Gene Expression and Antiviral Activity of Interleukin-35 in Response to Influenza A Virus Infection*

Interleukin-35 (IL-35) is a newly described member of the IL-12 family. It has been reported to inhibit inflammation and autoimmune inflammatory disease and can increase apoptotic sensitivity. Little is known about the role of IL-35 during viral infection. Herein, high levels of IL-35 were found in peripheral blood mononuclear cells and throat swabs from patients with seasonal influenza A virus (IAV) relative to healthy individuals. IAV infection of human lung epithelial and primary cells increased levels of IL-35 mRNA and protein. Further studies demonstrated that IAV-induced IL-35 transcription is regulated by NF-κB. IL-35 expression was significantly suppressed by selective inhibitors of cyclooxygenase-2 (COX-2) and inducible nitric-oxide synthase, indicating their involvement in IL-35 expression. Interestingly, IL-35 production may have suppressed IAV RNA replication and viral protein synthesis via induction of type I and III interferons (IFN), leading to activation of downstream IFN effectors, including double-stranded RNA-dependent protein kinase, 2′,5′-oligoadenylate synthetase, and myxovirus resistance protein. IL-35 exhibited extensive antiviral activity against the hepatitis B virus, enterovirus 71, and vesicular stomatitis virus. Our results demonstrate that IL-35 is a novel IAV-inducible cytokine, and its production elicits antiviral activity.

Seasonal influenza A virus (IAV) 2 is a common human infection, resulting in significant morbidity and mortality (1). It is a negative-sense RNA virus of the Orthomyxoviridae family (2). IAV attaches to membrane receptors and enters epithelial, macrophage, and dendritic cells (3). These cells release large quantities of antiviral and immunostimulatory cytokines during IAV infection. An extensive array of cytokines and chemokines are produced by host cells in response to IAV infection, including interleukin-1(IL-1), IL-2, IL-8, IL-10, IL-15, IL-18, IL-27, IL-32, TNF-␣, IFN-␣/␤/␥, MIP-1␣/␤ (macrophage inflammatory protein-1 ␣/␤) (4 -7), and other cytokines and related pathway factors (1,8). There remain many unknown cytokines induced by and involved in the response to IAV infection. Further investigation of factors activated by IAV infection is critical to better understand host-virus interaction and advance antiviral research.
IL-35 has been linked to various disease, including autoimmune encephalomyelitis (15), autoimmune diabetes (16), inflammatory bowel disorder (17), collagen II-induced arthritis (11), airway inflammation (18), allergic asthma (18,19), multiple sclerosis (20), and chronic and aggressive periodontitis (21). It also impacts colorectal cancer progression and prognosis (22) and is highly expressed in tumor tissue in lung and colon cancer and esophageal, hepatocellular, and cervical carcinomas (23). Regulatory T cells (Tregs) are a critical subpopulation of CD4ϩ T cells essential to immune response (24,25). IL-35 can convert naïve T cells into strongly suppressive Tregs (26), promote Treg expansion, and suppress proliferation of conventional T cells (27,28). It might be specifically produced by Tregs and may be required for maximal suppressive activity (27). Although studies report that IL-35 is expressed in patients infected with the hepatitis B virus (HBV) (29 -31), the role of IL-35 during infection, particularly, IAV infection is largely unknown.
Herein, we show that IL-35 expression is induced by IAV infection and is regulated by the transcription factor, NF-B. The inflammatory factors, cyclooxygenase-2 (COX-2) and inducible nitric-oxide synthase (iNOS), may also be involved in this signaling cascade. IL-35 also activated the IFN pathway and indirectly elicited antiviral activity in response to viral infection. Our results provide a basis for further investigation of the relationship between IAV and IL-35 and suggest that IL-35 might be a potential novel target for antiviral therapies.

IAV-induced IL-35 Expression in Different Cell
Types-Because high levels of IL-35 mRNA were observed in IAV-infected patients, we next assessed in vitro changes in IL-35 expression in response to IAV infection. A549 cells were infected with IAV at various doses. IAV-induced (m.o.i. Ͻ 1) FIGURE 1. Measurement of IL-35 in IAV-infected patients and healthy individuals. A, total RNA was extracted from freshly isolated PBMCs from healthy individuals (n ϭ 12) or IAV-infected patients (n ϭ 12). EBI3 (left) and p35 (right) mRNA was quantified using qPCR. B, throat swab EBI3 (left) and p35 (right) mRNA levels in controls (n ϭ 10) and patients (n ϭ 10). In A and B, the lowest value from controls was as assigned a value of 1. Data are expressed as -fold induction (folds) relative to the lowest value from controls. Data are expressed as the mean Ϯ S.E. from samples tested in triplicate. Boxplots illustrate medians with 25 and 75%, and error bars for 5 and 95% percentiles (**, p Ͻ 0.01). C, correlation analysis between NP mRNA and EBI3 (left, n ϭ 33) and p35 (right, n ϭ 33) mRNA levels in PBMCs from IAV-infected patients. Solid line, linear growth trend; r, correlation coefficient; Pearson correlation analysis was used to determine r values; Student's t test was used to determine p values.
IL-35 mRNA expression was positively correlated with the infectious dose of IAV ( Fig. 2A). Viral infection induced maximal expression at 1 m.o.i. These data suggest that IAV stimulates IL-35 mRNA expression in a dose-dependent manner (m.o.i. Ͻ 1).
An ELISA kit was used to measure IL-35 protein expression in culture supernatants. IL-35 expression was up-regulated ϳ2-fold by IAV infection compared with mock infection (Fig.  2D). Similar results were obtained in freshly isolated PBMCs and alveolar type II (AT II) cells (Fig. 2, E and F, respectively), wherein IL-35 mRNA levels were up-regulated by IAV infection. In the meantime, NF-B was activated by IAV infection with p-IB (36, 37) levels measured in A549 cells (Fig. 2G). These data indicate that IAV infection can induce IL-35 mRNA and protein expression.
Transcription factor binding sites at the IL-35 promoter were predicted by transcription factor assay databases, includ- ing JARSPAR DATABASE, ALGGEN, Biobase, Gene regulation, and Transgene. Diagrams showing comprehensive analysis of these sites at the IL-35 promoter are shown in Fig. 3, B and C. To identify potential regulatory sites, a series of truncated IL-35 promoters was constructed. Luciferase activity assays showed that elimination of the Ϫ150 to Ϫ450 region of pIL-35/ EBI3-Luc greatly decreased promoter activity (Fig. 3B). The results further suggest that the Ϫ360 to Ϫ347 and Ϫ233 to Our results indicate that the Ϫ1434 to Ϫ1444 NF-B/p50 binding site and the Ϫ1368 to Ϫ1378 NF-B/p65 binding site may be important for IAV-induced activation of the IL-35/p35 promoter. Four mutant reporters were constructed for the NF-B binding sites. These reporters encompassed site 1 (Ϫ360 to Ϫ347), site 2 (Ϫ233 to Ϫ242), and site 3 (Ϫ84 to Ϫ97) regions of pIL-35/EBI3-Luc. Site 3 is an important cis-regulatory element binding site in IL-18 and IL-1␤-induced IL-27/EBI3 promoter activity (38). Mutant reporters also encompassed the site 4 region (Ϫ1368 to Ϫ1378) on pIL-35/p35-Luc. Our results suggest that there was a decrease in IAV-induced mutant reporter luciferase activity compared with activity using the wild-type promoter (Fig. 3, B and C).
To confirm whether p50 or p65 regulates IL-35 promoter activity, overexpression and knockdown experiments of p50 or p65 were performed. Overexpression of p65, but not p50, significantly increased IL-35/EBI3 and IL-35/p35 luciferase activity in cells infected with IAV relative to control (Fig. 3D). Knockdown of NF-B with shRNA-p65, but not shRNA-p50, significantly decreased IAV-induced IL-35/EBI3 and IL-35/p35 luciferase activity relative to the shRNA control (Fig. 3E). These data suggest that NF-B/p65 might be a vital transcription factor regulating IL-35 promoter activity.
We also investigated whether CREB binding sites on the IL-35 promoter are important for IAV-induced promoter activity. Our results show that overexpression (  (6,39,40). We, thus, investigated the roles of COX-2 and iNOS in IAV-induced IL-35 expression. A549 cells were transfected with a COX-2 expression plasmid (pCMV-COX-2) or a vector for 24 h and then infected with IAV. IL-35 mRNA levels increased with COX-2 overexpression (data not shown). A549 cells were then treated with Etoricoxib, a selective COX-2 inhibitor, for 2 h at different doses (10, 50, or 100 M) before infection with IAV (m.o.i. ϭ 1). DMSO and IAV infection without inhibitor were used as solvent and positive controls, respectively. Our results show that IL-35 mRNA levels decreased in a dose-dependent manner with pretreatment with Etoricoxib (Fig. 4A). A549 cells were transfected with two different COX-2-specific siRNAs that modestly reduced levels of COX-2 mRNA and protein (Fig. 4B). IL-35 expressions were confirmed to be down-regulated by COX-2 silencing with siCOX-2-#2 (Fig. 4C). These data suggest that COX-2 positively regulates IAV-induced IL-35 expression. It is reported that overexpressed COX-2 contributed to prostaglandin (PG) overproduction (41); we incubated A549 cells with the dissolved prostaglandin E2 (PGE2) for 2 h, then infected with IAV (m.o.i. ϭ 1). The results showed that expression of IL-35 mRNA was significantly up-regulated by PGE2 (Fig. 4D).
We next determined whether IL-35 expression correlates with COX-2 levels in clinical samples. IL-35 and COX-2 mRNA levels were measured in PBMCs from IAV-infected patients then subjected to correlation analysis. There were statistically significant correlations between IL-35 and COX-2 mRNA levels (IL-35/EBI3 and COX-2: r ϭ 0.51; *, p Ͻ 0.05, n ϭ 20, Fig. 4E; IL-35/p35 and COX-2: r ϭ 0.47; *, p Ͻ 0.05, n ϭ 20, Fig. 4F; Pearson's correlation). Similarly, IL-35 mRNA levels decreased in a dose-dependent manner when A549 cells were treated with S-methylisothiourea sulfate (50, 100 M; Alexis Biochemicals, Grünberg, Germany), the selective iNOS inhibitor (Fig. 4G), and IL-35 mRNA decreased (Fig. 4I) with si-iNOS-#3 (Fig. 4H) transfected. Meanwhile, IL-35 mRNAs were up-regulated in A549 cells with sodium nitroferricyanide(III) dihydrate (SNP) incubation (Fig. 4J). To verify the relationship between IL-35 and iNOS expression during IAV infection, PBMCs were isolated from IAV-infected patients. IL-35 and iNOS mRNA levels were measured and analyzed using Pearson's correlation. A statistically significant correlation was found between IL-  5A), and Western blots with EBI3 and p35 antibodies were used to confirm the cellular overexpression IL-35 (Fig. 5B). However, overexpression of IL-35 in A549 cells did not strongly affect IAV replication (data not shown). Li et al. (42) shows that IL-32␥ does not directly affect HBV replication in HepG2.2.15 cell, and subsequently established protocols utilizing collection of supernatants from IL-32␥-treated PBMCs for indirect antiviral assays . Using the method mentioned above, Jurkat cells were first electroporated with pCMV-IL-35; after 36 h, the IL-35 protein level in the supernatants were measured by ELISA (Fig. 5C), and the supernatants were collected for antiviral experiments. A549 cells were incubated with the collected supernatants of Jurkat cells and then infected with IAV (m.o.i. ϭ 1). IAV NP RNA levels, including plus-sense RNA and minussense RNA were measured 3 hpi via qRT-PCR; supernatant IAV titers were measured using a hemagglutinin assay 24 hpi. This treatment significantly decreased NP RNA levels (Fig. 6A) and significantly decreased IAV titers (Fig. 6B). Our results indicate that the incubation of A549 cells with the supernatants of IL-35-transfected Jurkat cells can inhibit IAV replication effectively.
We next investigated whether secreted IL-35 can inhibit IAV replication using commercial rhIL-35. Previous studies demonstrated interactions between interleukins and IFN␥. IL-12 was positively regulated by IFN␥ (43,44), and IL-23 interacted with IL-12, IL-18, and IL-2 to promote IFN␥ production in NK cells (45). In turn, IFN␥ interacted with IL-27 to induce Treg proliferation, limiting pathology due to infection (46). Evidence also suggested that IFN␥ and STAT1-dependent expression of IL-12R␤2 were crucial for T cell activation (47). In human cancer cell lines, IL-35 expression can be induced after TNF-␣ and IFN␥ stimulation (23). IL-35 not only decreased production of IL-17 but can also increase IFN␥ production (11). Jurkat cells were incubated with 100 ng/ml of rhIL-35 (11) and 2 g/ml IFN␥ for 24 h, and the supernatants were collected for antiviral assays. A549 cells were subsequently incubated with these supernatants and then infected with IAV (m.o.i. ϭ 1). IAV NP RNA levels were measured via qRT-PCR (Fig. 6C), and IAV titers were measured using a hemagglutinin assay (Fig. 6D). Jurkat cell viability was not affected by 24-h incubation with rhIL-35 (at different doses) and with or without 2 g/ml IFN␥ (Fig. 6E). The activity of A549 cells was not affected by incubation with supernatants from Jurkat cells treated with 100 ng/ml rhIL-35 with or without 2 g/ml IFN␥ (Fig. 6F). The data sug-gest that the incubation of A549 cells with the supernatants of rhIL-35 and IFN␥-treated Jurkat cells can restrict IAV infection enormously.
Next, we investigated whether IL-35 has wide-ranging antiviral function. Rhabdomyosarcoma (RD) cells were incubated with the supernatants of Jurkat cells described in Fig. 5, A or C, and then infected with human enterovirus 71 (EV71) (m.o.i. ϭ 1). 12 h later, EV71 VP1 expression was measured using absolute qRT-PCR. Replication of EV71 was inhibited by IL-35 overexpression (Fig. 6G), and EV71 copy numbers were significantly reduced by this treatment of rhIL-35 and IFN␥ (Fig. 6H). Next, Huh7 cells were transfected with pHBV and incubated with supernatants from Jurkat cells described in Fig. 5, A or C. Our results show relatively lower HBeAg and HBsAg levels when cells were incubated with Jurkat cell-derived supernatants compared with control (Fig. 6, I and J). We then assessed the effects of the above protocol on the production of recombinant vesicular stomatitis virus carrying enhanced green fluorescent protein (VSV-eGFP) in A549 cells. Consistent with our other results, VSV-eGFP infection was significantly reduced when cells were incubated with Jurkat cell-derived supernatants compared with control (Fig. 6K), and the number of infected cells decreased from 75.94% to 38.97%, as measured by flow cytometry. However, there were no antiviral effects in Vero cells on VSV-eGFP expression (Fig. 6K). Because Vero cells lack functional IFN gene expression (48 -50), we inferred that any potential antiviral functions of IL-35 require IFN production.
Similar results utilizing the above protocol were obtained when PBMCs were used. Freshly isolated PBMCs from healthy donors were electroporated with pCMV-IL-35 or vector, and the supernatants were collected for antiviral assays. IAV NP RNA levels and virus titers (data not shown) were reduced in IAV (m.o.i. ϭ 1)-infected A549 cells incubated with the above PBMC-derived supernatants. HBeAg and HBsAg expression was reduced in pHBV-transfected Huh7 cells cultured with the same PBMC-derived supernatants (data not shown). EV71 copy numbers were also reduced in 1 m.o.i. EV71-infected RD cells cultured with these PBMC-derived supernatants (data not shown). These data indicate that treatment with the above PBMC-derived supernatants had antiviral functions not only in IAV infection, but also in VSV, HBV, and EV71 infection. Taking together, IL-35 has extensive antiviral activity to DNA and RNA viruses.
We next explored whether expression of downstream IFN effectors can be induced by IL-35. Jurkat cells were transfected with pCMV-IL-35, and the supernatants were collected after 36 h. A549 cells were subsequently incubated with these Jurkat cell-derived supernatants and infected with 1 m.o.i. IAV. qRT-PCR analysis and Western blots revealed that intracellular RNA-dependent protein kinase (PKR), 2Ј,5Ј-oligoadenylate synthetase (OAS), myxovirus resistance protein (Mx) mRNA, and protein levels increased with this treatment (Fig. 7, F and  G). Similar results were obtained using rhIL-35. The expression of IFN-␣, IFN-␤, and IFN-1 were all induced by rhIL-35 in a time-dependent manner (Fig. 7H). The PKR, OAS, and Mx were also up-regulated (Fig. 7I).
To investigate whether the IFNs mediated IL-35 anti-viral action, IFNAR1 or IFNLR1 were knocked down by shIFNAR1 or shIFNLR1 (Fig. 7J), and the anti-viral action of IL-35 was not detected in the presence of either shIFNAR1 or shIFNLR1 (Fig.  7K). Collectively, these data demonstrate that IL-35 can induce IFN production and stimulate expression of downstream IFN effectors, and the induced IFNs mediate IL-35 anti-viral activity.

Discussion
Herein, we investigated IL-35 expression during IAV infection. IAV is one of the most common causes of infection in humans (51) and can lead to high morbidity and significant mortality. Host cells secrete a variety of cytokines and chemokines in response to IAV infection (1, 4 -7). We present the first evidence from clinical samples that IL-35 expression may change in response to IAV infection, consistent with additional findings that IL-35 mRNA levels were positively correlated with IAV NP mRNA levels.
We used various cell lines in this study for different purposes. Lung epithelial cells are the primary targets of IAV infection (52). AT II cells constitute ϳ60% of alveolar epithelial cells and ϳ15% of all lung parenchymal cells. A549 cells are adenocarcinomic human alveolar basal epithelial cells commonly used in IAV research. PBMCs are widely used to investigate immune response to viral and microbial infections (53,54). Our findings suggest that IL-35 is significantly up-regulated in these three cell types during IAV infection in a time-and dose-dependent manner. Our results suggest that IAV infection can induce IL-35 expression via the transcription factor, NF-kB/p65, and the inflammatory factors COX-2 and iNOS. IL-35 can also activate the IFN pathway and may have potential antiviral activity. Consistent with these findings, others have shown that HBV infection can induce IL-35 expression. One report shows that IL-35 mRNA and protein are detectable in CD4 ϩ T cells in patients with chronic hepatitis B (55). Likewise, in patients chronically infected with HBV, CD4 ϩ T cells have significantly higher levels of EBI3 mRNA and protein compared with healthy individuals and patients in whom HBV infection had resolved. Furthermore, IL-35 has also been shown to suppress proliferation of HBV antigen-specific cytotoxic T-lymphocytes and IFN-␥ production in vitro (56).
NF-B is inactive in the cytoplasm due to IB masking its nuclear localization sequence (57). Upon viral infection, phosphorylated IKK can phosphorylate IB␣, which is then ubiquitinated (58,59), allowing NF-B to enter the nucleus (39). We show that binding of NF-B/p65 to the p65 transcription binding site initiates IL-35 transcription during IAV infection.
COX-2 is critical to the inflammatory response (60) and can be activated by IAV infection in A549 cells (61). It can regulate protein expression, including that of CRC3, IL-10, and IL-17 (62). iNOS is an inducible and calcium-independent isoform of FIGURE 6. The indirect antiviral activity of IL-35. A and B, Jurkat cells were transfected with vector or pCMV-IL-35. Thirty-six hours later the supernatants were collected for indirect antiviral assays. Antiviral assays were performed in 1 m.o.i. IAV-infected A549 cells incubated with the collected supernatants for 24 h. IAV RNAs (A) and IAV production (B) were analyzed using qRT-PCR and hemagglutination assays, respectively. C and D, Jurkat cells were treated with 100 ng/ml rhIL-35 and 2 g/ml IFN␥. Twenty-four hours later the supernatants were collected, and indirect antiviral assays were performed as in A (C) and B (D). E, Jurkat cells were treated with doses of rhIL-35 or co-treated with 2 g/ml IFN␥ for 24 h, and cell viability was measured using the MTT assay. F, A549 cells were incubated with the supernatants from rhIL-35-treated Jurkat cells, and cell viability were measured using the MTT assay. G and H, EV71-infected RD cells were cultured then incubated with the supernatants described in A (G) or C (H); the supernatants were then collected and used for EV71 copy testing using the absolute value of qRT-PCR. I and J, Huh7 cells transfected with pHBV were incubated with collected supernatants from Jurkat cells described in A (I) or C (J). HBe/s Ag was analyzed using ELISA. K, A549 cells or Vero cells were incubated with collected supernatants from Jurkat cells described in A, then infected with VSV-eGFP (m.o.i. ϭ 1) for 6 h. VSV-eGFP replication was visualized via fluorescence microscopy and measured using flow cytometry for eGFP expression. Bar graphs represent the mean Ϯ S.D. n ϭ 3; *, p Ͻ 0.05; **, p Ͻ 0.01.

IL-35 Mediates Antiviral Activity
NO synthetase. It plays a critical modulatory role in immune response, chronic inflammation, and carcinogenesis and can be stimulated by viral infection (6,63,64). Our findings suggest that COX-2 and iNOS may be important factors in IAV-induced IL-35 expression. Furthermore, both COX-2 and iNOS mRNA levels are positively correlated with IL-35 mRNA levels in IAV-infected blood samples.
IL-35 is detectable in peripheral CD4 ϩ T cells and is believed to play important functions in inhibition of the immune response during chronic HBV infection (56), suggesting it may be a potential therapeutic target to control HBV infection (29). Thus, we investigated the biological function of virus-induced IL-35 expression. Using the indirect approach, we demonstrate that overexpression of IL-35 and incubation with rhIL-35 may suppress IAV, EV71, VSV, and HBV replication and viral production.
IFN-␣/␤ is the host's central innate immune response to viral infection (65), although IFN-1 has also been shown to inhibit replication of a number of viruses (66). In this study, the indirect antiviral effect of IL-35 was abolished in IFN-deficient Vero cells (48), suggesting that IL-35 function during viral infection depends on expression of IFN. Our findings show that IL-35 can induce IFN-␣, IFN-␤, and IFN-1 mRNA expression and increases PKR, OAS, and Mx mRNA and protein expression. And we found up-regulated expression of IL-35 receptors (14), including gp130 and IL-12Rb2, concurrent with IFN production (data not shown). Previous reports show that signaling through the IL-35 receptor requires STAT1 and STAT4 (14).
Here, IL-35 may have stimulated IFN expression via activating IL-35 receptors, thereby presumably stimulating the Jak-STAT pathway, leading to phosphorylation of STAT1 and STAT4 and subsequent up-regulation of IFN.
In conclusion, we propose a hypothetical model of IAV-induced IL-35 production and its biological function (Fig. 8). IAV infection stimulates IL-35 expression via activation of its promoter by the transcription factor, NF-B/p65, and through activation of COX-2 and iNOS pathways. IL-35 then activates IFN and its downstream effectors, leading to inhibition of viral replication and production. Further studies are required to better understand the IL-35-related complex regulatory mechanisms of antiviral host response during virus infection. However, our findings provide evidence of a distinct role for IL-35 in this process and point to potential novel clinical uses for IL-35 in antiviral therapy.

Experimental Procedures
Ethics Statement-Peripheral blood samples and throat swabs were obtained from IAV-infected patients and healthy individuals. Clinical samples were collected in accordance to the Declaration of Helsinki from patients admitted to the Hubei Provincial Center for Disease Control and Prevention and were approved by the Institutional Review Board of the College of Life Sciences of Wuhan University in accordance to its guidelines for protection of human subjects. Written informed consent was obtained from all participants.
Cell and Virus-Human lung epithelial cells (A549) were cultured in F12K medium (Gibco). Human T cell lymphoblast-like cell (Jurkat) lines were cultured in RPMI 1640 medium (Gibco). RD cells were cultured in minimum Eagle's medium (MEM, Gibco). Huh7 cells were cultured in Dulbecco's modified Eagle's medium (DMEM, Gibco). All media were supplemented with 10% fetal bovine serum (FBS, Gibco). All cultures were maintained at 37°C in a 5% CO 2 incubator. The IAV/Hong Kong/498/97 (H3N2) strain used in these studies was provided by the China Center for Type Culture Collection. Recombinant VSV-eGFP was a gift from Mingzhou Chen (Wuhan University). EV71 was obtained from Xiangyang (GenBank TM accession number JN230523.1).
Isolation of PBMCs and Primary AT II Cells-PBMCs were isolated using density centrifugation diluted 1:1 in a solution of human lymphocytes (TBD-Science, Tianjin, China) as previously described (67). PBMCs were washed twice with PBS and cultured in RPMI 1640 medium at 37°C in a 5% CO 2 incubator. Human AT II cells were purchased (Wuhan PriCells Biotechnology & Medicine Co., Ltd, Wuhan, China) and cultured in RPMI 1640 medium (37°C, 5% CO 2 ).
Luciferase Assay-A549 cells were transfected with luciferase reporter plasmids containing the promoter of IL-35 along with pRL-TK infected with IAV. Luciferase activity was measured after 12 h or 24 h of serum starvation and normalized to Renilla luciferase activity. Results are expressed as relative luciferase activity.
MTT Assay-90-l Jurkat cell suspensions were plated on 96-well plates (10 4 /cm 2 ). rhIL-35 (0, 5, 10, 50, 100, or 500 ng/ml) was added to wells with or without IFN␥ (2 g/ml, total Ͻ10 l). Each group contained 5 wells. Each plate contained a zero-value well with DMSO but without cells. Plates were cultured at 37°C in a 5% CO 2 incubator for 24 h. MTT reagent (20 l, 5 mg/ml, methylthiazolyldiphenyl-tetrazolium bromide dissolved in PBS, pH 7.4, stirred at constant temperature for 30 min and passed through a 0.22-m microfiltration membrane, stored at Ϫ20°C) was added to each well, and cells were cultured for another 4 h. Triple solution (100 l, 10% SDS, 5% isobutyl alcohol, 0.012 M HCl) was added to each well, and plates were then cultured at 37°C overnight. Absorbance values for each well were determined at 570 nm.
A549 cells were plated on 96-well plates (10 4 /cm 2 ) and cultured in DMEM (200 l/well with 10% FBS). Cells were treated with Jurkat supernatants previously incubated with 100 ng/ml rhIL-35 with or without 2 g/ml IFN␥. After 24 h, MTT reagent (20 l) was added to each well, and cells were cultured for 4 h. Media were removed, and DMSO (150 l) was added to each well with shaking for 10 min to dissolve formazan. Absorbance values for each well were determined at 490 nm, adjusted to the zero-value well.
Statistical Analysis-Relationships between IL-35 expression and either IAV NP levels in clinical samples, COX-2 mRNA, or iNOS mRNA levels were analyzed using Pearson's correlation. All the experiments were performed in triplicate. Two-group comparisons were performed using Student's t test. Data are expressed as the mean Ϯ S.D., except for the clinical results, shown as mean Ϯ S.E. Results with p Ͻ 0.05 (*) and p Ͻ 0.01 (**) were considered statistically significant.