BACKGROUND. Morphological patterns of Paneth cells are a prognostic biomarker in Western Crohn’s disease (CD) patients, and are associated with autophagy-associated ATG16L1 and NOD2 variants. We hypothesized that genetic determinants of Paneth cell phenotype in other ethnic CD cohorts are distinct but also involved in autophagy.
METHODS. We performed a hypothesis-driven analysis of 56 single nucleotide polymorphisms (SNPs) associated with CD susceptibility or known to affect Paneth cell function in 110 Japanese CD patients who underwent ileal resection. We subsequently performed a genome-wide association analysis. Paneth cell phenotype was determined by defensin-5 immunofluorescence. Selected genotype–Paneth cell defect correlations were compared to a Western CD cohort (n = 164).
RESULTS. The average percentage of abnormal Paneth cells in Japanese CD was similar to Western CD (P = 0.87), and abnormal Paneth cell phenotype was also associated with early recurrence (P = 0.013). In contrast to Western CD, ATG16L1 T300A was not associated with Paneth cell defect in Japanese CD (P = 0.20). Among the 56 selected SNPs, only LRRK2 M2397T showed significant association with Paneth cell defect (P = 3.62 × 10–4), whereas in the Western CD cohort it was not (P = 0.76). Pathway analysis of LRRK2 and other candidate genes with P less than 5 × 10–4 showed connections with known CD susceptibility genes and links to autophagy and TNF-α networks.
CONCLUSIONS. We found dichotomous effects of ATG16L1 and LRRK2 on Paneth cell defect between Japanese and Western CD. Genes affecting Paneth cell phenotype in Japanese CD were also associated with autophagy. Paneth cell phenotype also predicted prognosis in Japanese CD.
FUNDING. Helmsley Charitable Trust, Doris Duke Foundation (grant 2014103), Japan Society for the Promotion of Science (KAKENHI grants JP15H04805 and JP15K15284), Crohn’s and Colitis Foundation grant 274415, NIH (grants 1R56DK095820, K01DK109081, and UL1 TR000448).
(A) The numbers of the ATG16L1 T300A risk allele did not correlate with the percentage of normal Paneth cells (R2 = 0.01717 and P = 0.20 by linear regression) in Japanese CD, while (B) in North American CD without the common CD-associated NOD2 risk allele, the numbers of the ATG16L1 T300A risk allele correlated with the percentage of normal Paneth cells (R2 = 0.04387 and P = 0.0395 by linear regression). In contrast, (C) the numbers of the LRRK2 M2397T risk allele negatively correlated with the percentage of normal Paneth cells in Japanese CD (R2 = 0.247, P = 3.62 × 10–4 by linear regression). (D) However, in North American CD, the numbers of the LRRK2 M2397T risk allele did not correlate with the percentage of normal Paneth cells (R2 = 0.02054, P = 0.76 by linear regression). Error bars represent ± SEM.