Genetics of Inflammatory Bowel Disease
Introduction
The recent description of the first susceptibility gene for Crohn's disease has ushered in a new era of investigation in inflammatory bowel disease (IBD).
Identification of additional susceptibility genes and correlation of genetic alterations with clinical characteristics of disease surely will lead to finding the causes of, and hopefully the cure for, IBD. Nothing in this field can be more exciting for investigators, clinicians, and patients alike.
Evidence of a Genetic Component for IBD
Genetic factors play an important role in the pathogenesis of the inflammatory bowel diseases, including both ulcerative colitis (UC) and Crohn's disease (CD).
Evidence includes epidemiological data demonstrating differences in IBD incidence among different races and ethnic backgrounds, familial aggregation, and high concordance for IBD type in monozygotic vs dizygotic twins. Incidence rates for IBD are highest among whites, and among individuals of Jewish background compared with other ethnic groups.
When a proband presents with IBD, the risk of a first-degree relative developing the disease is higher among Jewish vs non-Jewish populations. The familial risk is especially high among siblings, in whom age-adjusted cumulative incidence for Jewish compared with non-Jewish individuals are 16.8% vs 7.0%, respectively, for CD and 4.6% vs 0.9%, respectively, for UC; Within Jewish populations, Ashkenazi (Eastern European) Jews have higher rates of IBD than Sephardic (Middle Eastern, African, or Spanish) Jews. Reports of familial clustering of IBD date back to the 1930s.
A family history is the single greatest known risk factor for the development of IBD, with at least a 10-fold increased risk compared with the population-based incidence rates. Among IBD patients, up to 20% will have a family history of IBD. A positive family history is more common with CD than UC, and relatives of patients with CD have a higher risk of developing IBD than do relatives of patients with UC, suggesting a stronger genetic influence for CD. Twin studies demonstrate greater concordance for IBD among monozygotic twins compared with dizygotic twins. Furthermore, the rates of concordance among monozygotic twins are higher for CD than UC, again suggesting that the genetic influence is stronger in the case of the former.
The Search for IBD Susceptibility Genes
The search for specific IBD susceptibility genes has been difficult due to complex genetic factors, such as the absence of simple Mendelian inheritance patterns, incomplete penetrance, genetic heterogeneity, and the involvement of more than 1 susceptibility gene.
An oligogenic model (limited number of genes acting together) or a genetic heterogeneity model are the most likely modes of inheritance for IBD. A number of genetic linkage studies have been performed to identify IBD susceptibility loci. Linkage methods show chromosomal locations of susceptibility genes by studying the transmission of genetic markers of known chromosomal location within families. To date, 5 confirmed linkages (meaning significant linkage found in one study and replication of the linkage with a nominal P-value < .01 in an independent study) for IBD have been identified: linkage between CD and a locus on chromosome 16 (the IBD1 locus); linkage between IBD (especially UC) and a locus on chromosome 12q (the IBD2 locus); linkage between IBD (especially CD) and a locus on chromosome 6p (the IBD3 locus); linkage between CD and a locus on chromosome 14q (the IBD4 locus); and linkage between IBD (especially CD) and a locus on chromosome 3p21.
Other significant linkages to IBD that have not yet been confirmed in independent study populations include linkage between early age of onset CD and a locus on chromosome 5q (the IBD5 locus), linkage between IBD and a locus on chromosome 19p (the IBD6 locus), linkage between IBD and a locus on chromosome 1p (the IBD7 locus), and linkage between IBD and a locus on chromosome 3p25-26. Recently, 3 independent groups have found that the linkage between CD and the pericentromeric region of chromosome 16 (the IBD1 locus) is due to CD-associated genetic variants in the NOD2 (nucleotide-binding oligomerization domain 2) gene.
Three coding region variants in this gene (Leu1007insC, Gly908Arg, and Arg702Trp) are associated with CD, but not with UC. Patients carrying 1 high-risk allele have a 1.5- to 3.0-fold increased risk of developing CD, whereas those carrying 2 high-risk alleles (homozygous for the same risk allele or compound heterozygous for 2 different risk alleles) have a 14- to 44-fold increased risk of developing CD. However, NOD2 variants only account for 20% to 30% of all CD cases. The NOD2 protein is primarily expressed in monocytes and is involved in apoptosis and nuclear factor-kappa B activation. One of the coding region variants (Leu1007insC) leads to a frameshift mutation with the resultant production of a truncated protein.
Preliminary functional data demonstrate that this mutation may result in hyporesponsiveness to bacterial lipopolysaccharides, suggesting an association between the innate immune response to intestinal bacteria and the pathogenesis of CD.
Genotypic and Phenotypic Heterogeneity: CD as an Example
It is likely that IBD is not a single disease entity, but rather comprises several disorders that share clinical features. Evidence for genetic heterogeneity in IBD comes in part from the varying results of genetic linkage mapping studies. The detection of linkage to established IBD susceptibility loci in some but not all study populations is likely due at least to genetic heterogeneity. Furthermore, genetic heterogeneity within CD clearly exists at the NOD2 and the chromosome 5q loci, because only a minority of patients with CD have the identified CD-associated mutations.
Data from 3 published studies demonstrate that NOD2 variants are found in only 20% to 30% of CD patients. Similarly, Rioux and colleagues found that 5q locus mutations contributed to CD susceptibility principally in families with young age of CD onset.
What is the implication of these findings?
The implication is that there are likely other, as yet unidentified CD susceptibility genes that will account for the remaining cases of CD. Since the original description in 1932 by Crohn and colleagues of a disease characterised by subacute or chronic inflammation of the terminal ileum affecting primarily young adults, it has been shown that Crohn's disease is a heterogeneous disorder with varied clinical symptoms, intestinal sites of involvement, and complications.
One of the first attempts to classify subgroups of patients with CD dates back to 1975, when Farmer and coworkers stratified 651 CD patients on initial disease location and found that this phenotype was a major determinant of symptoms, clinical course, and prognosis. For patients with involvement of both ileum and colon, there were frequent perianal fistulas (38%), internal fistulas (34%), intestinal obstruction (44%), and need for surgery (73%).
For patients with isolated colonic disease, megacolon and arthritis were more frequent, and patients with small bowel disease were more likely to develop intestinal obstruction. In a follow-up evaluation of this cohort for a mean of more than 13 years, patients with ileocolonic disease were much more likely to have required surgical resection (92%) than those with either isolated small intestinal disease (66%) or colonic disease (58%); P < .001. Another example of phenotypic heterogeneity in CD relates to disease behaviour. CD "disease behaviour" generally can be classified as follows: fistulizing/perforating; structuring; or inflammatory. Such classifications may define different disease courses in patients with CD. For example, Greenstein and colleagues proposed that perforating (acute perforation, abscess, fistula) and nonperforating (obstruction, medical intractability, haemorrhage) indications for surgery define 2 distinct clinical forms of CD.
Patients with perforating indications for surgical intervention tend to manifest the same disease pattern when they require further surgery, whereas those with nonperforating disease retain a nonperforating clinical pattern at diagnosis of postoperative recurrence. Patients with a perforating indication for surgery also are more likely to develop an early postoperative recurrence. Such differences in disease behaviour among CD patients may be related to differences in the host immune response. Finally, age at diagnosis may also define different subgroups within CD. A study of 552 patients with CD examined the association between age at diagnosis and disease location and behaviour. Patients of younger age at diagnosis (less than 20 years old), when compared with those of older age at diagnosis (40 years or older), were more likely to have small bowel involvement (89% vs 58%; P < .0005), structuring disease (46% vs 29%; P < .005), surgery frequency (71% vs 55%; P < .001), and a family history of IBD (30% vs 14%; P < .005). Conversely, colonic disease and inflammatory behaviour were significantly more common in patients diagnosed with CD at the age of 40 years or older.
The study authors concluded that younger age at diagnosis represents a separate and more severe phenotype of CD, with a different genetic component. Based on the above discussion, defining more homogeneous groups of CD patients is clearly important in examining genetic determinants of the disease. Genetic studies are more likely to be productive if they are conducted on a phenotypically homogenous population.
To help standardise such phenotyping studies, there was an initial working party classification of patient subgroups in CD known as the "Rome classification and a subsequent simplified classification, the "Vienna criteria. The Rome classification defined disease location, disease behaviour, extent of disease, and operative history as important phenotypic endpoints. This system, however, could theoretically lead to as many as 756 subgroups of CD. In the subsequent Vienna classification system, 3 variables were chosen as primary phenotypic endpoints: (1) age at diagnosis (< 40 years or > 40 years); (2) disease location (terminal ileum, colon, ileocolon, or upper gastrointestinal tract); and (3) disease behaviour (nonstricturing, nonpenetrating, structuring, or penetrating). Other phenotypic markers were considered as "further data to be collected," and included sex, race, Jewish ethnicity, family history of IBD, and extraintestinal manifestations.
Conclusion
We have learned a great deal about the genetics of IBD in recent years. Further advances will come with the detection of additional susceptibility genes, elucidation of the function of these genes and their protein products, and the designing of therapies targeted at the pathophysiologic consequences of specific gene alterations. Indeed, we are in an exciting and new era.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Copied from an Article by Jean-Paul Achkar, MD, Bret A. Lashner, MD, Cleveland Clinic Foundation, Cleveland, Ohio
