TIMP1 and TIMP2 Downregulate TGFβ Induced Decidual-like Phenotype in Natural Killer Cells

SIMPLE SUMMARY: Cancer patients are characterized by NK cells with altered surface markers, such as CD56 brightness, CD9, CD49a (pro-angiogenic) and PD-1, and TIM-3 (exhaustion), that favor immune escape. Transforming growth factor-beta (TGFβ) is a major tumor-derived cytokine that favors cancer growth and supports pro-angiogenic activities in NK cells by inducing pro-angiogenic molecules. TIMP-1 and TIMP-2 play a crucial role in extracellular matrix (ECM) regulation, wound healing, pregnancy and cancer, and there is increasing evidence that they are immune-modulatory. We found that recombinant TIMP-1 and -2 can partially contrast the induction of pro-tumor/pro-angiogenic decidual-like polarization of NK cells by TGFβ. ABSTRACT: Natural Killer (NK) cells have been found to be anergic, exhausted and pro-angiogenic in cancers. NK cell from healthy donors, exposed to TGFβ, acquire the CD56(bright)CD9(+)CD49a(+) decidual-like-phenotype, together with decreased levels of NKG2D activation marker, increased levels of TIM-3 exhaustion marker, similar to cancer-associated NK cells. Tissue inhibitors of metalloproteases (TIMPs) exert dual roles in cancer. The role of TIMPs in modulating immune cells is a very novel concept, and the present is the first report studying their ability to contrast TGFβ action on NK cells. Here, we investigated the effects of TIMP1 and TIMP2 recombinant proteins in hindering decidual-like markers in NK cells, generated by polarizing cytolytic NK cells with TGFβ. The effects of TIMP1 or TIMP2 on NK cell surface antigens were determined by multicolor flow cytometry. We found that TIMP1 and TIMP2 were effective in interfering with TGFβ induced NK cell polarization towards a decidual-like-phenotype. TIMP1 and TIMP2 counteracted the effect of TGFβ in increasing the percentage of CD56(bright,) CD16(−), CD9(+) and CD49a(+), and restoring normal levels for TIMP 1 and 2 also inhibited decrease levels of the activation marker NKG2D induced by TGFβ and decreased the TGFβ upregulated exhaustion marker TIM-3. NK cell degranulation capabilities against K562 cells were also decreased by TGFβ and not by TIMP1 or TIMP2. TIMP1 treatment could partially restore degranulation marker CD107a expression. Treatment with recombinant TIMP-1 or TIMP-2 showed a trend, although not statistically significant, to decrease CD49a(+) and TIM-3+ expression and increase NKG2D in peripheral blood NK cells exposed to conditioned media from colon cancer cell lines. Our results suggest a potential role of TIMPs in controlling the tumor-associated cytokine TGFβ-induced NK cell polarization. Given the heterogeneity of released factors within the TME, it is clear that TGFβ stimulation represents a model to prove TIMP’s new properties, but it cannot be envisaged as a soloist NK cell polarizing agent. Therefore, further studies from the scientific community will help defining TIMPs immunomodulatory activities of NK cells in cancer, and their possible future diagnostic–therapeutic roles.