CAG expansion in the Huntington disease gene is associated with a specific and targetable predisposing haplogroup

Abstract

Huntington disease (HD) is an autosomal-dominant disorder that results from >or=36 CAG repeats in the HD gene (HTT). Approximately 10% of patients inherit a chromosome that underwent CAG expansion from an unaffected parent with <36 CAG repeats. This study is a comprehensive analysis of genetic diversity in HTT and reveals that HD patients of European origin (n = 65) have a significant enrichment (95%) of a specific set of 22 tagging single nucleotide polymorphisms (SNPs) that constitute a single haplogroup. The disease association of many SNPs is much stronger than any previously reported polymorphism and was confirmed in a replication cohort (n = 203). Importantly, the same haplogroup is also significantly enriched (83%) in individuals with 27-35 CAG repeats (intermediate alleles, n = 66), who are unaffected by the disease, but have increased CAG tract sizes relative to the general population (n = 116). These data support a stepwise model for CAG expansion into the affected range (>or=36 CAG) and identifies specific haplogroup variants in the general population associated with this instability. The specific variants at risk for CAG expansion are not present in the general population in China, Japan, and Nigeria where the prevalence of HD is much lower. The current data argue that cis-elements have a major predisposing influence on CAG instability in HTT. The strong association between specific SNP alleles and CAG expansion also provides an opportunity of personalized therapeutics in HD where the clinical development of only a small number of allele-specific targets may be sufficient to treat up to 88% of the HD patient population.

Figure 1

Genetic Diversity Is Observed in the Human HTT Region (A) Schematic representation of the location of 190 SNPs (red triangles) relative to the intron/exon structure of HTT. SNP 1 is located 5′ of the gene, whereas SNP 190 is the furthest 3′ of the gene. (B) Each of the 190 SNPs has a minor allele frequency (MAF) and the number of times each MAF occurs is plotted to demonstrate that there is a large amount of diversity in the HTT region in the HapMap populations. 23% of SNPs were very common (MAF > 0.20). Many (45%) rare SNPs (MAF < 0.05) were also detected. (C) The MAF of each SNP is aligned for each ethnic population in the HapMap cohort. The pattern is similar for each ethnic group, although the Yoruba have the most SNPs overall and tend to have higher MAFs. (D) Plot of linkage disequilibrium (LD) in the HTT region in the HapMap cohorts. Linkage across HTT is relatively high (red is high LD, white is low LD), enabling the use of tagging SNPs to sample the HD population. Chinese and Japanese have similar linkage and are pooled into an Asian grouping. As expected, in both the MAF and the LD plot, the greatest diversity is seen in Yoruba relative to Europeans and Asians.

Figure 2

Specific SNPs Are Highly Associated with CAG-Expanded Chromosomes (A) HD patient chromosomes were phased to allow comparison between the disease chromosome (>35 CAG) and control chromosome within each patient (total 65 individuals). tSNP is identified by number and its position indicated relative to HTT. Alleles are either A or B (major or minor). Allele counts are indicated (middle) and the frequency graphed (below). Twelve out of 22 tSNPs have a significantly different allele distribution between HD and control chromosomes (^∗chi-square p < 0.0023). (B) Allelic frequency on 27–35 CAG chromosomes is similar to disease chromosomes. Allele counts are indicated for phased control chromosomes (n = 116) and compared to 27–35 CAG chromosomes (n = 66) that contain an intermediate CAG tract size for HTT and may result in new mutations for HD in future generations. Eleven out of 22 tSNPs have a significantly different allele distribution between 27–35 CAG and control chromosomes (^∗chi-square p < 0.0023). These 11 associated tSNPs were found in both HD and 27–35 CAG chromosomes and appear to be common on CAG-expanded chromosomes. (C) There is no significant difference in the allele distribution between 27–35 CAG and HD chromosomes for any tSNPs.

Figure 3

CAG-Expanded Chromosomes Are Associated with Haplogroup A (A) Three major haplogroups (A, B, C) are defined with 22 tSNP positions. The nucleotide defining each haplogroup at each tSNP is shown. Variable tSNP positions are indicated (^∗). tSNPs with a gray box indicate nucleotide changes relative to haplogroup A. The amount of similarity between the haplogroups is indicated by a neighbor-joining tree (right). (B) Frequency distribution of haplogroups on HD (n = 65), 27–35 CAG (n = 66), and general population (n = 116) chromosomes. CAG-expanded chromosomes (>27 CAG) are enriched for haplogroup A relative to the general population. Chromosomes from the general population with <27 CAG phased for CAG size (right) demonstrate that high-normal CAG chromosomes also have an enrichment for haplogroup A relative to low-normal CAG chromosomes. The mean CAG tract size for each group is indicated. (C) CAG size distribution for all chromosomes on haplogroups A or C. In the chromosomes used in this study, the mean CAG sizes for haplogroup A are significantly larger (p < 0.00001, t test) than haplogroup C. The high odds ratio on haplogroup A is an indication that CAG expansion is much more likely to occur on haplogroup A chromosomes.

Figure 4

Specific Haplogroup A Variants Are Enriched on CAG-Expanded Chromosomes (A) To determine whether there are differences in haplogroup A chromosomes from CAG-expanded and normal chromosomes, haplogroup A was divided into five major variants by subtracting the common tSNPs (light gray text) and with differences at the 12 remaining tSNP positions (black text). The wildcard asterisk (^∗) is used for variable allele positions. Dark gray boxes indicate differences relative to the A1 variant. The relationship between the variants is shown by a neighbor-joining tree (right). (B) CAG-expanded chromosomes (HD, N = 62; 27–35 CAG carriers, n = 55) have similar haplogroup A variant distributions and are specifically enriched for A1 and A2 relative to chromosomes from the general population (n = 61). Phased chromosomes from the general population (right) demonstrates that large-normal chromosomes also have an enrichment for variant A1 and A2 relative to low-normal chromosomes. Variants A4 and A5 are almost absent from CAG-expanded chromosomes. (C) CAG size distribution of chromosomes in each of subgroup. Variants A1, A2, and A3 chromosomes have a broad CAG size distribution that extends from low normal (<16 CAG) to high (>50). For the chromosomes used in this study, the mean CAG size and odds ratio of each variant is indicated. The highest HD risk variants, A1 and A2, have significantly elevated mean CAG size and odds ratios >1. Variant A3 is a moderate HD risk haplotype, because it has a larger component of CAG sizes in the normal range and therefore a lower mean CAG size. Chromosomes with variant A4 or A5 are stable in the normal range.

Figure 5

Ethnic Groups that Have a Low Prevalence of HD Do Not Have HD Risk Haplotypes in Their General Population The prevalence of HD is much higher in Western European populations relative to Asia and Africa. Although the frequency of haplogroup A is similar between Europe and Asia (A), the frequencies of the high-risk variants of haplotype A, A1 and A2, are not found in the Asian populations (B). As expected, there is more genetic diversity in the Yoruba population, with a lower level of risk haplotypes and a relatively greater frequency of “other” haplotypes. Number of chromosomes assessed in each ethnic group is indicated in brackets.

Figure 6

Disease-Associated SNPs Can Be Efficiently Targeted for Allele-Specific Silencing of the Mutant HTT mRNA In an HD patient whose genotype is known, specific heterozygous alleles can be used to distinguish the CAG-expanded mRNA from nonexpanded mRNA (i.e., alleles that are 100% sensitive of the disease allele and 100% specific). Because of the expense of clinically testing and validating each target, it is important to maximize the patient coverage with a minimum number of targets. A theoretical maximum number of targetable patients (89%) exists because in this cohort, 7 of the 65 HD patients were not heterozygous at any tSNP and therefore could not be targeted. The maximum percent of the HD population in this study that could be treating with a single target (disease-associated allele) is 52%. The development of a therapy toward a second allele target would increase the patient coverage to 80%.

Figure 7

Pattern of Disease-Associated SNPs Is Consistent with a Model of Stepwise CAG Expansion on a Predisposing Haplogroup Schematic of possible models of CAG expansion. Arrows represent mutational events that result in a change in CAG tract size across generations on specific chromosomal haplogroups. (I) Stochastic model results in essentially unpredictable CAG expansion over many generations. If the primary factors influencing CAG instability are other than cis-elements (i.e., genetic trans-factors, environmental influences, or entirely stochastic processes), then CAG expansions should occur randomly on chromosomes with random genotypes. HD chromosomes should therefore have a distribution of haplogroups similar to the general population. (II) The expanded CAG founder (or large normal founder) model predicts a rare and large expansion in the CAG tract in a limited number of individuals. CAG instability then results because of the overall increase in CAG tract size. All HD individuals are related to the founder chromosome(s) and would be expected to have an identical haplogroup except for instances where recombination events have occurred. The founder population is genetically isolated from the general population and unique haplotypes should arise on CAG-expanded chromosomes only. (III) The predisposing haplogroup model predicts that CAG expansion occurs primarily on a haplogroup containing predisposing cis-elements. The cis-elements may interact with trans-factors or environmental influences, but the predisposing haplogroup (containing the cis-elements and polymorphisms in LD with the cis-elements) is required for instability to occur and is therefore the predominant haplogroup of disease chromosomes. The source of the predisposing haplogroup is not unique to intermediate-sized CAG alleles, so the origin of unstable CAG chromosomes is the general population carrying the predisposing cis-elements. The significant enrichment of HD on haplogroup A is most consistent with model III.