Regulation of Human Natural Killer Cell IFN-γ Production by MicroRNA-146a via Targeting the NF-κB Signaling Pathway

Abstract

Natural killer (NK) cells are one group of innate lymphocytes that are important for host defense against malignancy and viruses. MicroRNAs (miRNAs) play a critical role in regulating responses of immune cells including NK cells. Accumulating evidence suggests that miR-146a is involved in the regulation of immune responses. However, the mechanism by which miR-146a regulates NK cell function is largely unknown. In the current study, we found that miR-146a intrinsically regulated NK cell function. Forced overexpression of miR-146a decreased IFN-γ production, whereas downregulation of miR-146a by anti-miR-146a significantly enhanced IFN-γ production in the human NK-92 cell line and primary human NK cells upon stimulation with IL-12 or co-stimulation with IL-12 and IL-18. Mechanistically, miR-146a regulated IFN-γ production via NF-κB, as evidenced in NK-92 cells, by downregulation of NF-κB p65 phosphorylation when miR-146a was overexpressed but upregulation of NF-κB p65 phosphorylation when anti-miR-146a was overexpressed. miR-146a directly targeted IRAK1 and TRAF6, the upstream signaling components of the NF-κB signaling pathway. This direct targeting mechanism confirmed the above gain- and loss-of-function approaches. However, the potent IFN-γ-producing subset, CD56^bright NK cells, expressed higher levels of miR-146a than the lesser IFN-γ-producing subset, CD56^dim NK cells. We also observed that co-stimulation of IL-12 and IL-18 significantly increased miR-146a expression in bulk NK cells and in the CD56^bright subset in a time-dependent manner, correlating with augmented IFN-γ production. These data suggest that miR-146a plays a negative role in IFN-γ production by human NK cells and this miRNA may be critical in preventing NK cells from being super activated and overproducing IFN-γ.

Keywords: IFN-γ; IRAK1; NF-κB; TRAF6; miR-146a; microRNA; natural killer cells.

Figure 1

miR-146a overexpression decreases IL-12 plus IL-18–induced IFN-γ production in NK cells. NK-92 cells were infected with the lentivirus containing GFP (miR-vector) or GFP and miR-146a. Cells were then sorted for high GFP expression and analyzed for miR-146a expression (A). NK-92 cells (B) or primary human CD56⁺ NK cells (C) were infected using either miR-vector or miR-146a, sorted by FACS for GFP, and co-stimulated with IL-12 (10 ng/mL) plus IL-18 (100 ng/mL) for 24 h. Cell pellets were collected for RNA preparation and cDNA synthesis, followed by assessment of IFN-γ gene expression by quantitative (q)RT-PCR. Supernatants were collected to measure IFN-γ protein levels by enzyme-linked immunosorbent assay (ELISA). The data shown are representative of at least four experiments with similar results. Data are reported as mean ± SD. *p < 0.05; **p < 0.01. Error bars represent SD.

Figure 2

miR-146a knockdown enhances IL-12 plus IL-18–induced IFN-γ production in NK cells. NK-92 cells were infected with an empty vector (anti-miR-vector) or a miR-146a antisense encoding vector (anti-miR-146a), sorted for GFP⁺ and analyzed for miR-146a expression (A). The NK-92 stable cell line with miR-146 overexpression or an empty vector was co-stimulated with IL-12 (10 ng/mL) plus IL-18 (100 ng/mL) for 24 h. Cell pellets were collected for RNA isolation and cDNA synthesis, followed by assessment of IFN-γ mRNA expression quantified by qRT-PCR. Supernatants were collected to measure IFN-γ by enzyme-linked immunosorbent assay (ELISA) (B). This experiment is representative of at least four such experiments performed with similar results. Data shown are mean ± SD. *p < 0.05; **, p < 0.01. Error bars represent SD.

Figure 3

NF-κB signaling is negatively regulated by miR-146a in resting and cytokine-stimulated human NK cells. NK-92 cells expressing miR-146a or anti-mIR-146a were cultured in medium without IL-2 for 24 h, followed by treatment with or without IL-12 (10 ng/mL) plus IL-18 (100 ng/mL) for an additional 24 h. Cells were harvested to stain with antibodies, followed by flow cytometric analysis. The NF-κB p65 protein levels and its phosphorylation were measured by flow cytometric analysis (A,B,D,E) (n = 3). The ratio of p-NF-κB to NF-κB was calculated based on mean fluorescence intensity (MFI) of each molecule (C,F) (n = 3). *p < 0.05; **p < 0.01. Error bars represent SD.

Figure 4

Expression of TRAF6 and IRAK1 in NK-92 cells transduced with miR-146a or anti-miR-146a. NK-92 cells expressing miR-146a or anti-miR-146a were cultured in medium without IL-2 for 24 h. Cells were harvested to extract RNA, followed by cDNA synthesis. The expression of IRAK1 (A,B,E,F) and TRAF6 (C,D,G,H) at the mRNA level and the protein level were assessed by qRT-PCR and immunoblotting, respectively (n = 3). Data shown are mean ± SD. *p < 0.05, **p < 0.01. Error bars represent SD.

Figure 5

IRAK1 and TRAF6 are direct targets of miR-146a. Schematic representation of the IRAK1 and TRAF6 3′ UTR indicating the putative binding sites of miR-146a (A). Sequence alignment of miR-146a and its target sites in 3′ UTRs of TRAF6 (B) and IRAK1 (C). WT and MT indicate wild-type and mutant sequences, respectively. In the IRAK1-UTR-MT plasmid, both miR-146-binding sites were mutated as shown. HeLa cells (D) and 293T cells (E) were transiently co-transfected with either pGL3 luciferase vector containing a fragment of IRAK1 and TRAF6 3′ UTR harboring miR-146a binding sites or the corresponding mutant constructs. Luciferase activities were normalized to the activity of renilla luciferase. Data shown are representative of at least three experiments with similar results and are presented as mean ± SD. *p < 0.05; **p < 0.01. Error bars represent SD.

Figure 6

miR-146a expression in resting and IL-12 and/or IL-18-stimulated natural killer (NK) cells. Human NK cells were unstimulated or stimulated with IL-12 (10 ng/mL), IL-18 (100 ng/mL), or the combination of both for 24 h. Supernatants were harvested to be analyzed for IFN-γ production by enzyme-linked immunosorbent assay (ELISA) (A). Cell pellets were collected for RNA preparation and cDNA synthesis, followed by assessment of miR-146a expression by qRT-PCR (B). Human NK cells were unstimulated or co-stimulated with the combination of IL-12 (10 ng/ml) and IL-18 (100 ng/ml) for the indicated times. Supernatants were harvested to be analyzed for IFN-γ production by ELISA (C). Cell pellets were collected and used to prepare RNA and cDNA synthesis, followed by assessment of miR-146a expression by qRT-PCR (D). Data reported are representative of at least four experiments with similar results and are shown as mean ± SD. *p < 0.05; **p < 0.01. Error bars represent SD.

Figure 7

miR-146a and NF-κB expressions in natural killer (NK) cell subsets CD56^bright and CD56^dim NK cells were sorted by a FACS Aria II cell sorter based on CD56 cell surface density, followed by assessment of NF-κB p65 mRNA and protein levels by qRT-PCR and immunoblotting, respectively (A). CD56^bright and CD56^dim NK cells were gated based on CD56 cell surface density and NF-κB p65 protein levels and phosphorylation levels were assessed by flow cytometric analysis (B). miR-146a expression was quantified by qRT-PCR in NK cell subsets (C). CD56^bright NK cells were left unstimulated or co-stimulated with a combination of IL-12 (10 ng/ml) and IL-18 (100 ng/ml) for the indicated times, after which the supernatants were analyzed for IFN-γ production by enzyme-linked immunosorbent assay. Cell pellets were collected and used to prepare RNA and miR-146a expression and were quantified by qRT-PCR (D). This experiment is representative of at least four such experiments performed with similar results. Data shown are mean ± SD. *p < 0.05; **p < 0.01. Error bars represent SD.