The implications of recent advances in coeliac disease

Replication of celiac disease UK genome-wide association study results in a US population

Celiac disease is a common disease with a prevalence of approximately 1%. A recent genome-wide association study (GWAS) and follow-up study identified eight loci significantly associated with celiac disease risk. We genotyped the top 1020 non-HLA single nucleotide polymorphisms (SNPs) from the GWAS study that were genotyped in the previous follow-up study. After quality control assessments, 975 SNPs were analyzed for association with 906 celiac disease cases and 3819 controls, using logistic regression. Additional genotype data were generated by imputation and analyzed across the regions showing the strongest statistical evidence for association. Twenty SNPs were associated with celiac disease with P < 0.01 in the current study as well as in the previous follow-up study, of which 16 had P < 0.001 and 11 had P < 1 x 10(-11). Five of eight regions identified in the follow-up study were strongly associated with celiac disease, including regions on 1q31, 3q25, 3q28, 4q27 and 12q24. The strongest associations were at 4q27, the region most strongly associated in the GWAS and follow-up study and containing IL2 and IL21, and at 3q28 harboring LPP. In addition, we provide new evidence for an association, not previously reported, on 2q31 harboring a strong candidate gene, ITGA4. In conclusion, in this first follow-up study of celiac cases from the USA, we provide additional evidence that five of eight previously identified regions harbor risk alleles for celiac disease, and new evidence for an association on 2q31. The underlying functional mutations responsible for these replicated associations need to be identified.

Coeliac disease, epilepsy and cerebral calcifications

Coeliac disease, epilepsy and cerebral calcifications (CEC) syndrome is a rare clinical condition. One hundred and seventy-one patients have been reported in the literature. Patients are mostly from Italy, Spain, and Argentina, suggesting a geographically restricted condition. Epilepsy is more frequently characterized by occipital seizures. It may be benign or drug-resistant, sometime evolving into severe epileptic encephalopathy. Gluten free diet (GFD) efficacy seems to be inversely related to the duration of epilepsy and the young age of the patient. Patients with cerebral calcifications (CC) and coeliac disease (CD) without epilepsy are considered as having an incomplete form of CEC syndrome. Some patients with epilepsy and CC without CD are supposed to have a CEC syndrome with silent or latent CD. Whether CEC syndrome is a genetic condition, or whether epilepsy and/or CC are a consequence of an untreated CD is unknown yet. Since histopathological findings seem to be the expression of vascular calcified malformation, CEC syndrome may be considered a genetically determined entity, such as a type of Sturge-Weber-like phacomatosis. Moreover, CEC, as well as CD, is associated with HLA-DQ2 and HLA-DQ8 phenotype and genotype. The progressive growth and late occurrence of CC before beginning a GFD, the demonstration of anti-gliadin antibodies in the cerebro-spinal fluid and the association with HLA class II genes, suggest that an immune reaction originating from the jejunal mucosa, triggered by gliadin in gluten intolerance predisposed subjects (HLA phenotype) may be responsible for seizures and CC. Moreover, a long-lasting untreated CD folic acid deficiency may cause calcifications. Probably, CEC is considered a genetic, non-inherited, ethnically and geographically restricted syndrome associated with environmental factors.

HLA DQA1-DQB1 genotypes in Bedouin families with celiac disease

Celiac disease (CD) has a strong genetic association with human leukocyte antigens (HLA). The primary susceptibility for CD is HLA-DQA1*05 DQB1*02 (also known as DQ2), with the remainder of cases primarily HLA-DQA1*03 DQB1*03 (also known as DQ8). In a set of nine Bedouin multiplex celiac disease families and one simplex, we genotyped DNA samples at HLA DQA1 and DQB1. Nineteen celiac disease patients had at least one DQA1*05 DQB1*02 genotype (= DQ2), 4 affecteds had the second most common genotype of DQA1*03 DQB1*0302 (= DQ8), 9 were DQ2 and DQ8, and 4 had at least one copy of DQB1*02 without the DQA1*05 genotype. Using transmission disequilibrium testing, we observed a significant over-representation in affecteds of the DQA1*05 DQB1*02 genotype (p = 0.0089), as well as over-representation of the DQA1*03 DQB1*0302 genotype (p = 0.078). The HLA DQA1 DQB1 high-risk genotypes associated with celiac disease are similar in these Bedouin families with CD to what is observed in Northern and Southern Europeans.

Celiac disease and human leukocyte antigen genotype: accuracy of diagnosis in self-diagnosed individuals, dosage effect, and sibling risk

BACKGROUND

Celiac disease is an autoimmune disorder of the small intestine characterized by intolerance to gluten. Traditionally, diagnosis is made by intestinal biopsy. Testing for immunoglobulin (Ig) A endomysial antibodies in the serum also is used for diagnosis. Biopsy and serology revert to normal with adherence to a gluten-free diet. Often, after an index case is diagnosed, siblings with symptoms adhere to a gluten-free diet without biopsy or serologic confirmation. More than 90% of patients with celiac disease have the human leukocyte antigen (HLA) DQA1*0501-DQB1*0201 genotype. Non-HLA genes also have been implicated.

METHODS

One hundred ninety-five individuals with confirmed or suspected celiac disease were identified in 73 families affected by the disease. IgA endomysial antibody testing was performed for all symptomatic family members who did not have biopsy-confirmed diagnoses. DNA samples were genotyped at D6S276 and the HLA class II loci DQA and DQB.

RESULTS

At the time sampling was begun in families, 88 of 177 (49.7%) individuals were self-diagnosed and adhering to a gluten-free diet. Ninety percent (91/101) of confirmed cases (biopsy or serology) had at least 1 copy of the DQA1*0501-DQB1*0201 genotype, whereas only 67% (46/69) of cases self-diagnosed (adherence to gluten-free diet without confirmation) had at least 1 copy. Of confirmed cases, 61% carried two copies of DQB*0201. It is estimated that the HLA association and other unlinked genes contribute approximately equally to the sibling risk of celiac disease.

CONCLUSIONS

A dosage effect of DQB1*0201 may be associated with an increased risk of celiac disease. Self-diagnosis of celiac disease is as common as confirmed diagnosis in families in the United States. Diagnosis of celiac disease on the basis of clinical response to gluten restriction is inaccurate. With long-term adherence to a gluten-free diet, serologic test results are likely to be negative. Based on HLA genotype, approximately one third of self-diagnosed individuals are unlikely to have celiac disease. However, it is not possible to determine which individuals consuming a gluten-free diet have the disease. Therefore, before starting a gluten-free diet, serologic screening and biopsy confirmation are necessary.

Association of insulin-dependent diabetes mellitus and celiac disease: a study based on serologic markers

BACKGROUND

The association of celiac disease and insulin-dependent diabetes mellitus has been known for some time. In an attempt to clarify this association, the prevalence of celiac disease among diabetic children was determined, and the risk of insulin-dependent diabetes mellitus was defined in pediatric patients with celiac disease.

METHODS

Ninety-three children with diabetes were analyzed for the presence of celiac disease-related markers (antigliadin and antiendomysial antibodies) and characteristic alterations in the intestinal mucosa. In another group, 93 children with celiac disease were screened for pancreatic autoantibodies and pancreatic beta-cell function.

RESULTS

Among children with insulin-dependent diabetes mellitus, a 6.45% prevalence of celiac disease was observed, a value significantly higher than that found among healthy controls. In contrast, only three celiac disease patients showed potential autoimmunity toward the pancreatic beta cell, a proportion not significantly different from that in the general population. Additionally, no alteration of glucose metabolism was observed in the antibody-positive patients.

CONCLUSION

The increased risk of celiac disease among patients with diabetes requires a long follow-up to determine the presence of celiac disease markers among patients with diabetes, to avoid potential malignant disease derived from untreated celiac disease. In contrast, there is no evidence to support an increased risk of insulin-dependent diabetes mellitus among children with celiac disease. In accordance with the accepted influence of diet in the development of autoimmune diabetes, a hypothetical mechanism of protection against insulin-dependent diabetes mellitus that is mediated by environmental factors related to restricted diet is suggested in this population.

Epilepsy, cerebral calcifications and clinical or subclinical coeliac disease. Course and follow up with gluten-free diet

We have studied four patients (three male, one female, age range 15-25 years) with epilepsy, bilateral occipital calcifications and latent coeliac disease (CD). The epilepsy started at mean age 7 years, in three cases there were partial seizures and in one case generalized seizure. Three cases had symptoms suggesting malabsorptive syndrome during infancy and one case was diagnosed CD before the onset of seizures. In all cases serologic markers of CD were found, especially antiendomisium antibody, and intestinal biopsy indicated several grades of atrophy. The electroencephalograph (EEG) findings pointed to focal abnormalities in three patients and generalized abnormalities in one patient. In all cases computer tomography (CT) showed bilateral, almost symmetrical occipital calcifications in the cortical subcortical layers. The enhanced CT were unremarkable and magnetic resonance images (MRI) were normal. After diagnosis of CD, all patients followed a gluten-free diet and in three patients a significant reduction in seizure frequency was observed. CD should be ruled out in all cases of epilepsy, cerebral calcifications of unexplained origin and malabsorption syndrome in infancy.

Considering wheat, rye, and barley proteins as aids to carcinogens

The increased incidence of lymphoma in celiac sprue (CS) is well documented, and the risk of developing this malignancy is 40-100-fold greater than in the general population. The author believes that gluten may also be at the root of lymphomas in asymptomatic and latent celiac sprue, as well. Among the 20-30% of the population which has the HLA factors most common in celiac, increased intestinal permeability leads to absorption of macromolecular peptides with opioid activity, which derive from pepsin digests of wheat. The presence, in the bloodstream, of these peptides may increase the risk of lymphomas for the entire hereditary group, which includes CS. Several processes contribute to the effect that is herein hypothesized, including opioid attachment at the hypothalamic-pituitary-adrenal axis (HPA), and subsequent downregulation of production of natural killer cells. This may offer an explanation for our longstanding awareness that there is an 'impaired lymphocyte reactivity against tumor cells in patients with coeliac disease' which may also apply to first-degree relatives with the same HLA markers.

Clinical aspects of coeliac disease in children with insulin-dependent diabetes mellitus

Coeliac disease (CD) is heterogeneous in its clinical presentation and pathological expression. Silent, latent and potential forms represent the submerged part of the so-called "coeliac iceberg". The association of insulin-dependent diabetes mellitus (IDDM) and CD has been widely reported. For the screening of CD in diabetic patients, anti-reticulin R1 (ARA-R1) and anti-endomysium (AEA) antibodies are more reliable markers than anti-gliadin (AGA) antibodies. Recent studies have reported an increased prevalence of CD in children with IDDM. In our experience intestinal biopsy confirmed a diagnosis of CD in 6 out of 172 diabetic patients, with a prevalence of 3.5%. Only occasionally does CD precede the onset of IDDM; more often CD is diagnosed shortly or sometimes years after the onset of diabetes. Typical gastrointestinal complaints of CD (such as diarrhoea, abdominal distension) are rare in IDDM patients, while atypical isolated signs or symptoms of CD are more common, in particular sideropenic anemia, short stature, delayed puberty, epilepsy, hypertransaminasemia, dyspeptic symptoms, herpetiform dermatitis, and recurrent aphthous stomatitis. It is recommended that all diabetic children, even those asymptomatic, should be screened yearly for CD, using a combination of AGA plus ARA-R1 and AEA.

In vitro hydrolysis of gliadin and casein peptides: secondary defect in coeliac disease shown by organ culture

BACKGROUND

Small-intestinal organ culture was used as an in vitro model of coeliac disease, studying biopsy specimens from patients with coeliac disease, cow's milk allergy, and controls.

METHODS

Organ culture incubations were done using the pure gliadin peptide B3144 (amino acid sequences 3-56 of alpha-type gliadins) and a control peptide from casein (amino acid sequences 152-193 of alpha s1-casein). The importance of using negative controls was stressed by non-specific tissue damage. By reversed-phase high-performance liquid chromatography of organ culture supernatants, 27 specimens were further investigated.

RESULTS

There was good retrieval of peptide calibration peaks after culture. Qualitative and quantitative evaluation of chromatography runs showed degradation of at least 29% of B3144 and 37% of Cas-P. Normal mucosa (controls and coeliac patients on a gluten-free diet) was able to hydrolyse peptide fractions completely, whereas incubation with damaged mucosa (coeliac disease, cow's milk allergy) left initial peptides.

CONCLUSION

It is concluded, using a pure single gliadin peptide, that deficient peptide hydrolysis in coeliac disease was a secondary event.

HLA-linked genes acting as additive susceptibility factors in celiac disease

Susceptibility to developing CD is widely accepted to be primarily associated with a particular HLA-DQ alpha beta heterodimer encoded by the DQA1*0501 and DQB1*0201 alleles in cis position on the DR3,DQ2 haplotype or in trans position by DR5,DQ7/DR7,DQ2 heterozygotes. We performed genomic HLA-DR and -DQ typing of 100 unrelated Spanish celiac children and 180 ethnically matched controls. As expected, most (92 out of 100) celiac patients carried the HLA-DQ alpha beta heterodimer, and we selected these individuals for further studies. The results corroborate that although the DQA1*0501 and DQB1*0201 genes in single dosage appear sufficient for conferring disease susceptibility, individuals homozygotes for DQB1*0201 show an increased risk. Furthermore, our data also show that those carrying the genotype DR5,DQ7/DR7,DQ2 have a significantly increased risk of developing CD as compared to those that are non-DR7 positive, also carrying the CD-associated HLA-DQ alpha beta heterodimer. This strongly suggests that there is an MHC linked non-HLA-DQ gene primarily associated with CD present on DR7,DQ2 haplotype, which should either be DR7 or in strong linkage disequilibrium with it. Our data also indicate that, as has already been suggested, another HLA-associated CD susceptibility gene may be present on some DR4-carrying haplotypes.