Manganese boosts natural killer cell function via cGAS-STING mediated UTX expression

Abstract

Natural killer (NK) cells play a crucial role in both innate immunity and the activation of adaptive immunity. The activating effect of Mn²⁺ on cyclic GMP-AMP(cGAS)-stimulator of interferon genes (STING signaling has been well known, but its effect on NK cells remains elusive. In this study, we identified the vital role of manganese (Mn²⁺) in NK cell activation. Mn²⁺ directly boosts cytotoxicity of NK cells and promotes the cytokine secretion by NK cells, thereby activating CD8+ T cells and enhancing their antitumor activity. Furthermore, Mn²⁺ can simultaneously activate NK-cell intrinsic cGAS and STING and consequently augment the expression of ubiquitously transcribed tetratricopeptide repeat on chromosome X (UTX to promote the responsiveness of NK cells. Our results contribute to a broader comprehension of how cGAS-STING regulates NK cells. As a potent agonist of cGAS-STING, Mn²⁺ provides a promising option for NK cell-based immunotherapy of cancers.

Keywords: UTX; antitumor immunity; cGAS–STING; manganese; natural killer cells.

FIGURE 1

NK cells mediate the antitumor effects of Mn²⁺. (A) 2 × 10⁵ B16F10 cells were inoculated subcutaneously to the wild‐type mice with or without 5 mg/kg MnCl₂ intraperitoneal treatment. (B) Representative images (left) and tumor growth curve (right) of B16F10 tumor in control and Mn²⁺‐treated WT mice (n = 5 per group). (C) Representative FACS data and statistics of frequency of tumor infiltrating NK cells (left) or ILCs (right) of mice as in (B) (n = 5 per group). (C) Representative FACS data and quantification of the frequency of tumor infiltrating CD107a+ NK (left) or IFNγ+ NK cells of mice as in (B) (n = 5 per group). (E) 1 × 10⁵ B16F10 cells were implanted subcutaneously to the Rag^−/− mice with or without intraperitoneal injection of 5 mg/kg MnCl₂. (F) Images of tumors (left) and quantification of tumor sizes (right) in B16F10 tumor in Rag^−/− mice treated with or without MnCl₂ (n = 5 per group). (G) As mice in (F), the frequency of tumor infiltrating NK cells (left), and IFNγ+, granzyme B+, and perforin+ NK cells (right) were quantified by flow cytometry (n = 5 per group). (H) NK cells were depleted from WT mice by delivering anti‐NK1.1 antibody every 3 days. 2 × 10⁵ B16F10 cells were inoculated to the NK cell‐depleted mice with or without 5 mg/kg MnCl₂ intraperitoneal administration. (I) Representative images (left) and tumor growth curve (right) of B16F10 tumor in control and Mn²⁺‐treated WT or NK cell‐depleted mice (n = 5 per group). (J and K) Frequency of tumor infiltrating NK cells (upper) and expression of CD44 on CD8+ TILs (lower) of mice as in (I) were quantified by flow cytometry (n = 5 per group). Data represent analyses of the indicated n mice per group, means ± SEM. ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001; ns, not significant, p > 0.05.

FIGURE 2

Mn²⁺ enhances the activation of NK cells. (A–D) Primary murine and human NK cells were isolated from spleens of mice and PBMCs of humans, respectively, and then preactivated by IL‐2 and IL‐15. Preactivated murine NK cells were incubated with or without Mn²⁺ for 24 h in vitro. Apoptosis (A and C) and proliferation (B and D) of treated NK cells were evaluated by flow cytometry (left) and subjected to statistical analysis (right) (n = 5 per group). (C) As murine NK cells cultured in (A), the effect factors of murine NK cells, namely CD107a, IFNγ, granzyme B, and perforin, were assessed by flow cytometry (left) and performed statistical analysis (right) (n = 5 per group). (D) As ex vivo human NK cells in (B), Frequency of CD107a+, IFNγ+, granzyme B+, and perforin+ human NK cells were analyzed by flow cytometry (n = 5 per group). Data represent analyses of the indicated n mice per group, means ± SEM. ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001; ns, not significant, p > 0.05.

FIGURE 3

Mn²⁺ directly improves the cytotoxicity of NK cells against tumors. (A–D) NK92‐MI cell line or preactivated primary human NK cells were cocultured with target cell K562 at various effector–target ratios, in the presence or absence of Mn²⁺. The killing ability of NK cells against K562 cells were evaluated by flow cytometry (A and C) ( n  = 5 per group). The lysis frequencies of K562 cells were statistically analyzed (B and D) (n = 5 per group). (E–H) Primary murine NK cells were preactivated and coculture with target lymphoma cell TAP2‐deficient RMA‐S or wild‐type RMA cells with or without Mn²⁺. Representative FACS data (E and G) and statistics (F and G) of frequency represent the tumor killing ability of murine NK cells (n = 5 per group). Data represent analyses of the indicated n mice per group, means ± SEM. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001; ns, not significant, p > 0.05.

FIGURE 4

Mn²⁺ boosts the activation of CD8+ T cells by functioning on NK cells. (A–E) The B16F10 tumor models were established, as shown in Figure 1(A). The tumors were collected after three injections of Mn²⁺ for single‐cell sequencing analysis (n = 3 per group pool). The composition ratio (A), abundance difference (B), and quantity (C) of various tumor infiltrating immune cells were tested. KEGG (D) and GO (E) analysis were performed in NK cell cluster. (F) Primary murine NK cells were isolated from mice spleen and treated with MnCl₂ for 24 h. The supernatant was collected for cytokine analysis by LEGENDplex immunoassay (n = 3 per group). (G–I) Mice CD8+ T cells were isolated from mice PBMCs and incubated with preactivated murine NK cells or supernatant (conditioned medium) (G). Representative FACS data (H) and statistics (I) of frequency of effector factors expressed by CD8+ T cells (n = 5 per group). (J–L) OT‐1 CD8+ T cells were incubated with preactivated murine NK cells, and then cocultured with OVA‐expression target cells (J). Representative FACS data (K) and quantification (L) of cytotoxicity of OT‐1 CD8+ T cells against target cell (n = 5 per group). Data represent analyses of the indicated n mice per group, means ± SEM. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001; ns, not significant, p > 0.05.

FIGURE 5

Mn²⁺ activates NK cells through NK cell‐intrinsic cGAS–STING. (A) Western blot of cGAS–STING signaling molecules in primary murine NK cells with or without Mn²⁺ treatment (n = 3 per group). (B) qRT‐PCR of IFNb1, the major effector indicator of cGAS–STING, in murine NK cells, in the presence or absence of Mn²⁺ (n = 5 per group). (C and D) Apoptosis (C) and proliferation (D) analysis of STING^−/− murine NK cells after Mn²⁺ treatment in vitro (n = 5 per group). (E) NK cells were isolated from STING^−/− mice and cultured as demonstrated in Figure 2A. Representative FACS data (left) and quantification (right) of effector factors of NK cells CD107a, IFNγ, granzyme B, and perforin (n = 5 per group). (F) Immunoblot analysis of cGAS–STING pathway in primary murine NK cells inhibited cGAS with RU.521 (cGAS inhibitor) or STING with H‐151 (STING inhibitor) during Mn²⁺ treatment in vitro. (G and H) Flow cytometry data of Ki67+ (G) and IFNγ+ (H) murine NK cells treated with Mn²⁺ and inhibitors of cGAS or STING in vitro (n = 5 per group). Data represent analyses of the indicated n mice per group, means ± SEM. ^*** p < 0.001; ^**** p < 0.0001; ns, not significant, p > 0.05.

FIGURE 6

Mn²⁺ induces UTX expression depending on STING to stimulate NK cell activation. Spearman correlation between expression of KDM6A (UTX) and TMEM173 (STING, left) or MB21D1 (cGAS, right) employing the aforementioned scRNA‐seq data. (B) Correlation plot of KDM6A expression and abundance of tumor infiltrating NK cells in tumors with publicly available data from TCGA database. (C) Image of immunoblot (left) and quantification of expression of UTX (right) in primary NK cell from WT and STING^−/− mice, in the presence or absence of Mn²⁺ (n = 3 per group). (D) qRT‐PCR of KDMA6 in murine WT and STING^−/− NK cells with or without Mn²⁺ treatment (n = 5 per group). (E) Western blot of cGAS–STING pathway in primary murine NK cell with overexpression of UTX and treatment with inhibitors of cGAS or STING. (F) Immunoblot analysis of cGAS–STING pathway in primary murine NK cell with knockdown of UTX and treated with agonists of cGAS (G3‐YSD) or STING (2′3′‐cGAMP). (G and H) Flow cytometry data of CD107a+ (left) and IFNγ+ (right) NK cells with above mentioned (E and F) treatment (n = 5 per group). Data represent analyses of the indicated n mice per group, means ± SEM. ^* p < 0.05; ^*** p < 0.001; ^**** p < 0.0001; ns, not significant, p > 0.05.