JP2010522560A - Genetic variants associated with HIV disease restriction - Google Patents

Abstract

  The present invention relates generally to human immunodeficiency virus (HIV), and in particular to genetic variants associated with limiting HIV disease progression.

Description

  This application claims priority from US Provisional Application No. 60 / 907,271, filed on March 27, 2007, and US Provisional Application No. 60 / 929,432, filed on June 27, 2007. The contents of the application are incorporated herein by reference.

This invention was made with government support under grant number A10678501 awarded by the National Institutes of Health. The US government has certain rights in this invention.
TECHNICAL FIELD The present invention relates generally to human immunodeficiency virus (HIV), and in particular to genetic variants associated with limiting HIV disease progression.

  Humans show significant variability in vulnerability to infection by HIV-1 and particularly in clinical outcome after infection. One of the most striking differences is the viral plasma level (virus setpoint) in the asymptomatic phase prior to progression to AIDS. Some individuals manage to control the virus to a low or even undetectable level, while others have a much higher plasma viral load (VL) and patient-to-patient variability over 4-5 logs (bottom 10). Observed routinely (see Figure 3 of Telenti et al., Nat. Rev. Microbiol. 4: 865 (2006)). A small portion of this variability can be explained by demographic factors and variants of known genes such as chemokines, chemokine receptors and cytokines (about 15% of the variation in the dataset, again Telenti Nat. Rev. Microbiol. 4: 865 (2006), O'Brien et al., Nat. Genet. 36: 565 (2004), Telenti et al., Future Virology 1:55 (2006), Carrington et al., Annu. 54: 535 (2003), Nelson et al., J. Acquir.Immune Def .. Syndr. 42: 347 (2006), Breiber et al., J. Virol. 79: 12474 (2005). See).

  The HIV-1 setpoint is a particularly important phenotype, not only due to dramatic variability between individuals, but also to relative stability within the individual over time and disease progression and to infectivity. It is also due to the effect (Mellors et al., Science 272: 1167 (1996)). A better understanding of the causes of VL differences can provide suggestions for new vaccines and drugs that control the virus. However, in order to facilitate the discovery of any genetic variants that can affect virus control, it is essential to go beyond targeted candidate gene research that has been considered to work to date (Telenti et al., Nat. Rev. Microbiol. 4: 865 (2006)).

  The present invention is the first of the variability in host control of HIV-1 that at least partly focuses on determinants of VL setpoints and then progression to AIDS (measured by a decrease in CD4 positive cells). Arising from whole genome association studies. This study has resulted in the identification of three genetic variants (or groups of variants) that are associated with HIV burden and limiting disease progression.

Teleni et al., Nat. Rev. Microbiol. 4: 865 (2006) O'Brien et al., Nat. Genet. 36: 565 (2004) Telenti et al., Future Virology 1:55 (2006) Carrington et al., Annu. Rev. Med. 54: 535 (2003) Nelson et al. Acquir. Immune Defi. Syndr. 42: 347 (2006) Bleiber et al. Virol. 79: 112674 (2005) Mellors et al., Science 272: 1167 (1996).

  The present invention relates generally to HIV. More specifically, the present invention relates to genetic variants associated with limiting HIV disease progression and to methods of using such variants as prognostic markers.

  Objects and advantages of the present invention will be apparent from the description that follows.

1A and 1B. The HIV-1 viral load at setpoints is HCP5 rs23995029 genotype (T majority allele, G minor allele) (FIG. 1A) and HLA-C 5 ′ region rs9264942 genotype (T majority allele, C minor allele) ( It is highly correlated with FIG. 1B). Partial map of the HLA class I region (chromosome 6 p21.3). The p-value [-log (P)] of all genotyped SNPs annotated with the gene structure is shown. Two independent SNPs showing a significant genome-wide association with HIV-1 VL at setpoint are shown and marked in red. Draw graphs with WGA Viewer software (see website: genome.duke.edu/centers/pg2/index_html/downloads/AnnotationSoftware). 3A-3C. FIG. 3A. Non-linear effect of HLA-C expression level on HIV-1 VL at setpoint. Respectively, the number of patients associated with each point is shown in italics in the Sanger Genevar database for expression data and in the cohort for setpoint results. Figures 3B and 3C show the distribution of the respective data according to genotype. 4A-4J. (FIG. 4A) rs9264942; (FIG. 4B) rs6457374; (FIG. 4C) rs2395029; (FIG. 4D) rs9261174; 4I) rs2301753; and (FIG. 4J) rs2074479.

  The present invention arises from the identification of three genetic variants (or groups of variants) that are associated with important differences in HIV load, the most important prognostic marker of HIV disease progression. Together, these variants can account for a significant portion of inter-individual variability in HIV plasma levels. The identification of these variants sheds new light on HIV virulence and on interactions between the immune system and the virus, thereby revealing new therapeutic targets.

In a first aspect, the present invention relates to two single nucleotide polymorphisms located in the 5 ′ region of the HLA-C gene in the MHC class I region on chromosome 6: SNP reference numbers are rs9264942 and rs6457374 (this See ncbi.nlm.nih.gov/projects/SNP for polymorphic identity referred to in the specification). (FIG. 4) These genetic variants are associated with differences in HLA-C mRNA expression and its 5 ′ position appears to explain this effect (promoter / enhancer). HLA ^* CwA has been speculated to be associated with rapid HIV progression (Carrington et al., Science 283 (5408): 1748-1752 (1999)), but more recently this effect has been linked to HLA ^* B35-Px and this type of It was attributed to linkage (Carrington, Annu. Rev. Med. 54: 535-551 (2003)). In fact, a partial linkage with rs9264942 is a more likely explanation. This is the first observed quantitative effect of HLA class I protein on HIV viral load, independent of HLA type.

In a second aspect, the present invention relates to a single nucleotide polymorphism located in the putative coding region of the HCP5 gene in the MHC class I region on chromosome 6: the SNP reference number is rs23995029. The function of the HCP5 gene was previously unknown. However, based on structural similarity to human endogenous retroviruses, an association with antiretroviral immunity has been suggested (Kulski et al., Immunogenetics 49 (5): 404-412 (1999)). The genetic variants identified herein are located in a region that shares homology with the pol sequence of a particular retrovirus (human endogenous retrovirus). This is an asymptomatic change, that is, it is responsible for the predicted amino acid sequence change of the peptide resulting from this genomic sequence. This SNP is in high linkage disequilibrium with HLA ^* B5701 which has the strongest known host genetic effects on HIV disease (Carrington, Annu. Rev. Med. 54: 535-551 (2003)). It is not yet clear whether this only serves as a tag for the effects associated with this type or indicates a causal relationship. This is the first documented description of the association between HIV disease and HCP5, and raises the possibility that HCP5 is involved in HIV regulation and not ^* B5701.

  In a third aspect, the present invention relates to seven single nucleotide polymorphisms distributed among three genes in the MHC class I region on chromosome 6: HCG9, RNF39 and ZNRD1. The SNP reference numbers are rs 9261174, rs 2074480, rs 7785512, rs 9261129, rs 3869068, rs 2301753 and rs 2074479. These mutants are in complete linkage disequilibrium and therefore it is impossible to distinguish between them with respect to the causal relationship to the observed phenotype (better or worse control of HIV viral load) It is. These are associated with significant differences in the expression of the transcription-related gene zinc ribbon domain containing 1 (ZNRD1) gene. ZNRD1 is known to play a role in the multidrug resistance phenotype of gastric cancer cells through up-regulation of other genes (Shi et al., Cancer Biol Ther. 3 (4): 377-381 (2004)). Given its structure and function, it is expected to have a direct impact on the control of HIV transcription.

  Differential expression of HLA-C and ZNRD1 genes is predicted to play an important role in HIV infectivity. Furthermore, if HCP5 is involved in the observed effects on HIV load, the product (RNA, peptide) can be predicted to have a direct functional role in protection against retroviruses.

  It is known that genetic variants often occur in groups along the chromosome (haplotype block). Thus, the identification of SNPs referred to above as being associated with viral load may uniquely identify additional SNP (s) present in such haplotype blocks. Even if the SNP associated with the viral load is found not to be a critical causative SNP that produces its effect, nevertheless, are these SNPs often in linkage disequilibrium with the causative SNP? Or a valid marker directly in the haplotype block. Re-sequencing the genomic DNA region surrounding the SNP specifically identified herein may reveal other SNP (s) (ie related intervals) associated with viral load / disease progression, these Can be predicted based on their presence in the same haplotype or by being in linkage disequilibrium with the specific SNPs disclosed herein.

  Using a variety of genotyping techniques known in the art (eg, “CHIP” or SNP panel), the presence of any of the above polymorphisms in a sample (eg, a biological sample such as blood) can be determined Also good. All SNPs described herein are present on Illumina's HumanHap550 genotyping BeadChip (see illuma.com). Suitable techniques also include the use of polymerase chain reaction and extension primers, RFLP analysis and mass spectrometry (Ye et al., Hum. Mutat. 17 (4): 305 (2001), Chen et al., Genome Res. 10 : 549 (2000)).

  Screening for genetic variants allows the clinician to better stratify a given patient for therapeutic intervention. In addition, knowing the genetic variants allows the patient to select from the available therapeutic approaches in consultation with a physician in a more informed manner.

  The present invention also relates to kits that are suitable for use in testing for the presence of the polymorphisms identified herein. Such a kit may include, for example, reagents (eg, probes or primers) necessary to identify the presence of the polymorphism referred to above.

  Certain aspects of the present invention are described in more detail by the following non-limiting examples (see also US Provisional Application No. 60 / 907,271, incorporated herein) (Fellay et al., Science 317). : 944-947 (2007)).

Experimental details
Selection criteria Patients are: (i) Biological markers:
1. Documented positive test and date, and documented negative test within two years before the first positive test;
2. Alternatively, one or more biological criteria for primary infection: incomplete Western blot and / or positive p24Ag and / or high viremia (> 1 million copies per milliliter of blood) and consistent dynamic pattern of biological parameters Subjects may be included if they show (completion of Western blot, p24Ag negation, reduced viremia peak), and consistent clinical syndromes were considered supporting evidence;
3. The individual subset has long-term spontaneous viral load control of less than 1000 RNA copies / ml and is included regardless of the actual antibody seroconversion date, and an effective antibody seroconversion date estimate; and (ii) after antibody seroconversion Eligible for study if they had plasma HIV RNA measurements in the absence of antiretroviral therapy for 3 months to 3 years.

Setpoint definition first step: Eliminate outlier VL based on clinical or biological discussion (Only for the Euro-CHAVI cohort) Visual inspection of individual data and query address definitions for clinical centers and laboratories. The point of concern was then eliminated when a sufficient explanation was obtained, such as acute or acute exacerbation of a chronic or chronic disease, vaccination, immunomodulatory treatment, serious trauma or laboratory problems.

  B. Of the CD4 cell profile associated with HIV-1 disease progression (defined by a decrease in CD4 cells to less than 350 cells / ml) leading to elimination of coexisting and subsequent VL results (for both cohorts) Inspection.

Second step: elimination of VL that does not reflect a stable state through a computerized algorithm. Three types of outliers were identified that correspond to the three-phase development of HIV-1 viremia:
A. VL measured before reaching setpoint, part of the first peak of viremia observed during primary HIV infection: this must be measured during the first year after antibody seroconversion, And it has a value> 0.25 log10 higher than the average of the following VL.

  B. VL measured during the accelerated phase of the disease, reflecting the fact that rapid progressors can develop into disease progression in a short period of time: for patients with a markedly upward VL slope, for setpoint calculations Only the first three results were retained.

  C. VL measured during setpoint period but inconsistent with other available results; probably associated with unreported interference conditions, laboratory errors, transcription or data management errors: from the average of all remaining points Defined as VL> 0.5 log10 higher or lower.

Third step: Setpoint is calculated as the average of all remaining VL results.
The Cohort HIV-AIDS Vaccine Center (CHAVI) is headed by Barton Haynes (Duke University, Durham, NC). Its host genetics core is headed by David Goldstein (Duke University, Durham, NC, USA). CHAVI is established by the National Laboratory for Allergic Infectious Diseases (USA). The Euro-CHAVI Consortium Columbo (University of Lausanne, Switzerland) and J. Org. P. A. With the help of Ioannidis (University of Ioannina, Ioannina, Greece) Coordinated by Telenti (University of Lausanne, Switzerland). Participating cohorts / studies (Principal Investigators) are: Swiss HIV Cohort Study, Switzerland (P. Francioli); IrsiCaixa, Barcelona, Spain (B. Clotet); Clinics Hospital, Barcelona, Spain (J.M. Cohort, Denmark (N. Obel); Modena Cohort, Modena Italy (A. Cosalizza); San Raffaele del Monte Tabor Foundation, Milano (A. Castagna); CO. NA Cohort, Rome, Italy (A. De Luca); Royal Perth Hospital, Perth, Australia (S. Mallal); Guy Kings St. Thomas Hospital, United Kingdom (P. Easterbrook). All participating centers provided local facility review board approval for genetic analysis and each participant provided genetic informed consent.

Data flow quality control The following quality control steps were taken to ensure that genotyping was called correctly.

1. Infinium BeadStudio raw data analysis All samples are collected in a single BeadStudio file and a standard Illumina cluster file is used. Evaluate the clustering of all samples. For a random set of SNPs, if the majority of samples do not fit the cluster, create a separate BeadStudio file for these samples. Once all files have been created, any samples with very low intensity or a very low call rate (<95%) using Illumina clusters are deleted. Then, all SNPs with a call frequency of less than 100% are reclustered. After re-clustering, any samples with less than 98% call rate are deleted. Next, a “1% rule” is applied that deletes all SNPs with a call frequency of less than 99%. Remove any SNPs when more than 1% of the samples are not called or are called ambiguously. It has been shown that SNPs that are not called (or potentially miscalled) in many samples can lead to false positives in a statistical association (unpublished data).

  The reclustering process results in a SNP calling error (even with 1000 samples in the file), but a procedure has been identified in the final report to prevent erroneous calls from being released. SNP data is screened in BeadStudio by looking at two criteria. First, manually check all SNPs with a cluster separation value of less than 0.3 to ensure that the call is correct. Many of these SNPs can be fixed manually, but some must be deleted. Next, any SNPs (excluding X-chromosome SNPs) with Het excess values between -1.0 to -0.1 and 0.1 to 1.0 are evaluated, and raw data and normalized data are Determine whether to show a clean call. Any SNP clusters that do not appear normal are deleted. This includes SNPs that appear to show deletions (hemizygous and homozygous deletions). This is done because they can be artifacts from either chemical reactions during hybridization or interfering SNPs.

  These procedures resulted in success rates of genotyping calls ranging from 97.5% to 99% (13,709-5355 deleted SNPs). For quality purposes, 2 percent of the samples were randomly selected and genotyped twice independently. The concordance rate for duplex genotyping was 99.99%. In addition, 10 SNPs from different chromosomes were regenotyped using the TaqMan assay. There was 100% agreement between BeadChip and Taqman genotype calling. A total of 535 samples were run on a whole genome chip. A total of 49 samples were excluded. Some of these were complete genotype failures (ie, less than 98%), but most were due to high levels of calling stringency (1% rule).

2. Minority Allele Frequency (MAF) Check for Data Handling Accuracy This step performs a basic check of data accuracy for the data flow pipeline from the output of the Illumina genotyping facility to the analysis process. Using PipeQC software, the PLINK-derived MAF report was checked against the original locus report generated by the genotyping facility. A check was made to see if the two MAF reports matched exactly.

3. Gender Identification This step checks for gender identification obtained from a phenotype database using the observed genotypes of SNPs on chromosome X, and Y if available. Individuals marked as “male” but with significant amount of heterozygote X genotype (> = 1%), or marked as “female” but with high frequency homozygote X genotype (> = 80%) or all individuals with Y genotype readings were individually examined against the original data source. These individuals were excluded from further analysis if sufficient correction could not be obtained. Four patients were excluded at this step.

4). Potential relatedness
This process checks for potential syngeneic relationships among study participants. Estimation of genetic information sharing was performed by estimating identity by descendant (IBD) using PLINK software.

All pairs of DNA samples exhibiting (the estimated ratio of allele IBD) were examined individually and one sample in each pair was excluded from further analysis. Seven samples were removed in this process.

5). Genotype deficiency This step checks whether the genotype deficiency is not distorted towards high or low phenotypic values, and therefore can lead to erroneous associated p-values. This check was performed on the top SNPs discussed herein using PLINK software. None of the genotype data violated this check.

6). Low MAF
All SNPs with MAF <0.006 were removed. This criterion ensured that at least 6 individuals with rare genotypes were present in the data set and controlled errors in asymptotic p-value estimation.

7). Hardy-Weinberg equilibrium (HWE)
This step checks whether the observed genotype data deviates from the HWE. This check was performed using PLINK software against the top SNP. Deviations from HWE were defined using a P-value criterion of less than 0.05. None of the genotype data violated this check.

8). Rechecking genotyping quality Top SNPs showing significant association were subjected to a double check for their genotyping quality. This is an individual recheck on raw and normalized data to ensure that it is called correctly as described in the “Infinium BeadStudio Raw Data Analysis” process. None of the genotype data violated this check.

Modified EIGENSTRAT method to manage for stratification This method obtains the main components of the correlation between the gene variants and corrects for these correlations in related tests. Thus, in principle, the principal component under analysis should reflect the population ancestry. However, some of the leading axes appear to depend on other sources of correlation, for example, it has been noted that there are a set of mutants in close proximity to each other that show an expanded association. The potential for inversion to cause this effect has been documented, and this can also occur due to other causes of extended linkage disequilibrium. For this reason, for each induction axis, a SNP “load” test is performed, and these are as expected if the given axis reflects the effects of population ancestry or more localized linkage disequilibrium, respectively. Was dependent on many or relatively few SNPs. This analysis identified that some axes were clearly due to inversion and suggested that 17 axes should be retained for ancestry adjustment. Therefore, significance was evaluated using the 17 principal components resulting from the EIGENSTRAT analysis as covariates in a logarithmic regression model that also incorporates gender as a covariate.

The EIGENSTRAT axis was chosen for use as a covariate adjusting for ancestry in subsequent linear regression analysis as follows:
1. To find the EIGENSTRAT axis, we started with an autosomal SNP with MAF> 0.01.

  2. Upon examination of the SNP load for each PC axis (Price et al. (Nat. Genet 38: 904 (2006)) “gamma” coefficient), several top axes are all due to a small number of SNPs that map to the same region of the genome. It was found to be monopolized. For example, one axis was found to be dominated by SNP mapping to the chr8p22-23.1 region, consistent with known inversion polymorphisms.

  3. To correct for these LD effects and to ensure that the EIGENSTRAT axis only reflects the effect applied equally across the entire genome (as ancestral effects should be) (i) Exclude certain known high LD regions (chr8: 8000000 ... 12000000, chr6: 25000000.335500, chr11: 45000000.57000000, chr5: 44000000..5100000); (ii) 1500 window sizes (150 (Iii) Patterson et al. (Patterson et al.), Using the “--indep-pairwise” option in PLINK so that all SNPs in the step size) must have r2 <0.2. Atson, et al., PLoS Genet. 2: e190 (2006)), each SNP is regressed against the previous 5 SNPs, and the remainder is input to the PCA analysis for the reduced SNP set. The analysis was reapplied.

  4). Examination of SNP loading on all axes is considered significant by Patterson et al.'S Tracy-Widom method using Q-Q plots against expected values, where a single high LD region of the genome A monopoly axis did not appear.

  5. The Tracy Widom test designated the first 17 resulting PC axes as significant (p <0.05). Therefore, the first 17 PC axes were adopted as covariates in subsequent analyses.

Results Success in genetic association studies depends on precisely defined and measured phenotypes. In this study, strict selection criteria were applied to participant identification and the database was supervised at the level of individual patient files. Patients will have (i) effective seroconversion date estimation as evidenced by biological markers, and (ii) plasma HIV in the absence of antiretroviral therapy for 3 months to 3 years after seroconversion. Eligible for study when having RNA measurements. Importantly, infectious disease clinicians experienced in managing HIV-infected subjects individually examined all long-term VL data used in the study. This test may identify atypical VL measurements and query the appropriate clinic and laboratory to either explain the reading or remove the reading or subject It has become possible. Also, considering the pattern of change, it was determined whether the set point was actually achieved. (676 patients in the Euro-CHAVI sub-cohort and study were eligible according to this criterion. After testing for viral load results, queries were defined for 195 of them. Based on responses from the center, 42 patients were excluded due to inaccurate assessment of setpoints.) An algorithm was developed with the intention of matching the decisions made after the test and reapplied to all subjects (above) See experimental details).

  For patients with at least 4 CD4 cell count results, when the progressive phenotype is reduced to less than 350 CD4 cells per milliliter of blood or when antiretroviral treatment is initiated, whichever occurs first Defined. Patients who did not progress under this definition during the follow-up period were evaluated for evidence of CD4 reduction (significantly reduced CD4 slope determined by simple regression). For those exhibiting CD4 reduction, the estimated slope of reduction was used to extrapolate the point in time when the CD4 count drops below 350, and this was used as the time to progression (Douek et al., Annu. Rev. Immunol. 21: 265 (2003)). Subjects who did not progress and did not show significant CD4 reduction were classified as non-progressors and a separate case-control comparison was established as progressors versus non-progressors.

  The Euro-CHAVI cohort specially created for this study has eight European and one Australian cohorts that have agreed to participate in the Host Genetics Core Initiative of the HIV / AIDS Vaccine Immunology Center (CHAVI) Corresponds to a research consortium. CHAVI is a university and university hospital consortium established by the National Institute of Allergy and Infectious Diseases, part of the global HIV vaccine business. From these cohorts, 676 patients were selected based on the criteria described above. For the purposes of this analysis, only subjects with self-reported Caucasian ancestry were considered: after eliminating individuals with insufficient quality phenotype (see above) or genotype (see experimental details) Finally, 486 patients were included. Of these, 386 were eligible for progress analysis (309 progressors and 77 non-progressors).

  All samples were genotyped using Illumina's HumanHap550 genotyping BeadChip, with a total of 555,352 single nucleotide polymorphisms (SNPs). (I) Infinium BeadStudio raw data analysis quality; (ii) minority allele frequency consistency to assess data handling accuracy; (iii) clinical and genetically implied mismatches of gender; (iv) latency (V) genotype deficiency pattern; (vi) low frequency of minor alleles; (vii) Hardy-Weinberg equilibrium violations; and (viii) genotyping quality appearance test for top SNPs (see experimental details) A series of quality control (QC) methods were performed, including checks on Based on these QC methods, a total of 20,251 SNPs were removed. Designed to eliminate deletions and target copy number variation (CNV) in samples and apply methods to assess whether they affect phenotype (designed to eliminate failed SNPs) Using quality control techniques, long stretches of failed SNPs were identified as putative deletion polymorphisms and examined for differences according to phenotype; 42 of these regions were checked and the size was 700 bp (4 SNPs) It was in the range of ~ 90 kb (16 SNP)). Core association analysis focused on the single marker allele test of QC transit SNPs using linear regression. To control for possible false associations arising from population stratification, a modified EIGENSTRAT method was used (Price et al., Nat. Genet. 38: 904 (2006)): the principal component axis as a covariate in the regression model Integrated. Other covariates considered included gender, age, HIV mode of infection, and year of HIV acquisition: only gender had an independent effect on setpoints (after inclusion of the EIGENSTRAT covariate) and thus this Included in the linear regression model. Based on a straight Bonferroni correction that accounts for all of the 535,101 polymorphs considered here to assess significance, the association is compared to a conservative threshold (P cut-off = 9.3x10-8) Declared significance of scale. Using available 4-digit HLA class I allele determinations, an analysis incorporating both SNP mutations and classical HLA typing was performed on a sub-group of 156 patients.

  These analyzes identify two independently acting groups of polymorphisms, which are related to the classical HLA loci B and C and explain, at setpoint, 15% of the total variation of HIV-1 VL . A third set of polymorphisms upstream of the ZNRD1 gene encoding the RNA polymerase subunit accounts for 5.8% of the total variation in clinical progression.

HCP5-HLA-B ^* 5701
One polymorphism located in the HLA complex P5 (HCP5) gene accounts for 9.6% of the total variation in HIV-1 setpoint, despite the minor allele frequency of 0.05, Demonstrate the degree of influence of the associated haplotype (dbSNP rs23995029, p = 9.36e-12). The median VL at setpoint is 19344 copies (cp / ml) per milliliter of blood in patients who are homozygous for common alleles, whereas one copy of minority alleles (TC, N = 45) or In patients with 2 copies (TT, N = 2) it is 592 cp / ml (FIG. 1), and at p = 9.36e-12, the association is significantly significant at the genome scale.

The HCP5 gene is located closer to 100 kb centromere from HLA-B on chromosome 6 (FIG. 2), and the related mutant is known to be in high linkage disequilibrium (LD) with the HLA allele B5701 (de Bakker et al. Nat. Genet. 38: 1166 (2006)) (r ² = 1 in our dataset). This particular HLA-B allele has the strongest protective effect described against HIV-1 disease progression (Migueles et al., Proc. Natl. Acad. Sci. USA 97: 2709 (2000)) and low It has been associated with HIV-1 viral load (Altfeld et al., AIDS 17: 2581 (2003)). Thus, in the context of genome-wide effects, HLA-B5701 is indeed the strongest host genetic factor that limits HIV-1 infection through direct effects on the initial viral load (Altfeld et al., PLoS Med. 3: e403. (2006)).

  Considering the robust functional data supporting the role of HLA-B5701 in limiting HIV, the first hypothesis is that the association observed here is reflected in tagged SNPs within HCP5. It must be by effect (de Bakker et al., Nat. Genet. 38: 1166 (2006)). However, genetics does not allow elucidation of whether this effect is solely due to B5701 or whether HCP5 mutations also contribute to the control of HIV-1. Indeed, HCP5 itself is also a novel candidate and a strong candidate for contributing to HIV-1 regulation. HCP5 is a member of the human endogenous retrovirus family (HERV) with sequence homology to the retroviral Pol gene (Kulski et al., Immunogenetics 49: 404 (1999)). Expression of the gene has been recorded in lymphocytes (Vernet et al., Immunogenetics 38:47 (1993)), suggesting potential antisense activity against retroviruses (Kulski et al., Immunogenetics 49: 404 ( 1999)). In addition, HCP5 is predicted to encode two proteins, and related polymorphisms result in amino acid substitutions in one of these.

  A model in which the newly related HCP5 variant and the HLA-B5701 allele have a combined haplotype effect on the HIV-1 setpoint escapes the restriction that HIV-1 is mediated by cytotoxic T lymphocytes (CTLs) Consistent with the observation that suppression of viremia can be maintained in B5701 elite regulators even after undergoing mutations that allow (Bailey et al., J. Exp. Med. 203: 1357 (2006). )). Perhaps most interestingly, B5701 patients consistently show lower peak viremia and do not present symptoms more frequently during acute HIV-1 infection (Altfeld et al., AIDS 17: 2581 (2003)). This suggests that the control is performed before the maximum CTL response time (McMichael et al., Nature 410: 980 (2001)). Therefore, it is an urgent priority to evaluate whether HCP5 contributes to the control observed in B5701 positive patients and cell types.

HLA-C
The second most important result in the related study, rs9264942, is located in the 5 ′ region of the HLA-C gene 35 kb away from the start of transcription (FIG. 2). This SNP accounts for 6.5% of the observed variation in VL at setpoint (Figure 1) and also shows a significant genome-wide association (p = 3.77e-09).

  Surprisingly, this same SNP is associated with a dramatic difference in HLA-C expression levels in an independent population (p = 1.69e-07, The Sanger Institute Genevar generated for the HapMap sample. Expression database (Stranger et al., PLoS Genet. 1: e78 (2005)): www.sanger.ac.uk/humgen/genevar). A protective allele leads to a lower viral load and is associated with higher expression of the HLA-C gene. This robust and independent association with HLA-C expression levels suggests for the first time that genetic control of the expression level of the classical HLA gene affects viral control.

  Further support for this role of HLA-C expression levels in limiting HIV-1 has an even higher impact on HLA-C expression identified in the same database (p = 4.42e-08) Resulting from two nearby SNPs (FIG. 2). The second mutant, rs6457374, is located closer to 3 kb to the HLA-C gene (-32 kb in the 5 'region) and has an independent effect on HLA-C expression and also HIV Associated with -1 setpoint, but not independent of rs9264942.

  As mentioned above, two SNPs (rs9264942 and rs6457374) in the HLA-C 5 'region were found to be highly associated with HLA-C expression levels in the HapMap CEU sample. Based on linear regression on normalized HLA-C expression levels in 60 CEU parents, allele regression coefficients (for genotypes encoded [0, 1, 2]) and p-value for each separately fitted SNP Are: rs9264942 (beta = 0.35, p = 3.38e-07), rs6457374 (beta = -0.29, p = 4.42e-08). If rs9264942 is in the regression model, adding rs6457374 to the model significantly improves the fit (p = 0.0008, R-square is improved from 0.36 to 0.46). If rs6457374 is in the regression model, adding rs9264942 to the model significantly improves the fit (p = 0.005, R-square is improved from 0.39 to 0.46). Based on linear regression on the HIV-1 setpoint in the Euro-CHAVI cohort, using gender and significant eigenstrat axes as covariates, the allele regression coefficient and p-value for each separately fitted SNP is: rs9264942 (beta = -39, p = 3.77e-09), rs6457374 (beta = -0.32, p = 2.68e-04). If rs9264942 is in the regression model, adding rs6457374 to the model does not improve the fit (p = 0.27, R-square = 0.15 in either case). If rs6457374 is in the regression model, adding rs9264942 to the model significantly improves the fit (p = 1.46e-06, R-square is improved from 0.10 to 0.15).

  To better describe the relationship between HLA-C expression levels and viral control, we considered each possible diploid genotype, including rs9264942 and rs64457374, and the average mRNA for each genotype observed in the Genevar database Levels (CEU parent, N = 60) were compared to the average viral load observed for each genotype in the study. These data strongly suggest a non-linear effect of increased expression levels on VL. Lower HLA-C expression level fluctuations in expression do not change viral control. However, once the threshold expression level is reached, increased HLA-C expression drastically reduces the viral load at the setpoint.

  These data provide a strong example of the role responsible for HLA-C expression levels, but since there is extensive LD throughout the HLA region, it can be determined whether the actual causative variant is elsewhere. There is a need to test directly. Specifically, a nested regression model was used to assess whether the observed association could be caused by the previously described functional alleles of HLA-A, HLA-B, and HLA-C.

  Surprisingly, HLA-C expressing SNPs are significantly associated with HLA-B5701, B27, B35Px and also with the HLA allele groups Bw4 and Bw6. However, in each case, HLA-C expression variants may explain the effect of these alleles on the HIV-1 setpoint, but the reverse is not true. If the linear regression model first includes HLA alleles (or groups of alleles), the addition of rs9264942 results in a significant increase in the explained variability (see Table 1). On the other hand, except for HCP5 / B5701, none of the HLA alleles considered significantly added to the models that already incorporated the HLA-C mutant (Table 1).

  Table 1. The effect of the HLA-C 5 'expression polymorphism rs9264942 on the setpoint is independent of the association with the HLA-B allele previously associated with HIV-1 regulation. As shown in Table 1a, the addition of rs9264942 to the linear regression model significantly improves the fit for all HLA-B alleles or allele groups suspected of having an impact on HIV disease. In contrast, after considering the effect of rs9264942, only HLA-B5701 has an independent effect. HLA-C independence is also clearly seen in the mean of HIV-1 setpoints for each rs9264942 genotype (Table 1b): minority allele C is independent of all considered alleles and allele groups , Associated with a decrease in VL. The numbers refer to a sub-group of 156 patients with available 4-digit HLA class I allele results.

This observation is interesting in the overall context of the previously proposed role of HLA-C in HIV-1 pathogenesis. HIV-1 nef selectively down-regulates HLA-A and -B expression on the infected cell surface, but not HLA-C expression (Cohen et al., Immunity 10: 661 (1999)). Originally, this strategy is that HLA-A and -B present foreign (especially viral) epitopes to CD8 T cells, resulting in cell destruction, while HLA-C binds to self-peptides and avoids NK attack. Therefore, it was considered suitable for viruses to interact with natural killer cells (NK). However, HLA-C also has the ability to present viral peptides to cytotoxic CD8 + T cells and consequently restrict HIV-1 (Goulder et al., AIDS 11: 1884 (1997)); 10 HLA C-restricted CTL epitopes are described in the LANL database ( www.hiv.lanl.gov ). These observations suggest that above this, there may be an expression threshold, where HLA-C mediated viral restriction is an effective defense mechanism against HIV-1. In such a scenario, the inability of HIV-1 nef to naturally down-regulate HLA-C molecules is an important advantage with respect to the immune system.

Other setpoint variables None of the other single markers reached genomic significance after correction for a number of studies, and any copy number variation identified was not any of the HIV-1 VL at the setpoint Also showed no association. However, among those with very low p-values, other SNPs are also expected to have a real impact. Thus, SNPs with lower p-values of 100 and their respective genes (ie, the closest genes as annotated to Illumina's HumanHap550 SNP list) are reported in Table 2. In addition, specific gene variants were examined based on their identified or suspected association with HIV-1 biology: SNPs and genes of interest are listed in Table 3 (known or suspected HIV-1 Restriction genes), Table 4 (host cell factors that interact with HIV-1, Ref (Goff, Nat. Rev. Micro. 5: 253 (2007))) and Table 5 (gene groups according to gene ontology category) ). This approach reveals some potentially interesting candidates that have a lower association with HIV-1 VL at setpoints.

  Table 2. SNP with top 100 p-value in setpoint related research

  Table 3. SNP analysis in genes encoding products known to have an effect on HIV-1 disease

  Table 4. SNP analysis in genes encoding host cell factors involved in HIV-1 biology (Goff SP. Host factors exposed by retroviruses. Nat Rev Microbiol. 2007 Apr; 5 (4): 253-63).

  Table 5. Analysis of SNPs in genes according to the immunity and inflammation terms of gene ontology (GO)

The strongest associations with ZNRD1 and advanced clinical progression include seven polymorphic sets in and near the ring finger protein 39 (RNF29) and zinc ribbon domain-containing 1 (ZNRD1) genes, respectively (rs 9261174, rs38669068, rs2074480, rs7758512, rs9261129, rs2301753 and rs2074479). This polymorphism group accounts for 5.8% of the variation in clinical progression and is slightly lacking in genome-wide significance (p = 3.89e-07). This is also mildly associated with viral load at the setpoint (p = 7.11e-03). Again in the MHC region, these mutations are closer to> 1 MB centromere than the previous candidate SNP.

  Using the Genevar expression database, ZNRD1 expression was observed to be significantly associated with the identified SNPs (p = 2.0e-03), and in particular, two of these were putative regulatory 5 ′ regions It is located 25 and 32 kb away from the gene (rs38669068 and rs9261174, respectively). Since ZNRD1 encodes an RNA polymerase subunit, if this gene actually limits HIV-1, the potential interaction with HIV-1 during transcription is the most likely causal mechanism. Efficiency in proviral transcription varies greatly between individuals. In one study, differences in transcription efficiency alone account for 64-83% of the total changes in virus production that can be attributed to post-entry cellular factors (Ciuffi et al., J. Virol. 78: 10747 (2004)).

  The second gene, RNF39, is hardly characterized but cannot be excluded as a candidate because two of the related polymorphisms result in synonyms and amino acid changes (rs2301753 and rs2074479, respectively).

No other relevant associations were observed for quantitative progression phenotypes or for case-control comparisons of progressors versus non-progressors.
To confirm the effect of these polymorphisms on confirmatory virus control in independent samples, a replication cohort of 140 Caucasian patients was established. Representative polymorphisms were selected for each of the associations reported above (HCP5, rs23995029; HLA-C 5 ′ region, rs9264942; ZNRD1 5 ′ region, rs9261174) and these were genotyped in the replication cohort. For each reported association, replication in the same direction was observed (for HCP5: p = 0.014, accounting for 4% of the variation in set points in this data set; for HLA-C: p = 2. 8e-03, accounting for 5.6% of mutations at setpoint; for ZNRD1: p = 0.048, accounting for 3.5% of mutations in progression).

  In summary, the results of the first comprehensive genome-wide association study of human genetic effects on HIV-1 disease are provided above. HLA-C expression is shown to have a strong influence on the control of HIV-1. It is possible that HER V-derived genes may contribute to the regulation known to be associated with the HLA-B5701 allele. This finding confirms the central role of the MHC region in HIV-1 restriction and provides a new way to consider its mode of action: Perhaps it is time to go beyond classical allele discrimination in the HLA region Yes. It is also noteworthy that this genome-wide study of host determinants has two whole genome significant hits and has successfully replicated all three of the strongest associations with viral load and progression. . This indicates that host response determinants can often include major influencing genetic variants, and suggests that there is urgency to conduct similar studies on other infectious diseases.

  Finally, the above results suggest that there may be two directions for therapeutic intervention. First, they show the possibility that HCP5 may contribute to the control associated with HLA-B5701, which, if true, will present an immediate therapeutic opportunity. Perhaps more importantly, it is shown that HLA-C restriction can constitute an important part of HIV-1 regulation with sufficiently high expression levels of HLA-C. The latter result indicates that HIV-1 nef is involved in virulence determination through differential down-regulation of HLA class I molecules, limiting the function of HLA-A and -B alleles, but HLA-C is resistant to nef. This suggests that HLA-C is in the spotlight. Future vaccine strategies may target HLA-C restricted T cell responses.

  All documents and other information sources cited herein are hereby incorporated by reference in their entirety.

Claims (14)

  A method for determining the risk of an asymptomatic HIV patient progressing to AIDS comprising assaying DNA from said patient for the presence of a minor allele of the rs9264942 polymorphism, wherein the presence of said minor allele is a viral load set. The method, wherein the method is associated with reduced points and a reduced risk of the patient progressing to AIDS as compared to asymptomatic HIV patients who do not possess the minority allele.   A method of determining the risk of an asymptomatic HIV patient to progress to AIDS, comprising assaying DNA from said patient for the presence of a minor allele of the rs64457374 polymorphism, wherein the presence of said minor allele is a viral load set. The method, wherein the method is associated with reduced points and a reduced risk of the patient progressing to AIDS as compared to asymptomatic HIV patients who do not possess the minority allele.   A method for determining the risk of an asymptomatic HIV patient to progress to AIDS comprising assaying DNA from said patient for the presence of a minor allele of the rs23995029 polymorphism, wherein the presence of said minor allele is a viral load set. The method, wherein the method is associated with reduced points and a reduced risk of the patient progressing to AIDS as compared to asymptomatic HIV patients who do not possess the minority allele.   A method of determining the risk of an asymptomatic HIV patient to progress to AIDS, comprising assaying DNA from said patient for the presence of a minor allele of the rs9261174 polymorphism, wherein the presence of said minor allele is said minor allele. The method wherein the patient is associated with a reduced risk of progression to AIDS as compared to an asymptomatic HIV patient who does not possess the disease.   A method for determining the risk of an asymptomatic HIV patient to progress to AIDS comprising assaying DNA from said patient for the presence of a minor allele of the rs2074480 polymorphism, wherein the presence of said minor allele is said minor allele. The method wherein the patient is associated with a reduced risk of progression to AIDS as compared to an asymptomatic HIV patient who does not possess the disease.   A method for determining the risk of an asymptomatic HIV patient to progress to AIDS comprising assaying DNA from said patient for the presence of a minor allele of the rs77558512 polymorphism, wherein the presence of said minor allele is said minor allele. The method wherein the patient is associated with a reduced risk of progression to AIDS as compared to an asymptomatic HIV patient who does not possess the disease.   A method for determining the risk of an asymptomatic HIV patient to progress to AIDS, comprising assaying DNA from said patient for the presence of a minor allele of the rs926111129 polymorphism, wherein the presence of said minor allele is said minor allele. The method wherein the patient is associated with a reduced risk of progression to AIDS as compared to an asymptomatic HIV patient who does not possess the disease.   A method for determining the risk of an asymptomatic HIV patient to progress to AIDS comprising assaying DNA from said patient for the presence of a minor allele of the rs3866908 polymorphism, wherein the presence of said minor allele is said minor allele. The method wherein the patient is associated with a reduced risk of progression to AIDS as compared to an asymptomatic HIV patient who does not possess the disease.   A method for determining the risk of an asymptomatic HIV patient to progress to AIDS comprising assaying DNA from said patient for the presence of a minor allele of the rs2301753 polymorphism, wherein the presence of said minor allele is said minor allele. The method wherein the patient is associated with a reduced risk of progression to AIDS as compared to an asymptomatic HIV patient who does not possess the disease.   A method for determining the risk of an asymptomatic HIV patient to progress to AIDS comprising assaying DNA from said patient for the presence of a minor allele of the rs2074479 polymorphism, wherein the presence of said minor allele is said minor allele. The method wherein the patient is associated with a reduced risk of progression to AIDS as compared to an asymptomatic HIV patient who does not possess the disease.   A method of screening a compound for the ability to reduce viral load setpoints in a patient infected with HIV and to reduce the risk of the patient progressing to AIDS, comprising HLA in the presence and absence of the compound A compound that increases the expression level of HLA-C, comprising culturing cells capable of expressing -C and determining the expression level of HLA-C in the presence and absence of said compound; Wherein said method is a compound suitable for use in reducing viral load setpoint and reducing the risk that said patient progresses to AIDS.   A method of screening a compound for the ability of a patient infected with HIV to reduce the risk of progression to AIDS, culturing cells capable of expressing ZNRD1 in the presence and absence of said compound, and said compound Determining the expression level of ZNRD1 in the presence and absence of a compound, wherein the compound that increases the expression level of ZNRD1 is suitable for use in reducing the risk that the patient progresses to AIDS Said method.   12. A method for reducing the risk that a patient progresses to AIDS, comprising administering to said patient an amount of a compound identifiable by the method of claim 11 sufficient to reduce said risk. Said method.   13. A method of reducing the risk that a patient will progress to AIDS, comprising administering to said patient an amount of a compound identifiable by the method of claim 12 sufficient to reduce said risk. Said method.

Images