Genetic Restriction of HIV-1 Infection and Progression to AIDS by a Deletion Allele of the CKR5 Structural Gene

See allHide authors and affiliations

Science  27 Sep 1996:
Vol. 273, Issue 5283, pp. 1856-1862
DOI: 10.1126/science.273.5283.1856


  • Fig. 1.

    Mapping of CKR genes. (A) The location of the gene encoding fusin is shown on the map of chromosome 2 in relation to the interleukin-1α (IL1A) gene and the closely related interleukin-8 receptor A and B (IL8RA, IL8RB) and their pseudogene (IL8RBP). (B) The CKR5 gene was also typed in this RH panel and is located in the same cluster on chromosomes 3p21 as CKR1. CKR1, C-C chemokine receptor 1; CKRL1, CKR1-like gene; CKR2B, C-C chemokine receptor 2B. Flanking framework markers are shown above and below the mapped genes, and their centirad (cR) distance from the top of the chromosome is shown to the left. AFMA338ye5, CHLC.GATA8H, and AFMB362wb9 are CA repeats, and WI-6983 is the RII subunit of cyclic AMP-dependent protein kinase. Methods are described in (51).

  • Fig. 2.

    Genotypic markers 37 and HIV-1 infection [G test (39)]. The significance value of the genotype association for each marker is plotted in physical order along each chromosome. The dotted line corresponds to significance at the 1% level for individual tests. The right arrow corresponds to an experiment-wide 1% significance level, estimated with the Bonferroni procedure for multiple samplings [see (43)].

  • Fig. 3.

    Analysis of CKR5Δ32 genotypes with reference to progression to AIDS (43) based on previously described clinical disease categories. (A) Frequency of CKR5Δ32/+ heterozygotes in HIV-1-infected patients: DCG plus MACS, seroconversion patients who progressed to clinical AIDS in ≤7 years compared with patients that did not progress to AIDS in 7 years (8); SFCC, patients that progress to AIDS or CD4+ T cell counts <200 within 10 years of seroconversion compared with patients without AIDS and CD4 counts ≥200 for 10 or more years (35); MHCS, patients with clinical AIDS in ≤10 years of estimated seroconversion compared with patients without clinical AIDS for more than 10 years after estimated seroconversion dates (32); HGDS, patients with CD4 counts ≤200 before 10 years after estimated seroconversion date compared with patients that had CD4 counts ≥200 for 10 or more years after estimated seroconversion (31). (B) Kaplan-Meier survival distribution curves demonstrating the dependence of disease progression on CKR5 genotype in seroconverters from MHCS, SFCC, and DCG (8, 33, 34, 35). A total of 309 patients with known seroconversion dates and for which long-term data were available were followed for development of AIDS [as defined in 1993 definition (45)—progression to an AIDS-defining illness, CD4 counts ≤200, or death]. The HGDS was not included because all patients were seropositive on study entry. The curves were different significantly [χ2 = 8.1, 1 degree of freedom (df), P = 0.005]. Relative hazard equals 0.61 in a Cox proportional hazards model. (C) Kaplan-Meier survival distribution curves demonstrating the dependence of disease progression to AIDS (1993 definition) on the CKR5 genotype, among 148 HIV-1-seropositive members of the SFCC with well-characterized dates of seroconversion who were seen for the study after 1987 (χ2 = 3.9, 1 df, P = 0.05). A value of P = 0.05 would not be significant when a correction for multiple tests is applied. Relative hazard, controlling for age, was 0.59 in a Cox proportional hazards model. The endpoint was defined as the first of two consecutive CD4 counts <200 or diagnosis of AIDS, with followup censored at the beginning of April 1996. CD4 counts were ascertained at 6- to 12-month intervals from the beginning of recruitment in 1984; AIDS diagnoses have been ascertained by active local surveillance by the San Francisco Department of Public Health as well as regular matches with the Centers for Disease Control and Prevention registry of AIDS cases.


  • Table 1.

    AIDS cohorts studied. [A list of researchers and their affiliations for HGDS and MHCS are given in (52).]

    Study nameAcronymRisk groupEnrollmentParticipantsSites*Principal investigatorsReference
    Hemophilia Growth and Development StudyHGDSHemophiliacs1989-199033314E. D. Gomperts, M. W. Hilgartner, W. K. Hoots, S. M. Donfield(31)
    Multicenter Hemophila Cohort StudyMHCSHemophiliacs1985-1990247210J. J. Goedert(32)
    DC Gay Cohort StudyDCGHomosexual men and intravenous drug users19823073J. J. Goedert, R. J. Biggar(33)
    Multicenter AIDS Cohort StudyMACSHomosexual men198450004A. Munoz, J. P. Phair, R. Detels, C. R. Rinaldo Jr., A. J. Saah(34)
    San Francisco City Clinic CohortSFCCHomosexual men1978-198067041S. P. Buchbinder(35)
    AIDS Link to the Intravenous ExperienceALIVEIntravenous drug users1988-199029581D. Vlahov(36)
    • * Sites for HGDS: New York Hosp.-Cornell Med. Ctr.; Childrens Hosp., Los Angeles; Gulf States Hemophilia Ctr., Univ. of Texas Med. Sch., Houston, TX; Children's Mercy Hosp., Kansas City, MO; Louisiana Comprehensive Hemophilia Care Ctr., New Orleans; Univ. of California-San Diego Med. Ctr.; Indiana Hemophilia Comprehensive Ctr., Indianapolis; Children's Hospital of Oklahoma, Oklahoma City; Mt. Sinai Hemophilia Ctr., New York; Nebraska Regional Hemophilia Ctr., Omaha; Santa Rosa Med. Ctr., San Antonio, TX; Children's Hosp. of Michigan, Detroit; Hershey Med. Ctr., PA; and Dept. of Pediatrics, Univ. of Iowa, New England Research Insitutes, Incorporated, Watertown; Baylor College of Med. Sites for MHCS: Mt. Sinai Med. Ctr., New York; Cornell Univ. Med. Ctr.; Tulane Univ. Sch. Med.; Univ. of North Carolina; Cardeza Foundation Hemophilia Ctr., Philadelphia; Children's Hosp. of Philadelphia; Pennsylvania State Univ. Sch. of Med.; Hemophilia Ctr. of West Pennsylvania, Pittsburgh; Case Western Reserve Univ.; and Uniformed Services Univ. of the Health Sci., Washington, DC. Sites for DCG; Washington, DC, and New York. Sites for MACS: Howard Brown Memorial Clinic-Northwestern Univ.; UCLA Sch. of Public Health, Los Angeles; Univ. of Pittsburgh, Graduate Sch. of Public Health; Johns Hopkins Univ. Sch. of Hygiene and Public Health; and Univ. of Maryland Cancer Ctr. Sites for SFCC: Dept. of Public Health, San Francisco; and Center for Disease Control and Prevention, Atlanta. Sites for ALIVE: Johns Hopkins Univ. School of Public Health.

    • † Includes 62 individuals from an intravenous drug use study.

  • Table 2.

    CKR5 genotype distribution among HIV-1-seropositive and HIV-1-seronegative individuals in the same risk group. G tests (39) for departure of three observed CKR5 genotypes within HIV-1-infected versus uninfected compared with overall cohort genotypic frequencies are listed. G tests using only Caucasian individuals are given in parentheses in the G and P columns. NC, no Caucasians.

    CohortRisk groupHIV-1 antibody statusTotalNumber of patients (% of total) with CKR5 genotypeStatistics
    DCGHomosexualPositive137118 (86)19 (14)0 (0)
    menNegative212175 (83)33 (16)4 (2)
    Total3492935242.0 (1.6)0.38 (0.44)
    MACHomosexualPositive265201 (76)64 (24)0 (0)
    menNegative2414 (58)6 (25)3 (17)
    Total2892157039.9 (9.6)0.007 (0.008)
    SFCCHomosexualPositive150110 (73)40 (27)0 (0)
    menNegative4234 (81)5 (12)3 (7)
    Total1931454539.2 (7.9)0.01 (0.02)
    HGDSHemophiliacsPositive133106 (80)27 (20)0 (0)
    Negative10487 (84)17 (16)0 (0)
    Total2371934400.62 (0.1)0.73 (0.85)
    MHCSHemophiliacsPositive192156 (81)36 (19)0 (0)
    Negative191158 (82)26 (14)7 (4)
    Total3833146278.0 (8.9)0.02 (0.01)
    ALIVEIntravenousPositive466457 (98)9 (2)0 (0)
    drug usersNegative3939 (100)0 (0)0 (0)
    Total505496900.1 (NC)0.77 (NC)
    All cohortsHomosexualPositive13431148 (85)195 (15)0 (0)
    men andNegative612508 (83)87 (14)17 (3)
    hemophiliacsTotal195516562821735.0 (33.3)2.5 × 10−8 (5.8 × 10−8)
  • Table 3.

    Survival analysis for progression to AIDS among HIV-1-infected seroconverters of individual cohorts. The number of seroconverters is shown for each cohort, broken down into homozygous wild-type (+/+) and heterozygous (+/Δ32) individuals. The χ2 and uncorrected with 1 df P values for a difference between the +/+ and +/Δ32 survival curves were computed for each cohort using both 1987 and 1992 AIDS definitions (44, 45).

    1987 AIDS definition as survival outcome
    1992 AIDS definition as survival outcome

Stay Connected to Science

Editor's Blog