MicroRNA-223 Promotes Type I Interferon Production in Antiviral Innate Immunity by Targeting Forkhead Box Protein O3 (FOXO3)*

Effective recognition of viral infection and subsequent triggering of antiviral innate immune responses are essential for the host antiviral defense, which is tightly regulated by multiple regulators, including microRNAs. Previous reports have shown that some microRNAs are induced during virus infection and participate in the regulation of the innate antiviral response. However, whether the type I IFN response is regulated by miR-223 is still unknown. Here, we reported that vesicular stomatitis virus (VSV) infection induced significant up-regulation of miR-223 in murine macrophages. We observed that miR-223 overexpression up-regulated type I IFN expression levels in VSV-infected macrophages. We also demonstrated that miR-223 directly targets FOXO3 to regulate the type I IFN production. Furthermore, type I IFN, which is triggered by VSV infection, is responsible for the up-regulation of miR-223, thus forming a positive regulatory loop for type I IFN production. Our results uncovered a novel mechanism of miR-223-mediated regulation of type I IFN production in the antiviral innate immunity for the first time.

Host innate immune response is the first line of defense following viral infection, which recognizes viral components and produces proinflammatory cytokines and interferons (1)(2)(3)(4)(5)(6)(7)(8). The IFN family of antiviral cytokines, especially type I IFN, plays pivotal roles in the host antiviral innate immune response. They are known to inhibit viral replication and mediate protection against viral infection (1)(2)(3)(4). Type I IFN production during virus infection should be tightly controlled by multiple intracellular regulators to initiate an appropriate immune response for eliminating the invading pathogens and preventing the development of immunopathological conditions; principal among these is the family of interferon regulatory factors (9 -12). Among the interferon regulatory factors, IRF3 and IRF7 are crucial in the host response to virus infection, inducing the transcription of the type I IFN genes, including the Ifn-␤ gene as well as a number of Ifn-␣ genes (13)(14)(15)(16). It has been reported that IRF7 is regulated at three different levels, including the transcriptional, post-transcriptional, and post-translational levels. Transcription of the Irf7 gene is induced by viral infection or type I IFN stimulation (17), although the stability of IRF7 is regulated by FOXO3 (18).
MicroRNAs (miRNAs) 4 are small, non-coding, regulatory RNAs that range from 18 to 22 nucleotides in length. They are mainly encoded by gene introns but are also sometimes encoded by dedicated genes. The miRNAs regulate the expression of specific target proteins by either inhibiting translation or degrading the corresponding mRNA (19,20). They regulate between one-and two-thirds of the human genome and participate in most of the cell's main functions (including growth, proliferation, differentiation, signal transduction, apoptosis, metabolism, and aging) (21). It has been reported that, with respect to virus infection, miRNAs constitute a critical additional control mechanism in the regulation of both cellular and viral gene expression. In fact, some miRNAs may target cellular genes involved in the virus replication control (22), and miRNAs targeting viral open reading frames may cause degradation of viral mRNAs to exhibit a direct antiviral effect (23). Previous reports have shown that some miRNAs, such as miR-146a and miR-155, are induced during virus infection and participate in the regulation of the innate antiviral response (24 -26). What's more, it has also been reported that miR-466l can directly target the 3Ј-UTR sequences of IFN-␣ mRNA species, thus inhibiting IFN-␣ production and enhancing viral replication (27). However, whether the type I IFN response is regulated by other miRNAs is still unknown. miR-223 is in the first cadre of miRNAs discovered to be highly expressed in myeloid cells of the bone marrow, where it is mainly expressed in the myeloid, granulocytic, and monocytic compartments, and miR-223 has a key role in the development and homeostasis of the immune system (28,29). To date, miR-223 has been demonstrated to be involved in many types of cancers, inflammatory diseases, autoimmune diseases, and other pathological processes (19). However, there are few reports about the function of miR-223 in the regulation of type I IFN signaling and subsequent antiviral innate immunity. Our initial observation of significant up-regulation of miR-223 in murine macrophages upon VSV infection prompted us to study * This work was supported by National Natural Science Foundation of China the potential effect of miR-223 in antiviral innate immunity. We observed that miR-223 overexpression up-regulated type I IFN expression levels in VSV-infected macrophages. We also demonstrated that miR-223 directly targeted FOXO3 to regulate the type I IFN production. Furthermore, type I IFN, which is triggered by VSV infection, is responsible for the up-regulation of miR-223, thus forming a positive regulatory loop for type I IFN production. Our results uncovered an essential mechanism of miR-223-mediated regulation of type I IFN production in the antiviral innate immunity.

Results
VSV-triggered Type I IFN Induces the Increase in miR-223 Expression in Macrophages-Previous reports have shown that some miRNAs, such as miR-146a and miR-155, are induced during virus infection and participate in the regulation of the innate antiviral response (24 -26). To identify other microRNAs that are potentially involved in the regulation of the innate antiviral response, we selected several miRNAs (miR-21, -25, -34a, -106a, -125b, -146a, and -223) that are enriched in myeloid cells, and we analyzed their expression in VSV-infected RAW264.7 cells and primary macrophages via qPCR (Fig. 1A). We observed that one of those miRNAs, miR-223, was significantly increased upon VSV infection. We also found that miR-146a was up-regulated by VSV challenge, which is consistent with a previous report (26). In RAW264.7 cells, miR-223 was significantly increased at 8 h post-infection and was rapidly up-regulated during the next 4 h, reaching the maximum at 12 h, and miR-223 expression was induced by VSV in a dose-dependent manner (Fig. 1B). In primary peritoneal macrophages, the up-regulation for miR-223 by VSV infection was also observed, but the induction extent was weaker than that in RAW264.7 cells (Fig. 1C). The up-regulation of miR-223 upon VSV infection suggests that miR-223 may function as a regulator of VSV-associated signaling events in macrophages. We next investigated the underlying mechanism by which miR-223 was induced. First, we observed that VSV infection increased VSV-G protein level in macrophages (Fig. 1D), so we considered whether viral RNA or viral protein was responsible for the up-regulation of miR-223. It was reported that suramin could selectively inhibit the RNA-dependent RNA polymerase of virus, viral RNA synthesis, and viral protein synthesis (30,31). So we treated the infected macrophages with suramin, and we found the viral RNA and viral protein were significantly down-regulated; however, miR-223 up-regulation was still induced along with type I IFN production (Fig. 1, E and F). Furthermore, we observed that heat-or UV-inactivated VSV and influenza viruses were no longer able to induce miR-223 up-regulation compared with infectious virus, due to the impaired type I IFN production ( Fig. 1, G-I), indicating that miR-223 expression might be regulated by virus-triggered type I IFN signaling. It was reported previously that miR-223 could be induced by type I IFN in macrophages (32), and we further investigated the effect of type I IFN on VSV-induced miR-223 expression. The results showed that miR-223 was significantly induced after rmIFN-␤ treatment in macrophages (Fig. 1, J and  K). On the contrary, after blocking type I IFN receptors with special antibody, the effect of VSV infection-induced miR-223 up-regulation was counteracted (Fig. 1, J and K). Taken together, these results suggest that VSV infection up-regulates miR-223 expression in macrophages mainly through type I IFN-dependent pathway.
miR-223 Positively Regulates VSV-triggered Type I IFN Production in Macrophages-To further identify whether VSV-induced miR-223 expression could affect VSV-triggered response in macrophages, we investigated the role of miR-223 in type I IFN production after VSV challenge. As shown in Fig. 2, A and B, transfection of miR-223 mimics increased miR-223 expression in macrophages, whereas miR-223 inhibitors decreased its expression level. miR-223 overexpression promoted VSV-triggered IFN-␤ production at both the mRNA and protein levels ( Fig. 2, C and D), whereas inhibition of miR-223 inhibited VSVtriggered IFN-␤ production (Fig. 2, C and D). Positive regulation of VSV-triggered IFN-␣4 mRNA production by miR-223 was also observed, which was similar to positive regulation of IFN-␤ by miR-223 (Fig. 2, C and D). However, we found no significant differences in mRNA or protein levels of TNF-␣ and IL-6 ( Fig. 2E). Taken together, these results demonstrate that miR-223 positively regulates VSV-triggered type I IFN production but does not significantly affect TNF-␣ and IL-6 production in macrophages.
miR-223 Feedback Attenuates the Viral Replication in Macrophages-To investigate the biological significance of upregulated miR-223 during viral infection, we further examined the effect of miR-223 on VSV replication in macrophages. By measuring intracellular VSV RNA replicates of the infected macrophages, we found that overexpression of miR-223 suppressed VSV replication, whereas inhibition of induced miR-223 facilitated VSV replication (Fig. 3A). Furthermore, we also observed that miR-223-induced viral resistance is not only against VSV infection but also the infection of influenza A H1N1 (Fig. 3B). Consistent with the results of intracellular VSV RNA replicates, VSV plaque assay and 50% tissue culture infectious dose (TCID 50 ) assay in cultural supernatants also showed that VSV replication was inhibited by miR-223 overexpression and increased by miR-223 inhibition (Fig. 3, C and D). To further verify the effect of miR-223 on VSV replication, we challenged macrophages with a VSV strain expressing GFP (VSV-GFP). The VSV genome was modified by inserting the GFP transgene into the viral genome, and thus the presence of GFP in cells indicates VSV infection and replication (33). We quantified VSV-GFP expression by the percentage of VSV-GFP-positive cells, which should reflect the degree of viral infection across the population. As shown in Fig. 3, E and F, overexpression of miR-223 reduced the degree of VSV-GFP infection, whereas inhibition of induced miR-223 increased the degree of VSV-GFP infection. Thus, we conclude that induced miR-223 expression, as a positive feedback, attenuates the viral replication in virus-infected macrophages.
miR-223 Enhances IRF7 Expression by Targeting FOXO3-We next investigated the effect of miR-223 on RNA virus-activated downstream signal pathways in macrophages. We found that miR-223 overexpression did not significantly increase phosphorylation of p38, ERK, JNK, and IB␣ after VSV challenge (Fig. 4A). Moreover, we did not observe a significant difference in the expression of RIG-I and phosphorylation of IRF3 (Fig.  4B). The phosphorylation of the dsRNA-dependent protein kinase, a primary regulator of antiviral activities (34), did not significantly increase after miR-223 overexpression (Fig. 4C). Interestingly, however, we observed that IRF7 was increased at both the protein and mRNA levels by miR-223 overexpression but decreased by miR-223 inhibition (Fig. 4

, D and E).
We then further investigated the possible targets of miR-223 that could modulate the expression of IRF7. Because it has been reported previously that the stability of IRF7 is regulated by FOXO3 (18), we next examined whether FOXO3 is a potential target of miR-223. This presumption was confirmed by the observation that FOXO3 was decreased at both the protein and mRNA levels by miR-223 overexpression but was increased by miR-223 inhibition (Fig. 4, F and G). Then we combined two algorithms, Miranda and PicTar, to search for the genes with putative miR-223 targeting sites in their 3Ј-UTRs, and we found that FOXO3 with one evolutionarily conserved seed region for miR-223 binding in the 3Ј-UTR was predicted to be a potential target (Fig. 4H). To certify the possibility that FOXO3 was regulated posttranscriptionally by miR-223, we constructed

miR-223 Regulates Antiviral Innate Immune Responses
reporter plasmids by cloning the mouse FOXO3 3Ј-UTR into the pMIR-REPORT TM luciferase vector. By cotransfection of the reporter plasmids and internal control pRL-TK-Renilla luciferase plasmids with miR-223 mimics in HEK293 cells, we observed that miR-223 mimics markedly decreased the luciferase level (Fig. 4I). Furthermore, transfection of miR-223 mimics decreased FOXO3 expression in macrophages at both the protein and mRNA levels, whereas miR-223 inhibition increased FOXO3 (Fig. 4J), suggesting that FOXO3 expression could be inhibited by miR-223 via both translational inhibition and mRNA degradation. Together, the results show that miR-223 enhances IRF7 expression by targeting FOXO3, and mouse FOXO3 is a new target of miR-223.
Antiviral Function of miR-223 Is Mainly through Targeting FOXO3-To demonstrate the role of targeting FOXO3 in the antiviral function of miR-223, we examined the effects of knockdown of FOXO3 on VSV-triggered type I IFN production. It has been shown previously that FOXO3 functions as a negative regulator of basal IRF7 transcription, thus suppressing the IRF7-dependent antiviral response (18). Consistent with this, we have observed that knockdown of FOXO3 promoted IRF7 expression and markedly increased VSV-triggered type I IFN production in primary mouse peritoneal macrophages (Fig. 5, A and B). These data suggest that positive regulation of VSV-triggered type I IFN production by miR-223 could be mediated by inhibition of its target FOXO3.
Then, to further demonstrate the role of targeting FOXO3 in the antiviral function of miR-223, we carried out experiments of FOXO3 knockdown and overexpression. In murine peritoneal macrophages, FOXO3 knockdown attenuated VSV replication, which resembled the effect of miR-223 overexpression (Fig.  5C). The VSV replication increased by miR-223 inhibition was rescued by FOXO3 knockdown (Fig. 5D). These results suggest that FOXO3 knockdown phenocopied the antiviral effect of induced miR-223 and counteracted the effect of miR-223 inhibition.
Furthermore, we confirmed that the FOXO3-overexpressed cell clone of RAW264.7 macrophages, which expressed FOXO3 mRNA without its 3Ј-UTR sequence, had significantly elevated FOXO3 expression at both the mRNA and protein levels (Fig.  5E). Because the FOXO3-overexpressed RAW264.7 cell clones transcribed FOXO3 mRNA without its 3Ј-UTR, miR-223 should no longer target FOXO3 expression in these cells. We also did not find that miR-223 overexpression could influence VSV proliferation in FOXO3-overexpressed RAW264.7 cells, as documented by measuring intracellular VSV RNA replicates (Fig. 5F). Therefore, we concluded that induced miR-223 upon viral infection exhibits its antiviral function mainly through suppressing endogenous FOXO3 expression and subsequently promoting type I IFN signaling.

Discussion
miR-223 is highly expressed in myeloid cells of the bone marrow (28). It is expressed specifically in cells of the granulocytic lineage and its expression changes during maturation, becoming incrementally higher as granulocytes mature (35). To date, miR-223 has been demonstrated to be involved in the regulation of a broad range of important cellular processes (including cell cycle regulation) and the invasiveness of many different cell types, hematopoietic differentiation, and immune cell function (19). Specifically, it was reported that miR-223 played a role in the pathogenesis of certain infectious diseases, such as influenza and chronic hepatitis B virus infection, and an antiviral effect for miR-223 was also observed in severe acute respiratory syndrome (SARS)-coronavirus-infected bronchoalveolar stem cells and in freshly isolated HIV-1-infected monocytes (36 -40). Furthermore, miR-223 was found to inhibit dengue virus 2 (DENV2) replication in human endothelium-like EAhy926 cells by targeting microtubule-destabilizing protein stathmin 1 (STMN1) (41). However, less is known about the detailed roles of miR-223 in the regulation of the innate antiviral response.
In this study, we report a positive regulatory loop for type I IFN production. First, we found that VSV infection could induce miR-223 up-regulation in a murine macrophage cell line and in murine primary macrophages, which was accompanied by a significant up-regulation of type I IFN. The most plausible hypothesis is that type I IFN can induce miR-223 up-regulation after VSV infection. Consistent with this, we observed that miR-223 was significantly induced after rmIFN-␤ treatment in macrophages. On the contrary, after blocking type I IFN receptors with special antibody, the effect of VSV infection-induced miR-223 up-regulation was counteracted. Second, we have proved that miR-223 positively regulates VSV-triggered type I IFN production and inhibits VSV replication. Third, we have demonstrated that miR-223 enhances IRF7 expression by targeting FOXO3 and then promotes the type I IFN production. Therefore, we present a new model in which type I IFN, miR-  JULY 8, 2016 • VOLUME 291 • NUMBER 28   JULY 8, 2016 • VOLUME 291 • NUMBER 28

JOURNAL OF BIOLOGICAL CHEMISTRY 14711
223, FOXO3, and IRF7 form a positive regulatory loop for type I IFN production (Fig. 5G).
FOXO3 is a member of Forkhead box transcription factor O subfamily. FOXO3 together with other members, such as FOXO1, FOXO4, etc., participates in cell function regulations related to cell cycle arrest, induction of apoptosis, and oxidative and cellular stress (42)(43)(44). In general, FOXO3 is known to inhibit cell cycle progression and to promote cell death, so it has been thought that FOXO3 can be an important target to inhibit cancer cell progression (44). FOXO3 has also been shown to  10. Total protein levels of FOXO3 and IRF7 in lysates were detected by immunoblot at the indicated times. Band intensity was calculated using ImageJ software, and ratios of FOXO3/GAPDH or IRF7/GAPDH were determined. F and G, FOXO3 mRNA (F) and IRF7 mRNA expression (G) were measured by qPCR and normalized to the expression of ␤-actin in each sample. H, alignment of miR223 and its target sites in 3Ј-UTR of FOXO3 is shown. I, HEK293 cells were cotransfected with pMIR-REPORT TM FOXO3-3Ј-UTR luciferase reporter plasmid or empty pMIR-REPORT TM and pTK-RL plasmid, together with miR-223 mimics. After 24 h, firefly luciferase activity was measured and normalized by Renilla luciferase activity. J, 0.5 ml of 2 ϫ 10 5 mouse peritoneal macrophages were transfected with miR-223 mimics or inhibitors; after 48 h, FOXO3 mRNA and protein levels were detected by qPCR and immunoblot. Data are the mean Ϯ S.D. (n ϭ 3) of one representative experiment. Similar results were obtained in three independent experiments. ***, p Ͻ 0.1; **, p Ͻ 0.01; *, p Ͻ 0.05; ns, not significant. FIGURE 5. Antiviral function of miR-223 is mainly through targeting FOXO3. A and B, murine peritoneal macrophages were transfected with control siRNA and FOXO3 siRNA as indicated. After 48 h, FOXO3, IRF7, IFN-␣4, and IFN-␤ mRNA expressions and FOXO3 protein levels were detected. C, murine peritoneal macrophages were transfected with control (Ctrl), miR-223 mimics, and FOXO3 siRNA, respectively, as indicated. After 48 h, macrophages were infected by VSV at m.o.i. 10 for 1 h and washed; intracellular VSV RNA replicates at indicated time points were quantified using qRT-PCR, and the value at 4-h time point in control cells served as 1-fold control. The relative IFN-␤ mRNA expression at indicated time points was detected by qRT-PCR, and IFN-␤ in supernatants was measured by ELISA. D, murine peritoneal macrophages were transfected with control inhibitor, miR-223 inhibitor, miR-223 inhibitors combined with control siRNA, and miR-223 inhibitors combined with FOXO3 siRNA, respectively, as indicated. After 48 h, macrophages were infected by VSV at m.o.i. 10 for 1 h and washed; intracellular VSV RNA replicates at indicated time points were quantified using qRT-PCR, and the value at 4 h in cells transfected with control inhibitors served as 1-fold control. E, RAW264.7 cells were transfected with FOXO3 overexpression plasmid. After 48 h, FOXO3 mRNA and protein levels were detected by qRT-PCR (left) and immunoblot (right). F, RAW264.7cells were transfected with control (Ctrl) mimics combined with control plasmid, miR-223 mimics combined with control plasmid, or control mimics combined with FOXO3 overexpression plasmid, and miR-223 mimics combined with FOXO3 overexpression plasmid. play a role in immunity and inflammation (45)(46)(47). For the involvement of FOXO3 in the regulatory loop, a previous report presented that FOXO3 inhibited IRF7 transcription and negatively regulated innate immune response (18). Consistent with this, we have observed that knockdown of FOXO3 promoted IRF7 expression and markedly increased VSV-triggered type I IFN production in primary mouse peritoneal macrophages. Furthermore, FOXO3 knockdown attenuated VSV replication, which resembled the effect of miR-223 overexpression. So, we concluded that induced up-regulation of miR-223 upon viral infection exhibited its antiviral function mainly through suppressing endogenous FOXO3 expression and subsequently promoting type I IFN signaling.

miR-223 Regulates Antiviral Innate Immune Responses
miRNAs have been thought to target multiple mRNAs, named targetomes, to regulate gene expression. A single miRNA might tune protein synthesis from thousands of genes by direct or indirect effects (48). In this study, mouse FOXO3 was proved to be a novel target of miR-223 and contributed to the feedback-positive regulatory function of miR-223 in the regulation of type I IFN production. It is probable that we are far from unveiling the last target of miR-223, and some of the potential targets might also regulate type I IFN production. This presumption may raise interesting future work to reveal the entire functions of miR-223 in innate immune response.
Cell Culture and Transfection-HEK293 cell line was obtained from American Type Culture Collection (Manassas, VA) and cultured as described previously (49). RAW264.7 cell line was obtained from American Type Culture Collection and cultured in DMEM containing 10% FBS. Murine primary macrophages were obtained and cultured as described previously (50). 1 ϫ 10 4 cells were seeded into 96-well plates and incubated overnight. JetPRIME transfection reagents (Polyplus-transfection, Illkirch, France) were used for the cotransfection of plasmids and RNAs, according to the manufac-turer's instructions. 0.5 ml of growth medium containing 2 ϫ 10 5 cells was seeded into each well of 24-well plates, incubated overnight, and then transfected with RNAs using INTERFERin (Polyplus-transfection), according to the manufacturer's instructions.
Detection of Cytokine Production-Into each well of 24-well plates, 0.5 ml of growth medium containing 2 ϫ 10 5 cells was seeded, incubated overnight, and then transfected as described above. After 48 h, the cells were infected with VSV for the indicated time periods. The concentrations of IFN-␤ in culture supernatants were measured with ELISA kits (BioLegend) according to the manufacturer's protocols. TNF-␣ and IL-6 in the supernatants were measured with ELISA kits (eBioscience) according to the manufacturer's protocols.
3Ј-UTR Luciferase Reporter Assays-The wild-type mouse FOXO3 3Ј-UTR luciferase reporter vectors were constructed by amplifying the mouse FOXO3 mRNA 3Ј-UTR and cloning it into the pMIR-REPORT TM luciferase vector. HEK293 cells described above were cotransfected with 80 ng of luciferase reporter plasmid, 40 ng of thymidine kinase promoter-Renilla luciferase reporter plasmid, and the indicated miRNA mimics or controls (final concentration, 30 nM). After 24 h, luciferase activities were measured using the Dual-Luciferase TM reporter assay system (Promega), according to the manufacturer's instructions.
Immunoblotting-The cells were washed twice with cold PBS and lysed with cell lysis buffer (Cell Signaling Technology) supplemented with protease inhibitor mixture (Beyotime). Protein concentrations of the cell lysis extracts were measured with BCA assay (Pierce) and equalized with the extraction reagent. Equal amounts of the extracts were loaded and subjected to SDS-polyacrylamide gel, transferred into PVDF membrane, and blotted as we described previously (52).
VSV Virus Production and Infection-Wild-type VSV (Indiana serotype) and VSV-GFP were propagated in Vero cells after infection at m.o.i. ϭ 0.01. 48 h later, supernatant was collected after centrifugation and then filtered through a 0.45-m filter (Millipore) and stored at Ϫ80°C. Virus infections were performed in the absence of serum for 1 h and then replaced with fully supplemented growth medium.
VSV Yield Qualification-Macrophages were transfected and infected by VSV as indicated. 0.1 ml of the cultural supernatants were serially diluted on the monolayer of Vero cells, which were obtained from ATCC, and 1 ϫ 10 4 cells were seeded into 96-well plates 1 day before measurement. The TCID 50 was measured after 3 days. Plaque assays were done on Vero cells in 12-well plates at 2 ϫ 10 5 cells per well for VSV. Supernatants from infected cells were serially diluted and infected on Vero cells for 1 h. The cells were then covered with growth medium containing 0.6% low-melting point agarose. Plaques stained with 0.5% crystal violet (m/v) in 20% ethanol (v/v) were counted after 16 h for VSV. For indicated experiments, cells were mock-infected or infected with VSV-GFP under similar conditions, and GFP expression was analyzed by FACSCalibur TM flow cytometry.
Virus Inactivation-For heat inactivation, the viruses were incubated for 30 min at 56°C. For UV inactivation, the viruses were placed in a 12-well plate on ice and exposed to UV light for 45 min.
Statistical Analysis-All experiments were repeated three times. Statistical significance was determined by Student's t test, with p Ͻ 0.05 considered to be statistically significant.
Author Contributions-Q. W., Y. L., and L. C. designed the research, L. C., Y. S., L. H., X. W., L. L., F. D., and Y. L. performed the experiments and analyzed the data. L. C., Y. L., and Q. W. wrote the paper.