Transplanting the genetic susceptibility to Crohn’s disease

Crohn’s disease (CD) is one of the two most common forms of inflammatory bowel disease (IBD). The prevalence of CD has increased in Western countries over the past decades and mainly young patients are affected, with a peak incidence between 15 and 35 years.¹ The aetiology of IBD is still unclear and should be considered as multifactorial according to recent studies.² Genetic factors seem to play a pathogenic role as well as environmental, infectious, and immunological factors. All of these different aetiological aspects are reconciled in a paradigm, in which CD could result from disturbances of the intestinal barrier and pathological activation of the intestinal immune response towards luminal bacterial antigens in individuals with genetic susceptibility.

Immunological key players for the pathogenesis of CD have been identified, including cellular components such as lamina propria macrophages and CD4+ T lymphocytes as well as cytokines such as tumour necrosis factor α (TNF-α), interleukin (IL)-6, IL-12, IL-18, and others.^1–3 Identification of these pathogenetically relevant factors has been greatly facilitated by the availability of appropriate animal models, in particular genetically engineered knockout mice or transgenic mice, respectively. When SCID mice lacking functional B cells and T cells are reconstituted with a special subset of CD4+ T helper cells expressing the surface markers CD45Rbhigh or CD62L, they develop chronic colitis.⁴ These T helper cell subsets are thought to differentiate preferentially towards Th1 cells in the host producing those proinflammatory cytokines that are involved in the pathogenesis of CD, such as TNF-α.^5,6 To date, however, clinical and experimental evidence for the role of distinct mononuclear cell populations has been limited. There are some reports on long term remission of CD after bone marrow transplantation.⁷ Furthermore, human immunodeficiency virus (HIV) associated decrease in CD4+ T helper cell number seems to induce clinical remission of CD.⁸ In addition, it has recently been shown that the immunosuppressive drug of choice for the long term treatment of CD, azathioprine, promotes apoptotic cell death of lamina propria CD4+ T helper cells and one effect of the anti-TNF antibody Infliximab is the rapid induction of apoptosis of peripheral blood monocytes and lamina propria T cells.^9,10

In their case report in this issue of Gut, Sonwalkar and colleagues¹¹ report on a patient with Hodgkin’s lymphoma who developed fulminant colitis following non-myeloablative allogeneic stem cell transplantation (ASCT) [see pages 1518–21]. Although the clinical course with sudden onset of severe bloody diarrhoea and pancolitis sparing the terminal ileum were atypical, the colitis was classified as Crohn’s colitis based on the histological findings of patchy transmural inflammation and the presence of non-caseating epitheloid granuloma. In addition, thorough diagnostic testing ruled out tuberculosis, graft versus host disease, neutropenic colitis, vasculitis, HIV, herpes simplex virus, or cytomegalovirus colitis as potential differential diagnoses. This report is remarkable in so far as it strongly suggests—without proving definitely—for the first time directly in the human system that susceptibility to CD can be transferred via haematopoietic stem cells. This report might thus be considered as a proof of principle for the pivotal role of genetic factors in the pathogenesis of CD.

An aetiological role of genetic factors has long been discussed based on family and ethnic studies. The first molecular genetic evidence was provided by large linkage analyses by microsatellites, suggesting relevant genetic loci on chromosome 5 (IBD5), chromosome 6 (IBD3), and chromosome 16 (IBD1).¹² The IBD5 locus comprises a cluster of genes coding for cytokine genes involved in Th1/Th2 differentiation, and single nucleotide polymorphisms are associated with susceptibility to CD. Finally, certain HLA haplotypes located at IBD3 have been suggested to confer a slightly increased overall risk for the development of CD.

The recent cloning of the NOD2/CARD15 gene on the gene locus IBD1 and the identification of a large number of different NOD2 mutations in a subgroup of patients with CD^13,14 has raised new interest in genetics in CD. Most mutations are localised in a structurally characteristic C terminal domain of the NOD2 protein that resembles bacterial lipopolysaccharide binding toll-like receptors. In vitro studies showed that NOD2/CARD15 activates the transcription factor nuclear factor κB. There is evidence that NOD2/CARD15 is expressed in monocytes and intestinal epithelial cells. As a potential intracellular receptor for bacterial components, NOD2/CARD15 may be involved in the early innate immune response (including defensin production) that induces the physiological state of tolerance towards bacterial antigens from the gut lumen.¹⁵ This concept might help to explain why inactivation of NOD2/CARD15 increases susceptibility to CD. Finally, NOD2/CARD15 mutations in CD correlate with the development of ileal and fibrostenotic forms of CD.¹⁶

To support the idea that susceptibility for CD has been transferred by ASCT, the authors performed a detailed genetic analysis of the CD susceptibility loci in the patient who developed CD after ASCT and in the donor.¹¹ This included the NOD2/CARD15 gene on the IBD1 locus, including the 5′ UTR (chromosome 16) HLA haplotypes on the IBD3 locus (chromosome 6), with special focus on non-classical HLA class III gene haplotypes and three single nucleotide polymorphisms at the IBD5 locus.

Although the screening was negative for all 30 NOD2/CARD15 mutations described, there was a change in a 5′ UTR polymorphism of the NOD2/CARD15 gene at position −33 between donor and recipient. The donor and post ASCT recipient were homozygous for a T allele that may be associated with CD, while the pre ASCT recipient was homozygous for the wild-type G allele.

High resolution molecular typing confirmed that donor and pre ASCT recipient DNAs were matched for most of the HLA class I and II haplotypes except for HLA-DPB1 and HLA-B where a novel allelic variant was identified in the recipient. However, genotyping for 320 single nucleotide polymorphisms in 24 non-classical HLA class III genes at IBD3 between HLA-E and TAPBP revealed significant mismatches at several sites, including MICB, TNF, HSP70, NOTCH4, and LMP2 and a double haplotype mismatch at LMP7. The 8.1 HLA haplotype previously associated with CD was not found.

The authors conclude correctly that the findings of the HLA class mismatches at IBD3 and the CD associated polymorphism of the 5′ UTR of NOD2/CARD15 do not prove that these genetic variations are the underlying cause for an adoptive transfer of genetic susceptibility to CD from donor to recipient via ASCT. Nevertheless, their findings make the idea likely. This case thus nicely illustrates the meaning of the multifactorial pathogenesis of CD. The presence of genetic susceptibility factors by itself is not sufficient to elicit clinical CD as the donor never had symptoms of colitis. Development of clinical CD requires the coincidence of genetic susceptibility and a special microenvironment in the gut, depending on alterations of the intestinal epithelium and/or the intestinal flora. It may be speculated whether conditioning chemotherapy, including fludarabine and melphalan, altered the intestinal epithelium or whether the long term antibiotic and immunosuppressive therapy post ASCT might have altered the patient’s intestinal microenvironment such that colitis could develop from susceptibility factors carried over from the stem cell donor to the recipient.

Starting from the present case of a probable adoptive transfer of CD, the authors discuss whether ASCT donor selection should include screening for IBD. They suggest that formal questioning about IBD should be included during ASCT donor ascertainment. However, given the current paradigm of IBD having a multifactorial genesis,¹⁷ what consequence would such a screening have? In light of the efforts necessary, using word wide data bases to identify appropriate HLA matched donors for ASCT in a timely manner, would one really decline a potential donor only because of a family history of IBD? Given the weak correlation of most IBD linked genes with clinical development of the disease, should volunteers be kept from stem cell donation because of genetic susceptibility for IBD? Larger studies are warranted, including formal questioning about IBD in the family history of stem cell donors and recipients as proposed by the authors, complemented by molecular screening for relevant gene loci such as NOD2 and prospective follow up of these patients. This way, empiric data might be generated to address this particular question.

However, the issue raised by the authors should be put into a broader perspective. The molecular approach in medical research has led to the identification of numerous genes as potentially relevant for disease and their number is increasing rapidly. Apart from other ethical aspects, the impact for transplant medicine has to be discussed. How should the knowledge be handled about genes in organ donors and recipients that are implicated in the pathogenesis of disease in a non-monogenetic fashion with varying penetrance?

With regard to the present case the answer is easy: for the time being our understanding of the role of genetics in IBD is to preliminary to justify the exclusion of a patient with a positive family history for IBD or with proven genetic susceptibility factors from stem cell donation. In more general terms, however, there will be increasing need for debate of this issue in the future.