Introduction
The human CCR5 chemokine receptor is a key cell-surface co-receptor for macrophage-tropic strains of HIV-1 [1-3]. Over the last 5 years, many studies have suggested a link between the presence of naturally occurring genetic polymorphisms within the CCR5 gene and patterns of HIV-1 disease progression [4-11]. A 32 base pair (bp) deletion mutation (CCR5Δ32) in the gene encoding the CCR5 receptor results in reduced cell-surface expression of CCR5 [12]. CCR 5Δ32/Δ32 homozygotes are significantly less likely to be infected with HIV-1 [4,5] and a number of natural history studies have indicated that CCR5wt/Δ32 heterozygotes progress more slowly to AIDS after infection than those without this mutation [4,6-8]. More recently, a number of polymorphisms in the promoter region of the CCR5 gene have been identified and associated with acceleration or protection with respect to HIV-1 progression to AIDS [9-11,13,14]. For example, the homozygous allele combination 208G, 627C and 676A in the CCR5 promoter region has been linked to accelerated progression to AIDS in untreated individuals [9], and the 303G, 627T and 927T alleles have been associated with a protective effect in several separate studies [10,11,13,14]. Furthermore, the 927T allele in the CCR5 promoter has been shown to be in complete linkage disequilibrium with a mutation in the nearby CCR2 chemokine receptor (CCR2-64I), which is also associated with a favorable effect on HIV-1 disease progression [13,15]. Other studies, however, have found little or no effect of CCR5 polymorphisms, notably the Δ32 mutation, on progression to AIDS-defining illness [16,17].
These studies, however, have all been conducted on untreated individuals. It remains unclear whether the CCR5Δ32 mutation and/or promoter polymorphisms affect the responses of patients to antiretroviral therapy. One study of CCR5Δ32 in 147 antiretroviral-treated individuals found no significant association with response to therapy [18], but another study of 166 protease-inhibitor-naive individuals with advanced disease found that CCR5Δ32 was predictive of a greater immunological and virological response rate to highly active antiretroviral therapy (HAART) [19]. More recently, an investigation of the effect of CCR5Δ32 and the promoter polymorphism at position 303 on treatment outcome in 272 Caucasian patients reported mixed results [20]. To date, no published studies appear to have examined whether combined CCR5 promoter polymorphisms and CCR5Δ32 are associated with altered response to antiretroviral therapy.
To address these questions, we retrospectively determined human CCR5 promoter polymorphisms and CCR5Δ32 mutations in a group of 436 antiretroviral-naive individuals starting their first HIV therapy in British Columbia, Canada, between June 1997 and August 1998, and examined their time to virological and immunological failure.
Materials and methods
Study design and patients
In the province of British Columbia, antiretroviral drugs are distributed free of charge to eligible HIV-infected individuals through a centralized drug treatment program based at the British Columbia Centre for Excellence in HIV/AIDS (the Centre). All antiretroviral-naive individuals first seeking treatment at the Centre between June 1997 and August 1998 were eligible for analysis in this retrospective study (n = 436). With respect to treatment, more than 75% of the antiretroviral-naive patients were prescribed triple therapy including two nucleoside analogues and one protease inhibitor, consistent with treatment guidelines at the time.
Ethical considerations
Ethical approval for this study was obtained from the local ethics board at St. Paul's Hospital Vancouver. All patient samples were given new anonymous code identifiers before analysis of CCR5 genotype, and studied anonymously.
Plasma viral load and CD4 cell count
Plasma HIV-1 RNA levels (viral loads) were determined using the Roche Amplicor Monitor assay (Roche Diagnostics, Laval, Quebec, Canada). CD4 cell counts were measured by flow cytometry, followed by fluorescent monoclonal antibody analysis (Beckman Coulter, Inc., Missisauga, Ontario, Canada).
CCR5 delta-32 determination
Plasma nucleic acids were extracted and precipitated with guanidine thiocyanate followed by isopropanol and ethanol washes. Human DNA was amplified from extracted plasma in a single round of polymerase chain reaction (PCR) using primers CCR5F (5′-AG GTCTTCATTACACCTGCAGC-3′) and FAMTM- labelled (FAMTM; 5-carboxy-fluorescein; Applied Biosystems, Foster City, California, USA) CCR5R (5′- CCTCTCATTTCGACACCGAAGC-3′). The PCR fragments (either a 172 bp product for wt/wt individuals, a 140 bp product for Δ32/Δ32 individuals; or both, for wt/Δ32 individuals) were visualized by electrophoresis on a 3% agarose gel. In cases where a clear identification could not be made from the gel, FAM-labelled PCR products were analyzed by electrophoresis on an Applied Biosystems 3700 DNA sequencer with an internal standard (ROX-350) and detected by laser scanning. The lengths of the PCR products were established relative to the internal standard.
CCR5 promoter polymorphisms
After nucleic acid extraction from plasma (as above), the CCR5 promoter region was amplified by nested PCR. For the first round of PCR, the oligonucleotide primers CCR5F12 (5′-CTGGAGTGAAGAATCCT GCC-3′) and CCR5R8 (5′-AAGGATTGTTAGT TATTAAAATAC-3′) were used to generate a 1.8 kb product. This was followed by a round of nested PCR using the oligonucleotide primers CCR5F13 (5′-CCTGCCACCTACGTATCTGGCATA-3′) and CCR5R7 (5′-TTTAAAGTCTTTCACTCACAATC A-3′) to generate a 1.4 kb product. PCR products were sequenced directly in both the 5′ and 3′ directions using the BigDye dye terminator cycle sequencing kit (Applied Biosystems) and run on an ABI 3700 automated sequencer.
Statistical analysis
Virological success was defined as achieving a plasma viral load (pVL) ≤ 400 copies/ml and virological failure was defined by two consecutive pVL measurements > 400 copies/ml. The time to failure (in months) was the difference between the earliest time of therapeutic success to the earliest time of failure. If a person never reached therapeutic success, then their time to failure was defined as 0. If a patient reached therapeutic success and did not subsequently fail then they were censored at the time of their last viral load date (occurring prior to 1 December 2000). Similarly, immunological failure was defined as the time to register two successive CD4 cell counts below baseline. Time to virological and immunological failure was analyzed by Kaplan-Meier methods. Group differences were assessed using the log-rank test. In a separate analysis, Cox proportional hazard regression was used to model the effect of CCR5 genotype on virological and immunological failure while adjusting for baseline pVL, baseline CD4 cell counts, and type of antiretroviral therapy at initiation.
Results
The PCR products and CCR5Δ32 genotype were obtained for 423 of 436 individuals (97.0%); 360 (85.1%) were wt/wt homozygotes, 62 (14.7%) were wt/Δ32 heterozygotes, and interestingly, one individual (0.2%) was a Δ32/Δ32 homozygote. Allele frequencies for wt and Δ32 alleles were 0.92 and 0.08, respectively.
The PCR products and CCR5 promoter sequence data were obtained for 416 of 436 samples (95.4%). Common CCR5 promoter polymorphisms were observed at positions 208, 303, 627, 676, and 927 (corresponding to GenBank accession number U95626 positions 58934, 59029, 59353, 59402 and 59653 respectively). The allele frequencies and percentages of patients displaying each individual genotype at these five sites are listed in Table 1. Rare polymorphisms were observed at positions 630 and 714 (Genbank U95626 positions 69356 and 59440). At position 630, one individual was identified as a 630Y (C/T heterozygote), all others being 630C/C. At position 714, two individuals were identified as 714S (C/G), all others being 714 C/C. These rare allele differences were not considered in our analysis.
The CCR5 promoter variants showed strong linkage disequilibrium. More than 95% of individuals possessing the 208T allele also displayed the 303G, 627T, 676G and 927C alleles, whereas individuals with the 303A allele were more than 95% likely to display the 627C and 676A alleles. Finally, more than 95% of individuals homozygous for 927T were also homozygous for the 208G, 303A, 627C and 676A alleles.
Figures 1a-f are Kaplan-Meier analyses of the independent influences of the CCR5Δ32 and five individual promoter polymorphisms on time elapsed to virological failure. Note that complete CCR5 genotype and plasma viral load follow-up data were only obtained for 405 of 436 patients (86.2%). No significant difference was observed between the CCR5 wt/Δ32 heterozygotes and CCR5 wt/wt homozygotes (P = 0.7). With respect to the influence of the CCR5 promoter on virological outcome, no significant effects were observed for the 208, 303, 627 or 676 polymorphisms (P ≥ 0.3). There was a trend for decreased time to failure for the 927C/T heterozygotes (P = 0.07), but this not significant after adjustment for multiple comparisons. In addition, 81 of 405 individuals (20.0%) who maintained a viral load > 400 copies/ml over the course of the entire study (time to failure = 0) did not have statistically different CCR5Δ32 and promoter allelic composition than the patients who recorded at least one viral load measurement < 400 copies/ml (data not shown). Of the remaining 324 patients who achieved at least one pVL < 400 copies/ml, there was a trend towards an increased, although not statistically significant, time to failure (P > 0.05) associated with the CCR5wt/Δ32 and the promoter genotypes 303G/G and 927C/T. Finally, note that the baseline viral load measurements did not significantly differ between any of the allelic groups: the median baseline pVL for all groups was approximately 90 000 copies/ml (interquartile range 33 000-280 000 copies/ml).
Time to immunological failure was defined as a return to a CD4 cell count below baseline on two consecutive visits. Note that complete CCR5 genotype and CD4 follow-up data were obtained for 376 of 436 patients only (86.2%), and again that baseline CD4 counts did not differ significantly between allelic groups. The median baseline CD4 count for all groups was 295 × 106 cells/l (interquartile range 120 × 106 to 460 × 106 cells/l). Analysis of the independent influence of CCR5Δ32 and individual promoter polymorphisms on time to immunological failure revealed no significant effect of any allelic group on time for CD4 cells return to baseline (CCR5Δ32 P = 0.3, promoter polymorphisms P = 0.2-1.0, Kaplan-Meier plots not shown). Note that the median follow-up time for both virological and immunological analysis was 22 months.
Since the Δ32 and promoter polymorphisms show strong linkage disequilibrium, we felt it was necessary to analyze the effect of the complete CCR5Δ32 and promoter genotype on disease progression. The 'complete' CCR5 genotype for each patient was determined by examining the allelic variation at four sites in the CCR5 gene: positions 208, 303 and 927 for the promoter, as well as the CCR5Δ32. (For figure clarity and simplification, variation at promoter positions 626 and 676 were not included in the analysis because they are tightly linked to the polymorphism at positions 303.) Table 2 shows the distribution of the most common combined promoter and CCR5Δ32 genotypes in our study group. Note that both CCR5 promoter and Δ32 data were obtained for 407 out of 436 patients only (93.3%).
Figure 2 shows the influence of combined Δ32 and promoter polymorphisms on time to virological failure. Only the outcomes of the groups with > 30 patients at time = 0 were analyzed. We found no significant influence of combined CCR5 promoter and Δ32 genotype on time to virological failure (P = 0.892). Similarly, we found no significant influence of combined CCR5 genotype on time to immunological failure (P = 0.622), Kaplan-Meier plot not shown).
In a separate analysis, Cox proportional hazard regression was used to model the effect of the CCR5Δ32 and promoter allelic groups on time to virological failure while adjusting for baseline pVL, baseline CD4 cell count, and type of antiretroviral therapy at initiation (triple therapy with two nucleoside reverse transcriptase inhibitors and a protease inhibitor versus dual therapy). A similar analysis was carried out for time to immunological failure and in all cases, the results were not statistically significant (P > 0.1).
Finally, in order to control for non-adherence, a subset analysis was performed on only those patients who were most likely highly adherent to therapy. As the patients all initiated therapy at different time points, adherence or non-adherence was roughly measured by analyzing patients prescription records and determining the frequency of prescriptions dispensed on time throughout the first year of starting therapy. Patients were considered adherent if their prescriptions were dispensed on time for all 12 months of the year. According to this definition, 225 of 436 patients (57%) were identified as adherent. An independent subset analysis of the influence of both individual and combined CCR5 genotype on time to virological and immunological failure in the 225 adherent patients revealed no significant differences from the analyses carried out on the original group (P > 0.3, Kaplan-Meier plots not shown).
Discussion
This study examined the effect of the CCR5Δ32 and promoter polymorphisms on the virological and immunological response to first antiretroviral therapy in a large, community-based cohort.
The polymorphisms we observed in the CCR5 promoter region are consistent with previously reported results. The homozygous allele combination 208G, 627C and 676A has been linked to accelerated progression to AIDS in untreated individuals [9], whereas separate studies have suggested protective roles for alleles 303G [10,11], 627T [11], and 927T [13,14]. Of interest, an A/G polymorphism at position 29 of the promoter gene has been observed, and a disease-retarding effect for 29G reported [14]. Note that our analysis of the CCR5 promoter region did not cover position 29 due to its position outside the area covered by our oligonucleotide primers. Finally, note that rare variations at positions 612, 626, 647, 684 and 811 have also been reported [9], which were not observed in our analysis.
There have been considerably more studies investigating the role of the CCR5Δ32 mutation with respect to disease progression in untreated individuals. Some studies have linked CCR5Δ32 with a protective effect on HIV-1 infection and disease progression [4-8], whereas others suggest that there is no correlation between CCR5Δ32 and altered rates of disease progression [14,16,17,21,22].
Bratt et al. [18] investigated the influence of CCR5Δ32 on treatment outcome in 147 unselected advanced patients receiving HAART. Similarly to our study, these researchers analyzed their data on an intent-to-treat basis and found no influence of CCR5Δ32 genotype on treatment efficacy. Two recently published research letters, however, disagree with these findings [19,23]. One study of 113 patients undergoing HAART associated the CCR5 wt/Δ32 genotype with an improved response [23]. The results of this study, however, may have been influenced by the stringent patient selection criteria, which eliminated patients on the basis of race and other reasons. In order to represent a realistic HIV-infected population, we did not eliminate any individuals from our study on the basis of racial background. The second study, involving 166 protease-inhibitor-naive patients who received triple combination therapy, concluded that CCR5wt/Δ32 heterozygous patients were more likely to respond to HAART than CCR5wt/wt patients [19]. This study, however, chose different parameters for defining successful virological and immunological responses to those used in our study.
To date, there has been only one study investigating the role of CCR5Δ32 and a single promoter polymorphism (303A/G) in antiretroviral-treated individuals. The study involved 272 Caucasian patients. The researchers found no correlation between CCR5 genotype and time to first HIV RNA < 200 copies/ml, but did observe a correlation when virological failure was defined under different parameters [20].
To our knowledge, there have been no published studies investigating the influence of both the CCR5Δ32 mutation and all common CCR5 promoter polymorphisms on treatment outcome. Although one might hypothesize that alleles associated with a protective effect in untreated individuals may also be associated with an improved response in individuals undertaking antiretroviral treatment, our study suggests that this is not the case. We found little or no correlation between the CCR5Δ32 or any promoter genotype and time to treatment failure. Similarly, separate analyses controlling for baseline pVL, baseline CD4 counts and adherence revealed no correlation between CCR5 genotype and treatment outcome. Some observed trends (for example, 927Y being apparently associated with a more rapid time to failure) may have arisen simply by chance or may be partially explained through linkage disequilibrium between the 927 allele and another gene. The 927T allele has previously been shown to be in complete linkage disequilibrium with a mutation in the nearby CCR2 receptor (CCR2-64I). The CCR2-64I /CCR5 927T genotype has been associated with a favorable effect on HIV-1 disease progression [13,15]. However, based on this correlation one would expect to see this effect for both the 927 C/C and 927 C/T allelic combinations, not only the 927Y.
The major limitations of this study were its retrospective nature, and the fact that viral phenotyping was not performed. Also, although an attempt was made to control for treatment compliance, the measure of non-adherence used was only rough. Other potential limitations of this study include the fact that the patient group, although large, represents only a geographically isolated population, and that racial background information was not generally available.
This study is of particular significance because it is the first to examine the effect of both CCR5Δ32 and five different CCR5 promoter polymorphisms on antiretroviral treatment outcome in a large number of HIV-1 infected individuals initiating therapy. In contrast to previous studies suggesting that natural genetic variation in the CCR5 gene may play a role in initial infection with HIV and in the early stages of the disease, our research suggests that among individuals undertaking antiretroviral drug treatment, these natural factors may not have a significant impact on initial virological and immunological response to antiretroviral therapy.
References
1. Samson M, Labbe O, Mollereau C, Vassart G, Parmentier M. Molecular cloning and functional expression of a new human CC-chemokine receptor gene. Biochemistry 1996, 35: 3362-3367.
2. Deng HK, Liu R, Ellmeier W. et al. Identification of a major co-receptor for primary isolates of HIV-1. Nature 1996, 381: 661-666.
3. Dragic T, Litwin V, Allaway GP. et al. HIV-1 entry into CD4+ cells is mediated by the chemokine receptor CC-CKR-5. Nature 1996, 381: 667-673.
4. Huang Y, Paxton WA, Wolinsky SM. et al. The role of a mutant CCR5 allele in HIV-1 transmission and disease progression. Nat Med 1996, 11: 1240-1243.
5. Liu R, Paxton WA, Choe S. et al. Homozygous defect in HIV-1 coreceptor accounts for resistance of some multiple-exposed individuals to HIV-1 infection. Cell 1996, 86: 367-377.
6. Dean M, Carrington M, Winkler C. et al. Genetic restriction of HIV-1 infection and progression to AIDS by a deletion allele of the CKR5 structural gene. Hemophilia Growth and Development Study, Multicenter AIDS Cohort Study, Multicenter Hemophilia Cohort Study, San Francisco City Cohort, ALIVE Study. Science 1996, 273: 1856-1862.
7. Samson M, Libert F, Doranz, BJ. et al. Resistance to HIV-1 infection in Caucasian individuals bearing mutant alleles of the CCR5 chemokine receptor gene. Nature 1996, 382: 722-725.
8. Stewart G, Ashton L, Biti R. et al. Increased frequency of CCR-5 (32 heterozygotes among long-term non-progressors with HIV-1 infection. AIDS 1997, 11: 1833-1838.
9. Martin M, Dean M, Smith M. et al. Genetic acceleration of AIDS progression by a promoter variant of CCR5. Science 1998, 282: 1907-1911.
10. McDermott DH, Zimmerman PA, Guignard F, Kleeberger CA, Leitman SF, Murphy PM. CCR5 promoter polymorphism and HIV-1 disease progression. Lancet 1998, 352: 866-870.
11. Clegg AO, Ashton LJ, Biti RA. et al. CCR5 promoter polymorphisms, CCR5 59020 and ccr5 59353C, are under represented in HIV-1-infected long-term non-progressors. The Australian Long-Term Non-Progressor Study Group. AIDS 2000, 14: 103-108.
12. Benkirane M, Jin DY, Chun RF, Koup RA, Jeang KT. Mechanism of transdominant inhibition of CCR5-mediated HIV-1 infection by ccr5delta32. J Biol Chem 1997, 272: 30603-30606.
13. 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. Nat Med 1998, 4: 350-353.
14. Mummidi S, Ahuja S, Gonzalez E. et al. Genealogy of the CCR5 locus and chemokine system gene variants associated with altered rates of HIV-1 disease progression. Nat Med 1998, 4: 786-793.
15. Smith M, 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.
16. Wilkinson DA, Operskalski EA, Busch MP, Mosley JW, Koup RA. A 32 bp deletion within the CCR5 locus protects against transmission of parenterally acquired human immunodeficiency virus but does not affect progression to AIDS-defining illness. J Infect Dis 1998, 178: 1163-1166.
17. Morawetz RA, Rizzardi GP, Glauser D. et al. Genetic polymorphism of CCR5 gene and HIV disease: the heterozygous (CCR5/delta ccr5) genotype is neither essential nor sufficient for protection against disease progression. Swiss HIV Cohort. Eur J Immunol 1997, 27: 3223-3227.
18. Bratt G, Karlsson A, Leandersson AC, Albert J, Wahren B, Sandstrom E. Treatment history and baseline viral load, but not viral tropism or CCR5 genotype, influence prolonged antiviral efficacy of highly active antiretroviral treatment. AIDS 1998, 12: 2193-2202.
19. Guerin S, Meyer L, Theodorou I. et al. CCR5 delta32 Deletion and Response to HAART in HIV-1-Infected Patients. SEROCO/HEMOCO Study Group. AIDS 2000, 14: 2788-2790.
20. O'Brien TR, McDermott DH, Ioannidis JP. et al. Effect of chemokine receptor gene polymorphisms on the response to potent antiretroviral therapy. AIDS 2000, 14: 821-826.
21. Smith M, Dean M, Carrington M, Huttley G, O'Brien S. CCR5-Delta32 gene deletion in HIV-1 infected patients [Letter]. Lancet 1997, 350: 741.741.
22. Schinkel J, Langendam M, Coutinho R, Krol A, Brouwer M, Schuitemaker H. No evidence for an effect of the CCR5D32/+ and CCR2b 64I/+ mutations on human immunodefiency virus (HIV)-1 disease progression among HIV-1-infected injecting drug users. J Infect Dis 1999, 179: 825-831.
23. Valdez H, Purvis S, Lederman M, Filingame M, Zimmerman P. Association of the CCR5(32 mutation with improved response to Antiretroviral therapy. JAMA 1999, 282: 734-734.
© 2001 Lippincott Williams & Wilkins, Inc.