Enhanced Myeloid and T Cell Development and Improved Functionality of Cord Blood Xenografts In the NSGS Mouse.

Abstract 3729

The advent of NOD/SCID mouse strains lacking interleukin-2 receptor gamma (IL2RG) function (NSG and NOG mice) has greatly improved efforts to develop xenograft models of human hematopoiesis. Notably, these mice exhibit markedly reduced adaptive and innate immunity and are well suited for stable engraftment of human CD34⁺ cells. Importantly, these mice also allow for the development of human T cells, setting them apart from other immunodeficient strains. As a result, NSG and NOG mice are increasingly used for the building of model systems to study HIV infection, graft versus host disease, and immunity. While significant improvements have been realized, IL2RG knockouts lack robust myeloid components, and lymphoid function is likely to be suboptimal due to problems with B cell differentiation defects and delayed appearance of T cells. Recently, we have generated an NSG mouse strain with transgenic expression of several human myelo-supportive cytokines (SCF, GM-CSF, and IL-3), the NSGS mouse. In our initial characterization and study of this novel strain, we found a significant improvement in engraftment of AML cell lines and patient samples relative to NSG mice. In the current study we have extended these findings to include xenografts obtained using umbilical cord blood CD34⁺ cells (UCB). We have found the NSGS mouse to be equal to the NSG as a host for long-term stable engraftment of these cells when saturating cells doses are used, and superior when limiting numbers of CD34+ cells are injected. While CD34⁺ levels are much lower in established grafts in primary NSGS recipients compared to NSG mice, possibly as a result of continual mobilization of these cells by the cytokines, human cells were readily detected in secondary NSGS recipients, indicating maintenance of a primitive stem/progenitor cell compartment in vivo. Bone marrow of NSGS mice was predominantly composed of human myeloid cells, while the NSG mice show primarily CD19⁺ B cells. In contrast, the lineage of the human cells in the peripheral blood were very similar in these two strains, with both showing a gradual switch from myeloid to B cell dominance between weeks 3 and 7 post engraftment. Surprisingly, human T cells were found in the PB of transplanted adult NSGS mice as early as 8 weeks post engraftment, a full 8 weeks before T cells were detectable in NSG mice engrafted with the same UCB sample in parallel. These CD3⁺ cells presumably develop from the CD34⁺ stem cells and not from contaminating CD3⁺ T cells, since FACS-sorted CD3⁻CD34⁺ UCB samples produced the same result. Furthermore, normal donor human T cells did not exhibit any advantage in engraftment, cycling, or expansion in NSGS mice when compared to NSG mice in models of GVHD. Characterization of the T cells generated from human CD34⁺ xenografts revealed CD4⁺ and CD8⁺ subpopulations with phenotypes resembling activated, naïve, and memory T cell subsets. CD3⁺ spleen cells cultured ex vivo were responsive to activation by PHA/IL-2 stimulation and were susceptible to HIV-1 infection. Finally, humanized NSGS mice immunized with a toxoplasmosis extract were able to mount a response to virulent toxoplasmosis infection sufficient to significantly prolong survival while humanized NSG mice or non-humanized NSGS did not. This difference could not be attributed simply to T cell levels, because several of the NSG mice had comparable CD3+ populations at the time of exposure to antigen and subsequent challenge. While T cells are likely to be required for a response to toxoplasmosis challenge, the increased myeloid and dendritic cell populations generated in the NSGS mouse may prove to be equally critical for the functionality of UCB CD34⁺ xenografts.