Rare Hypomorphic Sucrase Isomaltase Variants in Relation to Irritable Bowel Syndrome Risk in UK Biobank

Abstract

Sucrase-isomaltase (encoded by the SI gene) is the major intestinal brush border enzyme primarily responsible for the digestion of starch and sucrose.1Dahlquist A. Semenza G. J Pediatr Gastroenterol Nutr. 1985; 4: 857-865Google Scholar As seen in patients with congenital SI deficiency (a rare form of sucrose intolerance due to recessive mutations in the SI gene), inactive SI leads to colonic accumulation of unabsorbed carbohydrates, causing osmotic diarrhea, abdominal pain, and bloating.2Treem W.R. J Pediatr Gastroenterol Nutr. 1995; 21: 1-14Google Scholar,3Gericke B. Amiri M. et al.Mol Cell Pediatr. 2016; 3: 2Google Scholar These symptoms are also typical of irritable bowel syndrome (IBS), and we recently reported (1) increased prevalence of rare (and the common Val15Phe) hypomorphic (defective) SI variants in patients with IBS, and (2) that IBS carriers of such variants are less likely to benefit from a low–fermentable oligosaccharides, disaccharides, monosaccharides, and polyols (FODMAPs) diet, which does not restrict sucrose and starch.4Henström M. et al.Gut. 2018; 67: 263-270Google Scholar, 5Thingholm L. et al.Gut. 2019; 68: 177-178Google Scholar, 6Garcia-Etxebarria K. Zheng T. et al.Clin Gastroenterol Hepatol. 2018; 16: 1673-1676Abstract Full Text Full Text PDF Scopus (35) Google Scholar, 7Zheng T. et al.Gut. 2020; 69: 397-398Google Scholar Preliminary evidence for the interaction between SI genotype, gut microbiota, and dietary carbohydrate intake in relation to IBS risk was also reported.5Thingholm L. et al.Gut. 2019; 68: 177-178Google Scholar Although this suggests carbohydrate malabsorption may contribute to IBS symptoms in some patients, these studies only tested known SI variants in up to 2200 IBS cases from tertiary centers (compared with publicly available data from ethnically matched populations), and the relevance of rare SI hypomorphic variants (hereafter referred to as rare pathogenic variants [RPVs]) toward IBS risk has not been investigated in large-scale studies at the general population level. UK Biobank is a large population-based cohort, rich in genotype, lifestyle, and healthcare data (Supplementary Material). Whole exome sequencing (WES) data recently became available, which allowed the prevalence of SI RPVs to be studied in relation to IBS. Although, for the whole of UK Biobank, IBS can be defined as self-reported condition at enrollment (touchscreen questionnaire data) or via linkage to hospital inpatient electronic medical records (EMRs), we focused our analyses on participants who filled a Digestive Health Questionnaire (DHQ), which includes questions apt to build an IBS classification according to the gold-standard Rome III Criteria. Therefore, SI RPVs were studied in a common set of asymptomatic controls and 4 alternative IBS definitions, namely, unprompted self-reported IBS (self-IBS; from touchscreen questionnaire data at enrollment), primary International Classification of Diseases 10th revision (ICD10) K58 diagnosis of IBS (ICD10-IBS; from EMRs), prompted self-reported IBS (DHQself-IBS; from DHQ), and Rome III Criteria IBS (DHQRome-IBS; also from DHQ) (Supplementary Methods, Supplementary Figure 1). UK Biobank phenotypic and genetic (WES) data are available via UK Biobank data access request as described (https://www.ukbiobank.ac.uk/enable-your-research/apply-for-access). The study was approved by the Stockholm Ethics Review Board (protocol number 2016/1620-31/2). Rare (minor allele frequency <0.01) SI variant genotypes were extracted from UK Biobank high-quality WES data, and the subset of RPVs identified was based on computational predictions and combined Mendelian Clinically Applicable Pathogenicity and Combined Annotation Dependent Depletion score annotations, an approach successfully adopted in previous studies and later corroborated at the experimental level.6Garcia-Etxebarria K. Zheng T. et al.Clin Gastroenterol Hepatol. 2018; 16: 1673-1676Abstract Full Text Full Text PDF Scopus (35) Google Scholar,8Husein D.M. et al.Nutrients. 2020; 13: 9Google Scholar SI RVPs were studied in cumulative collapsed analyses, comparing the aggregate number of carriers in cases versus controls to test the hypothesis they increase the risk of IBS (1-tailed test). This was done by adopting the Scalable and Accurate Implementation of Generalized (SAIGE) mixed model to control for population structure, relatedness, and case:control ratio imbalance (also including age, sex, and genotype-derived principal components as covariates). By these means, we studied 71,557 UK Biobank participants with WES, EMR, touchscreen, and DHQ questionnaire data available, which allowed the identification of 31,218 asymptomatic controls and a total of 17,643 cases, based on 4 alternative definitions of IBS including self-IBS (N = 2327), ICD10-IBS (N = 248), DHQself-IBS (N = 8901), and DHQRome-IBS (N = 13,212) (Supplementary Figure 1). Overlap across these definitions ranged from 0.9% (for DHQRome-IBS also classified according to ICD10-IBS) to 85.4% (for self-IBS also classified according to DHQself-IBS), with ICD10-IBS being by far the definition with the highest specificity (15.7% of ICD10 patients satisfying all IBS definitions, compared with only 0.3%–1.7% in other groups). Screening of WES data from these individuals led to the identification of 397 SI RPVs (Supplementary Table 1), which were tested for increasing risk of IBS as previously observed in tertiary case studies.4Henström M. et al.Gut. 2018; 67: 263-270Google Scholar,6Garcia-Etxebarria K. Zheng T. et al.Clin Gastroenterol Hepatol. 2018; 16: 1673-1676Abstract Full Text Full Text PDF Scopus (35) Google Scholar,7Zheng T. et al.Gut. 2020; 69: 397-398Google Scholar As shown in Table 1 where association results are reported, the cumulative prevalence of SI RPVs was associated with a statistically significant, markedly increased risk of IBS in the analysis of individuals with ICD10-IBS diagnoses (P = .00087; odds ratio [OR], 1.73; 95% confidence interval [CI], 1.10–2.72). Identical results were obtained when only unrelated individuals were included in the analysis (Ncase = 248, Ncontrol = ,30,690), using a logistic regression model adjusting for age, sex and top principal components (P = .10; OR, 1.57; 95% CI, 1.07–2.29 for ICD10-IBS). Carrying a SI RPV was associated with increased risk of ICD10-IBS also when looking at the whole of UK Biobank, which is including all participants with EMR and WES data available (sum of DHQ respondents and nonrespondents; 753 RPVs; Ncase = 651; Ncontrol = 180,355; P = .019; OR, 1.35; 95% CI, 1.02–1.78; Supplementary Methods). No other significant associations were detected, including when testing RPVs for increased prevalence in IBS split into constipation-predominant, diarrhea-predominant, or mixed subtypes according to Rome III Criteria and data from the DHQ questionnaire (not shown). Finally, carrying the common Val15Phe hypomorphic risk variant 4Henström M. et al.Gut. 2018; 67: 263-270Google Scholar,5Thingholm L. et al.Gut. 2019; 68: 177-178Google Scholar was also associated with increased risk of ICD10-IBS when tested on the whole of UK Biobank (54.8% carriers in cases vs 50.2% in controls; P = .0090; OR, 1.21; 95% CI, 1.03–1.41).Table 1Association of SI RPVs With IBS in UK BiobankIBS definitionNRPVsRPV carriers (%)PaVs asymptomatic controls.OR (95% CI)aVs asymptomatic controls.Asymptomatic controls31,2182972509 (8.0%)ICD10-IBS2482531 (12.5%).00871.73 (1.10–2.72)Self-IBS232774198 (8.5%).291.04 (0.89–1.22)DHQself-IBS8901165726 (8.2%).461.01 (0.92–1.10)DHQRome-IBS13,2122011053 (8.0%).750.97 (0.90–1.05)NOTE. Boldface indicates statistical significance.a Vs asymptomatic controls. Open table in a new tab NOTE. Boldface indicates statistical significance. The association between SI RPVs and increased risk of IBS in a large population-based cohort from the UK confirms and extends previous findings from case-control studies of patients with IBS from tertiary centers.4Henström M. et al.Gut. 2018; 67: 263-270Google Scholar,6Garcia-Etxebarria K. Zheng T. et al.Clin Gastroenterol Hepatol. 2018; 16: 1673-1676Abstract Full Text Full Text PDF Scopus (35) Google Scholar,7Zheng T. et al.Gut. 2020; 69: 397-398Google Scholar However, significant findings were only detected in UK Biobank participants with a primary diagnosis of IBS made during hospital inpatient admissions (as from EMRs), and not for definitions solely questionnaire-based. Although the latter may be more sensitive to response bias or over-reporting (as the respective prevalence may also suggest), it is conceivable that ICD10-IBS inpatient diagnoses correspond to more severe phenotypes (eventually more comparable with cases from tertiary centers). It is also tempting to speculate that the ICD10-IBS case group may represent a better reservoir of “organic” disease, including carbohydrate malabsorption as postulated to affect SI RPVs carriers. Altogether from this and previous studies, compelling evidence is accumulating for a role of SI variants in IBS, which holds potential for the management of a subgroup of patients via dietary modifications, possibly via starch- and sucrose-reduced diet9Nilholm C. et al.Nutrients. 2019; 11: 1662Google Scholar or similar specific carbohydrate diets. This research has been conducted using the UK Biobank Resource under Application Number 17435. Tenghao Zheng, MD, PhD (Formal analysis: Lead; Writing - Original Draft: Supporting). Leticia Camargo-Tavares, MSc (Formal analysis: Supporting; Writing - Original Draft: Supporting; Visualization: Lead). Ferdinando Bonfiglio, PhD (Formal analysis: Supporting; Writing - Review & Editing: Supporting). Francine Z. Marques, PhD (Resources: Supporting; Writing - Review & Editing: Supporting). Hassan Y. Naim, PhD (Conceptualization: Supporting; Writing - Review & Editing: Supporting; Funding acquisition: Supporting). Mauro D’Amato, PhD (Conceptualization: Lead; Resources: Lead; Writing - Original Draft: Lead; Funding acquisition: Lead). Download .pdf (.11 MB) Help with pdf files Supplementary Material Download .pdf (.15 MB) Help with pdf files Supplementary Table 1

Topics & Concepts