Engineering of human mesenchymal stem cells resistant to multiple natural killer subtypes.

Mesenchymal stem cells (MSCs) as a therapeutic promise are often quickly cleared by innate immune cells of the host including natural killer (NK) cells. Efforts have been made to generate immune-escaping human embryonic stem cells (hESCs) where T cell immunity is evaded by defecting ß-2-microglobulin (B2M), a common unit for human leukocyte antigen (HLA) class I, and NK cells are inhibited via ectopic expression of HLA-E or -G. However, NK subtypes vary among recipients and even at different pathologic statuses. It is necessary to dissect and optimize the efficacy of the immune-escaping cells against NK subtypes. Here, we first generated B2M knockout hESCs and differentiated them to MSCs (EMSCs) and found that NK resistance occurred with B2M-/- EMSCs expressing HLA-E and -G only when they were transduced via an inducible lentiviral system in a dose-dependent manner but not when they were inserted into a safe harbor. HLA-E and -G expressed at high levels together in transduced EMSCs inhibited three major NK subtypes, including NKG2A+ /LILRB1+ , NKG2A+ /LILRB1- , and NKG2A- /LILRB1+ , which was further potentiated by IFN-γ priming. Thus, this study engineers MSCs with resistance to multiple NK subtypes and underscores that dosage matters when a transgene is used to confer a novel effect to host cells, especially for therapeutic cells to evade immune rejection.

Human ESC-derived MSCs enhance fat engraftment by promoting adipocyte reaggregation, secreting CCL2 and mobilizing macrophages.

Mesenchymal stem cells (MSCs) derived from somatic tissues have been used to promote lipotransfer, a common practice in cosmetic surgery. However, the effect of lipotransfer varies, and the mechanism of action remains vague. To address these questions, we differentiated human embryonic stem cells, a stable and unlimited source, into MSCs (EMSCs). Then we subcutaneously transplanted human fat aspirates together with EMSCs or PBS as a control into the back of nude mice. Within 24 h of transplantation, EMSCs promoted aggregation and encapsulation of injected fat tissues. Afterward, all grafts gradually shrank. However, EMSC-containing grafts were larger, heavier and had fewer dark areas on the surface than the control grafts. Histologically, more live adipocytes, vascular cells, and macrophages and less fibrosis were observed in EMSC-containing grafts than in the controls. Some EMSCs differentiated into vascular cells and adipocytes in the EMSC-containing grafts. RNA sequencing revealed that human RNA was shown to decline rapidly, while mouse RNA increased in the grafts; further, human genes related to extracellular matrix remodeling, adipogenesis, and chemokine (including CCL2) signaling were expressed at higher levels in the EMSC-containing grafts than they were in the controls. CCL2 knockout reduced macrophage migration towards EMSCs in vitro and early macrophage recruitment to the grafts and the pro-engraftment effect of EMSCs in vivo. Treating mice with a macrophage inhibitor abolished the EMSC effects and converted the grafts to heavy masses of cell debris. Together, these data demonstrate that EMSCs promote fat engraftment via enhanced tissue reconstitution and encapsulation of implanted tissues, which was followed by increased angiogenesis and adipocyte survival and reduced fibrosis, in which stimulated CCL2 signaling and mobilized macrophages play pivotal roles.