A DNA bank of 92 unrelated coeliac families with 385 individuals has been collected, 100 of whom were biopsy-confirmed as having CD, and for the rest the serology (anti-tTG antibodies) results were available (data not shown). Seventy-six percent of the coeliac patients were females and 24% were males; they were classified into groups according to their age at diagnosis. The 100 biopsy-confirmed Maltese CD patients were recruited from the Maltese Coeliac Association and referred to the Molecular Genetics Clinic, Medical School, St Luke Hos­pital, G’Mangia, Malta. Informed consent was obtained from all participants. The study was approved by the Research Ethics Committee of the University of Malta. All participants answered a questionnaire about their medical conditions, use of medications, family history of CD and adherence to a gluten-free diet. Peripheral blood was obtained by venipuncture for all coeliac patients except two children from whom mouthwash samples were taken, for DNA analysis. Anonymous neonatal cord blood sam­ples (n = 187) were obtained from the University of Malta DNA Bank which is a member of Eurobiobank. The sam­ple size was calculated to give a confidence interval (CI) of 95% for the Maltese population (390,000).

      DNA was extracted and purified from peripheral blood leukocytes in all patients except two young children from whom we extracted DNA from cheek cells. Both tech­niques were based on a modification of the salting-out technique [3]. The genotypes of the –318 (C→T) and +49 (A→G) polymorphisms were analyzed utilizing restriction enzyme digestion of the respective polymerase chain reac­tion (PCR) product. The PCRs were performed utilizing 75 pM of each primer for the –318 (C→T) SNP (forward: 5'-TGA ATT GGA CTG GAT GGT T-3' and reverse: 5'-TTA CGA GAA AGG AAG CCG TG-3') and 50 pM of each primer for the +49 (A→G) SNP (forward: 5'-GCT CTA CTT CCT GAA GAC CT-3' and reverse: 5'-AAC CCA GGT AGG AGA AAC AC-3'), in a final 25 μL reaction mixture containing 1.5 mM MgCl₂, 50 mM KCl, 10 mM Tris-HCl (pH 9.0), 0.1% Triton X-100 and 200 μM of each dNTP (Qiagen Operon GmbH, Cologne, Germany). An initial “hot-start” cycle at a temperature of 95° C for 5 min. was performed, followed by addition of 1 U Taq DNA polymerase (Promega GmbH, Mannheim, Germany). The PCR conditions were 30 seconds at 95° C, 30 seconds at 56° C and 1 min. at 74° C for 30 cycles followed by a final extension at 74° C for 3 min., for both PCRs. The PCR products were digested overnight with the restriction enzymes, MseI for the –318 (C→T) and Fnu4HI for the +49 (A→G) (New England BioLabs Inc., Beverly, MA, USA) SNPs, and then analyzed on a 3% agarose gel and an 8% polyacrylamide gel electrophoresis, respectively. For the +49 (A→G) SNP, Fnu4HI cleaves the G allele at posi­tion +49, while MseI cleaves the T allele at position –318 (see Figure 1). Re-analysis of 20% of the samples was performed for both –318 (C→T) and +49 (A→G) to check for accuracy of genotyping. No discrepancies were found with the original genotyping.

      Statistical Analyses. Statistical analyses of the CTLA4 +49 (A→G) and –318 (C→T) genotypes were carried out using the SPSS for Windows Student Version 9.0 (SPSS, Chicago, IL, USA). The chi-square test was used to check for any significant difference in the distribu­tion of genotype frequencies and Fisher’s exact test for any significant difference in the distribution of alleles, for both SNPs, between the neonatal controls and the coeliac patients in the study. The Hardy-Weinberg equilibrium was calculated and a chi-square test was performed between the observed and expected genotype frequencies. Haplotype frequencies were constructed and tested for association (linkage disequilibrium) between alleles using the estimated haplotypes program and a chi-square test was performed to compare the haplotype frequencies between patients and controls. All statistical tests were considered two-tailed at a level of significance of 0.05.