Background Natural killer (NK) cells hold potential as one of the next generation adoptive cell therapy candidates due to their allogenecity, capability to lyse target without prior sensitization, and low risk of host-versus-graft disease. Thus, being able to utilize NK cell receptors to drive their cytotoxicity remains crucial. Additionally, for treatments of solid tumors, NK cells need to retain their viability against the suppressive tumor microenvironment. One method to modulate NK cell immunometabolism is to leverage the surface receptor TIM-3, which lies at the intersection of NK cytotoxic functions and metabolism – the mTOR-associated pathways.

Methods NK cells were isolated from healthy adult peripheral blood, and expanded in K562-based feeder media. Flow cytometry was used to phenotype TIM-3 expression and phosphorylation of mTOR, Akt, and ribosomal protein S6 (rpS6). CCK-8 assay was used to assess mTOR inhibition against patient-derived glioblastoma cells (GBM43). NK co-operative activities with mTOR inhibitors against GBM43 were evaluated through target lysis, degranulation, and cytokine secretion. TIM-3 knock-out (TIM-3 KO) NK cells were generated by electroporating CRISPR/Cas9 ribonucleopreotein complexes.

Results We first explored the link between TIM-3 expression on NK cells and mTOR proteins, including mTOR itself, rpS6, and Akt, which were linked to TIM-3 regulation in T cells. Previously, we had reported that TIM-3 downregulation was associated with lower IFN-γ production. As changes in phosphorylation of rpS6 was linked to lower IFN-γ mRNA levels, we hypothesized that TIM-3 expression leads to increased rpS6 phosphorylation, thus encouraging higher IFN-γ production. Here, we report changes in phosphorylation levels of Akt and rpS6 as consequences of TIM-3 KO, and how these changes can affect NK cell cytotoxicity against GBM43. We also cross-examine these findings using specific pharmacological inhibitors of mTOR, including rapamycin, JR-AB2-011, and torin-1, against mTORC1, mTORC2, and both complexes, respectively, to demonstrate that TIM-3 expression could provide some functional protection against activities these inhibitors through retention of NK cell activity. Lastly, to show that these inhibitions would not compromise NK cell functions, we show GBM43 lysis by NK cells with mTOR inhibitor adjuvants.

Conclusions We have found that TIM-3 expression can exert some control over the mTOR pathway on ex vivo NK cells, and this affects NK cells’ cytokine production capacity. Since the tumor microenvironment is highly immunosuppressive, the protection against mTOR inhibition by TIM-3 allowed tumor-infiltrating NK cells to remain metabolically and functionally viable, and affords an opportunity to leverage mTOR inhibitors as adjuvants with NK cell therapy against GBM.

Ethics Approval IRB #1804020540