Final Progress Report
Proposal No. IBD-0003R
Principal Investigator: Geoffrey W. Krissansen, Ph.D.
Applicant Organization: University of Auckland (New Zealand)
Project Title: Pattern recognition receptors as potential susceptibility loci for Crohn's disease
Period of Award: November 1, 2002 – March 31, 2005
A. Summary of project aims
1) To genotype Crohn’s disease (CD) patients in order to identify CD susceptibility genes in New Zealand and to determine whether CD patients carry multiple polymorphisms in susceptibility genes that contribute to CD.
2) To demonstrate that polymorphisms in compound heterozygotes who carry both the 3020insC and R702W polymorphisms in the NOD2 gene are located on separate alleles.
3) To define the functions of alternatively spliced natural NOD2 RNA transcripts lacking the LRR domain versus those of intact full-length NOD2 transcripts.
4) To characterize the molecular defect in patients who fail to express intact NOD2 transcripts containing the LRR domain.
5) To determine whether NOD2 binds directly to LPS and MDP via the LRR domain.
6) To identify intracellular ligands for NOD2.
7) To determine whether introduction of a wildtype NOD2 protein can restore responsiveness to MDP and lipid A deficient LPS in either compound heterozygotes who carry both the 3020insC and R702W polymorphisms, or homozygotes who carry the 3020insC polymorphism.
B. Accomplishments towards meeting those aims
1) A total of 185 CD patients in New Zealand and 187 ethnically matched controls were genotyped for polymorphisms (3020insC/L1007fsinsC, G908R and R702W) in the NOD2 gene, a functional polymorphism (–159C/T) in the promoter of the CD14 gene, and polymorphisms in 12 other genes. Whilst only the 3020insC/L1007fsinsC and a polymorphism in one other gene were found to be associated with CD in New Zealand, many patients were found to carry more than one potential susceptibility allele.
2) There were six compound heterozygotes who carried both the 3020insC and R702W polymorphisms, one who carried both the 3020insC and G908R polymorphisms, and three who carried both the G908R and R702W polymorphisms. The polymorphisms were found to reside on separate alleles.
3) Fifteen different mRNA splice variants of the NOD2 gene were identified, which are predicted to give rise to nine different NOD2 proteins with distinct structures. The activity of the most common variant was further characterized.
4) With further study, it was revealed that all CD patients express NOD2 transcripts containing the LRR domain.
5) We were not able to demonstrate a physical association between MDP and the NOD2 LRR by using biotinylated MDP immobilized on strepavidin magnetic beads to isolate NOD-WT or a stand-alone version of LRR. In agreement, Chen et al. (2005) could not detect an interaction between Sepharose immobilized MDP and NOD2, but rather MDP bound to calreticulin [Chen, D., Texada, D.E., Duggan, C., Liang, C., Reden, T.B., Kooragayala, L.M. and Langford, M.P. (2005) Surface calreticulin mediates muramyl dipeptide induced apoptosis in RK13 cells. J. Biol. Chem., (in press)]. Taken together, the data indicate that the LRR of NOD2 does not bind directly to MDP.
6) This aim was not pursued.
7) Introduction of increasing amounts of NOD2 gene bearing the 3020insC mutation had no affect on wildtype NOD2, indicating that introduction of a wildtype NOD2 protein into monocytes from CD patients should be able to restore responsiveness to MDP, i.e., alleles bearing the 3020insC mutation do not display dominant-negative properties. This work was done using transfected HEK293T cells (a human embryonic kidney cell line), and will need to be repeated using cells from CD patients.
C. A list of significant results (positive or negative)
Here we describe the results of the NOD2 and CD14 genotyping in more detail. Polymorphisms in the CARD15/NOD2 gene, which encodes a cytosolic protein involved in bacterial recognition, are associated with development of Crohn’s disease. Other potential susceptibility genes such as CD14 may compound the risk of developing CD. We examined the frequency of the three major CARD15 risk alleles (3020insC/L1007fsinsC, G908R and R702W), and a functional polymorphism (–159C/T) in the promoter of the CD14 gene in 185 CD patients in New Zealand and 187 ethnically matched controls. The frequencies of the 3020insC (8.1 vs. 0.8%, P < 0.0001), G908R (3.5 vs. 2.4%, P = 0.37) and R702W (7.3 vs. 5.1%, P = 0.21) alleles in CD patients and controls, respectively, were similar to those described in Australia, and the ancestral countries of Scotland, Ireland and the UK. Only the 3020insC polymorphism was found to be a significant risk factor for CD in our New Zealand cohort (odds ratio = 10.91 [95% confidence intervals 3.30–36.08]; P < 0.0001 for heterozygotes), but not a single patient was homozygous for the 3020insC polymorphism. The T allele (51 vs. 50%, P = 0.77) and TT genotype (26 vs. 24%, P = 0.84) frequencies of the –159C/T CD14 gene promoter polymorphism did not significantly differ between CD patients and controls. In summary, our findings provide evidence that the CARD15 3020insC risk allele influences disease susceptibility in a small proportion (<17%) of New Zealand CD patients, whereas there was no evidence that the CD14 –159C/T polymorphism is associated with CD.
D. Lay summary of this report
Recent research identified a gene called NOD2 that was damaged in ~15% of patients suffering from Crohn’s disease (CD). The NOD2 protein that is made from the NOD2 gene helps our immune systems to recognize and respond to gut bacteria, but the damaged NOD2 protein cannot do this correctly in CD patients. NOD2 is but one of several proteins that allow our immune systems to recognize gut bacteria. We have screened 185 CD patients from New Zealand to determine whether NOD2 and thirteen other genes that allow the immune system to either recognize or respond to gut bacteria are associated with CD. Certain mutations in the NOD2 gene and in one other gene were found to be associated with CD, but we could find no association with mutations in the other twelve genes. Mutations in eleven of the other twelve genes have been found to be associated with CD in other countries, indicating that ethnicity (genetic make-up) plays a key role in determining disease susceptibility.
While studying the NOD2 gene, we serendipitously discovered that it gave rise to an array of different forms of NOD2 protein differing in length. The most common truncated variant was characterized as regards its ability to sense bacteria. Studies of the variants are warranted, as the results may provide insights into the biology of NOD2. The biology and role of NOD2 is still poorly understood. It has previously been proposed that NOD2 binds directly to bacterial components, but our results and the recent results of one other group could provide no evidence that this is the case.
In the future, it may be possible to treat individuals who have a damaged NOD2 gene by introducing an undamaged NOD2 gene into their cells to restore NOD2 function and thereby prevent development of CD. A beneficial outcome would only be possible if the damaged NOD2 protein is neutral, i.e., it does not antagonize the undamaged NOD2 protein. Our results using molecular tools indicate that the damaged NOD2 protein does not antagonize the normal NOD2 protein, hence providing one option for preventative therapy.