Motoko Koyama, MD, PhD, Geoffrey Hill, MD, and their colleagues at Fred Hutchinson Cancer Center have long been interested in the precise mechanics of antigen presentation after allogeneic hematopoietic stem cell transplantation (alloHCT). Their recent work, published in the August edition of Immunity, elegantly builds on previous data from their group and others, shedding further light on the mechanisms controlling the development of graft-versus-host disease (GVHD) after alloHCT.1
A decade ago, a number of groups demonstrated that non-hematopoietic cells — namely the intestinal epithelial cells — play a key role in GVHD initiation via the upregulation of major histocompatibility complex (MHC) class II, a surface molecule through which professional antigen-presenting cells usually display antigens to CD4+ T cells.2-4 Indeed, epithelial cells were found to be sufficient to induce GVHD, which was contrary to previous paradigms in the field. Further work demonstrated that MHC class II upregulation by epithelial cells was dependent upon myeloid differentiation primary response 88 (MyD88) and TIR-domain-containing adapter-inducing interferon-β (TRIF), both of which are adaptor molecules critical for toll-like receptor (TLR) signaling. TLRs are responsible for transmitting “danger” signals from either pathogen- or damage-associated molecular patterns (DAMPs/PAMPs), the vast majority of which are produced by bacteria, fungi, and viruses that usually live in symbiosis with the human host.
The current work by Dr. Koyama and colleagues offers the next layer of insight. Intrigued by the observation that genetically identical mice from different vendors (i.e., the same strain but raised in different environmental conditions) had different levels of MHC II expression, they examined the microbiome of these mice. In line with the observations of others, Dr. Koyama and her team found marked differences in the microbiome of genetically identical mice that were dependent on the vendor of origin. Furthermore, they leveraged these microbial differences to define potential “inducers” and “suppressors” of MHC class II. Manipulation of the pre-transplant microbiome such that it was dominated by suppressive organisms (the anaerobes Bacteroides thetaiotomicron and Clostridium sporogenes) resulted in dramatic protection from lethal GVHD in an antigen-specific mouse model of GVHD (i.e., where transgenic donor cells recognize a specific host antigen).
Furthermore, alongside colleagues at Memorial Sloan Kettering Cancer Center, I was able to explore a clinical dataset to ask whether there were commonalities between these “inducer” and “suppressor” bacteria in mice and the bacteria found in people undergoing transplantation. Using clinical stool samples from the pre-transplant period, we compared the specific bacteria present in patients who went on to develop acute GVHD with the bacteria present in those who did not. Intriguingly, the pre-transplant abundance of members of the Clostridia order was associated with reduced GVHD, which was consistent with the mouse data.
In Brief
This study is very important for the field, as we move toward clinical studies that aim to manipulate the human microbiome for patient benefit. Strategies that ensure patients have a diverse, anaerobe-rich intestinal microbiome prior to transplantation are likely to make a useful contribution to GVHD reduction. Furthermore, the pre-transplant window has now been identified as a key time point in both pediatric and adult patients undergoing transplants5 — further highlighting that this may be the ideal window for microbiome optimization.
Competing Interests
Dr. Markey indicated no relevant conflicts of interest, though she was a co-author on Dr. Koyama’s study, contributing to the clinical microbiome analysis along with colleagues Drs. Miltiadous and van den Brink from Memorial Sloan Kettering Cancer Center.
