Colonic Diverticulosis. Is there a Genetic Component?

Abstract

Diverticular disease is a common condition among the elderly, with a steady increase in prevalence over time. Also, it has a temporal, geographic, and ethnic variation in incidence and prevalence, the genetic component having only recently been studied. For many years, environmental factors were the main link to the etiology. Recent studies estimated the heritability in this disease and identified genetic variants associated with diverticulosis. The interaction between structural changes of the colonic wall, diet, lifestyle, and genetic factors results in the development of diverticular disease. The purpose of this article is to review the existing data about genetic influence in this disease.

Keywords:colonic diverticulosis, genetics, GWAS, diet, ethnicity.

INTRODUCTION

Diverticulosis is a disease that has a high incidence, especially in the elderly population, and an increasing prevalence, reaching 71%-75% for individuals aged over 80 (1-3). It has a multifactorial etiology, including lifestyle factors, aging, structural changes of the colonic wall, and functional alteration (abnormal motility) (4, 5). In the case of diverticulosis, investigation at a genetic level has only recently begun (6). The idea that genetic factors are involved in underlying pathological mechanisms of diverticular disease had to build up over the years, and insinuating evidence emerged from many directions (5).

The first steps came with the observation of a temporal, geographic, and ethnic variation in both incidence and prevalence (7-9). Thus, an environmental etiology has been initially suggested, with the role of diet being emphasized (8, 10).

Over the years, more evidence came to light, indicating that these differences might be due, besides environmental factors, to a genetic influence or a combination of both (7, 11).

In the first part of the 20th century, the disease was rare in Europe, with a prevalence of 18% (7, 12, 13), but recent years have been witnessing an increasing incidence (5, 12). It is frequent in Western developed countries, especially in the USA (a prevalence of 55.5%) (14) and UK (47%) (5), less common in European developing countries (7) and rare in rural areas (8, 13). Short life expectancy and limited access to healthcare facilities might explain the low incidence in emerging countries from Africa (5). In South Africa, due to recent migration from rural to urban areas, the prevalence increased to 13.5% (if colonoscopy is the method of diagnosis) and 22.3% (on barium enema) (15). A variation in prevalence exists between Western (45%) and Asian countries (up to 25%) (2, 3). If early studies estimated a low prevalence in some parts of Asia, i. e., Japan, with only 1.6% in 1975, the situation has been changing. The tendency is for a steady increase, up to 23,6% in 2000 and 25% in 2010 (14). This group of high prevalence includes Philippine and Hong Kong (25-35%) (9), Singapore (45%) (16), Thailand (28.5%) (16), and Taiwan (13.5%) (17). For other countries such as China, the prevalence did not change very much over time, remaining low 1.97% (16). It is also the case for India, South Korea, with a maximum of 3% (9).

Given the increasing trend that was noticed for countries shifting to a Westernized diet, it becomes more evident that lifestyle factors can alter the incidence of the disease (2, 5).

Ethnicity and migrational studies

Conflicting results emerged from studying the risk of diverticular disease in migrant population. Some studies show a low incidence, indicating the role of genetic factors, while others revealed an increase in incidence, pointing towards environmental factors (18, 19).

Recent population-based studies have investigated migration from regions like Eastern Europe, Balkan, Middle East, Asia, and Africa, known to have a low risk for diverticular disease, to countries with high prevalence and high risk. Compared to the native population, the risk of being hospitalized as a result of diverticula in immigrants is low. After a period of naturalization, the risk of developing diverticular disease is increasing, and as this population is getting older, the incidence is higher, especially in women (20). Evidence shows that the rates of diverticular disease are changing with the cultural assimilation process and the influential role of environmental factors and diet (18, 20).

Other migrational and cross-sectional studies on ethnic groups living in high-risk countries for diverticular disease demonstrate different prevalence compared with the native population and also other ethnic groups. It is the case for immigrants of Turkish descent living in the Netherlands (12), for Indians of subcontinental Asia living in England (7), and the Bangladesh population establish in London (11). All these groups had a low prevalence of diverticular disease and a younger age at diagnosis, in contrast with the Caucasian population (7, 11, 12). One study has also assessed the prevalence rate among different ethnic groups: the lowest rate was for Bangladesh (2,7%), followed by Pakistani (9,9%), Oriental (34%), and Caucasian (36%) (11).

Ethnicity influences the predominant phenotype too. Proximal diverticulosis is more frequent in black people (2, 3) and Asian countries than in Europe (9, 14, 16). The proximal occurrence was not associated with the consumption of fruits or vegetables (7, 13), suggesting a genetic predisposition (7, 18, 21).

Also, diverticulitis rates are higher in Europeans living in the tropic compared with the native population (22).

Furthermore, an ethnic predilection was indicated by the variation in incidence among societies with a different cultural background sharing similar living conditions (23).

Inherited diseases and association with diverticular disease

Evidence suggesting a possible link between genetics and diverticular disease came from the observation that some inherited disorders of the connective tissue have a tendency to diverticula formation (18, 19, 24). In Ehlers-Danlos syndrome, the rate of diverticular disease and admissions for diverticulitis are higher than the general population (25, 26). The long term natural history of Williams-Beuren syndrome comprises multiple complications, including diverticular disease (27, 28). The prevalence of sigmoid diverticulitis is increasing in young adults with this syndrome in comparison with the general population of the same group age (below 40 years old) (29). Among patients with end-stage renal disease, the risk of diverticulitis is higher in those with renal failure due to polycystic kidney disease, and for half of them, the episodes are severe (30, 31).

Autopsy findings for patients with Coffin-Lowry syndrome include extensive intestinal diverticula (32). Generalized connective tissue disorder is a part of the syndrome (33). Patients with Marfan syndrome have an increased prevalence of colonic diverticulosis and a younger age at diagnosis. Diverticulosis is associated with the length and severity of active disease (high circulating concentrations of growth hormone and insulin-like growth factor-1) (34, 35).

Twin studies

These studies take us a step forward, estimating the heritability in diverticular disease. The advantage is that they assess the contribution of genetic factors, the importance of the environment, and possibly the relationship between them. They rely on the fact that monozygotic twins are genetically identical, the dizygotic share half of the genome, and the family environment is similar (36, 37). If the degree of similar phenotype in monozygotic is higher than the dizygotic pair, the genetic influence is suggested (19, 37). Advances in this direction were made in 2012, by Granlund, in Sweden (38) and in 2013, by Strate, in Denmark (10). They estimated a heritability of 40%-53% in the case of diverticular disease (19). However, there are some limitations. For example, Granlund’s study included only hospitalized patients and did not consider the extent of interplay between genes and environment, assuming that both monozygotic and dizygotic had similar environmental conditions (38). His results showed that the odds ratio of developing diverticulosis for the co-twin of the affected pair was 7.15 in monozygotic twins and 3.2 in dizygous twins (38). Strate did more extensive work. Besides the classical twin studies statistical genetics, she included both inpatients and outpatients, asymptomatic, and complicated cases for siblings in order to examine familial aggregation (10). The study showed that the relative risk for developing the disease in a sibling was 2.92 compared with the general population, and for twins, the relative risk for diverticulosis in one when the partner is affected is 14.5 in monozygotic and 5.5 in dizygous twins (10). Because monozygotic twins tend to be more exposed to the same environmental factors, the risk exists that this similarity might be due to them rather than genetics one. Strate’s findings for the longitudinal sibling cohort further support the twin studies results in both Sweden and Denmark populations (10).

Genetic association studies

Twin studies give compelling evidence of the genetic influence in diverticular disease. The next logical approach was to search for a specific gene. The observation that diverticulitis and inflammatory bowel disease (IBD) share some clinical traits and inflammation has a role in the pathophysiology of both diseases suggested a similar genetic susceptibility (1, 5, 39-41). Previous genetic studies in IBD associated TNFSF15 with Crohn disease and severe ulcerative colitis (5, 19, 42, 43). This information led to additional investigations on this gene role in patients with diverticulitis (44).

A single nucleotide polymorphisms (SNPs) rs7848647 was identified in the TNF SF15 gene from DNA samples obtained from colonic specimens in patients with complicated diverticulitis, requiring surgery. This study did not include mild forms of diverticulitis, and we should look at the SNPs as a marker for severity (45).

By what means the gene product influences the clinical course of diverticular disease can only be speculated at this moment (44). From research done for IBD and cancer we now know that the gene product of TNFSF15 is a cytokine (TL1A). It modulates inflammation (it is a pro-inflammatory molecule) and has an antiangiogenic activity (43, 46, 47). TL1A expression correlates to the degree of inflammation in IBD (43, 48).

The above mentioned study falls into the category of candidate gene association studies. It is a deductive approach, investigating a specific gene based on prior knowledge of its possible association with a disease (49). This fact implies a degree of subjectivity when selecting the candidate gene and represents a limitation. Also, in the case of complex polygenic disease, this type of study may not include all possible causative genes, but it is a fast way to determine the association of a specific gene variant with phenotype (50).

Another approach is genotype-first. Genome- wide association studies (GWAS) have the advantage of screening the entire genome, reading for all genetic variants (49). It is done without any prior presumption of the importance of a specific functional SNP (a common genetic variation which describes changes to a single DNA base) (49-51).

In 2017, a GWAS of the Icelandic and Danish populations discovered the first loci to associate with diverticular disease. The intronic variants were identified in three genes ARHGAP15, COLQ, and FAM155. The latter is associated with diverticulitis and the rest with uncomplicated diverticular disease (52).

Mutations in COLQ can cause muscle weakness by reducing acetylcholinesterase availability, which results in prolonged nerve to muscle signaling (1, 52). ARHGAP15 encodes a protein that affects the actin cytoskeleton and cell morphogenesis (1, 52). Information about FAM155 function are scarce, and the hypothalamus and pituitary gland are the tow places were it is mostly expressed (1, 52).

In the following year, Maguire performed a GWAS using clinical and genetic data from UK biobank. In the replication stage, they tested in the Michigan Genomics Initiative in order to validate the results. With a five times bigger discovery cohort and a higher number of case-control in both the discovery and replication stage, they reported and additional 39 novel loci associated with diverticular disease. In total, they were able to outline 99 genes for the 42 discovered loci. In the follow-up cohort, they were able to replicate only eight SNPs, all of them located in introns (53). Among the identified genes, some have roles in motility, cell adhesion, extracellular matrix or immunity, but others – i. e., CUTC and SPINT2 – have a role in transport of different molecules (like calcium, sodium or copper) (1, 5, 53).

This study is validating previously discovered risk-loci near ARHGAP15, FAM155A, and COLQ. They found no TNFSF15 association with diverticular disease, and Ehlers-Danlos genes were not detected (53).

The most recent study was also the largest GWAS and used the UK Biobank database. The authors were able to identify 48 loci associated with diverticular disease. Twelve of these variants were new and reported for the first time in 2019. Their findings were replicated in a large clinical cohort (6). The strength of this study is represented by the large number of replicated loci, 27 in total, of which 25 replicated for the first time (4). As the number of DNA samples grows, so is the number of identified loci, single nucleotide polymorphism (SNPs) located near, or in, genes (6).

Up to now, the replicated susceptibility loci for diverticular disease increases to 35 (4). The reported results validate all previously known risk loci in both the discovery phase and replication analysis (4).

Because GWAS identifies different variants in DNA associated with a disease, and each locus can contain multiple genes, detecting the causal gene is difficult (6, 51). In order to prioritize the candidate gene for each locus, they performed a series of downstream in silico analyze and were able to point out one gene per locus (4, 6).

There is no genetic overlap with Crohn’s disease, ulcerative colitis, or irritable bowel disease. There was no overlap for any of the known risk variants or lead candidate genes (4, 52). Twelve of the risk genes in diverticular disease are also involved in monogenic syndromes (neuromuscular, connective tissue stability disorders, and morphogenesis traits) (4).

This study suggests that polygenic diverticular disease is a disorder of impaired connective fibers support, intestinal neuromuscular, and mesenteric vascular smooth muscle function (4, 6). These findings offer a molecular basis for previously suggested pathogenic pathways of structural weakness of the intestinal wall and altered intestinal motility (4). The impact of these variants on the mechanisms is not fully clarified or understood (54). These new insights must be viewed for now as a hypothesis to be tasted (6)

Another GWAS was carried out on the Korean population; the intention was to assess their genetic susceptibility to diverticulosis. The study also focused on identifying genetic factors associated with the predominant phenotype in the Asian population. To increase the effect of genetic predisposition, authors set up a case group of patients with right-sided diverticulosis (defined as diverticula present from the cecum to the transverse colon) or bilateral and control group with no diverticulosis (55).

In total, they identified nine SNPs near tree genes WNT4, RHOU, and OAS1/3. For the right phenotype, the main associated SNP was rs22538787 (near WNT4 gene).

Unfortunately, their findings have a low statistical power and none of the SNPs meet the Bonferroni correction criteria. It is not possible to establish a convincing association between diverticulosis and these SNPs (55).

Overall, it is wise not to overestimate the genetic results. Because of genetic heterogeneity, variable expressivity, and penetrance, it is difficult to give a clear clinical interpretation of the variants found (6, 56). Also, we have to bear in mind the gene-environment interaction and that some genetic associations are relevant only in the presence of exposure factors (49).

CONCLUSION

At first glance, results in the published literature, to some degree, appear conflicting. Evidence is available for both genetic and environmental factors involved in the development of diverticulosis. For most studies, limitations exist, ranging from technical aspects to design and small number of patients, not to mention some variables, data with different statistical power. Nevertheless, knowledge accumulates with every new study, technology advances, allowing us to achieve in-depth exploration, comprehensive view and understanding of this complex disease.

Most likely, diverticular disease is the result of interactions between colonic motility, diet, lifestyle (smoking, reduced physical activity, obesity) and genetic factors (5, 19, 54, 57).

Taking into consideration all the evidence, genetics factors have a role in increasing the risk of diverticular disease in the presence of one or more known environmental risk factors (38).

Conflict of interests: none declared.

Financial support: none declared.