Introduction of the Kit Mutation into Immunodeficient Mice Enables Xenogeneic Reconstitution of Human Megakaryocyte-Erythroid Hematopoietic System

Immunodeficient mice are used to reconstitute human hematopoiesis and analyze human hematopoietic function in vivo by xenotransplantation of hematopoietic stem cells (HSCs). It is well recognized that in addition to disruption of the lymphoid system, the strain background such as NOD is critical for efficient human cell engraftment in xenotransplantation models. We have reported that the NOD-specific polymorphism of the signal regulatory protein-alpha (Sirpa) allows the mouse SIRPA to bind human CD47, preventing activation of recipient macrophages to engulf human hematopoietic cells (Takenaka et al., Nature Immunol., 2006). We then developed a C57BL/6.Rag2^nullIl2rg^null mouse line harboring the NOD-type Sirpa, named the BRGS mouse. The engraftment efficiency of human HSCs in BRGS mice was equal to that of NOD-Rag2^nullIl2rg^nullmice. This BRGS strain should be useful for further genetic modification because its background is simplified (Yamauchi et al., Blood, 2013).

One of the persistent problems with the current xenograft models is that reconstituted human hematopoiesis is B-lymphoid dominant, whereas the engraftment levels of human myeloid cells, including megakaryocyte/erythrocyte lineage cells are quite low. In mouse syngeneic and allogeneic transplantation, recipients bearing mutation in the receptor tyrosine kinase Kit can accept donor HSCs without irradiation preconditioning. These data suggest that the Kit mutation impairs endogenous murine HSC self-renewal and allows donor HSCs efficiently to access to niche space. To develop a more efficient xenograft model, we further introduced a loss-of-function Kit^Wvinto the BRGS strain to deteriorate self-renewal and reconstitution capability of host HSCs. Here we show that this new mouse strain, termed the BRGSK mouse, showed very efficient human cell engraftment including myeloid, erythroid and megakaryocyte lineage cells.

CD34 ⁺ cells obtained from human umbilical cord blood were transplanted into sublethally-irradiated BRGSK and BRGS mice at the age of 6-8 weeks by intrafemoral injection. At 8-12 weeks after transplantation, the average human hematopoietic chimerism defined by human CD45⁺ cells was more than 90% in the BM and this chimerism was maintained over 24 weeks in homozygous BRGSK recipients. Furthermore, the chimerism of human CD33⁺ myeloid cells in the BM was significantly improved (~5.7% in BRGS vs. ~32.5% in BRGSK). Surprisingly, CD71^-CD235a⁺ mature erythrocytes and CD41⁺ platelets were detected in the BM. All human erythroid series including mature erythrocytes, CD71⁺CD235a^- early erythroblasts and CD71⁺CD235a⁺ late erythroblasts were clearly seen (~0.1% in BRGS vs. ~42.2% in BRGSK). Immunohistochemical staining showed that these human erythroid cells were positive for human hemoglobin subunit alpha. Confocal immunofluorescence imaging revealed that human erythroid cells formed large erythroid islands, independently from mouse erythroid islands in the sternal marrow of recipient BRGSK mice. A number of CD41⁺ megakaryocytes were scattered in the BM. The BRGSK mice also well supported terminal differentiation of myeloid cells such as neutrophils, eosinophils, basophils, mast cells, monocytes, and dendritic cells. Thus, reconstitution and maintenance of human myelo-erythropoiesis is significantly improved in the new BRGSK strain. The new BRGSK xenograft model should be a strong tool to investigate normal and malignant human hematopoiesis.