Reevaluation of NOD/SCID Mice as NK Cell-Deficient Models

Objective. Natural killer (NK) cell-deficient mice are useful models in biomedical research. NOD/SCID mice have been used as a model of this type in research. However, the actual status of NK cells in NOD/SCID mice and CB17/SCID mice in comparison with that in BALB/c mice has not been sufficiently evaluated. Methods. Splenocytes from naive or poly(I:C)-treated mice were isolated for phenotyping and analysis of cytotoxicity-related molecules and inhibitory receptors; for cytotoxicity assay, purified NK cells were also used. Results. The proportion of splenic NK cells did not differ significantly between NOD/SCID and CB17/SCID mice. The perforin levels in NK cells were similar between the poly(I:C)-treated CB17/SCID and NOD/SCID mice, while the granzyme B and NKG2A/C/E levels in NK cells from NOD/SCID mice were significantly lower than those from CB17/SCID mice. Moreover, the NKG2D and Ly49A levels in NK cells from NOD/SCID mice were higher than those from CB17/SCID. The splenocytes from CB17/SCID mice showed higher cytotoxicity than those from NOD/SCID mice, while the cytotoxicity of purified NK cells basically did not differ between the two strains. After in vitro stimulation with cytokines, the splenocytes from CB17/SCID mice showed higher IFN-gamma production than those from NOD/SCID mice; however, NK cells did not. Conclusion. There was no significant difference in the proportion of splenic NK cells between CB17/SCID and NOD/SCID mice, and the function of NK cells was only partially compromised in NOD/SCID mice. Caution should be taken when considering the use of NOD/SCID mice as an NK-deficient model.