Allelic variants of human beta-chemokine receptor 5 (CCR5 ... - Nature

36 downloads 64 Views 150KB Size Report
of Microbiology, University of Alabama at Birmingham, Birmingham, AL 35294, USA; 5National ... ently regulated by different CCR5 promoter alleles.3 More.
Genes and Immunity (1999) 1, 20–27  1999 Stockton Press All rights reserved 1466-4879/99 $15.00 http://www.stockton-press.co.uk

Allelic variants of human beta-chemokine receptor 5 (CCR5) promoter: evolutionary relationships and predictable associations with HIV-1 disease progression J Tang1,2, C Rivers2, E Karita5, C Costello2, S Allen3, PN Fultz4, EE Schoenbaum6 and RA Kaslow1,2,3 1

Division of Geographic Medicine, Department of Medicine, University of Alabama at Birmingham, Birmingham, AL 35294, USA; Program in Epidemiology of Infection and Immunity, School of Public Health, University of Alabama at Birmingham, Birmingham, AL 35294, USA; 3Department of Epidemiology, University of Alabama at Birmingham, Birmingham, AL 35294, USA; 4Department of Microbiology, University of Alabama at Birmingham, Birmingham, AL 35294, USA; 5National AIDS Control Program, BP 780, Kigali, Rwanda; 6Department of Epidemiology and Social Medicine, Albert Einstein College of Medicine at Bronx Medical Center, Bronx, NY 10467, USA 2

Variability in the natural history of HIV-1 infection has been repeatedly associated with genetic variants in the betachemokine receptor 5 (CCR5) locus. While CCR5 coding sequences have demonstrated relatively limited variation, sequences of its promoter appear polymorphic in all major populations. Our studies revealed five major CCR5 promoter alleles with distributions that differed widely among the four distinct ethnic groups from Kigali, Rwanda and Bronx, New York. In particular, promoter allele P*0103 (G59029-T59353-T59356-A59402-C59653) was largely restricted to black subjects. The promoter allele P*0202 (A59029-C59353-C59356-A59402-T59653) was tightly linked to the slightly less frequent CCR2b-64I, a variant of the CCR2b gene, which is about 12.7 kbp upstream from the promoter region. Another closely related promoter allele P*0201 (A59029-C59353-C59356-A59402-C59653) exclusively carried the far less common CCR5-⌬32, a 32-bp deletion in the CCR5 coding sequence 2 kbp downstream from the promoter. The homozygous P*0201/*0201 genotype can be predicted as a risk factor for more rapid disease progression. Among human, chimpanzee, pig-tailed macaque, and sooty mangabey promoter allelic sequences, the apparent ancestral lineage of the promoter sequence (G59029-T59353-C59356A59402-C59653 = human P*0102) was highly conserved across the primate species analyzed here while P*0201 and P*0202 arose more recently than the other three major alleles. Further effort to establish the mechanism by which CCR chemokine receptor polymorphisms govern the initiation and pathogenesis of primate lentivirus infection apparently requires fully detailed genotypic characterization of the affected populations. Keywords: AIDS; CCR2b; CCR5; CCR5 promoter; HIV-1; genetic variation

Introduction Recent work in predominantly Caucasian AIDS cohorts has revealed associations between several polymorphisms in the CC (beta) chemokine receptor loci and either variable degrees of protection against HIV-1 transmission or variable evolution of the disease that follows.1–4 In addition, extensive heterogeneities have been described in the expression and splicing of CCR5, apparently regulated by different CCR5 promoter alleles.3 More importantly, two CCR5 promoter variants have been associated with contrasting rates of HIV-1 disease pro-

Correspondence: Dr RA Kaslow, Department of Epidemiology, School of Public Health, 220A Ryals Bldg 1665, University Blvd, University of Alabama at Birmingham, Birmingham, AL 35294, USA This work has been supported jointly by grants AI41951 and AI40591 from NIAID, and grant DA04347 from NIDA Received 22 February 1999; revised 29 March 1999; accepted 2 April 1999

gression.5,6 These findings highlight the importance of host genetic factors in HIV/AIDS and may guide development of new measures for the prevention and control of HIV-1-related diseases.7–9 The mechanisms and credibility of these specific genetic associations are still under debate,3,10 especially because some of the CCR variants are tightly linked to each other.2 In this report, we have analyzed polymorphisms at the CCR2b, CCR5, and CCR5 promoter loci in four ethnic groups, with a special emphasis on the relationships of CCR5 promoter allelic variants to other well-characterized markers that have been associated with different outcomes of HIV-1 transmission and disease progression.

Results and discussion Direct allele-level sequencing of PCR-amplified human genomic DNA revealed six CCR5 promoter alleles: five were common and one was observed in only a single

Genetic variations in the human CCR5 promoter J Tang et al

individual (Table 1). The five major allelic variants were defined by five dimorphic sites at nucleotide positions 59029, 59353, 59356, 59402, and 59653 when compared with GenBank sequence U95626. By this definition, allele sequences P*0101 and P*0201 also matched GenBank sequences U95626 and AF031237, respectively. The distributions of CCR2b, CCR5, and CCR5 promoter variants differed widely among distinct ethnic groups (Figure 1). In particular, CCR5-⌬32 was only found in ethnic Caucasians (non-Hispanics and Hispanics), while promoter allele P*0103 was largely restricted to ethnic Africans (African-Americans and Rwandans) (␹2 = 18.3, P ⬍ 0.0001). As a result, there were more CCR5 promoter genotypes (allele combinations) in ethnic Africans than ethnic Caucasians. Among the other variants CCR2b-64I and promoter alleles P*0102 and P*0202 were more frequent in ethnic Africans than in ethnic Caucasians, while the opposite was seen for promoter alleles P*0101 and P*0201 (P ⬍ 0.0001 for all comparisons). Multiple linkages among CCR2b, CCR5, and CCR5 promoter variants were consistent in all populations (Tables 2 and 3). For example, exclusive linkage of promoter allele P*0202, defined by the 59653T, to CCR2b-64I was consistent with an earlier observation in Caucasians.2 Meanwhile, promoter allele P*0201 was the only variant that was linked to CCR5-⌬32. Mutually exclusive linkages with P*0201 and P*0202 implied that CCR2b-64I and CCR5-⌬32 could not exist on the same chromosome. The negative linkage between CCR2b-64I and CCR5-⌬32 as revealed by our analysis confirmed this relationship (Table 2). Our characterization of CCR5 promoter allelic sequences provided simple explanations for previous associations with disease progression (Figure 2) and allowed further predictions. First, alleles P*0201 and P*0202 were the only promoter variants carrying the previously recognized disease-accelerating P1/P1 genotype.6 These two alleles carried exclusively 59029A and not the Table 1 Extended CCR5 downstream promoter alleles obtained by allelic sequencing of PCR products derived from 67 selectively sequenced samples Allele No. 59029b Combinations of other polymorphic 59653d designationsa sequence sites in the minimum promoterc sequence CCR5P*0101 CCR5P*0102 CCR5P*0103 CCR5P*0104 CCR5P*0201 CCR5P*0202 a

15 19 21 1 22 56

G G G G A A

T59353-C59356-G59402-G59648 (= P4) T59353-C59356-A59402-G59648 (= P2) T59353-T59356-A59402-G59648 (= P3) T59353-C59356-A59402-A59648 (= P2) C59353-C59356-A59402-G59648 (= P1) C59353-C59356-A59402-G59648 (= P1)

C C C C C T

Allelic sequences P*0101, P*0102, P*0103, P*0201, and P*0202 have been deposited in GenBank, with accession numbers AF109379, AF109380, AF109381, AF109382, AF109383, respectively. Allele *0104 was only observed once and may deserve further confirmation by additional techniques. The allele nomenclature reflects the two distinct lineages as well as findings in the phylogenetic analyses (Figure 3). bThe numbering of nucleotide positions is based on GenBank sequence U95626. Polymorphism at position 59029 corresponds to the classification in an earlier study.5 cThis region has also been studied previously6 and the alternative allele nomenclature is shown in parentheses. Clearly, P1/P1 homozygotes and 59029G/G represent mutually exclusive groups of promoter genotypes. dThe 59653T is in tight linkage disequilibrium with the disease-delaying CCR2b-64I.2

disease-delaying 59029G/G.5 As a result, P1/P1 homozygotes and 59029G/G-carriers represented two mutually exclusive groups of promoter genotypes. Second, as noted above, P*0202 (59653T) is exclusively linked to CCR2b-64I (Tables 2 and 3), which has been associated with slower disease progression in cohorts of seroconverters.2,4,11,12 Thus, the relationship between P1/P1 genotype and rapid HIV-1 disease progression must depend on P*0201 and not the CCR2b-64I-linked P*0202 (Figure 2b). Despite the reported associations between CCR5 promoter variants and contrasting rates of HIV-1 disease progression, the mechanisms for these effects are not clear. For example, the single dimorphic site (nucleotide position 59653) differentiating P*0201 from the CCR2b64I-linked P*0202 is actually downstream from the minimum promoter region (Figure 3). How the 59653 polymorphism mediates the CCR5 promoter activities is puzzling. Indeed, none of the polymorphic promoter sites map within any of the six putative transcription-factorbinding (TFB) elements,13,14 suggesting either that other unrecognized TFB elements exist in the promoter region or that other unknown variants in linkage disequilibrium at neighboring positions are involved. Sequences from non-human primates did reveal differences in two of the six putative TFB elements (Figure 4a). As would be predicted, pair-wise genetic distance (Figure 4b) and phylogenetic analyses (Figure 4c) placed the chimpanzee CCR5 promoter sequence closer to all human CCR5 promoter alleles than to alleles from other primate species. The ancestral lineage of human CCR5 promoter alleles appeared to be P*0102 (Figure 4b), which shared identical sequences with other non-human alleles at nucleotide positions 59029, 59353, 59356, 59402, and 59653 that define all major promoter alleles in humans. Moreover, P*0201 and P*0202 apparently diverged from the common ancestral sequence more recently than did other alleles (Figure 4c). By the same analyses, polymorphisms at two positions, the G at 59029 and the T at 59353 appeared ancestral to any other lineages. The nomenclature used here deliberately reflects the apparent phylogenetic relationships among these major promoter alleles. HIV-1 represents a close relative of simian immunodeficiency virus (SIV) that is naturally found in African primates including chimpanzees and sooty mangabeys.15 Recent studies have proposed that HIV-1 originated in the chimpanzee subspecies Pan troglodytes troglodytes from Central West Africa.16 Both HIV and SIV prefer CCR5 as their coreceptor for penetrating CD4+ cells;17–19 that preference suggests that CCR5 is commonly expressed in both human and non-human primates. The TFB elements highly conserved in human and nonhuman promoter sequences may represent a critical mechanism for regulating transcription of the CCR5 gene. They may therefore serve as appropriate targets for intervention experiments. Finally, in all human and non-human alleles, exclusive linkage between the G-to-A nucleotide substitution at position 59029 to the T-to-C change at position 59356 presumably indicates either gene conversion or simultaneous point mutations. If the former is true, other upstream (CCR2b coding region) or downstream (CCR5 coding region) polymorphisms are likely to be linked in the same fashion. The CCR2b-64I-59653T linkage served as one example, and CCR5-⌬32-P*0201 linkage served as

21

Genetic variations in the human CCR5 promoter J Tang et al

22

Figure 1 Allelic frequencies of CCR2b-64I, CCR5-⌬32, and CCR5 promoter variants in four ethnic groups from Bronx, New York (nonHispanic Caucasians, Hispanic Caucasians and African-Americans) and Kigali, Rwanda (Africans). The CCR2b-64I is invariably linked to CCR5 promoter allele P*0202 in all four groups, while the ethnic Caucasian-specific CCR5-⌬32 is exclusively linked to CCR5 promoter allele P*0201 (see Tables 3 and 4). The frequency of CCR5-⌬32 in the non-Hispanic Caucasians studied here was lower than those reported elsewhere. Table 2 Linkage among CCR2b, CCR5, and CCR5 promoter variants in subjects from Bronx, New Yorka Variant 1 CCR2b-64I CCR2b-64I

CCR5-⌬32

Variant 2

Nb

1+/2+

1+/2−

1−/2+

1−/2−

OR

95%CI

P

CCR5-⌬32 CCR5 promoter P*0101 P*0102 P*0201 P*0202c 59029Gd promoter *0201

193 =

1

53

11

128

0.22

0.03–1.74

0.118

228 = 228 = 228 = 228 = 228 = 193 =

20 10 23 62 37 12

42 52 39 0 25 0

109 46 99 7 138 97

57 120 67 159 28 84

0.25 0.50 0.40 23.7 0.3 1.87

0.13–0.46 0.24–1.07 0.22–0.73 11.5–49.0 0.16–0.58 1.63–2.14

0.001 0.071 0.002 0.001 0.001 0.002

Subjects from Bronx included non-Hispanic Caucasians (n = 58), Hispanic Caucasians (n = 135), and African-Americans (n = 35). bAfricanAmericans (n = 35) are not included in analysis involving CCR5-⌬32 because the ⌬32 variant is not present in this ethnic group. cPromoter allele P*0202 exclusively carries 59653T (Table 1). d59029G is exclusively linked to 59353T, and both occur at equal frequency in this cohort regardless of the ethnic group.

a

another. Thus, a disease association with an individual polymorphic site can actually signal relationships with one or more additional polymorphisms at multiple interlinked positions, and a combination of genetic, functional and epidemiologic analyses will usually be required to resolve these interrelationships.

Subjects and methods Human subjects We studied human subjects from two distinct populations. The first consisted of all 278 Rwandan women

originally recruited consecutively from the antenatal clinic of the Central Hospital of Kigali, with a specimen available either at the time of death from HIV infection or upon resuming follow-up care after the genocide.20,21 Women of all Rwandan ethnic and geographic origins were seen at the hospital.22 The second population consisted of relatively heavily exposed HIV-infected and uninfected subsets of injecting drug users from three selfidentified ethnic groups in Bronx, New York, including 58 non-Hispanic Caucasians, 135 Hispanic Caucasians, and 35 African-Americans.23,24 Studies in both cohorts conformed to the procedures for informed consent

Genetic variations in the human CCR5 promoter J Tang et al

Table 3 Linkage analysis of CCR2b-64I and individual CCR5 promoter alleles and polymorphic sites in a cohort of 278 Rwandan womena Variant 1 CCR2b-64I

a

Variant 2

1+/2+

1+/2−

1−/2+

1−/2−

OR

95% CI

P

Promoter variants P*0101 P*0102 p*0201 P*0202 59029Gb

15 33 33 129 67

115 97 97 1 63

38 87 73 19 126

110 61 75 129 22

0.4 0.2 0.4 875.8 0.2

0.20–0.73 0.14–0.40 0.21–0.58 115.5–6639.7 0.10–0.33

0.003 0.001 0.001 0.001 0.001

CCR5-⌬32 is absent in this cohort. b59029G is exclusively linked to 59353T, and both occur at equal frequency in this cohort.

Figure 2 Major CCR5 promoter alleles from humans and their reported and predicted relationship to HIV-1 disease progression. (a) The five major CCR5 promoter alleles (deposited in GenBank with accession numbers AF109379, AF109380, AF109381, AF109382, AF109383, respectively) found in various human ethnic groups are defined by five polymorphic sites in the region spanning nucleotide positions 59029 to 59653 (U95626). (b) Relationship between previously recognized CCR2b, CCR5 promoter genotypes and the CCR5 promoter alleles specified in (a). The + and − indicate rapid and slow HIV-1 disease progression, respectively, as reported in several studies.4–6 The predicted effect (+, rapid progression) of the P*0201/*0201 genotype can be inferred from the reported studies.

approved by local and/or sponsoring institutional review boards. Genotyping materials DNA samples used throughout the study were extracted from whole buffy coats or precipitates of cervicovaginal fluids, using the standard salting out procedures25,26 and QIAamp blood kit (QIAGEN Inc, Chatsworth, CA, USA), respectively. Genotyping procedures Genotyping of CCR5-⌬32 by PCR amplification size polymorphism and CCR2b-64I by PCR-RFLP (restriction fragment length polymorphism) was done according to previously established procedures.1,4 The CCR5-⌬32 and CCR2b-64I are approximately 2 kbp downstream and 12.7 kbp upstream, respectively, from the CCR5 downstream promoter that spans from nucleotide 59052G (in the extended loci as defined by GenBank sequence U95626) to 59530C.13

Typing of CCR5 promoter variants was initially achieved through automated sequencing of PCR-amplified products corresponding to nucleotide position 59012G to 59943G (Figure 3). Briefly, the design of allele-specific PCR amplification primers took advantage of the dimorphic site at nucleotide position 59029 (Genbank accession number U95626), 20 bp upstream from the first nucleotide of the minimum CCR5 promoter segment.14 Two separate PCR reactions were performed for each sample: the first using the 50029G-specific forward primer (nucleotides 59012G → 50029G) and common reverse primer (CCR5P-COM3N, 59925G ← 59943G), the second using the 59029A-specific forward primer (59912G → 50029A) and primer CCR5P-COM3N. PCR conditions were optimized to allow sequence-specific amplification determined by the 59029G/A-specific primers: each 12.5 ␮l PCR consisted of 1 × buffer C (60 mm Tris-HCI, pH 8.5, 15 mm (NH4)2SO4, 2.5 mm MgCl2), 0.3 units of AmpliTaq polymerase, together with 0.5 ␮m of each primer, 80 ng genomic DNA, 0.2 mm each of dGTP, dCTP, dTTP and dATP. The PCR mix was subjected to 10 cycles of denaturing at 95°C for 25 s, annealing at 59°C for 40 s, and extension at 72°C for 50 s, followed by 23 cycles of denaturing at 95°C for 25 s, annealing at 55°C for 40 s, and extension at 72°C for 50 s. Samples yielding a 932-bp amplicon were diluted 1:40 in TE buffer (10 mm Tris-HCl, pH 8.0, 2 mm EDTA) and 1.0 ␮l of the diluted product served as template for a 50-␮l second-round PCR with a reverse biotinylated primer (CCR5P-COMR, 59709T ← 59729T) and a forward M13 (18mer)-tailed primer (M13-CCR5P, 59012G → 59028G). After a further 22 cycles of denaturing at 95°C for 25 s, annealing at 59°C for 40 s, and extension at 72°C for 50 s, the resulting products were bound to streptavidin-coated combs and single-stranded templates were generated and sequenced using a Cy5-labelled universal M13 primer (5′-ACA ggA AAC AgC TAT gAC-3′), a Cy5-labelled internal primer (CCR5PSBT, 59291T → 59310A), and the ALFexpress autoload sequencing procedures (Pharmacia Biotech, Inc, Piscataway, NJ, USA). The allelic sequencing strategy outlined above resolved all known polymorphic sites in the constitutive downstream CCR5 promoter region without cloning and its attendant problems due to PCR-introduced nucleotide substitution.27 The reliability of data generated by automated sequencing was further confirmed by additional sequencing of the complementary strand. To expedite the genotyping procedures, 15 pairs of sequence-specific primers (SSP) were used to type the five major CCR5 promoter alleles simultaneously with the CCR5 and CCR2b variants (Tables 4 and 5). The finalized conditions for SSP-based typing were as fol-

23

Genetic variations in the human CCR5 promoter J Tang et al

24

Figure 3 Genomic organization of the polymorphic downstream CCR5 promoter. Filled triangles indicate nucleotide 59052 and 59530 (Based on Genbank accession number U95626), which enclose the region associated with basal and induced promoter activities. The two circled nucleotides are believed to be the major transcription starting sites.34 The highlighted (underlined bold face) fragments are predicted transcription factor binding elements (D-element-like = albumin D-element-like motif; Oct = octamer element; TATA-like = TATA box-like AT-rich sequences; SIE-like = sis/platelet-derived growth factor-inducible element-like; TCF1␣ = T cell factor-1␣ binding site). Sequences in underlined italics represent restriction enzyme cutting sites. The L-shaped arrows denote the boundaries of the putative enhancer (59045 to 59287) and silencer (59287 to 59451).14 The intron-exon dinucleotide boundary sequences are doubly underlined. Among the exon sequences shown in uppercase13 only the underlined 43-bp region has been consistent in multiple studies.14 Based on the various features outlined above, primers (boxed) are designed for 59029G-specific and 59029A-specific amplification followed by sequencing. The arrow below each primer name indicates the polarity (5′ to 3′). The alternative sequences at dimorphic positions are shown immediately above or underneath the individual nucleotides. Alternative numbering (in parentheses) of the polymorphic sites based on GenBank sequence AF031236 is also shown.

lows: (1) Each 10 ␮l PCR solution contained 1 × buffer C, 50–70 ng of genomic DNA, 0.3 units of AmpliTaq polymerase (Fisher Scientific, Norcross, GA, USA), 120 nm of each control primer, 250 nm each of specific primer, 0.4 mm each of dGTP, dCTP, dTTP and dATP, 10% (v/v) glycerol, and 0.02% cresol red; (2) PCR cycle reactions were performed on an UNO-thermoblock (Biometra, Inc, Tampa, FL, USA) first for 10 higher-stringency cycles of denaturing at 94°C for 25 s, annealing at 60°C for 45 s, and extension at 72°C for 45 s, followed by 21 more lower-stringency cycles of denaturing at 94°C for 25 s, annealing at 57°C for 40 s, and extension at 72°C for 40 s; (3) all PCR reactions were done in 96-well microtiter plates, each capable of typing six individual samples; (4) before each cycle reaction, PCR began with a denaturing step at 95°C for 2.5 min; (5) all PCR reactions included a final extension step at 72°C for 6 min; (6) half of each PCR reaction product was loaded directly to 1.7% agarose gel for electrophoresis. The SSP-banding patterns were recorded on photographs of ethidium bromide-stained gels. The use of PCR-SSP was limited to the identification of known genetic variants; novel alleles that may exist in different populations could either give rise to novel reaction patterns or be misclassified by our scheme. The 67 samples from Kigali and New York typed by both sequencing and PCR-SSP demonstrated 100% consistency

between the two methods (Table 1). Thus, new alleles were not encountered even when the number of genotyped samples increased in our studies. Linkage analysis of genetic variants at the CCR2b, CCR5, and CCR5 promoter loci Linkage between allelic variants at different loci are typically estimated with data from families,28 which were not available in the cohorts studied here. Instead, ‘apparent’ 2-locus linkage was determined by ␹2 tests for a 2 × 2 table containing the numbers of individuals with both, one without the other, or neither of the two markers in question.29 A P value documenting a significant association of two alleles (ie, suggesting linkage disequilibrium) did not eliminate the possibility of their coincident occurrence but on opposite chromosomes. Conversely, fully reliable two-locus haplotypes could be established only when alleles at one of the two loci were homozygous, and only such haplotypes were accepted for analysis. Primate species Non-human primates, including two chimpanzees (Pan troglodytes), two pig-tailed macaques (Macaca nemestrina), and two sooty mangabeys (Cercocebus atys) were also studied for their CCR5 promoter polymorphisms. Their CCR5 promoter sequences were amplified by PCR and

Genetic variations in the human CCR5 promoter J Tang et al

25

Figure 4 Phylogenetic relationships among CCR5 promoter sequences from humans (see Table 1 and Figure 1), chimpanzees (Ptr), pigtailed macaques (Mne), and sooty mangabeys (Cat). (a) CCR5 promoter sequences from non-human primate species demonstrate nucleotide substitutions in two putative transcription-factor-binding (TFB) elements: SIE-like (sis/platelet-derived growth factor-inducible elementlike) and TATA-like (TATA box-like AT-rich sequence). (b) Pair-wise genetic distance matrix calculated using Kimura’s two-parameter method.32 The alleles from non-human primates (GenBank accession numbers AF109384, AF115963, AF115964) are identical to human allele P*0102 at all five positions found to be polymorphic in humans. (c) Phenogram based on maximum likelihood method. Numbers next to branches indicate the frequencies supported by bootstrap re-analyses of the sequence data using both neighbor-joining method33 in the PHYLIP package30 and parsimony method in the PAUP package.31 Grouping of human promoter allele P*0102 with P*0103 is not supported by neighbor-joining or parsimony analysis.

Table 4 Oligo primers used for typing major polymorphisms in CCR2b, CCR5, and the CCR5 downstream promoter region Oligo name

CCR2b-5/1S CCR2b-5/2S CCR2b-3/1G CCR5P-5/1S CCR5P-5/2S CCR5P-3/1S CCR5P-3/2S CCR5P-3/3S CCR5P-3/4S CCR5P-3/5S CCR5P-3/6S CCR5P-3/7S CCR5P-5/3S CCR5P-5/4S CCR5P-3/8G CCR5-SP4G CCR5-PM6G a

Specificitya

CCR2b-64V-specific CCR2b-64I-specific general 59029G-specific 59029A-specific 59353T-specific 59353C-specific 59353T-59356T 59356C-specific 59356T-specific 59402G-specific 59402A-specific 59653C-specific 59653T-specific general general general

5′ → 3′ sequences (underlined = polymorphic) TGGGCAACATGCTGGTCG TGGGCAACATGCTGGTCA TGGAAAATAAGGGCCACAGAC GAGTGGAGAAAAAGGGGG GAGTGGAGAAAAAGGGGA AGAATAGATCTCTGGTCTGAAA AGAATAGATCTCTGGTCTGAAG GAGAATAGATCTCTGGTCTAAAA TAGAGAATAGATCTCTGGTCG TAGAGAATAGATCTCTGGTCTA AGAATCAGAGAACAGTTCTTCC AGAATCAGAGAACAGTTCTTCT CAGGAAACCCATAGAAGAC CAGGAAACCCATAGAAGAT GTGGGCACATATTCAGAAG TCATTACACCTGCAGCTCTC TGGTGAAGATAAGCCTCAC

The numbering of nucleotide positions is based on GenBank sequence U95626.

Primer annealing positionsa

46278 → 46295 46278 → 46295; 61619 → 61636 46670 ← 46690 59012 → 59029 59012 → 59029 59353 ← 59374 59353 ← 59374 59353 ← 59375 59356 ← 59377 59356 ← 59377 59402 ← 59423 59402 ← 59423 59635 → 59653 59635 → 59653 59925 ← 59943 62004 → 62023 62182 ← 62200

Genetic variations in the human CCR5 promoter J Tang et al

26

Table 5 Primer mixes used in PCR-SSP-based typing of major polymorphisms in the CCR2b, CCR5, and the CCR5 promoter Primer mixes 1a = CCR2-5/1S + CCR2-3/1G 1b = CCR2-5/2S + CCR2-3/1G 2a = CCR5P-5/1S + CCR5P-3/1S 2b = CCR5P-5/1S + CCR5P-3/2S 2c = CCR5P-5/2S + CCR5P-3/1S 2d = CCR5P-5/2S + CCR5P-3/1S 2e = CCR5P-5/1S + CCR5P-3/3S 3a = CCR5P-5/1S + CCR5P-3/6S 3b = CCR5P-5/1S + CCR5P-3/7S 3c = CCR5P-5/2S + CCR5P-3/6S 3d = CCR5P-5/2S + CCR5P-3/7S 4a = CCR5P-5/3S + CCR5P-3/8G 4b = CCR5P-5/4S + CCR5P-3/8G 5a = CCR5P-5/1S + CCR5P-3/4S 5b = CCR5P-5/1S + CCR5P-3/5S CONTROL = CCR5-SP4G + CCR5PM6G

Specificitya

OATa

Size of amplicon

PCR artifactsb

CCR2b-64Val CCR2b-64Ile and CCR5 59029G-59353T 59029G-59353C 59029A-59353T 59029A-59353C 59029G-59353T-59356T 59029G-59402G 59029G-59402A 59029A-59402G 59029A-59402A 59653C 59653T 59029G-59356C 59029G-59356T CCR5 32-bp deletion region

55°C 55°C 50°C 50°C 50°C 50°C 50°C 50°C 50°C 50°C 50°C 50°C 50°C 50°C 50°C 50°C

413 bp 413 bp and 585 bp 363 bp 363 bp 363 bp 363 bp 364 bp 412 bp 412 bp 412 bp 412 bp 309 bp 309 bp 367 bp 367 bp ⌬32 = 165 bp WT = 197 bp

— +/− +/− +/− +/− +/− — — — — — — +/− — — — —

a

The numbering of nucleotide positions is based on GenBank sequence U95626. Optimal annealing temperature (OAT) is calculated based on the composition of primer sequences and the GC content of the expected amplicon sequences. bThe PCR artifacts considered here are primer-dimer formation and mispriming; several reactions may produce weak false positive or negative amplifications when suboptimal PCR conditions were used. Adjusting PCR conditions helps eliminate artifacts, but sequencing serves as the final tool to resolve any remaining ambiguous SSP-based typings.

analyzed by automated sequencing using the same procedures outlined above. Phylogenetic analyses of CCR5 promoter sequences The phylogenetic relationships among allelic sequences from human and non-human primates species were analyzed using software in the PHYLIP30 and PAUP31 packages. All CCR5 promoter sequences were aligned manually, with gaps introduced to increase alignment of sequences from primates. The final data set (not shown) contained sequences spanning from nucleotide position 59029 (in the extended loci as defined by GenBank sequence U95626) to 59940. Note added in proof More recent functional analysis of the human CCR5 downstream promoter sequence has revealed additional regulatory elements in the regions we studied (Liu R, Zhao XQ, Gurney TA, Landau NR. Functional analysis of the proximal CCR5 promoter. AIDS Res Hum Retroviruses 1998; 14: 1509–1519). Again, the nucleotide positions polymorphic in humans fall beyond these additional elements.

References 1 Dean M, Carrington M, Winkler M et al. Genetic restriction of HIV-1 infection and progression to AIDS by a common deletion allele of the chemokine receptor 5 structural gene. Science 1996; 273: 1856–1862. 2 Kostrikis LG, Huang Y, Moore JP et al. A chemokine receptor CCR2 allele delays HIV-1 disease progression and is associated with a CCR5 promoter mutation. Nature Med 1998; 4: 350–353. 3 Mummidi S, Ahuja SS, Gonzalez E et al. Genealogy of the CCR5 locus and chemokine system gene variants associated with altered rates of HIV-1 disease progression. Nature Med 1998; 4: 786–793. 4 Smith MW, Dean M, Carrington M et al. Contrasting genetic influence of CCR2 and CCR5 variants on HIV-1 infection and disease progression. Science 1997; 277: 959–965.

5 McDermott DH, Zimmerman PA, Guignard F et al. CCR5 promoter polymorphism and HIV-1 disease progression. Lancet 1998; 352: 866–870. 6 Martin MP, Dean M, Smith MW et al. Genetic acceleration of AIDS progression by a promoter variant of CCR5. Science 1998; 282: 1907–1911. 7 Fauci AS. Host factors and the pathogenesis of HIV-induced disease. Nature 1996; 384: 529–534. 8 Chen J-D, Bai X, Yan A-G et al. Inactivation of HIV-1 chemokine co-receptor CXCR-4 by a novel intrakine strategy. Nature Med 1997; 3: 1110–1116. 9 Simmons G, Clapham PR, Picard L et al. Potent inhibition of HIV-1 infectivity in macrophages and lymphocytes by a novel CCR5 antagonist. Science 1997; 276: 276–279. 10 Michael NL, Louie L, Rohrbough AL et al. The role of CCR5 and CCR2 polymorphisms in HIV-1 transmission and disease progression. Nature Med 1997; 3: 1160–1162. 11 Ioannidis JPA, O’Brien TR, Rosenberg PS et al. Genetic effect on HIV disease progression. Nature Med 1998; 4: 536. 12 Garred P. Chemokine-receptor polymorphisms: clarity or confusion for HIV-1 prognosis. Lancet 1998; 351: 2–3. 13 Mummidi S, Ahuja SS, McDaniel BL, Ahuja SK. The human CC chemokine receptor 5 (CCR5) gene. Multiple transcripts with 5′end heterogeneity, dual promoter usage, and evidence for polymorphisms within the regulatory regions and noncoding exons. J Biol Chem 1997; 272: 30662–30671. 14 Guignard F, Combadiere C, Tiffany HL, Murphy PM. Gene organization and promoter function for CC chemokine receptor 5 (CCR5). J Immunol 1998; 160: 985–992. 15 Mindell DP, Schultz JW, Ewald PW. The AIDS pandemic is new, but is HIV new? Syst Biol 1995; 44: 77–92. 16 Gao F, Bailes E, Robertson DL et al. Origin of HIV-1 in the chimpanzee Pan troglodytes troglodytes. Nature 1999; 397: 436–441. 17 Chen Z, Zhou P, Ho DD et al. Genetically divergent strains of simian immunodeficiency virus use CCR5 as a coreceptor for entry. J Virol 1997; 71: 2705–2714. 18 Hoffman TL, Doms RW. Chemokines and coreceptors in HIV/SIV-host interactions. AIDS 1998; 12 (Suppl A): S17-S26. 19 Ward SG, Bacon K, Westwick J. Chemokines and T lymphocytes: more than an attraction. Immunity 1998; 9: 1–11. 20 Lindan CP, Allen S, Serufilira A et al. Predictors of mortality among HIV-infected women in Kigali, Rwanda. Ann Intern Med 1992; 116: 320–328.

Genetic variations in the human CCR5 promoter J Tang et al

21 Lifson AR, Allen S, Wolf W et al. Classification of HIV infection and disease in women from Rwanda. Ann Intern Med 1995; 122: 262–270. 22 Allen S, Lindan C, Serufilira A et al. Human immunodeficiency virus infection in Urban Rwanda. Demographic and behavioral correlates in a representative sample of childbearing women. JAMA 1991; 266: 1657–1663. 23 Gourevitch MN, Klein RS, Schoenbaum EE. Neurosyphilis in patients with human immunodeficiency virus infection. N Engl J Med 1995; 332: 1170. 24 Hartel DM, Schoenbaum EE. Methadone treatment protects against HIV infection: two decades of experience in the Bronx, New York City. Pub Health Rep 1998; 113 (Suppl 1): 107–115. 25 Miller SA, Dykes DD, Polesky HF. A simple and efficient nonorganic salting out procedure for extracting DNA from human nucleated cells. Nucl Acids Res 1988; 16: 1215. 26 Grimberg J, Nawoschik S, Belluscio L et al. A simple and efficient non-organic procedure for the isolation of genomic DNA from blood. Nucl Acids Res 1989; 17: 8390. 27 Tang J, Unnasch TR. Discriminating PCR artifacts using DHDA (directed heteroduplex analysis). BioTechniques 1995; 19: 902– 905.

28 Klitz W, Stephens JC, Grote M, Carrington M. Discordant patterns of linkage disequilibrium of the peptide-transporter loci within the HLA class II region. Am J Hum Genet 1995; 57: 1436–1444. 29 Ronningen KS, Undlien DE, Ploski R et al. Linkage disequilibrium between TAP2 variants and HLA class II alleles; no primary asociation between TAP2 variants and insulin-dependent diabetes mellitus. J Immunol 1993; 23: 1050–1056. 30 Felsenstein J. PHYLIP (Phylogeny Inference Package). Department of Genetics, University of Washington: Seattle, WA, 1993. 31 Sworford DL. PAUP. Ilinois Natural History Survey: Champaign, Ilinois, 1991. 32 Kimura M. A simple model for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. J Mol Evol 1980; 16: 111–120. 33 Saitou N, Nei M. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Molecular Biology and Evolution 1987; 4: 406–425. 34 Moriuchi H, Moriuchi M, Fauci AS. Cloning and analysis of the promoter region of CCR5, a coreceptor for HIV-1 entry. J Immunol 1997; 159: 5441–5449.

27