Celiac Disease, Susceptibility To, 1

A number sign (#) is used with this entry because a susceptibility to celiac disease-1 (CELIAC1) has been found to be genetically determined by possession of specific HLA-DQ alleles (see 146880 and 604305) on chromosome 6p21.3; this region has been designated CELIAC1.

Description

Celiac disease, also known as celiac sprue and gluten-sensitive enteropathy (GSE), is a multifactorial disorder of the small intestine that is influenced by both environmental and genetic factors. It is characterized by malabsorption resulting from inflammatory injury to the mucosa of the small intestine after the ingestion of wheat gluten or related rye and barley proteins (summary by Farrell and Kelly, 2002). Long regarded as gastrointestinal disorder of childhood, the disease is now considered to be a chronic systemic autoimmune disease and is more often diagnosed in adults than in children (Monsuur et al., 2005).

For a discussion of genetic heterogeneity of celiac disease, see MAPPING.

Clinical Features

In individuals with celiac disease, ingestion of gluten leads to inflammation and tissue remodeling of the intestinal mucosa, resulting in malnutrition and severe complications (summary by Monsuur et al., 2005).

Because of anecdotal accounts of an association between celiac disease and epilepsy with cerebral calcifications that resembled those of Sturge-Weber syndrome (185300), Gobbi et al. (1992) studied 43 patients: 31 patients with cerebral calcifications of unexplained origin and epilepsy underwent intestinal biopsy, and 12 patients with celiac disease and epilepsy underwent computed tomography. Of the first series, 24 were identified as having CD on the basis of flat intestinal mucosa; 15 of 22 had a high concentration of serum antigluten. Of the 12 patients in the second series, 5 showed cerebral calcifications, giving a total of 29 cases with the combination of CD, epilepsy, and cerebral calcifications (CEC). In 27 of the 29, calcifications were located in the parietooccipital regions. Only 2 patients of the first series had gastrointestinal symptoms at the time of intestinal biopsy; however, most patients had recurrent diarrhea, anemia, and other symptoms suggestive of CD in the first 3 years of life. The epilepsy in CEC patients was poorly responsive to antiepileptic drugs. Gluten-free diet beneficially affected the course of epilepsy only when started soon after epilepsy onset. Cases of 'atypical Sturge-Weber syndrome' (characterized by serpiginous cerebral calcifications and epilepsy without facial port-wine nevus) should be reviewed for possible celiac disease, and CD should be ruled out in all cases of epilepsy and cerebral calcifications of unexplained origin.

Maki and Collin (1997) noted that in some cases of celiac disease, extraintestinal symptoms such as dermatitis herpetiformis (DH; 601230) and neurologic symptoms are also present.

Fasano (2003) listed the atypical clinical manifestations of celiac disease, which include osteoporosis, chronic fatigue, irritable bowel, and miscarriage.

Reviews

Farrell and Kelly (2002) reviewed all aspects of celiac sprue.

Diagnosis

Of 30 asymptomatic children with a genetic risk for celiac disease and positive transglutaminase antibodies studied by Hoffenberg et al. (2000), 21 had definite and 4 possible small bowel biopsy evidence for celiac disease, yielding a positive predictive value of 70 to 83% by antibody detection. Maki et al. (2003) screened serum samples from 3,654 Finnish students, aged 7 to 16 years, and screened them for endomysial and tissue transglutaminase antibodies. In 56 (1.5%) the antibody tests were found to be positive, despite the fact that none of the subjects had received a clinical diagnosis of celiac disease (although 10 who had positive tests for both antibodies and serum in early stages of the study later received a diagnosis of celiac disease). In 27 of 36 subjects with positive antibody assays who agreed to have intestinal biopsy, 27 had evidence of celiac disease. Thus, the estimated biopsy-proved prevalence was 1 case in 99 children. All but 2 of the antibody-positive subjects had either the HLA-DQ2 or the HLA-DQ8 haplotype. The prevalence of the combination of antibody positivity and an HLA haplotype associated with celiac disease was 1 in 67.

Fasano (2003) urged increased awareness of the celiac disease on the part of health care professionals, especially primary care physicians, because of the high rate of morbidity related to untreated celiac disease and the typical delay in diagnosis. He also urged a low threshold for the use of serologic tests as pivotal both to alleviate the social and personal costs of the disease and to increase the quality of life for the many persons affected by celiac disease.

Population Genetics

Trier (1991) pointed to prevalence rates of celiac disease ranging from 1:300 in western Ireland (Mylotte et al., 1973) to between 1:1,000 and 1:2,000 in other regions of Europe.

In the United States, the disease is exceedingly rare when the criteria for diagnosis rely on classic symptoms such as diarrhea and short stature (Rossi et al., 1993), but by broadening the clinical indication Hill et al. (2000) found that antibody screening indicated a higher prevalence, similar to that in Europe.

Greater awareness of the presentation of celiac disease and the availability of accurate serologic tests led to the realization that the disorder is relatively common, affecting 1 of every 120 to 300 persons in both Europe (Johnston et al., 1998; Catassi et al., 1996) and North America (Not et al., 1998).

Maki et al. (2003) estimated that the prevalence of celiac disease among Finnish schoolchildren is at least 1 case in 99 children.

With a prevalence close to 1% (Dube et al., 2005), celiac disease is the most common food intolerance in general Western populations.

Zhernakova et al. (2010) assessed signatures of natural selection for 10 confirmed celiac disease loci in several genomewide datasets comprising 8,154 controls from 4 European (British, Dutch, Italian and Finnish) populations and 195 individuals from a North African (Saharawi) population. They found consistent signs of positive selection for celiac disease-associated alleles at 3 loci, with the strongest signatures of selection observed for rs17810546 from the IL12A locus (CELIAC10; 612008). Consistent signs of positive selection were also seen for rs3184504 at SH2B3 (CELIAC13; 612011) in the European populations, and rs917997 from the IL18RAP locus (CELIAC8; 612006) showed a borderline-significant signature for the European and North African populations. Functional investigation of the SH2B3 genotype in response to lipopolysaccharide and muramyl dipeptide showed that carriers of the SH2B3 rs3184504 risk allele showed stronger activation of the NOD2 recognition pathway, suggesting that SH2B3 plays a role in protection against bacterial infection and providing a possible explanation for the selective sweep.

By serum antiendomysial antibody (AEA) analysis and histologic confirmation, Catassi et al. (1999) determined that CD prevalence in a Saharawi population living in Algeria was 5.6%, more than 5 times that in any European population. Catassi et al. (1999) speculated that celiac enteropathy may confer protection from intestinal infections/parasites in this population.

Pathogenesis

Falchuk et al. (1972) found that a particular HLA type (A8) showed an abnormally high frequency in patients with this disorder. They interpreted this as indicating the presence of an abnormal 'immune response gene,' leading to the production of pathogenic antigluten antibody, or, alternatively, a particular membrane configuration leading to binding of gluten to mucosal cells with subsequent tissue damage.

Trier (1991) reviewed all aspects of celiac sprue, including genetic factors. Concordance for celiac sprue in identical twins approximates 70% as compared with 30% concordance in HLA-identical sibs. The disease had been found to be associated with HLA-DR3 and HLA-DQw2 in all populations tested. The proportion of gamma-delta TCR (T cell receptor)-bearing intraepithelial lymphocytes is increased in the jejunum of patients with active celiac disease.

By studying healthy first-degree relatives of celiac patients, Holm et al. (1992) found that 45 of 109 (41%) had an increased density of gamma-delta T cells in their mucosa and 66% had an increased density of alpha-beta T cells. By contrast with alpha-beta T cells, the density of gamma-delta cells was significantly associated with genetic markers for celiac disease susceptibility: DR3, DQA (146880), and DQB (604305). They also found a dosage effect of DQA and DQB genes on the number of intraepithelial gamma-delta T cells.

IgA antibodies to endomysium, a structure of the smooth muscle connective tissue, are particularly specific indicators of celiac disease, suggesting that this structure contains one or more target autoantigens that play a role in the pathogenesis of the disease. Dieterich et al. (1997) identified tissue transglutaminase (190196) as the endomysial autoantigen involved.

Karell et al. (2002) evaluated the role of the HLA-DQ locus in 25 families in which both classic CD and dermatitis herpetiformis (601230) occurred in sibs. By using a family-based approach, they assumed that within each family, variation in environmental factors was substantially lower than in the standard case-control setting, and that the problems related to population stratification could be avoided. Results from Finnish family material comprising 25 discordant and 85 concordant sib pairs, and from case-control material comprising 71 unrelated Hungarian dermatitis herpetiformis and 68 classic CD patients, together indicated that the HLA-DQ locus did not differ between the 2 major outcomes of gluten-sensitive enteropathy. Karell et al. (2002) concluded that the non-HLA-DR;DQ factors are crucial for the different clinical manifestations of gluten sensitivity.

McManus and Kelleher (2003) reviewed the pathogenic mechanisms underlying celiac disease. Tissue transglutaminase, which is the target of endomysial autoantibody (Dieterich et al., 1997), is expressed on the subepithelial layer of intestinal epithelium, where it deamidates the glutamine residues in gliadin, resulting in glutamic acids. Deamidated peptides adhere strongly to the binding grooves of HLA-DQ2 and DQ8 molecules and elicit strong T-cell responses (Molberg et al., 1998).

Gianfrani et al. (2003) found that CD8 (see 186910)-positive T cells from peripheral blood of CD patients on gluten-free diets, but not cells from patients on gluten-containing diets or healthy controls, produced IFNG (147570) in response to a gliadin-derived peptide. However, gliadin peptide-specific cells were isolated from small intestinal mucosa of patients on either diet, and these cells were capable of recognizing full-length gliadin in an MHC class I-restricted manner.

Vader et al. (2003) noted that although HLA-DQ2-bound peptides are thought to cause CD, only HLA-DQ2.5, and not HLA-DQ2.2, predisposes to disease. They compared gluten-specific T-cell responses in the context of HLA-DQ2.5 and HLA-DQ2.2 molecules and found that HLA-DQ2.5 could present a large repertoire of gluten peptides, whereas HLA-DQ2.2 could present only a subset of these peptides. Peptide presentation was superior in HLA-DQ2 homozygous antigen-presenting cells compared with HLA-DQ2/non-DQ2 cells. HLA-DQ2.5/DQ2.2 heterozygous cells induced intermediate levels of proliferation and cytokine secretion. Vader et al. (2003) concluded that there is an HLA-DQ2 gene dose effect in the development of CD. They suggested that, since HLA-DQ expression is regulated by MHC2TA (600005), which is in turn influenced by IFNG, local responses to infection may lead to increased gluten-specific T-cell responses, particularly in HLA-DQ2.5 homozygotes.

Kim et al. (2004) presented results indicating that the HLA association in celiac disease can be explained by a superior ability of DQ2 to bind the biased repertoire of proline-rich gluten peptides that have survived gastrointestinal digestion and that have deamidated by tissue transglutaminase. Furthermore, they showed that surface-exposed proline residues in the proteolytically resistant ligand could be replaced with functionalized analogs, thereby providing a starting point for the design of orally active agents for blocking gluten-induced toxicity.

HLA-DQ2 has a key role in the disease by presenting gluten peptides to CD4+ cells in the lamina propria of the intestine (Arentz-Hansen et al., 2000). Much of the research on celiac disease was focused on the activation and regulation of gluten-reactive T cells. The etiologic steps that preceded this T-cell activation were poorly understood; for example, YRB antigenic gluten peptides are resistant to further breakdown in the intestinal lumen, and how do the gluten peptides pass through the intestinal barrier? Evidence that the integrity of the intestinal barrier is impaired in active celiac disease (Schulzke et al., 1995; Van Elburg et al., 1993) implies that the epithelial cell barrier has evolved in the early pathogenesis of the disease.

Meresse et al. (2004) studied intestinal CTLs obtained from healthy subjects and patients with CD. They noted that, in CD patients, intraepithelial CTLs are constitutively exposed to high levels of IL15 (600554). Meresse et al. (2004) found that, in the effector stage, CD8-positive CTLs from healthy subjects could be armed for NKG2D (602893)-mediated lysis by IL15. Intraepithelial CTLs from CD patients, unless they were on a gluten-free diet, also expressed high levels of NKG2D, as well as MIC (MICA; 600169) and DAP10 (604089), and exhibited significant NK-like activity accompanied by ERK (see 176948) activation.

Using immunohistopathologic and flow cytometric analyses, Hue et al. (2004) found increased expression of MICA at epithelial cell surfaces in CD patients exposed to gliadin. Patient biopsy specimens exposed to an IL15-inducing gliadin fragment did not express MICA in the presence of anti-IL15. Cytotoxicity assays showed that NKG2D played primarily a costimulatory role on intraepithelial lymphocytes (IELs), with a TCR-mediated signal required for complete activation. However, in refractory celiac sprue patients, NKG2D mediated a direct activating signal. ELISA detected soluble MICA in serum in half of untreated CD patients, but in few patients on gluten-free diets; the presence of soluble MICA was independent of MICA genotype. Hue et al. (2004) proposed that villous atrophy in CD may be ascribed to IEL-mediated damage to enterocytes involving NKG2D-MICA interaction after gliadin-induced expression of MICA on gut epithelium.

Shan et al. (2002) identified a 33-mer peptide that is glutamine and proline rich that has several characteristics suggesting it is the primary initiator of inflammatory response to gluten in celiac sprue patients. In vitro and in vivo studies in rats and humans demonstrated that this 33-mer is stable toward breakdown by all gastric, pancreatic, and intestinal brushborder membrane proteases. The peptide reacted with tissue transglutaminase, a major autoantigen in celiac sprue, with substantially greater selectivity than natural substrates of this extracellular enzyme. The 33-mer was a potent inducer of gut-derived human T cell lines from 14 of 14 celiac sprue patients. Homologs of this peptide were found in all food grains that are toxic to celiac sprue patients but were absent from all nontoxic food grains. The peptide could be detoxified in in vitro and in vivo assays by exposure to a bacterial prolyl endopeptidase, suggesting a strategy for oral peptidase supplement therapy for celiac sprue.

Hovhannisyan et al. (2008) showed that the HLA-DQ8 beta-57 polymorphism (lacking the canonical aspartic acid residue at position 57) promotes the recruitment of T cell receptors bearing a negative signature charge in the complementary determining region 3-beta (CDR3-beta) during the response against native gluten peptides presented by HLA-DQ8 in celiac disease. These T cells showed a cross-reactive and heteroclitic (stronger) response to deamidated gluten peptides. Furthermore, gluten peptide deamidation extended the T cell receptor repertoire by relieving the requirement for a charged residue in CDR3-beta. Thus, Hovhannisyan et al. (2008) concluded that the lack of a negative charge at position beta-57 in MHC class II was met by negatively charged residues in the T cell receptor or in the peptide, the combination of which might explain the role of HLA-DQ8 in amplifying the T cell response against dietary gluten.

In mice, DePaolo et al. (2011) found that in conjunction with IL15, a cytokine greatly upregulated in the gut of celiac disease patients, retinoic acid rapidly activates dendritic cells to induce JNK (also known as MAPK8, 601158) phosphorylation and release the proinflammatory cytokines IL12p70 (see 161561) and IL23 (see 605580). As a result, in a stressed intestinal environment, retinoic acid acted as an adjuvant that promoted rather then prevented inflammatory cellular and humoral responses to fed antigen. DePaolo et al. (2011) concluded that their data showed an unexpected role for retinoic acid and IL15 in the abrogation of tolerance to dietary antigens.

Inheritance

Celiac disease has a strong heritable component, although the inheritance is complex and multifactorial (summary by Monsuur et al., 2005).

McDonald et al. (1965) suggested that the mechanism of inheritance is autosomal dominant with incomplete penetrance. Frezal and Rey (1970) reviewed the subject and concluded that mendelism is unlikely. Familial aggregation was undoubted, however. Of 3 pairs of carefully studied identical twins, only one pair was concordantly affected. The authors thought this made a single gene hypothesis unlikely, especially in view of the invariable anatomic relapse on reexposure to gluten, even without clinical or biochemical signs.

Robinson et al. (1971) concluded that celiac disease is multifactorial, the causative genetic component being polygenic and interacting with environmental factors.

David and Ajdukiewicz (1975) found that 13 of 141 cases (9%) of overt, biopsy-proven celiac disease had a definitely affected relative. Greenberg et al. (1982) presented additional data supporting the hypothesis that GSE results from the interaction of 2 loci: one linked to HLA and associated with recessive inheritance and the other a non-HLA-linked GSE-associated B-cell alloantigen also exhibiting recessive inheritance ('a recessive-recessive 2-locus model'). Greenberg and Lange (1982) rejected a dominant-recessive 2-locus model, but could not reject a double recessive model. Furthermore, they concluded that affected sib pair HLA data are inconsistent with single-locus dominant or recessive models with environmentally-caused reduced penetrance.

Weiss et al. (1983) found antigliadin antibody in GSE patients on a gluten-free diet only when they had the IgG immunoglobulin heavy chain allotype marker G2m(n). Antibody occurred in these persons regardless of whether HLA-B8 and/or HLA-DR3 antigen was present.

Analyzing published pedigrees, Tiwari et al. (1984) concluded that a gene 'with a frequency of 0.022, which is nearly recessive on the penetrance scale,' is responsible; that less than one-eighth of DR3 and DR7 haplotypes carry a determinant for celiac disease and that the determinant is linked to HLA.

Simoons (1981) observed a negative correlation between the frequency of antigen HLA-B8 and the length of time that wheat farming has been practiced in various parts of Europe. He put forward the hypothesis that this gene was selected against because of associated gluten intolerance (celiac disease).

Lin et al. (1985) stated that 'available evidence is most consistent with the hypothesis that the genetic predisposition to GSE is due to disease alleles at two unlinked loci.' One of the loci is linked to HLA. They reviewed the high concordance in monozygotic twins. The concordance for celiac disease in monozygotic twins has been estimated to be 71%. However, Salazar de Sousa et al. (1987) reported an instance of identical twins in whom the diagnosis of celiac disease was made in one at age 2.5 years and in the other at age 10.5 years, intestinal biopsy at age 3 years and 10 months having been normal. Salazar de Sousa et al. (1987) questioned whether discordance for celiac disease is ever permanent.

Susceptibility to celiac disease has a strong genetic component, demonstrated by a high prevalence rate (10%) among first-degree relatives and a high concordance rate (approximately 70%) among monozygotic twins (Sollid and Thorsby, 1993).

Howell et al. (1988) described a RFLP haplotype of the HLA-D region that is associated with celiac disease. Mannion et al. (1993) reported that 1 extended MHC haplotype accounted for 50% of haplotypes from celiac patients and only 27% of MHC haplotypes in 'nontransmitting' parents.

Genes outside the HLA region are likely to be involved in the genetic susceptibility as the concordance rate among monozygotic twins (70%) differs from that among HLA-identical sibs (30%) (Strober, 1992).

Dermatitis herpetiformis (601230) is frequently associated with duodenojejunal villous atrophy similar to that found in gluten-sensitive enteropathy (Brow et al., 1971). Like celiac disease, dermatitis herpetiformis shows a high frequency of HLA-A8 (Falchuk et al., 1972; Katz et al., 1972). DH and CD are gluten-sensitive diseases that have a common immunogenetic background; both disorders are associated with HLA alleles DQA1*0501 and B1*0201. Reunala (1996) reported on the familial incidence of DH in a prospective study started in 1969 in Finland. A total of 1,018 patients with DH were diagnosed and questioned for positive family histories. Of the 999 unrelated DH patients, 105 (10.5%) had 1 or several affected first-degree relatives. Disease in the relatives was either DH (4.4%) or CD (6.1%). Analysis of the 105 families showed that 13.6% of parents, 18.7% of sibs, and 14% of children were affected, a segregation pattern that fitted well to a mendelian dominant mode of inheritance. Gender may also be important because the first-degree relatives affected with DH were more often females and those affected with CD twice as often females as males.

Based on a sib recurrence risk of 10% and a population prevalence of 0.0033, the overall sib relative risk for celiac disease is 30. Bevan et al. (1999) examined haplotype-sharing probabilities across the MHC region in 55 families with celiac disease. Based on these probabilities, they calculated the sib relative risk of celiac disease associated with the MHC region to be 3.7. They combined this result with published data and estimated the sib relative risk associated with the MHC region to be 3.3. Under the assumption of a multiplicative interaction between HLA- and non-HLA-linked loci, Bevan et al. (1999) concluded that MHC genes contribute no more than 40% of the sib familial risk of celiac disease and that non-HLA-linked genes are likely to be the stronger determinant of celiac disease susceptibility.

Hervonen et al. (2000) studied 6 monozygotic twin pairs found among 1,292 prospectively collected patients with dermatitis herpetiformis in Finland. Three of the 6 twin pairs were concordant for dermatitis herpetiformis. Two other twin pairs were partially discordant: 1 of each pair had dermatitis herpetiformis and celiac disease, whereas the other had only celiac disease. Only 1 pair was found to be discordant for gluten sensitivity. All the pairs had typical risk alleles for gluten sensitivity, i.e., either HLA-DQ2 or -DQ8. These results demonstrated that the genetic component in gluten sensitivity as broadly defined is very strong (5/6 concordant). Hervonen et al. (2000) concluded that genetically identical individuals can have clearly distinguished phenotypes, either dermatitis herpetiformis or celiac disease, suggesting that environmental factors determine the exact phenotype of this multifactorial disease.

Bourgain et al. (2000) developed a method, the maximum identity length contrast (MILC) statistic, for the study of multifactorial diseases in isolated populations. The major characteristic of MILC is that, unlike most previously proposed methods for the study of complex diseases in founder populations, it does not make the assumption that all carriers of a susceptibility allele inherited it from a single common ancestor. Although this assumption may be true for the rare mutations involved in monogenic diseases, it is likely to be irrelevant for many genetic risk factors involved in complex diseases. The MILC method relies on the fact that, even though not all affected individuals carrying the susceptibility allele had received it from the same single ancestor, they have a stronger kinship coefficient at the disease locus than do unaffected individuals. Affected individuals are thus more likely to share common haplotypes in the vicinity of the disease locus than are control subjects. The MILC approach compares the identity length of parental haplotypes that are transmitted to affected offspring with the identity length of those that are not transmitted to affected offspring. Bourgain et al. (2001) used the MILC approach to analyze the role of HLA in celiac disease and showed that the effect of HLA may be detected with the MILC approach by typing only 11 affected individuals who were part of a single large Finnish pedigree.

Greco et al. (2002) stated that the genetic contribution to celiac disease had previously been inferred from case series and anecdotal case reports of concordant twin pairs. By crossmatching a registry of individuals with celiac disease with a national twin registry, Greco et al. (2002) were able to study 47 twin pairs with at least 1 affected twin. Zygosity was confirmed in all same-sex cases using standard techniques. Each individual was typed for HLA class II DRB1 (142857) and DQB1. Concordance rates for MZ twins were higher than for DZ twins (0.86 vs 0.20). A logistic regression model, corrected for age, sex, number of shared HLA haplotypes, and zygosity showed that genotypes DQA1*0501/DQB1*0201 and DQA1*0301/DQB1*0302 conferred to the non-index twin a relative risk of contracting the disease of 3.3 and 1.4, respectively. The relative risk of being concordant for celiac disease for the non-index twin of an MZ twin pair was 17 (95% CI, 2.1-134), independent of the DQ at-risk genotype. This study provided evidence for a very strong genetic component to celiac disease, only partially due to HLA genotype.

Naluai et al. (2008) reviewed the genetics of celiac disease with an emphasis on the non-HLA genetic component, and discussed the difficulties of searching for susceptibility genes in disorders with complex inheritance patterns.

Mapping

CELIAC1 on Chromosome 6p21.3

Liu et al. (2002) performed a genomewide scan of CD in 60 Finnish families and identified strong evidence for linkage to the HLA region at chromosome 6p21.3. The role of HLA-DQ was studied in more detail by analysis of DQB1 alleles (604305) in all 98 families. All but 1 patient carried 1 or 2 HLA-DQ risk alleles, and 65% of HLA-DQ2 carriers were affected.

Using pooled data from the European Genetics Cluster on Coeliac Disease, Babron et al. (2003) performed meta- and megaanalyses of genotype data from 2,025 individuals, 1,056 of whom had celiac disease. They confirmed the association to the HLA region.

CELIAC2 on Chromosome 5q31-q33

See CELIAC2 (609754) for a celiac disease susceptibility locus on chromosome 5q31-q33.

CELIAC3 on Chromosome 2q33

See CELIAC3 (609755) for a celiac disease susceptibility locus on chromosome 2q33. This locus is associated with variation in the CTLA4 gene (123890).

CELIAC4 on Chromosome 19p13.1

See CELIAC4 (609753) for a celiac disease susceptibility locus on chromosome 19p13.1. This locus is associated with mutation in the MYO9B gene (602129).

CELIAC5 on Chromosome 15q11-q13

See CELIAC5 (607202) for a celiac disease susceptibility locus on chromosome 15q11-q13.

CELIAC6 on Chromosome 4q27

See CELIAC6 (611598) for a celiac disease susceptibility locus on chromosome 4q27 within a linkage disequilibrium (LD) block encompassing the KIAA1109 (611565), TENR (ADAD1), IL2 (147680), and IL21 (605384) genes.

CELIAC7 on Chromosome 1q31

See CELIAC7 (612005) for a celiac disease susceptibility locus on chromosome 1q31. This locus may be associated with variation in the RGS1 gene (600323).

CELIAC8 on Chromosome 2q11-q12

See CELIAC8 (612006) for a celiac disease susceptibility locus on chromosome 2q11-q12 within a linkage disequilibrium block encompassing the IL18RAP (604509) and IL18R1 (604494) genes.

CELIAC9 on Chromosome 3p21

See CELIAC9 (612007) for a celiac disease susceptibility locus on chromosome 3p21 within a linkage disequilibrium block encompassing a cluster of chemokine receptor genes.

CELIAC10 on Chromosome 3q25-q26

See CELIAC10 (612008) for a celiac disease susceptibility locus on chromosome 3q25-q26 within a 70-kD linkage disequilibrium block near the IL12A gene (161560).

CELIAC11 on Chromosome 3q28

See CELIAC11 (612009) for a celiac disease susceptibility locus on chromosome 3q28 within a linkage disequilibrium block near the LPP gene (600700).

CELIAC12 on Chromosome 6q25

See CELIAC12 (612010) for a celiac disease susceptibility locus on chromosome 6q25 within a linkage disequilibrium block encompassing the TAGAP gene (609667).

CELIAC13 on Chromosome 12q24

See CELIAC13 (612011) for a celiac disease susceptibility locus on chromosome 12q24. This locus may be associated with variation in the SH2B3 gene (605093).

Genomewide Linkage and Association Studies

Zhong et al. (1996) presented evidence for linkage of at least 1 non-HLA locus to celiac disease. They commented that the HLA component of celiac disease (a specific HLA-DQ heterodimer) is largely established and relatively uncomplicated; furthermore, the environmental component (gluten and related grain storage proteins in the diet) is well understood. Previous work had suggested that at least 1 non-HLA locus might be a stronger genetic factor than HLA, and that it may operate as an autosomal recessive. Zhong et al. (1996) used a 3-step genome screening protocol to identify loci that contribute to celiac disease in the western counties of Ireland, a region with the highest prevalence of celiac disease in the world (Mylotte et al., 1973). The most significant of several possible non-HLA loci that they found was on chromosome 6p about 30 cM telomeric from HLA. It had a multipoint maximum lod score of 4.66 (compared with 4.44 for HLA-DQ) and appeared to have a recessive mode of inheritance.

In linkage analysis of 28 celiac disease families, Houlston et al. (1997) could find no support for a predisposition locus telomeric to HLA on chromosome 6p. There also was no significant evidence in favor of linkage to 8 other chromosomal regions that had previously been suspect. There was, however, excess sharing of alleles close to D15S642. The maximum nonparametric linkage score was 1.99 (P = 0.03). Although the evidence for linkage of celiac disease to 15q26 was not strong, the well established association between celiac disease and insulin-dependent diabetes mellitus, together with the mapping of an IDDM susceptibility locus (IDDM3; 600318) to 15q26, provided indirect support for this as a candidate locus conferring susceptibility to celiac disease in some families.

In an independent study, Brett et al. (1998) were also unable to replicate the results of Zhong et al. (1996). They granted that the difficulties involved in analyzing complex diseases mean that one cannot be certain that the regions pointed to by Zhong et al. (1996) did not, in fact, harbor susceptibility loci, at least in some families.

King et al. (2001) reviewed various studies on linkage to non-HLA loci for celiac disease. Their own genomewide linkage study (King et al., 2000) genotyped 16 highly informative pedigrees for 400 microsatellite markers spanning the entire genome with an average marker spacing of 10 cM. They identified 17 potentially linked regions with lod scores significant at p less than 0.05. In a later study, King et al. (2001) investigated these 17 regions in a larger set of pedigrees using more finely spaced markers. The study involved 34 additional highly informative pedigrees to a total of 50 multiply affected families. By 2-point and 3-point linkage analysis using classic and model-free methods, they identified 5 potential susceptibility loci with heterogeneity scores greater than 2.0. The most significant was a heterogeneity lod of 2.6 at D11S914 on 11p11. This marker mapped to a position implicated in 1 of the 2 previous genome scans; taken together, these results provided strong support for the existence of a susceptibility locus in that region.

Holopainen et al. (2001) studied 102 Finnish families with affected sib pairs who had gluten sensitivity and found evidence of linkage to 11q (maximum lod score, 1.37) but no linkage to 5q. Heterogeneity between subgroups was suggested: families with only intestinal disease showed linkage mainly to 2q33, whereas families with dermatitis herpetiformis showed linkage to 11q and 5q, but not to 2q33. Linkage in all 3 non-HLA loci was strongest in families with predominantly male patients; HLA-DQ2 conferred much stronger susceptibility to females than males. Holopainen et al. (2001) noted that possible linkage heterogeneity suggested that there are genetic differences between intestinal and skin manifestations and for the gender-dependent effect of HLA-DQ2.

Naluai et al. (2001) performed a genomewide scan including 398 microsatellite markers on 106 Swedish and Norwegian families with at least 2 affected sibs with CD. By nonparametric linkage (NPL) analysis, 8 chromosomal regions besides the HLA-region on 6p were found to have nominal P values below 0.05. The regions were 2q11-q13, 3p24, 5q31-q33, 9p21, 11p15, 11q23-q25, 17q22, and Xp11, among which 5q and 11q both had been suggested in previous linkage studies. The finding on chromosome X was interesting due to the excess of affected females in CD.

In 60 Finnish families, Liu et al. (2002) identified suggestive evidence for linkage at 6 non-HLA chromosome regions. They analyzed linkage for 3 of these regions in an additional 38 families and found that the chromosome 4p15 region gave a lod score of 3.25 on multipoint analysis with dense markers.

To identify non-HLA loci for CD, Neuhausen et al. (2002) performed a genomewide search on 62 families with at least 2 cases of CD. Accounting for multiple testing, they found genomewide intermediate linkage evidence at 18q (hlod = 3.6) and at 3p (hlod = 3.2), and suggested linkage at 5p (hlod = 2.7). They did not find a good consensus between 2-point and multipoint evidence, and after genotyping with new markers in the regions stated, the results were inconclusive.

To identify risk variants contributing to celiac disease susceptibility other than those in the HLA-DQ region, Hunt et al. (2008) genotyped 1,020 of the most strongly associated non-HLA markers identified by van Heel et al. (2007) in an additional 1,643 cases of celiac disease and 3,406 controls. The most significant SNP outside the HLA-DQ region and the previously identified IL2/IL21 region (see CELIAC6, 611598) was rs2816316 (p overall = 2.58 x 10(-11)), located on chromosome 1q31 (see CELIAC7, 612005). Six additional regions showed significant association (see CELIAC8, 612006; CELIAC9, 612007; CELIAC10, 612008; CELIAC11, 612009; CELIAC12, 612010; and CELIAC13, 612011).

In an Italian cohort involving 538 patients with celiac disease and 593 healthy controls, Romanos et al. (2009) analyzed 9 SNPs tagging 8 celiac disease loci previously reported by Hunt et al. (2008) and confirmed association with 6 of them, but found no association with the CELIAC8 locus on chromosome 3p21 or CELIAC9 locus on chromosome 2q11-q12. The authors noted that this was the first celiac disease association study in a southern European cohort, and suggested that there may be population differences across Europe regarding the loci contributing to celiac disease.

Animal Model

GSE in Irish setter dogs was proposed by Batt et al. (1984, 1987) as an animal model for human celiac disease. As in the human, the gluten fraction of wheat induces various degrees of villus atrophy, and withdrawal of wheat from the diet leads to recovery of the normal villus structure and a remission of clinical signs. Polvi et al. (1998) studied the canine major histocompatibility complex (MHC) in 2 large families of gluten-sensitive Irish setter dogs. They could detect no linkage between GSE and the MHC class II gene cluster.