5 polymorphism. AIDS Res Hum Retroviruses 1997; of new receptors used by simian and human immunodeficiency 13: viruses. Nature 1997;388: Mackewicz CE, Barker E, Greco G, Reyesteran G, Levy JA. Do b-chemo- 12. Farzan M, Choe H, Martin K, et al. Two orphan seven-transmembrane kines have clinical relevance in HIV infection? J Clin Invest 1997;100: segment receptors which are expressed in CD4-positive cells support simian immunodeficiency virus infection. J Exp Med 1997;186: 31. Bjorndal A, Deng HK, Jansson M, et al. Coreceptor usage of primary human immunodeficiency virus type 1 isolates varies according to bio- 13. Cocchi F, DeVico AL, Garzino-Demo A, Arya SK, Gallo RC, Lusso P. logical phenotype. J Virol 1997;71: Identification of RANTES, MIP-1a, and MIP-1b as the major HIVsuppressive 32. Connor RI, Sheridan KE, Ceradini D, Choe S, Landau NR. Change in factors produced by CD8 / T cells. Science 1995;270: coreceptor use correlates with disease progression in HIV-1 infected individuals. J Exp Med 1997;185: Paxton WA, Martin SR, Tse D, et al. Relative resistance to HIV-1 infection 33. Scarlatti G, Tresoldi E, Björndal A, et al. In vivo evolution of HIV-1 of CD4 lymphocytes from persons who remain uninfected despite multiple co-receptor usage and sensitivity to chemokine-mediated suppression. high-risk sexual exposure. Nature Med 1996;2: Nature Med 1997;3: Liu R, Paxton WA, Choe S, et al. Homozygous defect in HIV-1 coreceptor 34. Misrahi M, Teglas JP, N Go N, et al. CCR5 chemokine receptor mutant accounts for resistance of some multiply-exposed individuals to HIV- allele in HIV-1 mother-to-child transmission and disease progression in 1 infection. Cell 1996;86: children. JAMA 1998;279: Samson M, Libert F, Doranz BJ, et al. Resistance to HIV-1 infection in 35. Buseyne F, Blanche S, Schmitt D, Griscelli C, Rivière Y. Detection of caucasian individuals bearing mutant alleles of the CCR-5 chemokine HIV-specific cell-mediated cytotoxicity in the peripheral blood from receptor gene. Nature 1996;382: infected children. J Immunol 1993;150: Rana S, Besson G, Cook DG, et al. Role of CCR5 in infection of primary 36. Burgard M, Mayaux MJ, Blanche S, et al. The use of viral culture and macrophages and lymphocytes by macrophage-tropic strains of human p24 antigen testing to diagnose human immunodeficiency virus infection immunodeficiency virus: resistance to patient-derived and prototype in neonates. The HIV Infection in Newborns French Collaborative Study isolates resulting from the DCCR5 mutation. J Virol 1997;71: Group. N Engl J Med 1992;327: Centers for Disease Control and Prevention revised classification 18. Dean M, Carrington M, Winkler C, et al. Genetic restriction of HIV-1 system for human immunodeficiency virus infection in children less infection and progression to AIDS by a deletion allele of the CKR5 than 13 years of age. MMWR Morb Mortal Wkly Rep 1994;43(RR structural gene. Science 1996;273: ): Huang Y, Paxton WA, Wolinsky SM, et al. The role of a mutant CCR5 38. Mulder J, Resnick R, Saget B, et al. A rapid and simple method for allele in HIV-1 transmission and disease progression. Nature Med 1996; extracting human immunodeficiency virus type 1 RNA from plasma: 2: enhanced sensitivity. J Clin Microbiol 1997;35: Zimmerman PA, Buckler-White A, Alkhatib G, et al. Inherited resistance 39. Hedrick JA, Zlotnik A. Chemokines and lymphocyte biology. Curr Opin to HIV-1 conferred by an inactivating mutation in CC chemokine recep- Immunol 1996;8:343 7.