Genome Variation and Donor-Recipient Compatibility in Graft-Versus-Host Disease

Abstract SCI-12

Graft-versus-host disease (GVHD) after hematopoetic stem cell transplantion (HSCT) arises from differences between the genomes of donor and recipient. The fact that GVHD is common when HSCT involves HLA-identical siblings but almost never occurs when HSCT involves identical twins has long indicated that the human genome harbors histocompatibility loci outside the HLA locus. An important, unresolved question has involved what specific genomic loci might have potent effects on transplantation outcomes. Although GVHD is a disease of genome compatibility, it was until recently impossible to know all the genomic differences between any specific donor and recipient, or among the individuals in a population. 1. An ongoing revolution in genome analysis methods is making it possible to characterize genome variation almost completely. I will first describe what the complete sequencing of large numbers of human genomes in the 1000 Genomes Project is revealing about how antigen repertoire varies among the individuals in a population, or between any particular candidate donor and recipient. These results have surprising implications for research strategies to identify the genetic basis of histocompatibility barriers in human transplantation. 2. In this and earlier work, we have found a surprising number of genome polymorphisms that remove entire protein-coding genes from people's genomes. Many of these gene deletion alleles are sufficiently common that individuals inherit them from both parents and therefore lack completely an entire protein that is expressed in other individuals. The immune system of such an individual may tend to recognize epitopes encoded by that gene as alloantigens. Colleagues from the Dana Farber Cancer Institute, the Helsinki University Central Hospital and Finnish Red Cross, the University of Michigan, the Fred Hutchinson Cancer Research Center and the Broad Institute recently came together to evaluate the relationship of six of these gene deletions to GVHD in clinical cohorts. We found that donor-recipient mismatch for one of these gene deletions – deletion of the UGT2B17 gene – associated in multiple clinical cohorts with the occurrence of acute GVHD after HSCT. This same gene has been found to give rise to histocompatibility antigens in individual HSCT patients, offering a molecular and cellular mechanism for its association with GVHD. 3. Finally, I will describe a strategy for identifying the genetic bases of histocompatibility barriers among individuals, based on our emerging understanding of how the human genome varies in populations.