The interplay between intestinal microbiota and graft-versus-host disease

From the Hill Lab, Translational Science and Therapeutics Division

“Allogeneic hematopoietic stem cell transplantation is a curative therapy for hematopoietic malignant and non-malignant disease. However, the immunological complication known as graft-versus-host disease (GVHD) is a serious complication,” noted Dr. Motoko Koyama, a Senior Staff Scientist in Dr. Geoffrey Hill’s lab, part of the Translational Science and Therapeutics Division at Fred Hutch, while discussing the Hill Lab’s recent research. GVHD occurs when cells from the transplant donor recognize healthy cells of the transplant recipient as foreign pathogens and attack them, leading to poorer outcomes and high-risk health scenarios for transplant recipients. Explaining how previous research in the Hill lab has focused on better understanding GVHD biology, Dr. Koyama explained that “acute GVHD occurs in skin, liver and gastrointestinal (GI) tract and is mediated by donor T cells. In our previous study (published in Immunity in 2019), we had shown that the presence of commensal bacteria in the mouse GI tract induced MHC class II (MHC-II) expression by intestinal epithelial cells (IEC), resulting in enhanced CD4+ T cell-mediated GVHD. However, we did not know which bacterial species were pivotal for the induction of MHC-II expression.” This previous work formed the basis for their recent follow-up study in Immunity which aimed to elucidate both the types of intestinal microbiota present in transplant recipients as well as the roles these microbiota may play in GVHD.

The authors’ first set of experiments compared genetically identical mice obtained from different vendors to determine if there were differences in IEC MHC-II expression and correlate these observations with the types of intestinal microbiota present. Notably, MHC-II was not expressed in mice derived from one specific vendor, Jackson Laboratories, but was present at similar levels in all mice from other vendors. The authors also observed that although intestinal bacterial load was comparable across the different mice, those mice lacking MHC-II expression had a distinct microbiota composition. Subsequent bone marrow transplant experiments in mice lacking MHC-II expression resulted in less severe GVHD than those mice with MHC-II expression. Explaining the rationale behind these first experiments, Dr. Koyama noted “the Hill lab moved from Australia to Seattle over this time and noted that IEC from mice sourced from the Jackson Laboratory in the USA unexpectedly did not express MHC-II. This initial finding is what led us to compare mice and commensal bacteria from various vendors. “We demonstrated that genetically identical mice of differing vendor origins have dramatically different intestinal microbiota and ileal MHC-II expression, resulting in significantly discordant GVHD severity,” she continued.


Microbiota present pre-transplant can dictate graft-versus-host disease severity in the post-transplant gastrointestinal tract.
Microbiota present pre-transplant can dictate graft-versus-host disease severity in the post-transplant gastrointestinal tract. Figure provided by Dr. Koyama

Next the authors utilized cohousing (whereby MHC-II negative mice were housed with MHC-II positive mice) and antibiotic treatment to characterize the bacterial taxa positively and negatively associated with MHC-II expression. “Here we refer to these bacteria as MHC-II inducers and MHC-II suppressors, respectively,” explained Dr. Koyama. These co-housing experiments caused MHC-II expression to be induced in mice previously lacking expression, with the MHC-II inducer strains present at significantly higher levels than MHC-II suppressor strains. Previous research had demonstrated the effectiveness of broad-spectrum antibiotics on reducing MHC-II expression in IEC, and— to confirm this finding—the authors treated their mice with antibiotics added to drinking water. A four-antibiotic cocktail successfully reduced MHC-II expression, and single antibiotic treatment led to the finding that “a large proportion of MHC-II inducers are vancomycin-sensitive and peri-transplant oral vancomycin administration attenuated CD4+ T cell-mediated GVHD,” said Dr. Koyama. The authors next hypothesized that MHC-II suppressors could be utilized post antibiotic treatment to maintain low MHC-II expression, and thus weaken GVHD severity. Confirming this hypothesis, mice treated with suppressor bacteria after antibiotic treatment were successfully colonized by the treated suppressors and demonstrated significantly decreased MHC-II expression. “We demonstrated that MHC-II suppressors actively suppress MHC-II expression by IEC whereby GVHD lethality is attenuated. In addition, we depicted significant overlap between bacteria that correlated with MHC-II expression and bacteria that correlated with interferon gamma (IFNg) production by intestinal T cells, suggesting that the effects of MHC-II inducers and suppressors were dependent on their ability to induce IFNg secretion in the ileum,” described Dr. Koyama.

Lastly, the authors extended their research to a cohort of patients undergoing allogenic stem cell transplantation. Analysis of pre-transplant stool samples established distinct microbiota composition in those patients who went on to develop severe GVHD post-transplant compared to those who did not, as well as distinct differences in patients who experienced transplant related mortality within 2 years versus those who did not. The authors were also able to determine a significant overlap in the types of bacteria associated with severe GVHD in humans and their mouse models. Summarizing these findings, Dr. Koyama said “we show a similar correlation between pre-transplant microbes and GVHD in a large clinical SCT [stem cell transplantation] cohort. These data highlight the mechanisms by which pre-transplant microbiota control the development of GVHD independently of genetic disparity and define therapeutic strategies to prevent disease.”

Overall, more research is needed to aid in future therapeutic decision making for this patient population. “While these findings confirm that perturbation in microbiota prior to transplant directly impacts transplant outcome, there is still much to learn in regard to the cellular and molecular mechanisms mediating this effect. Better understanding of these pathways should allow rational therapeutic intervention,” concluded Dr. Koyama. 

This work was funded by the National Institutes of Health, the Biostatistics Shared Resource of the Fred Hutch/University of Washington/Seattle Children’s Cancer Consortium, the National Cancer Institute, MSK Cancer Center Support Grant/Core Grant, the National Institute on Aging, Tri Institutional Stem Cell Initiative, the Lymphoma Foundation, the Susan and Peter Solomon Family Fund, the Solomon Microbiome Nutrition and Cancer Program, Cycle for Survival, Parker Institute for Cancer Immunotherapy, Paula and Rodger Riney Multiple Myeloma Research Initiative, Starr Cancer Consortium, Seres Therapeutics, the American Society of Clinical Oncology, Hyundai Hope on Wheels, the Cycle for Survival Equinox Innovation Award, the Collaborative Pediatric Cancer Research Program, a Michael Goldberg Fellowship, and a Tow Center for Developmental Oncology Career Development Award.

Fred Hutch/University of Washington/Seattle Children's Cancer Consortium members Drs. William Grady, Neelendu Dey, Timothy Randolph, Kate Markey, David Fredricks, and Geoffrey Hill contributed to this work.

Koyama M, Hippe DS, Srinivasan S, Proll SC, Miltiadous O, Li N, Zhang P, Ensbey KS, Hoffman NG, Schmidt CR, Yeh AC, Minnie SA, Strenk SM, Fiedler TL, Hattangady N, Kowalsky J, Grady WM, Degli-Esposti MA, Varelias A, Clouston AD, van den Brink MRM, Dey N, Randolph TW, Markey KA, Fredricks DN, Hill GR. Intestinal microbiota controls graft-versus-host disease independent of donor-host genetic disparity. Immunity. 2023 Aug 8;56(8):1876-1893.e8. doi: 10.1016/j.immuni.2023.06.024. Epub 2023 Jul 21. PMID: 37480848.