New NSG models further graft-versus-host-disease (GVHD) research

One of the key therapies for many human cancers and a number of genetic disorders is allogeneic stem cell transplantation – the transplantation of hematopoietic stem cells from a closely related donor. The efficacy of this therapy, however, is limited by allogeneic graft-versus-host-disease (allo-GVHD), an adverse reaction by the donor's immune cells to proteins (antigens) in the recipient, resulting in damage to the recipient's lungs, skin, gut, and liver. A University of Massachusetts Medical School research team led by Michael Brehm, Ph.D., in collaboration with Jackson Laboratory professor Lenny Shultz, Ph.D., recently developed two new mouse models for studying GVHD (Covassin et al. 2011).

Allo-GVHD is a tough problem for medical professionals because the very T cells that mediate it also improve engraftment, help re-establish the recipient's immune system and improve anti-tumor responses. To ameliorate allo-GVHD, the factors responsible for the adverse reactions by the donor's T cells to the recipient's body must be understood.

Much of our current understanding of GVHD derives from xenograft studies of human cells transplanted into animals. Indeed, all previous studies of GVHD in immunodeficient mice – including the development by Shultz and his colleagues of a model of human peripheral blood mononuclear cell (PBMC)-induced xenogeneic-GVHD (xeno-GVHD) using the NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ mouse (005557, NSG) – characterize how human T cells respond to mouse antigens. The Brehm team realized that a more relevant model is needed – one that allows researchers to study allogeneic responses of human cells to human antigens. So, together with Shultz, Brehm and his colleagues developed two new NSG models. The first, NOD.Cg-Prkdcscid H2-Ab1tm1Doi/SzJ (005589, NSG-Abo), available from The Jackson Laboratory, is an NSG mouse that lacks murine major histocompatibility complex (MHC) class II. The second, NOD.Cg-PrkdcscidIl2rgtm1WjlH2-Ab1tm1Gru/Tg(HLA-DRB1)31Dmz/Sz (NSG-Abo DR4), is maintained in the laboratory of Dr. Shultz. It lacks murine MHC class II and expresses human DRB1-0401 [DR4], a human leucocyte antigen (HLA).

The Brehm team first determined that isolated human CD4+ T cells (which recognize antigens presented by class II molecules) can initiate a strong xeno-GVHD (human versus mouse) in the absence of CD8+ T cells. They injected some NSG mice with isolated CD4+ T cells and some with whole PBMC (which contains CD8+ T cells and other components). They found that the mice injected with isolated CD4+ T cells lose weight, produce abnormally low numbers of platelets and die early, although not as early as those injected with enriched CD4+ T cells or whole PBMCs. Nevertheless, these data demonstrate that CD4+ T cells alone can mediate a strong xeno-GVHD response in the absence CD8+ T cells.

Brehm and his colleagues then determined that NSG-Abo mice and NSG-Abo DR4 mice can both engraft purified CD4+ T cells isolated from a human DR4-negative donor, demonstrating that human CD4+ T cells can engraft and survive in NSG mice that lack murine MHC class II, whether or not they express human DR4. Human DR4-negative CD4+ T cells do not to mediate xeno-GVHD in NSG-Abo mice, however, suggesting that the expression of murine MHC class II is required to efficiently induce xeno-GVHD by human CD4+ T cells. This suggested that the NSG-Abo mouse can be used to study other aspects of human CD4+ T cell function, such as the rejection of allogeneic skin, in the absence of the confounding effects of xeno-GVHD.

Finally, the researchers determined that human CD4+ DR4-negative T cells can induce GVHD responses in NSG-Abo DR4 mice, demonstrating that these mice can be used to establish a human CD4+ T cell-mediated allo-GVHD (in the absence of xeno-GVHD) driven by a human T cell receptor that recognizes the DR4 binding region.

In summary, the NSG models developed by the Brehm and Shultz team can be used to dissect the immunobiology of both xeno- and allo-GVHD and to test novel therapies that target human CD4+ T cells in vivo. The first model, NSG-Abo, can be used to study human CD4+ T cell-mediated alloimmunity, autoimmunity and viral immunity in the absence of xeno-GVHD. The second model, NSG-Abo DR4, can be used to study allo-GVHD, in which human CD4+ T cells recognize and react to allogeneic HLA class II molecules instead of murine MHC molecules.