Mutation, selection, and evolution of the Crohn disease susceptibility gene CARD15.

Three common mutations in the CARD15 (NOD2) gene are known to be associated with susceptibility to Crohn disease (CD), and genetic data suggest a gene dosage model with an increased risk of 2-4-fold in heterozygotes and 20-40-fold in homozygotes. However, the discovery of numerous rare variants of CARD15 indicates that some heterozygotes for the common mutations have a rare mutation on the other CARD15 allele, which would support a recessive model for CD. We addressed this issue by screening CARD15 for mutations in 100 CD patients who were heterozygous for one of the three common mutations. We also developed a strategy for evaluating potential disease susceptibility alleles (DSAs) that involves assessing the degree of evolutionary conservation of involved residues, predicted effects on protein structure and function, and genotyping in a large sample of cases and controls. The evolutionary analysis was aided by sequencing the entire coding region of CARD15 in three primates (chimp, gibbon, and tamarin) and aligning the human sequence with these and orthologs from other species. We found that 11 of the 100 CD patients screened had a second potential pathogenic mutation within the exonic and periexonic sequences examined. Assuming that there are no additional pathogenic mutations in noncoding regions, our study suggests that most carriers of the common DSAs are true heterozygotes, and supports previous evidence for a gene dosage model. Four novel nonsynonymous mutations were detected, one of which would produce premature termination of translation c.2686C>T (p.Arg896X). Two potential DSAs--c.2107C>T (p.Arg703Cys) and g.2238T>A (c.74-7T>A)--were significantly associated with CD in the case control sample. Analysis of the evolution of CARD15 revealed strong conservation of the encoded protein, with identity to the human sequence ranging from 99.1% in the chimp to 44.5% in fugu. Higher primates possess an open reading frame (ORF) upstream of the putative initiation site in other species that encodes a further 27 N-terminal amino acids, while four regions of high conservation are observed outside of the known domains of CARD15, indicative of additional residues of functional importance. The strategy developed here may have general application to the assessment of mutation pathogenicity and genetic models in other complex disorders.

Genetic evidence for interaction of the 5q31 cytokine locus and the CARD15 gene in Crohn disease.

A common haplotype spanning 250 kb in the cytokine gene cluster on chromosome 5q31 has recently been reported to be strongly associated with Crohn disease (CD) in Canadian families. We have replicated this finding by both the transmission-disequilibrium test (TDT) (P=.016) and in a case-control association study (P=.008) in a large European cohort of patients with CD, although the increase in disease risk was small (odds ratio 1.49 for homozygotes, 95% CI 1.11-2.0). No association was detected in families or individuals with ulcerative colitis (UC). Stratification of offspring with CD in the TDT sample by mutation status in the CD susceptibility gene CARD15 showed that the association with the 5q31 risk haplotype was present only in offspring with at least one of the known CARD15 disease susceptibility alleles (P=.044). The 5q31 risk haplotype frequency was 53.1% in unrelated individuals with CD who had one or two CARD15 mutations versus 43.7% in control subjects (P=.0001) but was not significantly elevated in individuals with CD who had no CARD15 mutations (45.4%, P=.41). Kaplan-Meier survival analysis of age at disease onset showed a significantly earlier onset in homozygotes for the 5q31 risk haplotype (P=.0019). These findings suggest that genetic variants at the 5q31 (IBD5) locus may hasten the onset of Crohn disease and cooperate with CARD15 in disease causation.

A chromosome 3-encoded repressor of the human telomerase reverse transcriptase (hTERT) gene controls the state of hTERT chromatin.

Telomerase is crucial for human carcinogenesis. The limiting component of telomerase activity is telomerase reverse transcriptase (hTERT), undetectable in differentiated somatic cells but present in most tumor cells. There is evidence that hTERT transcription is shut down by a repressor in normal cells, but the mechanisms that turn on or maintain expression in tumor cells are not understood. To identify cis-acting regulatory elements, we scanned the hTERT gene for nuclease sensitive sites. In tumor cells and in in vitro transformed fibroblasts that contain hTERT mRNA, we detected a pattern of nuclease-sensitive sites in the second intron different from that in normal fibroblasts. To test whether the chromatin state characterized by the increased nuclease sensitivity plays a role in tumor-specific hTERT expression, we used a telomerase-positive breast carcinoma line, 21NT. Introduction of a normal chromosome 3 into these cells is known to down-regulate hTERT expression, probably through transcriptional silencing. 21NT cells displayed a similar pattern of micrococcal nuclease (MNase) sensitivity to other telomerase-positive lines, whereas the hTERT chromatin of the chromosome 3-hybrids resembled that of normal fibroblasts. In segregants that had lost the normal chromosome 3, the MNase sensitivity pattern characteristic of telomerase-positive cells was restored, and some (but not all) re-expressed the hTERT gene. The simplest model compatible with these results, and with data on the mapping of an hTERT repressor on chromosome 3, is that hTERT expression in tumor cells depends on an open state of intron 2-chromatin. We propose that, during the development of the breast carcinoma from which the 21NT cell line was derived, loss of function of this repressor led to chromatin remodeling necessary (but probably not sufficient) for expression of the hTERT gene. An improved understanding of the precise mechanism of hTERT dysregulation in human cancer may well find applications in the development of antitelomerase cancer therapy.

A mortality gene(s) for the human adenocarcinoma line HeLa maps to a 130-kb region of human chromosome 4q22-q23.

Human chromosome 4 was previously shown to elicit features of senescence when introduced into cell lines that map to complementation group B for senescence, including HeLa cells. Subsequently, a DNA segment encoding the pseudogene Mortality Factor 4 (MORF4) was shown to reproduce some of the effects of the intact chromosome 4 and was suggested to be a candidate mortality gene. We have identified multiple MORF4 alleles in several cell lines and tissues by sequencing and have failed to detect any cancer-specific mutations in three of the complementation group B lines (HeLa, T98G, and J82). Furthermore, MORF4 was heterozygous in these lines. These results question whether MORF4 is the chromosome 4 mortality gene. To map other candidate mortality gene(s) on this chromosome, we employed microcell-mediated monochromosome transfer to introduce either a complete copy, or defined fragments of the chromosome into HeLa cells. The introduced chromosome 4 fragments mapped the mortality gene to a region between the centromere and the marker D4S2975 (4q27), thus excluding MORF4, which maps to 4q33-q34.1. Analysis of microsatellite markers on the introduced chromosome in 59 immortal segregants identified a frequently deleted region, spanning the markers BIR0110 and D4S1557. This defines a new candidate interval of 130 kb at 4q22-q23.

Functional evidence for a squamous cell carcinoma mortality gene(s) on human chromosome 4.

Squamous cell carcinoma (SCC) immortality is associated with p53 and INK4A dysfunction, high levels of telomerase and loss of heterozygosity (LOH) of other chromosomes, including chromosome 4. To test for a functional cancer mortality gene on human chromosome 4 we introduced a complete or fragmented copy of the chromosome into SCC lines by microcell-mediated chromosome transfer (MMCT). Human chromosome 4 caused a delayed crisis, specifically in SCC lines with LOH on chromosome 4, but chromosomes 3, 6, 11 and 15 were without effect. The introduction of the telomerase reverse transcriptase into the target lines extended the average telomere terminal fragment length but did not affect the frequency of mortal hybrids following MMCT of chromosome 4. Furthermore, telomerase activity was still present in hybrids displaying the mortal phenotype. The MMCT of chromosomal fragments into BICR6 mapped the mortality gene to between the centromere and 4q23. Deletion analysis of the introduced chromosome in immortal segregants narrowed the candidate interval to 2.7 Mb spanning D4S423 and D4S1557. The results suggest the existence of a gene on human chromosome 4 whose dysfunction contributes to the continuous proliferation of SCC and that this gene operates independently from telomeres, p53 and INK4A.

The contribution of NOD2 gene mutations to the risk and site of disease in inflammatory bowel disease.

BACKGROUND & AIMS: Mutations in the NOD2 gene are strongly associated with susceptibility to Crohn's disease (CD). We analyzed a large cohort of European patients with inflammatory bowel disease to determine which mutations confer susceptibility, the degree of risk conferred, their prevalence in familial and sporadic forms of the disease, and whether they are associated with site of disease. METHODS: Individuals were genotyped for 4 NOD2 mutations: P268S, R702W, G908R, and 3020insC. Allelic transmission distortion to 531 CD- and 337 ulcerative colitis-affected offspring was assessed by the transmission disequilibrium test. Association was also tested in an independent cohort of 995 patients with inflammatory bowel disease and 290 controls. Cases were stratified by disease site and compared across NOD2 genotypes. RESULTS: R702W, G908R, and 3020insC were strongly associated with CD but not with ulcerative colitis. Linkage disequilibrium was observed between P268S and the other mutations, forming 3 independent disease haplotypes. Genotype relative risks were 3.0 for mutation heterozygotes and 23.4 for homozygotes or compound heterozygotes. The frequency of NOD2 mutations was higher in cases from families affected only with CD and was significantly increased in ileal-specific disease cases compared with colon-specific disease (26.9% vs. 12.7%, P = 0.0004). CONCLUSIONS: The R702W, G908R, and 3020insC mutations are strong independent risk factors for CD and are associated particularly with ileal disease.