Association of the matrix metalloproteinase-3 (MMP-3) promoter polymorphism with celiac disease in male subjects.

Matrix metalloproteinases (MMPs), such as stromelysin-1 (MMP-3), are a family of enzymes important in resorption and remodeling of the extracellular matrix whose degradation may play a role in the villous atrophy characteristic of celiac disease (CD). We investigated the association between the polymorphism at position -1171 of the MMP-3 gene and susceptibility to CD in 225 Italian patients and 170 controls previously typed for human leukocyte antigen (HLA) class II genes. We also evaluated sex differences in the effect of this polymorphism on disease risk. A male-specific association of the 5A/6A polymorphism with CD was observed. The frequencies of 6A allele and 6A/6A genotype in affected male subjects were significantly above those observed both in male controls (p = 4.1 x 10(-3) and p = 3.4 x 10(-3); odds ratio = 2.4, 95% confidence interval 1.3-4.3) and in female patients (p = 2.7 x 10(-4) and p = 6.2 x 10(-4)). This is the first demonstration of a sex-specific association between the MMP-3 promoter polymorphism and CD. Homozygosity for the 6A allele appears as a risk factor for CD only in men, which is different from the HLA susceptibility alleles that confer a higher risk in women.

CTLA-4 +49 A/G dimorphism in Italian patients with celiac disease.

The chromosome region 2q33, which contains the cytotoxic T lymphocyte antigen-4 (CTLA-4) gene, has been reported in linkage and association with celiac disease (CD). In the present work we have tested the association between the polymorphism of the CTLA-4 exon 1 and susceptibility to CD in an Italian population, using case-control and family-based approaches. The +49 A/G dimorphism was analyzed in 86 patients, 144 ethnically matched controls, and 113 nuclear families by the polymerase chain reaction-restriction fragment length polymorphism method. A significantly higher frequency of the CTLA-4 +49A allele was observed in patients when compared with controls (p = 3 x 10(-2)). The segregation analysis in the 113 trios showed a preferential transmission of the A allele to the probands (chi(2)(TDT) = 4.85). When the patients were stratified according to the presence/absence of the high-risk human leukocyte antigen-DQ2 heterodimer, a significant difference was observed between the two groups, that is, the A allele was increased in the subjects without the DQ2 heterodimer (88.9% vs 73.5%, p = 8.3 x 10(-3)). The A allele was transmitted from heterozygous parents to eight of nine DQ2-dimer-negative patients. These data support CTLA-4 as a predisposing gene for CD in an Italian population with a prominent role in patients not carrying the high-risk human leukocyte antigen-DQ2 molecules.

Serologic and genetic markers of celiac disease: a sequential study in the screening of first degree relatives.

OBJECTIVES: The prevalence of celiac disease (CD) among the relatives and the complications of an undiagnosed CD prompted us to identify a useful disease screening strategy. METHODS: We studied 441 first degree relatives of 208 CD patients by immunoglobulin (Ig)A antiendomysium antibodies (EMA) and radioimmunoprecipitation assay (RIA) IgA antitransglutaminase autoantibodies (TGAA). Of these, 364 were typed for human leukocyte antigen-DRB1, -DQA1, and -DQB1 genes by the polymerase chain reaction sequence specific primers method. It was suggested to the autoantibody-positive subjects that they should undergo intestinal biopsy. RESULTS: TGAA were positive in 46 of 439 relatives, EMA in 38; intestinal lesions related to CD were present in 40 subjects. We also found two immunodeficient fathers with duodenal villous atrophy. In three serology-positive subjects, permission for intestinal biopsy was refused; for another three serology-positive cases, duodenal mucosa was normal. Thus, the strict CD prevalence resulted 9.5%, the enlarged prevalence 10.9%. The DQ2/DQ8 heterodimers were carried in 231 of 364 subjects and in 38 of 40 biopsy-proven celiac patients. Three DQ2-positive parents became positive to the serology during a long-lasting follow-up. CONCLUSIONS: On the basis of a carefully conducted study, CD prevalence in our series was seen as very high. These data suggest an accurate algorithm to select candidates for intestinal biopsy among CD high-risk subjects. First, an evaluation of the sensitive RIA TGAA and of total IgA (in IgA deficiency RIA IgG anti-tissue transglutaminase assay) should be performed. Then, an evaluation of the TGAA and the genetic study would be advisable 2 to 3 years later in negative subjects. Those carrying the DQ2/DQ8 heterodimers should continue the serologic follow-up; the others need a clinical follow-up.

Antiendomysial antibody detection in biopsy culture allows avoidance of gluten challenge in celiac children.

OBJECTIVE: Antiendomysial antibody (EMA) production has been induced in vitro by the small bowel mucosa of celiac disease (CD) patients in clinical remission cultured in the presence of gliadin peptides. The aim of the present study was to use this in vitro system to determine whether it could be used to predict the clinical or histologic relapse to gluten challenge in CD children on a gluten-free diet (GFD). METHODS: Enrolled were 32 CD children and adolescents on GFD (group 1), and 80 controls (group 2) who underwent in vitro gliadin challenge. Subsequently, 24 group 1 CD children underwent in vivo gluten challenge to confirm the diagnosis. Biopsy cultures, with and without gliadin, morphometric analysis, immunoglobulin (Ig)A and IgG1 EMA detection, both in sera and culture supernatants, were performed. RESULTS: Of the 32 group 1 CD patients, 23 were IgA EMA positive in culture supernatants. The other nine were IgG1 EMA positive. All 24 children who had in vivo gluten challenge showed clinical or histologic relapse. All culture supernatants from disease controls belonging to group 2 were both IgA and IgG1 EMA negative, irrespective of gliadin challenge. CONCLUSIONS: Organ culture with in vitro gliadin challenge is able to reproduce the results of in vivo challenge. This system could reduce the need for gluten challenge in celiac children.

Tissue transglutaminase autoantibody detection in human saliva: a powerful method for celiac disease screening.

OBJECTIVE: To test the possibility of detecting tissue transglutaminase autoantibodies (tTG-Abs) in saliva with a novel sensitive fluid-phase radioimmunoassay (RIA). STUDY DESIGN: Paired saliva and serum samples from 39 patients with celiac disease (CD), at the first biopsy (Group 1: 28 females, mean age 11.5 +/- 11.1 years); 32 controls with a normal duodenal mucosa (Group 2: 18 females, mean age 8.1 +/- 3.6 years); and 32 healthy volunteers (Group 3: 21 females, mean age 31.7 +/- 9.8 years) were studied for tTG-Ab presence. Limit of positivity for salivary assay was calculated according to the 99th percentiles of Group 2 control children and was expressed as an autoantibody (Ab) index. RESULTS: Salivary tTG-Abs were found in 97.4% of the patients with CD and in 100% of the corresponding serum samples. All Group 3 subjects were negative with both saliva and serum assays. A correlation between saliva and serum tTG-Ab titers was found (r=0.826, P=.0014). CONCLUSIONS: This study demonstrates that it is possible to detect salivary tTG-Abs in CD with a non-invasive, simple to perform, reproducible and sensitive method.

Patchy villous atrophy of the duodenum in childhood celiac disease.

OBJECTIVES: Patchy villous atrophy of the duodenal mucosa has been described in adults with untreated celiac disease (CD) but not in children. The authors evaluated the presence and the distribution of villous atrophy in children with celiac disease to see whether this histologic pattern exists in children. METHODS: We studied 95 children at diagnosis (Group 1) and seven during gluten challenge (Group 2). We measured anti-endomysium antibodies (EMA) by immunofluorescence on monkey esophagus, antihuman-tissue transglutaminase autoantibodies (anti-tTG Abs) by radioimmunoprecipitation, and HLA-DQ2/DQ8 heterodimers by polymerase chain reaction using specific primers. During upper intestinal endoscopy, at least five duodenal biopsy samples were obtained, one from the duodenal bulb and four from the distal duodenum. RESULTS: Thirteen of 95 (13.7%) patients in Group 1 and in 3 of 7 (42.9%) in Group 2 had patchy villous atrophy of the duodenum. In all 16 patients, villous atrophy of the bulb was present. In four children from Group 1, villous atrophy was observed only in the bulb samples. EMA, anti-tTG Abs, and HLA-DQ2/DQ8 heterodimers were present in all patients. Fourteen of 16 had symptomatic CD, and two were silent, detected during screening in subjects at risk for CD. CONCLUSIONS: This is the first study demonstrating that children with CD may have patchy villous atrophy of the duodenum. The bulb mucosa may be the only duodenal area involved, both at diagnosis and after gluten challenge. Therefore, multiple endoscopic biopsies should always be performed, not only in the distal duodenum, but also in the bulb.

Evaluation of stool antigen test, PCR on ORAL samples and serology for the noninvasive detection of Helicobacter pylori infection in children.

BACKGROUND: Endoscopy represents the gold standard for the diagnosis of Helicobacter pylori infection. We evaluated three noninvasive tests in a group of children: the immunoassay for detection of H. pylori stool antigen, the polimerase chain reaction for identification of bacterial DNA on the oral cavity and the serum specific antibodies. MATERIALS AND METHODS: One hundred and ninety children underwent endoscopy for various gastrointestinal symptoms. H. pylori stool antigen and anti-H. pylori antibodies were assayed by commercial kits. The bacterial DNA on saliva and oral plaque was detected by a seminested PCR. RESULTS: Based on the positivity of culture or urease rapid test and histology, infection was detected in 47 patients. The statistical analysis showed that, for the detection of the infection, stool antigen assay is more effective in sensitivity and negative predictive value (91.5% and 96.5%), whereas specificity and positive predictive values appear slightly better in serology (89.6% and 76.0%). Correlations between serum IgG both with patients' age (r = 0.21, p < .05) and H. pylori stool antigen (r = 0.47, p < .01) were found. The search for bacterial DNA on oral samples proved to be very specific (99.1% on saliva and 98.2% on plaque), but insensitive (22.2% and 25.7%). CONCLUSIONS. In children H. pylori stool antigen represents a sensitive test, suitable for detecting H. pylori infection. Serum IgG proved to be more specific; the PCR on the oral cavity resulted as being a very specific, but insensitive test.