Molecular and anatomical roadmap of stroke pathology in immunodeficient mice

Stroke remains a leading cause of disability and death worldwide. It has become apparent that inflammation and immune mediators have a pre-dominant role in initial tissue damage and long-term recovery. Still, different immunosuppressed mouse models are necessary in stroke research e.g., to evaluate therapies using human cell grafts. Despite mounting evidence delineating the importance of inflammation in the stroke pathology, it is poorly described to what extent immune deficiency influences overall stroke outcome.

Here, we assessed the stroke pathology of popular genetic immunodeficient mouse models, i.e., NOD scid gamma (NSG) and recombination activating gene 2 (Rag2^-/-) mice as well as pharmacologically immunosuppressed mice and compared them to immune competent, wildtype (WT) C57BL/6J mice three weeks after injury. We performed histology, gene expression, blood serum and behavioural analysis to identify the impact of immunosuppression on stroke progression.

We detected changes in microglia activation/macrophage infiltration, scar-forming and vascular repair in immune-suppressed mice three weeks after injury. Transcriptomic analysis of stroked tissue revealed the strongest deviation from WT was observed in NSG mice affecting immunological and angiogenic pathways. Pharmacological immunosuppression resulted in the least variation in gene expression compared with the WT. These anatomical and genetic changes did not affect functional recovery in a time course of three weeks. To determine whether timing of immunosuppression is critical, we compared mice with acute and delayed pharmacological immunosuppression after stroke. Mice with delayed immunosuppression (7d) showed increased inflammatory and scarring responses compared to animals acutely treated with tacrolimus, thus more closely resembling WT pathology. Transplantation of human cells in the brains of immunosuppressed mice led to prolonged cell survival in all immunosuppressed mouse models, which was most consistent in NSG and Rag2^-/- mice.

We detected distinct anatomical and molecular changes in the stroke pathology between individual immunosuppressed mouse models that should be considered when selecting an appropriate mouse model for stroke research.