Introduction

Coeliac disease (CD) is an autoimmune disorder characterised by chronic inflammation of the small intestine in response to dietary gluten. Genetic predisposition, including the presence of human leukocyte antigen (HLA) haplotypes, HLA-DQ2 or HLA-DQ8, is widely recognised in CD pathogenesis.^1,2 However, these haplotypes are common and non-specific, and perhaps half of the disease heritability remains unexplained despite GWAS studies. Epigenetics, as an interface between genome and environment, may also partly explain disease susceptibility, as well as identify novel biological processes, therapeutic targets and biomarkers.³

Aims & Methods

Previously, altered DNA methylation profiles have been found in the intestinal tissue and saliva of patients with CD.⁴ Our objective is to compare the DNA methylation profiles of whole blood between children with CD and healthy controls (HC) to identify unique DNA methylation profiles, an approach that has proven to be fruitful in inflammatory bowel diseases (IBD). Most studies have focused on epigenetic alterations in adult CD populations despite it being more commonly diagnosed in children and young adults. Moreover, we aim to determine whether CD-associated age acceleration exists, as observed in adult IBD patients.⁵
Paediatric patients with CD (n=36) and HCs (n=35) well-matched for age and sex were recruited at the John Radcliffe Hospital, Oxford, UK. DNA methylation was assayed at 850,000 CpGs with HumanMethylationEPIC arrays (Illumina). After pre-processing in minfi, epigenome-wide methylation changes were analysed with limma, using age, sex, Marsh scores, and cell counts as covariates, and with Benjamini-Hochberg correction for multiple testing. Epigenetic age was determined using general (Horvath⁶) and paediatric-specific (Wu⁷) clock models. Genotyping was assayed using Global Screening Arrays (Illumina), with data prepared in GenomeStudio and PLINK. Cis-meQTLs within 300kbp were evaluated using linear regression.

Results

Fifty-three CpGs were significant genome-wide (FDR <0.05) between CD and HCs. Among these, 23 CpGs exhibited hypermethylation while 30 were hypomethylated, with top loci including protein tyrosine phosphatase PTPN6 (p=4.79×10^-9); CDC42 (1.28×10^-9), involved in cell migration, structure, and phagocytosis; and collagen gene COL4A1 (9.95×10^-8). The significant probes were found to be associated with ontology terms concerning cytidine metabolism (p=6.68×10^-5), collagen metabolism (1.0×10^-3) and filopodium assembly (8.6×10^-4). No significant CpGs were found in the HLA region in this study. Differentially methylated regions containing clusters of significant CpGs were identified within the genes APOBEC3B (3 CpGs), ASGR1 (2), and CCNG1 (2). The significant CpG with the strongest evidence of genetic regulation in a meQTL was NOP10 (p=4.48×10⁶). There was some evidence of genetic regulation of methylation at other significant CpGs including CDC42 and PTPN6 but these did not survive correction for multiple testing due to low power in this small cohort. No age acceleration was observed in CD patients with either epigenetic clock.

Conclusion

Paediatric patients with coeliac disease showed altered blood DNA methylation compared to controls, identifying novel targets for further investigation, although HLA region methylation was not altered in this cohort in contrast to some reports in adult CD. Age acceleration was not observed in paediatric CD patients, perhaps reflecting a short disease duration and mild systemic inflammation.

References

1. Karell, K. et al. HLA types in celiac disease patients not carrying the DQA1 *05-DQB1 *02 (DQ2) heterodimer: Results from the European genetics cluster on celiac disease. Hum Immunol 64, 469–477 (2003).
2. Sollid, L. M. et al. Evidence for a primary association of celiac disease to a particular HLA-DQ alpha/beta heterodimer. J Exp Med 169, 345–350 (1989).
3. Anderson, R. P. et al. A novel serogenetic approach determines the community prevalence of celiac disease and informs improved diagnostic pathways. BMC Med 11, 1–13 (2013).
4. Hearn, N. L., Chiu, C. L. & Lind, J. M. Comparison of DNA methylation profiles from saliva in Coeliac disease and non-coeliac disease individuals. BMC Med Genomics 13, 1–11 (2020).
5. Kalla, R. et al. Analysis of Systemic Epigenetic Alterations in Inflammatory Bowel Disease: Defining Geographical, Genetic and Immune-Inflammatory influences on the Circulating Methylome. J Crohns Colitis 17, 170–184 (2023).
6. Horvath, S. DNA methylation age of human tissues and cell types. Genome Biol 14, (2013).
7. Wu, X. et al. DNA methylation profile is a quantitative measure of biological ageing in children. Ageing 11, 10031–10051 (2019).

Introduction

Coeliac disease (CeD) is a chronic immune-mediated enteropathy driven by dietary gluten proteins. Previous studies have indicated an increased risk of cardiovascular disease (CVD) in individuals with CeD [1-3]. However, results are inconsistent and specific guidelines on the management of CVD risk and comorbidities for individuals with CeD do not exist. In this study, we aimed to compare the distribution of traditional CVD risk factors in CeD patients and non-CeD controls from a population-based cohort.

Aims & Methods

We performed a CeD screening of 54,541 serum samples by anti-transglutaminase 2 immunoglobulin A and G serology from in total 56,042 adult participants in the fourth wave of the Norwegian Trøndelag Health Study (HUNT4) [4]. All seropositive participants were invited to upper endoscopy with duodenal biopsies to confirm a CeD diagnosis. Seronegative participants with previously diagnosed CeD were identified by linkage to hospital records and the Norwegian Patient Registry. We calculated SCORE2 and SCORE2-OP as predictors for 10-year risk of first-onset CVD [5, 6]. The groups “no CeD”, previously diagnosed CeD (“known CeD”) and CeD diagnosed in HUNT4 (“new CeD”), as well as “CeD” (known + new cases) were examined in this study. To compare group characteristics, we used the Wilcoxon rank sum test. Associations between groups and SCORE2(-OP) were assessed using ordinal logistic regression analysis.

Results

468 (0.80%) individuals received a CeD diagnosis in HUNT4, while 309 (0.57%) seronegative individuals had known CeD. Individuals with newly diagnosed CeD were younger, better educated, and had lower blood lipids compared to controls. Individuals with CeD smoked significantly less, had lower BMI and waist-hip-ratio, glycated hemoglobin levels, serum creatinine, and hsCRP levels compared to controls without CeD (n=52,438; p<0.05 for all variables mentioned, see table). We found a significantly lower SCORE2 for CeD cases (n=589) compared to controls (n=39,747; p<0.05). Adjustment for alcohol consumption did not affect group characteristics. A significantly lower SCORE2 was observed in CeD cases also after adjustment for level of education (OR .84, 95% CI .72-.98), high sensitivity C-reactive protein (hsCRP) and serum creatinine (OR .85, 95% CI .78-.99), but not for the combination of all three variables (OR .90, 95% CI .76-1.05).

Conclusion

Individuals with CeD appear to have a lower cardiovascular risk based on traditional risk factors like smoking habits, body fat, diabetes, renal function and inflammation compared to control individuals without CeD. In line with that, SCORE2 and SCORE2-OP as predictors for 10-year risk of first-onset CVD were lower in individuals with CeD.

References

1. Bathrellou, E., M.D. Kontogianni, and D.B. Panagiotakos, Celiac disease and non-celiac gluten or wheat sensitivity and health in later life: A review. Maturitas, 2018. 112: p. 29-33.
2. Ludvigsson, J.F., et al., Nationwide cohort study of risk of ischemic heart disease in patients with celiac disease. Circulation, 2011. 123(5): p. 483-90.
3. Figtree, G.A., et al., Mortality in STEMI patients without standard modifiable risk factors: a sex-disaggregated analysis of SWEDEHEART registry data. Lancet, 2021. 397(10279): p. 1085-1094.
4. Asvold, B.O., et al., Cohort Profile Update: The HUNT Study, Norway. Int J Epidemiol, 2023. 52(1): p. e80-e91.
5. SCORE2 working group and ESC Cardiovascular risk collaboration, SCORE2 risk prediction algorithms: new models to estimate 10-year risk of cardiovascular disease in Europe. Eur Heart J, 2021. 42(25): p. 2439-2454.
6. SCORE2-OP working group and ESC Cardiovascular risk collaboration, SCORE2-OP risk prediction algorithms: estimating incident cardiovascular event risk in older persons in four geographical risk regions. European Heart Journal, 2021. 42(25): p. 2455-2467.

Disclosure

None.