Protective Effect of Interleukin-4 -589T Polymorphism on Human Immunodeficiency Virus Type 1 Disease Progression: Relationship with Virus Load

Emi E. Nakayama 2 3 Laurence Meyer 2 6 Aikichi Iwamoto 1 2 Anne Persoz 2 6 Yoshiyuki Nagai 0 2 Christine Rouzioux 2 Jean-Francois Delfraissy 1 2 Patrice Debre 2 4 Dorian McIlroy 2 4 Ioannis Theodorou 2 4 Tatsuo Shioda () 2 3 the SEROCO Study Group 2 5 0 National Institute of Infectious Diseases , Tokyo , Japan 1 Institute of Medical Science, University of Tokyo 2 Received 12 July 2001; revised 27 November 2001; electronically published 20 March 2002. Financial support: Human Science Foundation, Ministry of Education , Culture, Sports, Science , and Technology, Japan; Ministry of Health, Labor , and Welfare , Japan; Japan Society for the Promotion of Science; Organization for Pharma- ceutical Safety and Research , Japan; Agence Nationale de Recherches sur le Sida , France 3 Research Institute for Microbial Diseases, Osaka University , Osaka 4 INSERM U543 H opital Pitie - Salpetrie`re , Paris , France 5 Members of the SEROCO Study Group are listed after the text. Research Institute for Microbial Disease, Osaka University , 3-1 Yamada-oka, Suita-shi, Osaka 565-0871 , Japan 6 Epidemiology Department, INSERM U292, AP/HP, Hopital de Bicetre , Kremlin- Bicetre The interleukin (IL)- 4 2589T allele bears a single nucleotide polymorphism at position 2589 upstream from the open-reading frame of the IL-4 gene. To determine the influence of this allele on human immunodeficiency virus (HIV) type 1 disease, disease progression and serum virus load were assessed by IL-4 genotype in 427 white patients with known seroconversion dates who were followed in the French SEROCO cohort between 1988 and 1996. Serum virus load was 0.20 log lower during the 6- 24-month plateau phase after seroconversion in patients with IL-4 2589T than in those without this allele (P .02). Kaplan-Meier analysis survival curves showed a slower progression to clinical AIDS in carriers of IL-4 2589T (P .04). Adjustment for early serum virus load greatly diminished the strength of this association. These results suggest that IL-4 2589T protects against HIV-1 disease progression by reducing virus load. - Interleukin (IL) 4 has multiple immune response modulating functions, including induction of IgE production in B lymphocytes and the differentiation of precursor T helper (Th) cells toward the Th2 subset that mediates humoral immunity [1]. With respect to human immunodeficiency virus (HIV) type 1 infection, IL-4 differentially regulates 2 major HIV-1 coreceptors, CCR5 and CXCR4 [2 4]. CCR5 is a coreceptor used by nonsyncytia-inducing/macrophage-tropic/R5 viruses, which are more likely to be transmitted through sexual contact [5] and are present early after infection in nearly all HIV-1 infected persons. In contrast, CXCR4 is a coreceptor used by syncytia-inducing T cell line tropic/X4 variants, which are usually present in the latter stages of infection [6]. IL-4 down-regulates CCR5 expression and thus inhibits replication of R5 HIV-1 in human T cells and macrophages [2, 3], while it up-regulates the expression of CXCR4 and enhances the replication of X4 variant [4]. Recently, polymorphisms in the HIV-1 coreceptor, their natural ligand, and cytokine genes were shown to modify HIV-1 transmission and disease progression [7, 8]. In the IL-4 gene, Rosenwasser et al. [9] reported a polymorphism with a C to T exchange at position 2589 upstream of the open-reading frame of the IL-4 gene, IL-4 2589T, that is associated with increased promoter activity for IL-4 transcription and elevated levels of serum IgE in asthmatic families [9]. Previously, we reported that IL-4 2589T is associated with increased rates of the acquisition of X4 variants and elevated serum IgE levels in HIV-1 infected Japanese persons [10]. We also showed a lower frequency of IL4 2589T in persons infected with HIV-1 who reported heterosexual contact as a risk factor than in uninfected persons. These results suggested that IL-4 2589T has a protective effect against replication of R5 virus via CCR5 down-modulation. To evaluate the impact of the IL-4 2589T allele on HIV-1 disease progression, we analyzed IL-4 genes of 427 HIV-1 infected subjects with a documented date of seroconversion who were enrolled in the French SEROCO cohort. Patients and Methods Patients. Since January 1988, the multicenter French SEROCO cohort has included 1516 HIV-1 infected adults without hemophilia. The 6 monthly follow-up visits included a full physical examination and collection of blood samples that were stored at 2180 C. A detailed description of the SEROCO cohort is given elsewhere [11]. In all, 454 patients had a documented date of infection within an interval between their last negative and first positive HIV antibody test of , 24 months or incomplete/complete Western blot results. The date of infection is defined as the date of an incomplete Western blot less 1 month, the date of primary symptomatic infection less 15 days, or the midpoint between the 2 tests. These patients were enrolled in the cohort , 2 years after infection (median, 8.0 months). Stored peripheral blood mononuclear cell samples were available for 428 patients (median follow-up from seroconversion until September 1996, 116 months). DNA of 1 patient was not amplified for this study. IL-4 genotyping. Polymerase chain reaction (PCR) amplification of the IL-4 promoter was done by using genomic DNA extracted from cryopreserved lymphocytes. The IL-4 2589T mutation was detected by PCR restriction fragment length polymorphism analysis as described elsewhere [10]. HIV-1 RNA quantification. Serum HIV-1 RNA levels were measured on samples taken at baseline by using a PCR-based assay with a lower limit of detection of 200 HIV RNA copies/mL (Amplicor HIV Monitor assay; Roche) according to the manufacturers instructions. Statistical analysis. We used the x 2 test to compare qualitative variables and Students t or Wilcoxon rank sum tests to compare continuous variables. The Kaplan-Meier analysis method was used to construct survival curves, which were compared by the log-rank test. Crude and adjusted relative risks (RRs) and their 95% confidence intervals (CIs) were calculated by Cox proportional hazards models. The cutoff date was 15 September 1996. The frequency of the IL-4 2589T allele was 0.15 in 427 patients (14 T homozygotes, 98 C/T heterozygotes, and 315 C homozygotes), and this proportion was not different from that of a healthy French population (0.15; 2 T homozygotes, 21 C/T heterozygotes, and 63 C homozygotes). Since there were few homozygous patients and the disease progression to AIDS or death did not differ between heterozygotes and homozygotes (data not shown), these 2 categories were considered together in the analysis. As shown in table 1, the frequency of patients bearing IL-4 2589T did not differ significantly by sex. The risk factor for HIV infection, age at infection, or year of infection did not differ significantly between patients with and without IL-4 2589T. The median period between (...truncated)