The National Institute of Diabetes & Digestive & Kidney Diseases

March 10, 2000 - The Genetics of Inflammatory Bowel Disease

Meeting Summary

1:00 - 3:00 p.m.
National Institutes of Health
Building 31, Room 6C6
Bethesda, MD

Participants

Welcome

Jay H. Hoofnagle, M.D., Director of the Division of Digestive Diseases and Nutrition at the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) and Chairman of the Digestive Diseases Interagency Coordinating Committee (DDICC), welcomed attendees to the meeting.

Research into the Genes of Inflammatory Bowel Disease

Steven R. Brant, M.D., Assistant Professor in the Department of Medicine Gastroenterology Section at the Johns Hopkins Medical Institutions and Director of the Meyerhoff Inflammatory Bowel Disease Center Genetics Laboratory, was the first speaker. Dr. Brant spoke on recent discoveries in the genetics of inflammatory bowel disease (IBD). He reviewed the two most common forms of IBD, ulcerative colitis (UC) and Crohn's disease (CD). UC tends to be localized to the colon, to consist of ulcers to the mucosal layer, and to not have fistulas present. CD, on the other hand, can be present beyond the colon, is transmural, and is often accompanied by fistulas. CD is associated with ocular complications such as uveitis, whereas UC is associated with autoimmune-implicated complications such as sclerosing cholangitis.

Dr. Brant discussed the epidemiology of IBD and how it is primarily a disease of Western, industrialized countries, where the IBD prevalence runs between 0.1 and 0.5%. Approximately 50% of IBD is CD and about 50% is UC, with a minor component grouped as either UC/CD or as unclassified IBD. There is evidence of a strong genetic component to IBD, mostly revealed through studies of twins, that point out a complex, non-Mendelian mode of inheritance involving the interactions of several gene loci. The strength of the genetic component can be represented by _s, the relative sibling risk as compared with the prevalence in the population at large. For IBD, this is quite substantial--around 35 for CD and 16 for UC (as compared with a _s of 15 seen in such putatively complex genetic diseases as type I diabetes).

Dr. Brant outlined the two principal methods for gene identification in diseases with complex genetic components such as IBD. They are (1) positional cloning and (2) the candidate gene approach. Positional cloning projects entails the identification of large sets of affected families and the analysis of affected relative pairs (sibling pairs, etc.) through the use of genomewide screens using large sets of chromosomal markers. The candidate gene approach begins with a basic understanding of the cellular phenotype of IBD tissue, and then looks for disease-correlated polymorphisms in genes that exhibit significant physiological roles in the IBD phenotype.

The first two IBD genetic susceptibility loci that were identified in humans were IBD1 on chromosome 16 (significant in CD patients) and IBD2 on chromosome 12 (significant in both CD and UC patients). The third loci described, now called IBD3, is on chromosome 1p, and it was first identified by Dr. Brant and associates in 1998.

Dr. Brant described a study that he and his colleagues conducted where they looked for IBD susceptibility gene loci by performing a family-based, genomewide screening of 594 people with IBD (297 pairs of relatives). About 34% of these affected relative pairs were Ashkenazi Jews, who have previously been shown by other investigators to have a twofold to eightfold higher incidence of IBD when compared with similarly situated non-Ashkenazi whites.

Dr. Brant's study looked at evidence for linkage in 377 genetic markers, distributed across the 22 human autosomal chromosomes. Dr. Brant and colleagues were able to replicate the findings of others for a susceptibility locus in the IBD1 of chromosome 16, finding a significant multiple logarithm of the odds (MLod) score of 1.69 when looking at patients with CD. The MLod scores were considerably less (0.61) if they looked only at non-Ashkenazi whites, and they were not present at all in families where the form of IBD was exclusively UC. Dr. Brant and associates also described novel IBD gene susceptibilities on chromosome 1p (IBD3: both CD and UC patients, MLod = 2.65), chromosome 3q (both CD and UC patients, MLod = 2.29), and chromosome 4q (seen in mixed relative pairs where one is CD and the other is UC, MLod = 2.76). The investigators were not able to identify the IBD2 CD susceptibility locus on chromosome 12 that other groups have seen when looking at European IBD populations.

Judy H. Cho, M.D., Assistant Professor at the Emma Getz Inflammatory IBD Research Center and the Department of Medicine at the University of Chicago Hospitals was the second speaker. Dr. Cho and Dr. Brant collaborate on research projects and have co-authored publications on IBD genetics. Dr. Cho discussed several strategies for refining searches for genetic susceptibility markers in complex genetic diseases like IBD. In fact, IBD is a heterogeneous collection of diseases that includes those with CD, UC, CD/UC, and unclassified IBD. Dr. Cho pointed out that one way to get better genetic susceptibility marker data, as reflected by higher MLod scores, is to reduce this heterogeneity, especially by studying well-defined homogeneous subsets of IBD patients. Another way to reduce heterogeneity is by looking at well-defined "special populations"--a good example being the Ashkenazi Jews mentioned above.

Dr. Cho identified other potential difficulties with the genetic study of IBD. It is expected that IBD will be multigenic, with several gene variants contributing to the IBD phenotype, but with no single gene being able to individually cause IBD. Also, IBD susceptibility gene variations might not be with the protein-coding open reading frames, rather might be either intronic or within a promoter. Such genetic variations might not result in changes in protein phenotype. There also may be a large amount of confounding nondisease-causing polymorphism within these candidate IBD susceptibility regions.

Dr. Cho discussed the evidence for epistasis between the IBD3 locus on chromosome 1p and the IBD1 locus on chromosome 16. If one focuses only on those CD patients who have positive nonparametric linkage for chromosome 1p, this subset of CD patients has an enhanced IBD1 genetic susceptibility MLod score at chromosome 16.

Dr. Cho presented further data on the IBD3 locus; it has been localized to the 1p35-6 chromosomal region, where there is a MLod score of 3.83 at the D1S562 locus.

Dr. Cho recounted that another potential source of IBD gene information is the Chaldean community, which is a Christian Iraqi community centered in Detroit, where IBD tends to be familial. Dr. Cho referred to the Detroit studies by Dr. Jason Bodson. There is suggestive evidence for a 200 kilobase IBD genetic susceptibility region centered on D1S436 on chromosome 1. This is a very focused region that is readily amenable to fine-structure analysis and final gene identification through DNA sequencing.

Dr. Cho reviewed the candidate gene approach for identifying IBD gene susceptibility genes. She pointed out several genes that have been considered as viable IBD candidates. They include (1) tumor necrosis factor-alpha, located within the major histocompatability complex on chromosome 6p; (2) guanine nucleotide binding protein alpha inhibiting activity polypeptide 2, located at chromosome 3p21; and (3) chemokine C-C motif receptor 5, also located at chromosome 3p21. The gene for interleukin-4 receptor is a potential candidate on chromosome 16 at position 16p12.1-p11.2, near the location of the IBD1. Dr. Cho also mentioned the multiple drug resistance-1 gene, located at chromosome 7q. All of these candidate genes are located near where positional genes have been sited, and they have phenotypic roles in colon physiology.

Dr. Cho remarked that IBD is often seen together with other genetically linked immune disorders, such as asthma and familial psoriasis, both of which are also complex gene disorders where several genes are expected to be implicated.

Dr. Cho mentioned another method for identifying IBD susceptibility--the use of expressional microarrays, where one can simultaneously monitor the gene expression of numerous proteins of interest to IBD phenotype bowel physiology and, thus, identify key proteins involved in the various IBD diseases. Dr. Cho is chairperson of the IBD Genotype/Phenotype Bank, a project of the Crohn's & Colitis Foundation of America.

An example of IBD pharmacogenomics that Dr. Cho recounted is the polymorphism seen in the thiopurine S-methyltransferase gene that is associated with azathioprine toxicity during treatment of IBD with azathioprine.

Dr. Cho summarized by pointing out that, at present, the primary goal of IBD genetic studies is to identify specific risk alleles for IBD. With the completion of the human genome sequence and the identification of all expressed genes, positional approaches and candidate gene approaches can be viewed as being largely complementary in nature. Understanding locus-locus interactions will be critical not only in identifying disease genes but, perhaps more important, in understanding pathogenesis. Population-specific differences will provide critical clues, and, within the U.S. population, a currently understudied population is African-Americans. There is strong clinical support for the concept that specific phenotypic patterns predominate among African-Americans with IBD. Gene-environment interactions may include an understanding of the role tobacco plays in the immune response as well as in intestinal epithelial function. More broadly, understanding the genetic basis for response to historically important infectious pathogens may provide clues to the pathogenesis of chronic inflammatory disorders. A more refined goal would be to understand the genetic basis for refined phenotypes: there is some evidence that specific extra-intestinal manifestations of IBD (e.g., primary sclerosing cholangitis associated especially with ulcerative pancolitis) may be more HLA-restricted. The ultimate goal of this work would be to ultimately reclassify the disease based on a specific, molecular understanding of disease pathogenesis. This will open up new therapeutic modalities and allow tailoring of therapeutic approaches for individual patients with IBD.

NIDDK Support of IBD and Genetics

Frank Hamilton, M.D., M.P.H., Chief of the Digestive Diseases Programs Branch of NIDDK, presented an overview of their IBD portfolio and NIDDK's interest in the genetics of IBD. Dr. Hamilton mentioned the congressionally mandated IBD Strategic Plan of NIDDK and the role of IBD research in it. Overall, most NIDDK support for IBD uses the K08 (Mentored Clinical Scientist Development Award) and the R01 (Research Project Grant) mechanisms. Dr. Hamilton pointed out funding by NIDDK on IBD had increased from $3 million in the late 1980s to more than $19 million a year currently.

Dr. Cho suggested that NIDDK funds might wisely be spent on efforts to redefine the IBD phenotypes beyond the simple dichotomy of CD and UC. It was also pointed out that basic research into intestinal T cell physiology would be likely to yield dividends in the understanding of IBD.

Dr. Hoofnagle stressed the importance of the further development and study of IBD animal models.

Reports and Conclusions

Dr. Hoofnagle reminded the audience of the recent appointment of a new NIDDK Director, NIH intramural scientist Allen Spiegel, M.D., in this, the 50th anniversary year of NIDDK.

Dr. Hoofnagle drew the attendees' attention to the 10th International Symposium on Viral Hepatitis and Liver Disease that is being sponsored by the Centers for Disease Control and Prevention and this year's Digestive Disease Week in San Diego. Mention was made of NIDDK initiatives in Hepatitis B and Hepatitis C virus research, drug-induced liver injury, and the living donor liver transplant concept.

Thomas F. Kresina, Ph.D., brought to the attendees' attention an upcoming NIAAA AIDS meeting.

Jorge Gomez, M.D., Ph.D., mentioned recent initiatives in colorectal cancer that NCI is undertaking.

Dennis Lang, Ph.D., recounted recent NIAID efforts toward understanding the microbial etiology of CD, especially the role of mycobacterium avium complex (MAC) and E. coli.

Phyliss Sholinsky, M.S.P.H., of the National Heart, Lung, and Blood Institute (NHLBI) reported that, in conjunction with NHGRI, NHBLI has started a genetic epidemiologic multicenter, multiethnic study of iron overload and hereditary hemochromatosis in 100,000 primary care patients. The study will focus on prevalence, genetic and environmental determinants, penetrance, and ethical, legal, and social implications of genetic testing and screening.

Ms. Sholinsky reminded the group that there are two genome scan genotyping services available across NIH--the NHLBI Mammalian Genotyping Service (MGS), which is free to all, and the NHGRI CIDR, which is free to grantees of contributing Institutes. Some of the linkage findings Dr. Cho referred to during her presentation resulted from genotyping performed by the NHLBI MGS.

Finally, mention was made of the interest of NIDDK in the new technology of DNA microarrays, and how it can be used to conduct molecular phenotyping of clinical IBD specimens.

Adjournment

Dr. Hoofnagle noted that the next DDICC meeting is scheduled for June 2000. The meeting adjourned at 3:00 p.m. on March 10, 2000.