Eomes and T-Bet Expression Are Required By Mature Primary Human NK Cells for Anti-Leukemia Responses In Vivo

Natural Killer (NK) cells are innate lymphoid cells that respond to hematologic cancers via cytotoxicity (perforin/granzyme and death receptors) and cytokine/chemokine production, yet the molecular determinants underlying their proliferation, function, and persistence are poorly understood. There are promising reports of pre-clinical and clinical NK cell responses to leukemia and lymphoma, which represent a nascent cellular therapy for these blood cancers. The T-box transcription factors (TFs) Eomes and T-bet are expressed by NK cells throughout their lifespan, and are required for development as evidenced by NK cell loss in Eomes and T-bet deficient mice. However, the roles of these TFs in mature human NK cell molecular programs and functions remain unclear.

We hypothesized Eomes and T-bet, which are the only T-box TFs expressed in NK cells, are critical regulators of NK cell homeostasis and functionality, and are necessary for proper mature NK cell responses. To address this, we utilized the CRISPR-Cas9 system to genetically delete both Eomes and T-bet in primary human NK cells isolated from healthy donors, and investigated their role beyond guiding NK cell development, specifically in the anti-leukemia response. Gene-editing of primary human NK cells has been technically challenging, thus most reports that modified NK cells were performed with cell lines, in vitro-differentiated, or highly expanded NK cells that likely do not reflect primary human NK cell biology.

Here, we introduced Cas9 mRNA and sgRNA targeting T-bet and Eomes by electroporation into unexpanded primary human NK cells isolated from healthy donors using the MaxCyte GT system. We observed highly efficient reductions of Eomes and T-bet protein expression, quantified by flow cytometry (p < 0.0001, Fig A-B) without viability differences between control (sgRNA targeting TRAC, an unexpressed locus in NK cells), and Eomes/T-bet double CRISPR-edited (DKO) cells after one week in vitro. To study Eomes and T-bet in NK cell anti-leukemia response, control or DKO primary human NK cells were engrafted into NSG mice, supported with human IL-15, and challenged with K562 leukemia cells. Utilizing bioluminescent imaging to visualize leukemia burden, we observed that NK cells lacking both TFs were unable to suppress leukemia growth in vivo.

To understand the mechanism responsible for impaired leukemia control, we investigated in vivo persistence and proliferation, cytotoxic effector molecule expression, as well as ex vivo degranulation and cytokine production of DKO NK cells compared to control NK cells. DKO or control human NK cells were transferred into NSG mice and supported with human IL-15. After 2-3 weeks, significantly fewer (<30%) DKO NK cells persisted compared to control NK cells: spleen (5-fold decrease, control 240e3±65e3 vs DKO 47e3±15e3 NK cells, p<0.01, Figure C), blood (6-fold decrease, p<0.01), and liver (4-fold decrease, p<0.05). Using intracellular flow cytometry, double T-bet/Eomes CRISPR-edited NK cells that lacked both Eomes and T-bet protein after in vivo transfer were identified. A proliferative defect was evident in flow-gated DKO (62±6% undivided), compared to unedited (WT) NK cells (4±2% undivided) assessed by CellTrace Violet dilution (Figure D). In addition, there were marked reductions in granzyme B and perforin protein (p<0.001) in flow-gated DKO NK cells compared to controls. To assess DKO NK cell functional capacity, we performed an ex vivo functional assay on NK cells from spleens of the NSG mice as effectors, and K562 targets or IL-12/15/18 stimulation for 6 hours. Degranulation to K562 targets was impaired (p<0.05), and IFN-γ production was reduced (p<0.0001) after cytokine stimulation in flow-gated DKO NK cells (Figure E).

Thus, CRISPR-editing of unexpanded, primary human NK cells revealed that Eomes and T-bet are required by mature human NK cells for their function and homeostasis, distinct from their role in development. This is translationally relevant, as defects in proliferation and function of human DKO NK cells manifested markedly reduced response against human leukemia cells in vivo in xenografts. These findings expand our understanding of key molecular regulators of mature NK cell homeostasis and function, with the potential to provide new avenues to enhance NK cell therapy.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal