CARD15 variants in both sporadic and familial cases of Crohn's disease across Italy. An Italian Group for Inflammatory Bowel Disease study

CARD15 variants in both sporadic and familial cases of Crohn's disease across Italy. An Italian Group for Inflammatory Bowel Disease study

Digestive and Liver Disease 36 (2004) 121–124 Alimentary Tract Frequency of NOD2 /CARD15 variants in both sporadic and familial cases of Crohn’s dis...

56KB Sizes 0 Downloads 28 Views

Digestive and Liver Disease 36 (2004) 121–124

Alimentary Tract

Frequency of NOD2 /CARD15 variants in both sporadic and familial cases of Crohn’s disease across Italy An Italian Group for Inflammatory Bowel Disease study V. Annese a,∗ , O. Palmieri a,b , A. Latiano a , S. Ardizzone c , F. Castiglione d , M. Cottone e , R. D’Incà f , P. Gionchetti g , C. Papi h , G. Riegler i , M. Vecchi j , A. Andriulli a a

g

Department of General and Specialist Medicine, Unit of Gastroenterology, CSS-IRCCS Hospital, Viale Cappuccini, 1, 71013 San Giovanni Rotondo, Italy b Div. Genetics, Thomas Jefferson University, Philadelphia, PA, USA c Unit of Gastroenterology, “L. Sacco” University Hospital Milan, Milan, Italy d Unit of Gastroenterology, 1 University of Naples, Naples, Italy e Institute of Internal Medicine & Gastroenterology, “Cervello” Hospital, Palermo, Italy f Unit of Gastroenterology, University of Padua, Padua, Italy Institute of Internal Medicine & Gastroenterolgy, “S. Orsola” University Hospital, Bologna, Italy h Unit of Gastroenterology, “S. Filippo Neri” Hospital, Rome, Italy i Unit of Gastroenterology, II University of Naples, Naples, Italy j Unit of Gastroenterology, “Maggiore” Hospital, IRCCS, Milan, Italy Received 30 July 2003; accepted 22 October 2003

Abstract Background. Three variants of the NOD2 /CARD15 gene are strongly associated with susceptibility to Crohn’s disease; however, striking racial and geographic differences of their frequency have been described. Aims. We have compared the allele frequencies of familial cases of Crohn’s disease recruited in a multicentre study across Italy, in order to disclose possible geographic heterogeneity. Moreover, we also compared the allele frequencies in sporadic cases of Crohn’s disease and healthy controls from Southern Italy with those reported in other two populations from Central and Northern Italy. Subjects and Methods. A total of 731 subjects were genotyped for the polymorphism of three main variants (R702W, G908R and 1007 fs): 152 patients were familial cases of Crohn’s disease, 183 were healthy first-degree relatives, 180 were sporadic cases of Crohn’s disease, and 216 were unrelated healthy subjects. Results. The frequency of the frameshift mutation (1007 fs) was significantly higher in both familial and sporadic cases of Crohn’s disease (P = 0.000001), and healthy first-degree relatives (P = 0.0001) compared to controls. At least one risk allele was found in 44% of familial Crohn’s disease patients, compared to 7% of healthy controls (OR = 4; CI = 2–6.5). Two risk alleles were found in 14% of familial Crohn’s disease, compared to less than 1% of controls (OR = 26: CI = 4–129). Conclusions. Our data confirm the strong correlation between the 1007 fs variant and Crohn’s disease, in both familial and sporadic cases. Moreover, no significant difference of allele frequencies was detected in familial cases, sporadic cases and healthy controls among different geographic areas of Italy. © 2003 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved. Keywords: CARD15 ; Crohn’s disease; Genetics; Inflammatory bowel disease; NOD2

1. Introduction



Corresponding author. Tel.: +39-0882-410235-335; fax: +39-0882-411879/411705. E-mail address: [email protected] (V. Annese).

Three major polymorphisms of the NOD2 /CARD15 gene, a frameshift mutation (1007 fs) encoding for a truncated protein and two missense mutations (R702W and G908R), have been found associated with susceptibility to Crohn’s disease

1590-8658/$30 © 2003 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.dld.2003.10.010

122

V. Annese et al. / Digestive and Liver Disease 36 (2004) 121–124

(CD) [1,2]. Heterozygous carriage of any of the three major risk alleles increases the susceptibility for CD two- to fourfold, whereas homozygous or compound heterozygous carries a 20- to 40-fold increase in the risk. Approximately 8–17% of CD patients carry two copies of the risk alleles (compared to less than 1% in control populations), whereas 27–32% of CD patients exhibit one risk allele (compared to less than 20% of controls). Patients with sporadic or familial CD seem to have a comparable allele frequency [3]. Higher (up to 50–60%) frequency of CARD15 variants has been found in patients with ileal location, earlier disease onset and stricturing phenotype (reviewed in [1,2]). The NOD2 /CARD15 gene, mainly expressed in peripheral blood monocytes, is believed to act as an intracellular sensor, mediating the resistance to microbial pathogens. All three major variants of the CARD15 gene determine a reduced NF-␬B activation in response to bacterial components and, more specifically, an impaired recognition of muramyldipeptide (MD), the minimal bioactive peptidoglycan motif, shared by both Gram-positive and -negative bacteria [4–8]. Previous studies, however, mainly apply to a subgroup of patients within the CD populations, those with a positive family history. As only a minority of CD patients have a positive family history, the question arises whether there are differences in the frequency of NOD2 /CARD15 variants between familial and sporadic cases of CD. Furthermore, the NOD2 /CARD15 variants identified in Caucasian population are absent or very rare in other populations, such as Japanase, Korean and African-American CD patients [9–12]. No data are yet available from Italy, where a genetic heterogeneity might exist due to the past migration of several populations from the North of Europe and North of Africa. To explore these issues, we carried out an association study aimed at comparing allele frequencies of the three major NOD2 /CARD15 variants across Italy in both familial and sporadic patients with CD.

2. Materials and methods In a multicentre study, 99 IBD families (53 CD and 46 mixed families) were consecutively recruited from 15 University or General Hospitals across Italy. A total of 152 CD patients (82 males, mean age 34 ± 15 yrs) and 183 healthy first-degree relatives (90 males, mean age 48 ± 22 yrs) were investigated. Only families with at least two affected and two non-affected members available for DNA genotyping were included in the study. A firm diagnosis of CD and UC was established at the same hospital according to accepted clinical, endoscopic, radiological and histological criteria [13,14].We also studied two other different series of cases, both of them recruited at a single institution (S. Giovanni Rotondo “C.S.S. Hospital, located in Southern Italy). The first series included 180 sporadic cases of CD (96 males, mean

age 33 ± 15 yrs), while the second one consisted of 216 unrelated healthy subjects (115 males, mean age 32 ± 12 yrs), who served as controls. CD cases were defined as sporadic when no other CD or ulcerative colitis subjects were reported within the families, among first- and second-degree relatives. Polymorphisms of SNPs 8 (Arg702Trp), 12 (Gly908Arg), and 13 (Leu1007fsinsC) of the CARD15 gene were investigated; primers for PCR amplification were designed on the basis of the Genebank sequence. After amplification of genomic DNA, sequence products were analysed on the ABI PRISM 377. With the intent to explore possible geographic variation in the allele distribution among controls, we compared data from our controls from Southern Italy to those from controls reported in recently published studies from both Central and Northern Italy [15,16]. Similarly, we compared the allele frequencies in CD patients collected from Southern, Central and Northern Italy. Finally, frequencies of the whole series of familial CD patients were compared to both sporadic CD and controls from Southern Italy. For all comparisons, the χ2 - or Fisher tests were used, as appropriate. After applying the Bonferroni’s correction for multiple comparisons, significant values were taken as those showing a P-value <0.005.

3. Results 3.1. Allele frequencies in familial cases The frequencies of the three major CARD15 variants are shown in Table 1. The frameshift mutation (Leu1007fsinsC) was significantly more frequent in both CD patients (P = 0.000001) and their first-degree healthy relatives (P = 0.0001), compared to the healthy controls. An increased frequency of the other two variants was also found, but after the Bonferroni’s correction, only the Arg702Trp variant was significantly more frequent in the whole group of CD patients in comparison to controls (P = 0.002). At least one risk allele was found in 44% of familial CD patients, compared to 7% of healthy controls (P ≤ 0.000001). Two risk alleles were found in 14% of familial CD, compared to less than 1% of controls (P ≤ 0.000001). Homozygous variants were rather infrequent: six out of 152 familial CD patients (4%), 1 out of 183 healthy relatives (0.5%), 1 out of 180 sporadic CD patients (0.5%), and 1 out of 216 healthy controls (0.4%). 3.2. Geographic differences No geographic difference in the alleles distribution in CD patients was identified, with the exception of an increased incidence of at least one risk allele in CD families from Central Italy, which approached the statistical significance (P = 0.0051). The allele frequency among healthy relatives of different geographic areas was not different (data not shown). To better disclose this issue, however, a direct comparison of

V. Annese et al. / Digestive and Liver Disease 36 (2004) 121–124

123

Table 1 Allele frequencies of different CARD15 variants in familial cases of CD, their unaffected first-degree relatives and controls SNP8, Arg702Trp

SNP12, Gly908Arg

SNP13, Leu1007finsC

1 Allele

2 Alleles

TOTAL (99 fam., 152 pts)

0.10∗

0.07

0.10∗∗

0.22∗∗∗

0.07∗∗

North (n = 35 fam., 56 pts) Center (n = 27 fam., 40 pts) South (n = 37 fam., 56 pts)

0.09 0.11 0.09

0.05 0.09 0.06

0.08 0.14 0.09∗∗

0.20∗∗ 0.39∗∗ 0.16

0.06 0.09∗∗∗ 0.07∗∗∗

Healthy relatives (n = 183) Healthy controls (n = 216)

0.11 0.04

0.08 0.02

0.15∗∗∗ 0.007

0.30∗∗ 0.067

0.04 0.002



P = 0.002 vs. controls. P = 0.00001 vs. controls. ∗∗∗ P = 0.000001 vs. controls. ∗∗

allele frequencies with controls from the same geographic area would have been more appropriate. Nevertheless, no difference of the CARD15 variants was found between our controls from Southern Italy and two other control populations from Central and North of Italy, recently reported at the National Meeting of Gastroenterology in Florence, Italy in 2003 (Table 2) [15,16]. 3.2.1. Allele frequencies in sporadic cases Table 2 summarises the data obtained in healthy controls and sporadic cases of CD in three different geographic areas in Italy. Similar to familial CD cases, in sporadic cases an increased frequency of the frameshift mutation (Leu1007finsC) was found. In our population, the frequency of the Gly908Arg was also significantly increased (P = 0.003). Single and double copy of risk alleles of the CARD15 gene were significantly more frequent in either sporadic (OR = 3.3, CI 1.9–5.8; OR = 22, CI 2.9–109, respectively) or familial cases of CD (OR = 4, CI 2–6.5; OR = 26, CI 4–129, respectively) compared to controls. No significant difference of allele frequencies was detected between sporadic cases of CD among different geographic areas, and between familial and sporadic cases of CD.

4. Discussion Our data confirm the strong correlation between the 1007 fs variant and CD, previously reported in Caucasian population from USA and Europe (reviewed in ref. [1,2]). In contrast, the frequency of the other two variants was also increased, similar to other studies [1,17]; however, this value did not reach the statistical significance after Bonferroni’s correction. A lower frequency of CARD15 variants has been recently described in Finnish population [18], but this discrepancy probably reflects the characteristics of the genetic isolate of that population. The frequency of 1007 fs mutations was similar among familial cases and sporadic cases of CD. Moreover, it was significantly higher than that in controls among unaffected first-degree relatives of familial cases, although more often only one copy of the risk allele was found. This finding, not specifically investigated in other studies, is expected from a genetic point of view and further supports the hypothesis that multiple genes contribute to CD susceptibility. Moreover, it underlines that the CARD15 variants are neither necessary nor sufficient for disease expression. Furthermore, a gene–gene and a gene–environmental (i.e. smoking) interactions are probably involved in the genetic predisposition.

Table 2 Distribution of allele frequencies of CARD15 variants among sporadic cases of CD and healthy controls across Italy SNP8, Arg702Trp

SNP12, Gly908Arg

SNP13, Leu1007finsC

1 Allele

2 Alleles

References

North (n = 171 CD pts) Healthy controls (n = 177)

0.10 0.06

0.04 0.02

0.07∗

0.17 –

0.03 –

[15]

0.02

Center (n = 165 CD pts) Healthy controls (n = 125)

– –

– –

0.09∗∗ 0.012

– –

0.03∗∗ 0

[16]

South (n = 180 CD pts) Healthy controls (n = 216)

0.08 0.04

0.07∗ 0.02

0.07∗∗∗ 0.007

0.19∗∗∗ 0.067

0.05∗∗ 0.002

Present series

P North vs. Center vs. South P North vs. Center vs. South

0.61 0.98

0.09 0.98

0.53 0.71

0.72 –

0.46 0.94

CD patients Controls

–: data not available. ∗ P ≤ 0.005 vs. respective controls. ∗∗ P = 0.001 vs. respective controls. ∗∗∗ P = 0.00001 vs. respective controls.

124

V. Annese et al. / Digestive and Liver Disease 36 (2004) 121–124

Because of the low risk carried by the presence of one risk allele in asymptomatic subjects (three- to fourfold), no further investigations should be recommended at this stage. The frequency of patients carrying two risk alleles (whether homozygous or compound heterozygous) was significantly higher in comparison to controls, with similar figures in both familial and sporadic cases. This finding, if further confirmed, should exclude that CARD15 variants are involved into the predisposition to familial clustering of CD. No clear geographic difference in allele frequencies of the CARD15 variants was found in CD families across Italy. While comparisons with controls of the same geographic areas would have been more appropriate for this purpose, we did not find any difference of allele frequencies between our controls (from Southern Italy), and that reported in other two series of controls from Center and Northern Italy. Similarly, the frequency of CARD15 variants in our CD sporadic cases did not differ significantly from that of sporadic cases from Center and Northern Italy. Conflict of interest statement None declared.

Acknowledgements The authors sincerely thank Prof. Paolo Fortina from the Division of Genetics, Thomas Jefferson University, Philadelphia, USA for the remarkable scientific support to the study. The Authors thank Miss De Santo Ermelinda, Mr Corritore Giuseppe, and Miss Latiano Tiziana for their skilful technical support. The authors thank all the other contributing centres of the IG-IBD (Italian Group for Inflammatory Bowel Disease): G. Iaquinto, (U.O. di Gastroenterologia, Avellino), M. Campieri, (Clinica Medica, Policlinico S. Orsola, Bologna), D. Valpiani, E. Ricci (Divisione di Medicina, Forli’), G. Frieri, M.T. Pimpo (Cattedra di Gastroenterologia, Universita’ L’Aquila), S. Giaccari, L. Grasso (Divisione di Gastroenterologia, Galatina), M Saibene, R. de Franchis (Gastroenterologia, Ospedale Maggiore IRCCS, Milano), G. Bianchi Porro (Gastroenterologia, Ospedale Sacco, Milano), F. Morace (Cattedra di Gastroenterologia, IIa Universita’ di Napoli), S. Cucchiara, O. Borrelli (Dipartimento di Pediatria, IIa Universita’ di Napoli), G. Mazzacca (Cattedra di Gastroenterologia, II Universita’ Napoli), C. Prantera, A. Andreoli (Divisione di Gastroenterologia, Ospedale N. Regina Margherita, Roma), L. Capurso (Gastroenterologia, S. Filippo Neri, Roma), GC. Sturniolo (Cattedra di Gastroenterologia, Universita’ di Padova), M. Oliva (Gastroenterologia, Ospedale “Cervello”, Palermo), G. Lombardi, S. Fiorella, G. Napoletano (Gastroenterologia, Ospedale CSS-IRCSS, San Giovanni Rotondo), M. Rizzetto, M. Astegiano, F. Bresso (Dipartimento di Gastroenterologia, Ospedale Molinette, Torino), A. Pera, M.T. Fiorentini (Divisione di Gastroenterologia, Ospedale

Mauriziano, Torino). This work was supported by a grant of the National Health Minister No GastroGARC15.

References [1] Annese V, Latiano A, Andriulli A. Genetics of inflammatory bowel disease. The beginning of the end or the end of the beginning? Dig Liver Dis 2003;35:442–9. [2] Bonen DK, Cho JH. The genetics of inflammatory bowel disease. Gastroenterology 2003;124:521–36. [3] Vermeire S, Wild G, Kocher K, Cousineau J, Duffresne L, Bitton A, et al. CARD15 genetic variation in a Quebec population: prevalence, genotype-phenotype relationship, and haplotype structure. Am J Hum Genet 2002;71:74–83. [4] Ogura Y, Bonen DK, Inohara N, Nicolae DL, Chen FF, Ramos R, et al. A frameshift mutation in NOD2 associated with susceptibility to Crohn’s disease. Nature 2001;41:603–6. [5] Bonen DK, Ogura Y, Nicolae DL, Inohara N, Saab L, Tanabe T, et al. Crohn’s disease-associated NOD2 variants share a signalling defect in response to lipopolysaccharide and peptidoglycan. Gastroenterology 2003;124:140–6. [6] Girardin SE, Boneca IG, Viala J, Chamaillard M, Labigne A, Thomas G, et al. NOD2 is a general sensor of peptidoglycan through muramyl dipeptide (MDP) detection. J Biol Chem 2003;278:8869– 72. [7] Inohara N, Ogura Y, Fontalba A, Gutierrez O, Pons F, Crespo J, et al. Host recognition of bacterial muramyl dipetide mediated through NOD2: implication for Crohn’s disease. J Biol Chem 2003;278:5509– 12. [8] Chamaillard M, Dana P, Girardin SE, Zouali H, Lesage S, Chareyre F, et al. Gene/environment interaction modulated by allelic heterogeneity in inflammatory bowel diseases. Proc Natl Acad Sci USA 2003;100:3455–60. [9] Inoue N, Tamura K, Kinouchi Y, Fukuda Y, Takahashi S, Ogura Y, et al. Lack of common NOD2 variants in Japanase patients with Crohn’s disease. Gastroenterology 2002;123:86–91. [10] Yamazaki K, Takazoe M, Tanaka T, Kazumori T, Nakamura Y. Absence of mutation in NOD2/CARD15 gene among 483 Japanase patients with Crohn’s disease. J Hum Genet 2002;47:469–72. [11] Crouncher PJP, Mascheretti S, Hampe J, Hise K, Frenzel H, Stoil M, et al. Haplotype structure and associations to Crohn’s disease of CARd15 mutations in two ethnically divergent populations. Eur J Hum Genet 2003;11:6–16. [12] Bonen DK, Nicolae DL, Moran T, Turkylmaz MA, Ramos R, Karaliukas R, et al. Racial differences in NOD2 variation: characterization of NOD2 in Africans-Americans with Crohn’s disease. Gastroenterology 2002;122:A29. [13] Lennard-Jones JE. Classification of inflammatory bowel disease. Scand J Gastroenterol 1989;24:2–6. [14] Fiocchi C. Inflammatory bowel disease: dogmas and heresies. Dig Liver Dis 2002;34:306–11. [15] Giachino D, Van Duist M, Salacone P, Bardessono M, Sapone N, Scaglione N, et al. CARD15/NOD2 mutations in Italian IBD patients: preliminary R720W, G908R, and 3020insC. Dig Liver Dis 2003;35:CO64. [16] Vavassori P, Borgian P, Biancone L, D’Apice MR, Del Vecchio Blanco G, De Nigris F, et al. CARD15 gene mutations in Italian patients with Crohn’s disease. Dig Liver Dis 2003;35:CO21. [17] Zheng CQ, Hu GZ, Zeng ZS, Lin LJ, Gu GG. Progress in searching for susceptibility gene for inflammatory bowel disease by positional cloning. World J Gastro 2003;9:1646–56. [18] Heliö T, Halme L, Lappalainen M, Fodstad H, Paavola-Sakki P, Turunen U, et al. Card15/NOD2 gene variants are associated with familially occurring and complicated forms of Crohn’s disease. Gut 2003;52:558–62.