Human Lymphocyte Engraftment and Function in HU-PBL-SCID Mice: a Dissertation

Abstract

The immune system is responsible for defending a host animal from a wide variety of threats. Manipulation of the immune system can result in beneficial outcomes such as immunity to pathogens, or deleterious outcomes such as autoimmunity. Advances in our understanding of how the immune system develops and functions have benefited greatly from studies in animals, particularly in mice where the genetics are well known and a multitude of reagents are readily available for experimental use. Although much has been learned from animal experimentation, it must be cautioned that animals are not humans. Unforeseen outcomes and complications often arise when translating research obtained in animal models to treatment of human patients. A small animal model in which the human immune system can be established and manipulated experimentally in vivo would be valuable for the study of human immune responses in infectious diseases, transplantation, and autoimmunity, and ultimately for translation of these findings to the human patient. Contribution to this model is the overall goal of this thesis. The severe combined immunodeficiency (Prkdcscid, termed scid) mutation was discovered in 1983 in the C.B-17 strain of mice. Scid mice lack functional T and B lymphocytes and are unable to mount immune responses or reject allogeneic or xenogeneIc grafts. However, C.B-17-scid mice do develop normal or even elevated innate immune function. Based on the ability of scid mice to accept xenografts, human peripheral blood mononuclear cells have been injected to generate "Hu-PBL-scid mice." These Hu-PBL-scid mice have been proposed as an in vivo model of the human immune system. Although this model was described over 12 years ago, there are a number of obstacles that impede its ability to recapitulate human immunity. The Hu-PBL-scid mouse model established using C.B-17-scid mice as recipients is hindered by 1) low levels of human leukocyte engraftment, 2) engraftment of predominately memory/activated human T cells with specificity directed against host murine MHC antigens, 3) rapid transition of engrafted human lymphocytes to a functionally anergic state, and 4) a paucity of knowledge providing an understanding of the mechanisms that underlie engraftment and function of the human lymphocytes in the immunodeficient hosts. This thesis was initiated at the time the NOD-scid mouse became available for establishment of Hu-PBL-scid mice, and this model has subsequently been termed "Hu-PBL-NOD-scid." The NOD-scid mouse was designed as an improved recipient for human PBL based on its innate immune characteristics. Defects in innate immunity in wild-type NOD/Lt mice include reduced NK cell activity, defects in macrophage development and function, and a lack of hemolytic complement. NOD-scid mice retained these characteristics, and engraft human cells at higher levels compared to C.B-17-scid mice. However, the ratio of CD4+ to CD8+ T cells in Hu-PBL-NOD-scid mice remained skewed towards the CD8+ population, similar to that of the Hu-PBL-C.B-17-scid mouse. My thesis was based on the overall hypothesis that additional genetic manipulation of the recipient strain would enhance further the engraftment and function of human cells. Based on the engraftment results obtained in NOD-scid mice, we performed these genetic manipulations using the NOD-scid strain as the reference. We first hypothesized that removal of MHC expression would lower human anti-mouse xenoreactivity and enhance engraftment of naïve T cells, a cell population not readily detectable in Hu-PBL-C.B-17-scid or Hu-PBL-NOD-scid mice. Towards this goal, the NOD-scid mouse expressing a targeted mutation for beta-2 microglobulin (B2mnull) mouse was created by Dr. Leonard Shultz at the Jackson Laboratory. Because B2m is required for expression of MHC class I and for development of functional NK cells, we predicted that NOD-scid-B2mnull mice would first exhibit a normalized CD4:CD8 T cell ratio resulting from reduced CD8 engraftment due to decreased human anti-mouse MHC class I reactivity. Since MHC class I molecule expression is required for the development of NK cells, we further predicted that there would be a reduction of NK cell activity, permitting enhanced engraftment. Data presented in this thesis demonstrates that human PBL engraft in NOD-scid-B2mnull mice at levels higher than NOD-scid mice, and with an increased CD4:CD8 T cell ratio. The mechanism(s) responsible for the increased engraftment of human cells in these mice became a major focus of this thesis. This thesis is composed of three Specific Aims. Specific Aim 1 was to determine the mechanism(s) underlying human cell engraftment and function in Hu-PBL-scid mice. These data are contained in Part 1 of the Results. Specific Aim 2 was to elucidate the costimulation interactions between human T cells and murine host APCs that control the level of engraftment and activation state of the human lymphocytes. These data are contained in Part 2 of the Results. Specific Aim 3 was to utilize these fundamental observations to initiate studies into the induction of primary human immune responses in Hu-PBL-scid mice. Although the goal of these latter studies was not attained, the data from the experiments performed is provided in Part 3 of the Results, and is discussed relative to future directions for this arm of the project. In Part 1 of the Results, we utilized in vitro and in vivo techniques to study the underlying mechanisms regulating human cell engraftment and function in scid mice. We observed that the absence of mouse MHC class I antigens in NOD-scid-B2mnull mice does not reduce the stimulatory capacity of APCs toward human T cells in vitro. We further demonstrated that naïve human T cells persist for at least 2 weeks in the peritoneal cavity of Hu-PBL-NOD-scid and Hu-PBL-NOD-scid-B2mnull mice. However, in both strains of recipients, the human cells progress to an anergic phenotype as documented by cell surface molecule expression and by functional activity. We further documented that the increased engraftment of human CD4+ T cells observed in NOD-scid-B2mnull mice is due predominantly to the ablation of host NK cell activity. Increased engraftment of human CD4+ cells in NOD-scid-B2mnull mice can be recapitulated in NOD-scid mice by antibody-mediated depletion of residual host NK cells. Finally, we demonstrated that expression of MHC class II molecules by recipient mice facilitated stimulation and engraftment of human cells. However, mouse MHC class II expression is not required for human cell engraftment into scid mice. In Part 2 of the Results, we addressed the cellular and costimulatory interactions of engrafted human cells in scid mice. Costimulatory interactions between T cells and APC are now known to be critical for the proper activation and functioning of cells in the immune system. Understanding the role of costimulatory interactions between human T cells, human APCs, and mouse APCs became a major focus of this thesis. We demonstrated that human CD4+ T cells are required for the engraftment of human CD8+ T cells. The mechanism by which human CD4+ cells mediate this "helper" activity requires expression of CD154. Antibody-mediated blockade of CD154 in vivo abrogates human cell engraftment in scid mice. The role of host APCs in the engraftment of human lymphocytes was demonstrated by blocking host CD40 with antibody. Preventing human T cell CD154 interaction with host CD40 on murine APCs blocked human cell engraftment in scid mice, demonstrating the importance of "trans-costimulation" in human T cell activation. This "trans-costimulation" appeared to be mediated by B7 expression on mouse APCs. We further demonstrated that in vivo depletion of human CD8+ T cells in Hu-PBL-NOD-scid mice leads to increased levels of human CD4+ T cells, elevated human immunoglobulin in the serum, and increased incidence of EBV-related lymphoproliferative disorders. These observations suggested that human CD8+ T cells are able to regulate human CD4+ T cell help and provide "surveillance" activity for EBV-related lymphoproliferative disorders. In Part 3 of the Results, we used the Hu-PBL-scid mouse to initiate experiments designed to generate primary human immune responses in vivo. These experiments were based on our observation that few if any human APCs survive in scid mice, and on reports that dendritic cells (DC) are required for activation of naïve T cells and initiation of a primary immune response. We used recombinant viruses expressing HIV-1 proteins that are being developed as potential vaccines for HIV as our immunizing reagent. For APCs, we used DCs from NOD-scid mice expressing human MHC class II molecules as the source of APCs presenting antigen to the engrafted human T cells in the scid mice. Our attempts to induce a primary immune response using DC from human MHC-transgenic NOD-scid mice were unsuccessful, as were direct immunization protocols. The results section, however, does highlight deficiencies that could be approached experimentally in future studies. In summary, the results presented in this thesis advance our understanding of the fundamental mechanisms controlling human cell engraftment in scid mice. This information supports the long-term goal of establishing a functional human immune system in a small animal model. We have identified many of the cell interactions and factors that regulate human cell engraftment and function in scid mice, and we have provided insights into host characteristics that will provide optimized engraftment of naïve human T cells. The studies led to the novel observations of the regulation of human CD4+ T cells by human CD8+ T cells, B cell activation, and progression of latent EBV infection to lymphoproliferative disorders in vivo. These studies further provide new information regarding "trans-costimulation", a previously unrecognized mechanism of T cell activation. These results provide data on the fundamental mechanisms that underlie obstacles to the goal of achieving engraftment of a functional human immune system in scid mice.