An essential role of PI3K in the control of West Nile virus infection

The phosphatidyl-inositol-3 kinases (PI3K) pathway regulates a variety of cellular processes, including cell proliferation, RNA processing, protein translation, autophagy, apoptosis and antiviral immunity. Many viruses depend on PI3K signaling for replication. However, its role in flaviviral infection has not been clearly defined. Here we report that PI3K signaling is critical for the control of West Nile virus (WNV) infection by regulating type I IFN (IFN-I) response. Inhibition of PI3K activity by 3-methyl adenine (3-MA), Wortmannin (WM) and LY294002 (LY) increased viral titers by 3–16 folds in primary mouse macrophages, embryonic fibroblasts and human cell lines. Both 3-MA and LY repressed IFN-I mRNA and protein expression significantly. Surprisingly, WM enhanced the mRNA expression of IFN-I and TNF-α, and TNF-α protein production modestly, while dramatically decreased the secreted IFN-I. Further studies showed that the catalytic subunit p110δ of class I PI3K played a role in induction of antiviral immune responses. Lastly translocation of interferon regulatory factor 7(IRF7) from the cytosol to the nuclei was effectively blocked in the presence of PI3K inhibitors. Our results clearly define an antiviral role of PI3K by modulating immune responses and demonstrate differential mode of action of three PI3K inhibitors on IFN-I.

WNV replication, and simultaneously repressed IFN-I protein production. Furthermore, we demonstrated that the inhibitor of the catalytic subunit p110δ of class I PI3K decreased IFN-I significantly. Lastly translocation of interferon regulatory factor 7(IRF7) from the cytosol to the nuclei was effectively blocked in the presence of PI3K inhibitors. Our results show that PI3K signaling is critical for the control of WNV infection by modulating IFN-I response.

PI3K signaling is critical for the control of West Nile virus infection by modulating type I IFNs response.
On one hand, PI3K/Akt signaling is critical for many viruses to complete their life cycle, including cellular entry, replication and egress 16 . On the other hand, the PI3K/Akt pathway also induces antiviral immunity, particularly type I interferon (IFN-I) response 10,11 . However, the exact role (favorable or unfavorable for virus infection) of the PI3K/Akt pathway may be virus-specific. We tested West Nile virus (WNV), a reemerging flavivirus of significant public health concern in the United States. Treatment of cells with a PI3K inhibitor, 3-methyladenine (3-MA), enhanced WNV titers by 3-16 folds in a dose dependent manner (Fig. 1). Two other pan-PI3K inhibitors, Wortmannin (WM) and LY294002 (LY) also increased viral infection by more than 4 times. These data clearly demonstrate that the PI3K/Akt pathway is essential for prevention of WNV infection. We next asked how PI3K controls WNV infection. PI3K may prevent viral infection by inducing cell-intrinsic antiviral mechanisms such as IFN-I response 18 and autophagy 16 . Indeed, both secreted TNF-α and IFN-β protein concentrations in the cell culture medium decreased dramatically after 3-MA treatment in a dose dependent manner ( Fig. 2A). At the mRNA level, Ifna was most dramatically repressed by 3-MA treatment (4 and10 times at 1 mM and 2 mM respectively); Ifnb1 was down-regulated modestly by 2-3 times. However, Tnfa, Tlr3, Ddx58 and Dhx58 mRNA expression was not significantly influenced by 3-MA at any concentrations (Fig. 2B). In the case of Wortmannin treatment, IFN-β protein concentration was reduced by 4 times (Fig. 2C), but Ifnb1, Ifna and Tlr3 mRNA expression was up-regulated slightly; Ddx58 and Dhx58 remained unchanged (Fig. 2D). Surprisingly, TNF-α protein abundance and mRNA expression was increased in the presence of WM in a dose dependent manner (Fig. 2C,D). Both TNF-α and IFN-β protein concentrations were also reduced significantly by LY294002 treatment, while IFN-β mRNA level was modestly downregulated (Fig. 2E,F). To see if the anti-WNV function of PI3K is universal or cell type-specific, we examined WNV infection with/without 3-MA treatment in primary mouse embryonic fibroblasts (MEFs) and two human cell lines, H1975 (lung epithelial cells) and THP-1 (macrophage-like monocytes). IFN-β mRNA induction by WNV was dramatically repressed in the presence of 3-MA and viral replication was enhanced at 12 and 24 h after infection in all the cell types ( Fig. 3A-C). The Ddx58 mRNA level was 5-fold lower in 3-MA than mock-treated MEFs at 12 h after infection (Fig. 3A). Together, these results suggest that PI3K can inhibit WNV replication through modulating IFN-I response.
PI3K regulates IRF7 nuclear translocation. The family of PI3Ks is grouped into 4 classes, with class I being extensively studied 16 . The class I PI3K catalytic p110 subunit has 4 isoforms: α, β, γ and δ. To determine which isoform was important for IFN-Is induction after WNV infection, we examined the mRNA levels of all the four p110 subunits. Accompanied by increasing viral replication and immune response (Infb1 and Tnfa), p110α and p110β mRNA expression increased by 2-3 fold between 2-6 h after infection; while p110γ and p110δ remained unchanged during the course of infection (Fig. 4A). However, only p110δ inhibitor enhanced WNV replication by 4-fold and decreased TNF-α by 1.56 and IFN-β production by 1.8 fold at 1 µM, to a lesser extent than the pan-PI3K inhibitor LY290042 though ( Fig. 4B-D). To validate the pharmacological results, we next used siRNA to knock down p110δ expression. We used two unique siRNA against p110β, p110δ and Vps34, all of which achieved >60% silencing efficiency (Fig. 5A). The p110β and Vps34 siRNA did not influence p110δ expression, suggesting specificity of the silencing effect (Fig. 5B). Consistent with the pharmacological inhibition (Fig. 4), p110δ knockdown enhanced viral replication by ~6 fold (Fig. 5C) and simultaneously decreased IFN-β production by 4-11 times (Fig. 5D), compared to the negative (scrambled siRNA) controls. These results suggest that class I p110δ is the dominant catalytic subunit for IFN-I induction by WNV. To further investigate the mechanism of interferon induction by PI3K, we examined three transcription factors critical for IFN-I expression, interferon regulatory factors (IRF) 3,−7 and NF-κB. We noted that WNV infection increased the protein amounts of IRF3, −7 and NF-κB in the nuclear fraction when compared to non-infected nuclear extracts. The amounts of nuclear IRF7 were reduced by 62% in the presence of LY294002, while IRF3 and NF-kB p65 was modestly decreased by 17% (Fig. 6A,B). We confirmed IRF7 nuclear translocation deficiency by immunofluorescence microscopy: the intensity of nuclear IRF7 was lower in LY294002-treated WNV-infected cells than DMSO-treated WNV-infected cells (Fig. 6C).

Discussion
The PI3K/Akt pathway regulates a variety of cellular processes, including cell proliferation, RNA processing, protein translation, autophagy and apoptosis 6 . In order to replicate in host cells, viruses develop strategies to stimulate basal PI3K activity 16 . As an adaptive strategy, PI3K may activate IFN-I response and autophagy to counteract viral invasion. However, autophagy may be hijacked by viruses for their efficient replication, likely in a cell-type and virus specific manner 20 . Autophagy is induced by WNV infection in many cell types including fibroblasts 21 , neurons 22 and macrophages (our unpublished data); its role in WNV replication however is still controversial. One study showed that autophagy inhibited WNV replication in mouse embryonic fibroblasts infected with a low multiplicity of infection (MOI = 0.01) 21 . While other studies demonstrated that autophagy played no role in WNV replication in cell lines derived from kidney, liver, skin and 23 , and in IFN-I-deficient Vero/BHK and primary neuron cultures 22 . Inhibition of PI3K with 3-MA repressed WNV replication in primary neurons 22 , suggesting PI3K signaling favors viral replication. In contrast to this report, our study shows that PI3K is critical for control of WNV replication and type I IFN response in primary macrophage, an important immune cell type that restricts WNV infection in vivo 24 and several other cell types. Macrophages may control WNV infection by direct effector mechanisms such as the production of nitric oxide intermediates 25 or production of IFN-I 26 . Our study indicates that PI3K positively regulates IFN-β protein expression.
Furthermore, we found that the modes of action of three pan-PI3K inhibitors on type I IFNs are different. 3-MA and LY294002 interfere with IFN-I expression at both the transcriptional and translational level. Interestingly, although 3-MA reduces the TNF-α protein concentration in the culture media dramatically, it has no effect on TNF-α mRNA expression. A possible explanation may be that 3-MA interferes with TNF-α translation or secretion. In macrophages, TNF-α is transported to the cell surface where it is cleaved by TNF converting enzyme and released into extracellular milieu via constitutive exocytosis and recycling endosomes [27][28][29] involving PI3Ks. Wortmannin does not influence type I IFN transcription significantly, but reduces protein abundance in the culture media. It is plausible that Wortmannin preferentially targets PI3Ks that are involved in IFN-I secretion, which underlying mechanism is currently unknown. Wortmannin may also inhibit other kinases such as myosin light chain kinase (MLCK) and phosphatidylinositol 4-OH kinase (PI4K) that potentially play a role in exocytosis 30,31 . Surprisingly Wortmannin enhances both TNF-α protein and mRNA expression, which may be   a secondary result of increased viral replication. Our results suggest that Wortmannin selectively interferes with IFN-I response and is a helpful tool for study of the molecular pathway governing type I IFN secretion.
Lastly, we identify the catalytic isoform of class I PI3K required for IFN-I induction. Inhibition of p110δ, but not other isoforms down-regulates type I IFN expression. However, the class I p110δ inhibitor is less potent than the pan-PI3K inhibitors to repress WNV-induced IFN-I expression, suggesting that other classes of PI3Ks also play a role. Consistent with our findings, it was recently shown that p110δ is essential for IFN-α production by plasmacytoid dendritic cells (pDCs) in response to TLR stimulation 32 .
In summary, our results clearly define an antiviral role for PI3K by modulating IFN-I immune responses and demonstrate differential mode of action of three PI3K inhibitors on IFN-I. Future work is to understand the molecular pathways governing IFN-I secretion using Wortmannin.

Materials and Methods
Cell culture and viral infection. All animal protocols were approved by the Institutional Animal Care & Use Committee at New York Medical College adhering to the National institutes of Health recommendations for the care and use of laboratory animals. Bone marrow derived macrophages (BMDMs) were differentiated as described previously 33 . Briefly, Bone marrows were isolated from the tibia and femur bones of 6 weeks old female C6/57BL (The Jackson Laboratory, # 000664. https://www.jax.org/strain/000664) and then differentiated into macrophages (BMDMs) in L929 (ATCC, #CCL-1) conditioned medium (RPMI1640, 20%FBS, 30% L929 culture medium, 1x antibiotics/antimycotics) in 10-cm Petri-dishes at 37 °C for 5 days. The attached BMDMs were dislodged by pipetting and counted for plating. BMDMs were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium, Vero cells in DMEM, containing 10% (volume/volume) fetal bovine serum (FBS, Invitrogen, Carlsbad, CA), 100U/ml penicillin and 100 μg/ml streptomycin (Invitrogen) and maintained at 37 °C and 5% CO 2 . The WNV strain was a CT2741 isolate of NY1999 34 and propagated in C6/36 cells to generate a stock of ~1 × 10 7 PFU/ml 35 . After differentiation, BMDMs were cultured in RPMI medium overnight, washed once with prewarmed fresh medium and then infected with WNV at a multiplicity of infection (MOI) of 10. After 1 hr at 37 °C, the inoculum was replaced with fresh culture medium and cells were incubated at 37 °C. Cells and culture supernatants were collected at the indicated time points for further analyses.
For preparation of mouse embryonic fibroblasts (MEFs), day 14 embryos were decapitated and eviscerated then digested with trypsin for 10 min at 37 °C rotating. Fibroblasts were filtered through a 100 μM filter, cultured in RPMI1640 medium (Life Technologies, NY 14072 USA) supplemented with 10% fetal bovine serum and antibiotics/antimycotics, propagated for 2 passages and then frozen. MEFs were infected with WNV at a MOI of 1.
H1975 (human lung epithelial cell, Cat# CR5908) and THP-1 (human monocytes, Cat# TIB202) were purchased from American Type Culture Collection (ATCC) (Manassas, VA20110, USA). These cells were cultured in RPMI1640 medium and infected with WNV at a MOI of 1 and 5 respectively.  Plaque forming assay. The detailed procedures of plaque forming assay were described previously 36 . Briefly, serially diluted culture supernatants from infected cells were inoculated into monolayers of type I IFN deficient Vero cells (ATCC # CCL-81). After 1 h of incubation at 37 °C, the inoculums were removed, and overlay medium (DMEM containing 10% FBS and 1% Seaplaque agarose) was added and incubated at 37 °C for 4 days. Plaques were visualized after Neutral Red (Sigma-Aldrich) staining on a white transilluminator.

ELISA.
Commercial enzyme-linked immunosorbent assay (ELISA) kit were used to measure levels of secreted TNF-α (Cat # 88-7324, Affimetrix, San Diego, CA 92121USA) and INF-β (Cat# 42400, PBL Assay Sciences, Piscataway Township, NJ 08854 USA) proteins in the cell culture supernatants. The procedures were exactly the same as described in the product manuals.
Quantitative RT-PCR. RNA was extracted following the QIAGEN manual exactly, reverse-transcribed into cDNA using the BIO-RAD iScript ™ cDNA Synthesis Kit. qPCR was performed with gene-specific primers and 6FAM-TAMRA (6-carboxyfluorescein-N,N,N,N-tetramethyl-6-carboxyrhodamine) probes, or primers (See Table 1 siRNA transfection and infection. BMDMs were transfected with siRNA and a mouse macrophage transfection Kit (Lonza #VPA-1009) on a Nucleofector 2. Briefly, differentiated BMDMs were dislodged gently from Petri dishes and counted using trypan blue staining. 1 × 10 6 cells were centrifuged at 200 × g for 10 minutes at room temperature. The resultant supernatant was discarded completely and the cell pellet was washed once with 1 ml sterile PBS. PBS was completely removed. The cell pellet was re-suspended carefully in 100 µl room temperature Nucleofector Solution VPA-1009 and 200 pmol siRNA per sample. The mix was transferred into a certified cuvette (sample must cover the bottom of the cuvette without air bubbles) and electroporated with a Nucleofector II using program Y-001. ~500 µl of the pre-equilibrated culture medium was added to the cuvette and then gently transferred to a 12-well cell culture plate. 48 h after transfection, the cells were infected with WNV for 24 h. Nuclear extraction. 5 × 10 6 cells were scraped off from culture dishes and washed with ice cold phosphate buffered saline (PBS) twice. The cells were then resuspended gently in 500 μl 1x Hypotonic Buffer (20 mM Tris-HCl, pH 7.4, 10 mM NaCl, 3 mM MgCl 2 , plus protease inhibitors) by pipetting up and down several times and then incubated on ice for 15 minutes. 25 μl detergent (10% NP40) was added to the cell suspension, which was then vortexed for 10 seconds at highest setting. The homogenate was centrifuged for 10 minutes at 3,000 rpm at 4 °C, resulting in supernatant (cytosol) and pellet (nuclei). The nuclei were lyzed in 50 μl complete Cell Extraction Buffer (10 mM Tris, pH 7.4, 2 mM Na 3 VO 4 , 100 mM NaCl, 1% Triton X-100, 1 mM EDTA, 10% glycerol 1 mM EGTA, 0.1% SDS, 1 mM NaF, 0.5% deoxycholate, 20 mM Na4P 2 O 7 , complete protease inhibitors) for 30 minutes on ice with intermittent vortexing every 10 minute. The lysate was then centrifuged for 30 minutes at 14,000 × g at 4 °C, resulting in nuclear lysate (supernatant).
Western blotting. For immunoblotting, cells were harvested at indicated time points after WNV infection, lysed in lysis buffer buffer [50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 5 mM EDTA, 1% Triton X-100] supplemented with Complete Protease Inhibitor Cocktail (Roche Diagnostics, Indianapolis, IN). Cell lysates were resolved on a 4-to 15%-gradient SDS-PAGE gel, and the separated proteins were transferred to a nitrocellulose membrane. The proteins of interest on the membrane were probed with specific antibodies, followed by HRP-conjugated secondary antiserum, and detected with an enhanced chemilumiscence system from GE Healthcare (Port Washington, NY 11050 USA).