Abstract
To better understand the biology of graft-versus-host disease (GvHD) after hematopoietic stem cell transplantation (HSCT) and develop novel treatment strategies, accurate and clinically relevant experimental animal models are indispensable. The majority of our understanding of this potentially lethal complication is based on mouse models of bone marrow transplantation in major histocompatibility complex antigens (MHC) and/or minor histocompatibility antigen mismatched settings. These mouse models of GvHD provide us with an immeasurable wealth of information; however, many findings obtained from mouse models are not necessarily correlated with clinical GVHD in humans. In addition, these mouse models are not suitable for predicting the effectiveness of recent human-specific therapies such as monoclonal anti-human antibodies and adaptive cell immunotherapies. To overcome these limitations, xenogeneic GvHD models using severe immunodeficient mice are currently being used in worldwide and allow us to investigate in vivo human immune reactions.
The major aim of our study is to clarify the precise mechanism of immune response in vivo and to determine the principle components responsible for xenogeneic GvHD. We first observed that immunodeficient NOG mice receiving either no or sublethal irradiation consistently showed gradual body weight loss and eventual severe GvHD following injection of human unmanipulated peripheral blood mononuclear cells (PBMCs). Histopathology performed at the late phase of GvHD showed extensive infiltration of human CD45+ mononuclear cells, most of which were T-cells, and tissue destruction in lungs, bone marrow, liver, and spleen, with a smaller number detected in gut and skin unlike in human GvHD. We identified the infiltration of human T-cells exclusively in lungs and spleen before the onset of GvHD. Flow cytometric analysis showed a marked reduce of naïve (CD45RA+CCR7+) and central memory (CD45RA-CCR7+) T-cells and an increase of effector memory (CD45RA-CCR7-) T-cells within CD4+ subset in lungs, suggesting that xenogeneic response by human cells in lungs occur preceding systemic inflammation. Of note, the distribution of human cells was similar between intraperitoneal and intravenous injection. These results indicated that the acute lung injury is not associated with cell trapping within mouse pulmonary microvasculature by intravenous delivery.
We next subdivided human lymphocyte subsets by magnetic cell sorting before adoptive transfer to characterize human cells responsive for xenogeneic GvHD development. CD4+ T-cells mediated more rapid and severe GvHD than did an equal number of CD8+ T-cells, whereas neither natural killer cells nor γδT cells caused any symptoms of GvHD. All antigen presenting cells (APCs)-depleted PBMCs mediated GvHD as well, suggesting that human T-cells are activated independently of their own APCs in this model.
CFSE cell division assay showed more rapid proliferation of CD4+ rather than CD8+ T-cells. It is a noteworthy that the proliferation of and the expression of early activation marker CD69 on CD8+ T-cells was accelerated in the presence of CD4+ T-cells, indicating that CD4+ T-cells are required for the rapid and more efficacious response of CD8+ T-cells. We also found a strong increase of human IL-2, IFN-γ, and TNF-α in serum of mice injected with whole T-cells. In contrast, we observed very low levels of these inflammatory cytokines in mice injected with isolated CD8+ T-cells alone.
Taken together, these results suggest that the early activation of human naïve CD4 T-cells by recognition of foreign antigen on mouse APCs following cytokine production play a critical role in triggering systemic inflammation in xenogeneic GvHD model. This in vivo response was confirmed in vitro by demonstrating that human CD4 T-cells rapidly proliferated and expressed CD69 with significantly increase of IL-2 and IFN-γ in response to mouse dendritic cells. These findings are helpful for adequate evaluation of immune response and creation of optimal experimental conditions in xenogeneic GvHD model.
Kanda:Otsuka Pharmaceutical: Honoraria, Research Funding.
Author notes
Asterisk with author names denotes non-ASH members.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal