Toll-like receptor 3 and RIG-I-like receptor activation induces innate antiviral responses in mouse ovarian granulosa cells

Molecular and Cellular Endocrinology 372 (2013) 73–85

Contents lists available at SciVerse ScienceDirect

Molecular and Cellular Endocrinology journal homepage: www.elsevier.com/locate/mce

Toll-like receptor 3 and RIG-I-like receptor activation induces innate antiviral responses in mouse ovarian granulosa cells Keqin Yan, Weiwei Zhu, Lili Yu, Nan Li, Xiaoyan Zhang, Peipei Liu, Qiaoyuan Chen, Yongmei Chen ⇑, Daishu Han ⇑ Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences, School of Basic Medicine, Peking Union Medical College, Beijing, PR China

a r t i c l e

i n f o

Article history: Received 9 December 2012 Received in revised form 30 March 2013 Accepted 30 March 2013 Available online 6 April 2013 Keywords: RIG-I-like receptor Ovary Granulosa cell Antiviral response

a b s t r a c t Viral infections of the ovary can cause pathological conditions. However, innate antiviral responses in the ovary are poorly understood. In this study, we demonstrate that Toll-like receptor 3 (TLR3), retinoic acid-inducible gene I (RIG-I) and melanoma differentiation-associated gene 5 (MDA5) are constitutively expressed in the mouse ovary and predominantly located in granulosa cells. Polyinosinic-polycytidylic acid [poly(I:C)], a common agonist of TLR3, MDA5 and RIG-I, induced innate antiviral responses in ovarian granulosa cells. Poly(I:C) up-regulated pro-inflammatory cytokines, including TNF-a and IL-6, and type I interferons (IFN-a/b). Moreover, poly(I:C) induced the expression of antiviral proteins, including 20 -50 -oligoadenylate synthetase, Mx GTPase 1 and IFN-stimulating gene 15, in granulosa cells. In contrast, P450 aromatase expression was inhibited by poly(I:C). The poly(I:C)-induced antiviral responses in TLR3 knockout (TLR3–/–) ovarian granulosa cells were reduced, and completely abolished by blocking of MDA5/ RIG-I signaling. Further, the poly(I:C)-induced cytokine expression in TLR3–/– cells was reduced by knockdown of MDA5 or RIG-I. Data suggest that TLR3, MDA5 and RIG-I cooperate in mediating innate antiviral responses in granulosa cells, which may contribute to the defense of the ovary against viral infections. Ó 2013 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Virus infections have been found in the ovaries of different species. Vaccinia viruses preferentially infect the ovary and impair ovarian functions (Zhao et al., 2011). Hepatitis B viruses can infect human oocytes and ovarian granulosa cells, which may represent one of the mechanisms underlying virus vertical transmission (Hu et al., 2011; Jin et al., 2011). The mechanisms underlying innate antiviral responses in the ovary remain poorly understood. The roles of pattern recognition receptors (PRRs) in mediating inflammatory responses in the ovary and regulating ovarian functions have been revealed (Sheldon and Bromfield, 2011). However, PRR-mediated innate antiviral responses in the ovary have yet to be investigated. PRRs are broadly expressed in many cell types and recognize highly conserved components specific to microbes, termed pathogen-associated molecular patterns (PAMPs) (Kumar et al., 2011). Recognition of PAMPs by PRRs initiates innate immune responses

⇑ Corresponding authors. Addresses: Department of Human Anatomy, Histology & Embryology, PUMC & CAMS, 5 Dong Dan San Tiao, Beijing 100005, PR China (Y. Chen), Department of Cell Biology, PUMC & CAMS, 5 Dong Dan San Tiao, Beijing 100005, PR China (D. Han). Tel.: +86 10 69156457; fax: +86 10 69156466. E-mail addresses: [email protected] (Y. Chen), [email protected] (D. Han). 0303-7207/$ - see front matter Ó 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.mce.2013.03.027

and subsequently restructures adaptive immunity (Iwasaki and Medzhitov, 2010). Three families of PRRs in cells have been recognized thus far: Toll-like receptors (TLRs), retinoic acid-inducible gene I (RIG-I)-like receptors (RLRs) and nucleotide-oligomerization domain-like receptors (NLRs). TLRs belong to a family of transmembrane proteins that include 13 members in mammals (Li et al., 2009). RLRs are a family of cytosolic viral sensors that include RIG-I, melanoma differentiation-associated gene 5 (MDA5) and laboratory of genetics and physiology 2 (LGP2) (Yoneyama and Fujita, 2007). NLRs represent a large family of intracellular sensors that detect a broad spectrum of pathogens and stress signals (Martinon et al., 2009). Different PRRs may recognize the same PAMPs. For example, TLR3, MDA5 and RIG-I recognize viral double-strained RNA (dsRNA) and a synthetic dsRNA analog, polyinosinic-polycytidylic acid [poly(I:C)]. Activation of TLR3, MDA5 and RIG-I triggers antiviral responses by inducing the production of type I interferons (IFN-a/b) (Takeuchi and Akira, 2008; Thompson et al., 2011). IFN-a/b subsequently induce synthesis of antiviral proteins to inhibit viral replication within cells, and facilitate adaptive immunity against viruses (Samuel, 2001; Yoneyama et al., 2005). TLR3, MDA5 and RIG-I trigger different signaling pathways that converge at the activation of nuclear factor kappa B (NF-jB) and IFN regulatory factor 3 (IRF3), thereby inducing pro-inflammatory cytokines and IFN-a/b (Matsumoto et al., 2011; Takeuchi and Akira, 2008).

74

K. Yan et al. / Molecular and Cellular Endocrinology 372 (2013) 73–85

Tissue-resident macrophages are believed to be important innate immune cells that may construct the first line of defense against microbial infections. Some macrophages are located in the stroma of the normal ovary and regulate ovarian functions through the secretion of inflammatory cytokines (Takaya et al., 1997; Wu et al., 2004). The intra-follicular environment where numerous granulosa cells encompass an oocyte is devoid of macrophages in mice (Petrovska et al., 1996). These observations lead researchers to question whether or not the granulosa cells within a follicle possess innate immune capabilities to protect the oocyte against invading pathogens (Sheldon and Bromfield, 2011). Several recent studies have shown the expression and function of TLRs in the ovary. TLR1-9 are expressed in human normal ovary and ovarian cancer (Zhou et al., 2009). Human granulosa cells from in vitro fertilization express TLR4 (Serke et al., 2009). TLR4-initiated inflammatory responses in bovine granulosa cells perturb oestradiol synthesis (Herath et al., 2007). In mouse, the expression and function of TLR2 and TLR4 in cumulus cells have been investigated (Shimada et al., 2006, 2008). A previous study demonstrated that TLRs regulate ovulation in response to both exogenous and endogenous stimuli in mice (Liu et al., 2008). TLR signaling pathways have been recently analyzed in human granulosa cell lines, bovine and hen primary granulosa cells (Bromfield and Sheldon, 2011; Price et al., 2012; Woods et al., 2009). Most studies on TLR-mediated innate immune responses in ovarian granulosa cells focus on TLR2 and TLR4. With the exception of a few studies on the expression of TLRs that recognize viruses in granulosa cells (Price et al., 2012; Zhou et al., 2009), the antiviral roles of PRRs in ovarian cells have yet to be exclusively investigated. The current study aims to evaluate the expression and function of the viral sensors in mouse ovary. We provide evidence that

mouse ovarian granulosa cells constitutively express TLR3, RIG-I and MDA5, all of which participate in mediating innate antiviral responses in granulosa cells. 2. Materials and methods 2.1. Animals C57BL/6J mice were obtained from the Laboratory Animal Center of Peking Union Medical College (Beijing, China). TLR3 knockout mice (B6/129S1-Tlr3tm1Flv/J) were purchased from Jackson Laboratories (Bar Harbor, ME, USA). The mice were bred under specific pathogen-free conditions with a natural day/night cycle, fed freely with food and water, and handled in accordance with the Guidelines for the Care and Use of Laboratory Animals established by the Chinese Council on Animal Care. 2.2. Major reagents and antibodies Poly(I:C) (tlrl-pic), BX795 (tlrl-bx7), and BAY11-7082 (tlrl-b82) were purchased from InvivoGen (San Diego, CA, USA). The antibodies used in this study are listed in Table 1. Small interfering RNA (siRNA) targeting mouse RIG-I (sc-61481), MDA5 (sc-61011), and control siRNA (sc-37007) were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). 2.3. Isolation and culture of granulosa cells Mural ovarian granulosa cells were isolated based on previously described procedures with modifications (Vanderhyden et al.,

Table 1 Antibodies used in this study. Vendor

Antibody

Host species

Catalog number

Use

Working dilution

Sigma–Aldrich, St. Louis, MO, USA

Anti-b-actin

Mouse

A5316

WB

1:4000

Abcam, Cambridge, MA, USA

Anti-F4/80 Anti-ISG15 Anti-OAS1 Anti-RIG-I

Rat Rabbit Rabbit Rabbit

ab6640 ab131119 ab86343 ab45428

Anti-TLR3

Mouse

ab13915

Anti-MDA5

Rabbit

ab69983

IF WB WB WB IF IHC WB IF IHC WB IF IHC

1:500 1:500 1:1000 1:1000 1:200 1:200 1:500 1:200 1:200 1:1000 1:500 1:500

Anti-FSHR Anti-PKR Anti-MX1 Anti-IRF3

Goat Rabbit Rabbit Rabbit

sc-7798 sc-708 sc-50509 sc-9082

IF WB WB WB IF

1:200 1:500 1:200 1:1000 1:200

Anti-p65

Rabbit

4764

Anti-p-p65 Anti-p-IRF3

Rabbit Rabbit

3031 4947

WB IF WB WB

1:1000 1:200 1:1000 1:1000

Rat IgG Rabbit IgG Mouse IgG Goat IgG Mouse IgG

Goat Goat Goat Rabbit Goat

ZF-0315 ZF-0311 ZF-0312 ZF-0314 ZB-2305

Rabbit IgG

Goat

ZB-2301

IF IF IF IF WB IHC WB IHC

1:200 1:200 1:200 1:200 1:4000 1:200 1:4000 1:200

Santa Cruz Biotechnology, Santa Cruz, CA, USA

Conjugation

Cell Signaling, Beverly, MA

Zhongshan Biotechnology Co., Beijing, China

WB, Western blot; IHC, immunohistochemistry; IF, immunofluorescence staining; FITC, fluorescein isothiocyanate; HRP, horseradish peroxidase.

FITC FITC FITC FITC HRP HRP

K. Yan et al. / Molecular and Cellular Endocrinology 372 (2013) 73–85

1992). Briefly, 3-wk-old female mice were intraperitoneally injected with 5 IU pregnant mare serum gonadotropin (Beijing Chief-East Tech Co., Beijing, China). At 45 h after injection, the follicles were punctured with a 25-gauge needle to release both oocyte-cumulus cell complexes and clumps of mural granulosa cells. The granulosa cells were collected in F12/DMEM (Life Technologies, Grand, NY, USA) supplemented with 10% fetal calf serum (FCS; Life Technologies) and ITS (5 lg/ml insulin, 5 lg/ml transferrin, and 5 lg/ml selenium) (Sigma–Aldrich, St. Louis, MO, USA). Cells were dispersed by gentle drawing in and out of a pipette and then centrifuged for 3 min at 250. The granulosa cells were resuspended in F12/DMEM medium containing 10% FCS and ITS and then cultured in a humidified atmosphere containing 5% CO2 at 37 °C for transfection. 2.4. Transfection The granulosa cells were seeded in 6-well plates at a density of 5  105 cells/well and cultured for 24 h. The medium was changed by serum-free F12/DMEM for 2 h and the cells were transfected with 2 lg/ml poly(I:C) using 2 ll of LipofectamineÒ RNAiMAX Reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. For gene silence with siRNA, 2  105 granulosa cells per well were seeded in 6-well plates. At 24 h after culture, the cells were transfected with 100 nM siRNA. Twentyfour hours later, the cells were transfected with poly(I:C). 2.5. MTT assay Cell viability was assessed using 3-(4,5-dimethylthiazolyl-2)2,5-diphenyltetrazolium bromide (MTT) assay kit (ATCC, Manassas, VA, USA) according to the manufacture’s instructions. Briefly, granulosa cells were cultured in 96-well microplates at 2  104 cells/well. After treatment with the same conditions as those for the innate immune response induction, the cells were incubated with 10 ll MTT solution for 2 h. After removal of medium, 100 ll detergent reagent was added each well to lyse the cells. Absorbance at 570 nm was recorded by a microplate reader (BioTek, Winooski, VT, USA). 2.6. Isolation of macrophages Peritoneal macrophages were isolated based on a previously described procedure (Chong et al., 2005). Briefly, the peritoneal cavities were lavaged with 5 ml of ice-cold 1  phosphate buffered saline (PBS). The peritoneal cells were cultured in RPMI 1640 medium (Life Technology) supplemented with 10% FCS and antibiotics (100 U/ml penicillin, 100 mg/ml streptomycin) in a humidified atmosphere with 5% CO2 at 37 °C. After 24 h, non-adherent cells were removed by washing with PBS and the adherent cells were determined to be the macrophages. 2.7. Immunofluorescence staining and immunohistochemistry The cells cultured on Lab-Tek chamber slides (Nunc, Naperville, IL, USA) were fixed with pre-cold methanol at –20 °C for 3 min and permeabilized with 0.2% TritonX-100 in PBS for 15 min. The cells were blocked by pre-incubation with 10% normal goat serum in PBS at room temperature for 30 min and then incubated with the primary antibodies at 37 °C for 1 h. After washing thrice with PBS, the cells were incubated with the appropriate fluorescein isothiocyanate-conjugated secondary antibodies (Zhongshan Biotechnology Co., Beijing, China) for 30 min. The cells were washed twice with PBS and mounted with a mounting solution for observation under a fluorescence microscope (IX-71; Olympus, Tokyo, Japan).

75

Ovaries from 5-wk-old mice were used for immunohistochemistry studies according to previously described procedures (Wang et al., 2005). Briefly, after 24 h of fixation in Bouin’s solution, the ovaries were embedded in paraffin and cut into 5 lm thick sections. The sections were incubated for 15 min with 1  PBS containing 3% H2O2 to inhibit endogenous peroxidase activity, soaked in citrate buffer, and then microwaved at 100 °C for 10 min to retrieve antigens. After blocking with 5% normal rabbit sera in PBS for 1 h at room temperature, the sections were incubated with the primary antibodies overnight at 4 °C. After rinsing with PBS, the sections were incubated with biotinylated secondary antibodies at room temperature for 30 min. The streptavidin-peroxidase activity was visualized using the diaminobenzidine method. Negative controls were incubated with pre-immune sera instead of the primary antibodies. Sections were counter-stained with hematoxylin and mounted with a mounting solution.

2.8. Real-time quantitative RT-PCR (qRT-PCR) Total RNA was extracted using Trizol™ reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. The RNA was treated with RNase-free DNase 1 (Invitrogen) to remove potential contamination by genomic DNAs. The absence of genomic DNA was confirmed by PCR to amplify b-actin using the RNA obtained before reverse transcription. RNA (1 lg) was reverse transcribed into cDNA using Moloney murine leukemia virus reverse transcriptase (Promega, Madison, WI, USA). PCR was performed with Power SYBRÒ Green PCR master mix kit (Applied Biosystems, Fostercity, CA, USA) using an ABI PRISM 7300 real-time cycler (Applied Biosystems). Transcription levels of the target genes were normalized to b-actin using the comparative threshold cycle method as described in Applied Biosystems User Bulletin No. 2 (P/N 4303859). The sequences of primer pairs used for PCR are listed in Table 2.

2.9. Western blot analysis Cells were lysed using lysis buffer (Applygen Technologies Inc., Beijing, China). Protein concentrations of the cell lysates were determined using a bicinchoninic acid protein assay kit (Pierce Biotechnology, Rockford, IL, USA). Equal amounts of proteins were separated on 10% SDS-PAGE gel and subsequently electrotransferred onto polyvinyl difluoride membranes (Millipore, Bedford, MA, USA). The membranes were blocked for 1 h in Tris-buffered saline (TBS, pH 7.4) containing 5% nonfat milk and incubated with the primary antibodies overnight at 4 °C. The membranes were washed twice with TBS containing 0.1% Tween-20 and then incubated with horseradish peroxidase-conjugated secondary antibodies (Zhongshan Biotechnology Co., Beijing, China) at room temperature for 1 h. Antigen–antibody complexes were visualized using an enhanced chemiluminescence detection kit (Zhongshan Biotechnology Co.).

2.10. Enzyme-linked immunosorbent assay The granulosa cells were cultured in 6-well plates at 37 °C at a density of 5  105 cells/well and then transfected with poly(I:C). At 24 h after transfection, the concentration of cytokines in the culture medium was measured using ELISA kits according to the manufacturer’s instructions: kits for IL-6 (BMS603/2), TNF-a (BMS607/3) and IFN-a (BMS6027) were purchased from eBioscience (San Diego, CA, USA) and the kit for IFN-b (42400) was purchased from R&D Systems (Minneapolis, MN, USA).

76

K. Yan et al. / Molecular and Cellular Endocrinology 372 (2013) 73–85 Table 2 Primers used for real-time PCR. Target genes

b-actin FSHR F4/80 TLR3 TLR7 TLR8 TLR9 TNF-a IL-6 IFN-a IFN-b RIG-I MDA5 MX1 PKR OAS1 ISG15 P450 aromatase

Primer pairs (50 –30 ) Forward

Reverse

GAAATAGTGCGTGACATCAAAG GGCGGCAAACCTCTGAACT AATCGCTGCTGGTTGAATACAAG TTGTCTTCTGCACGAACCTG GGAAATTGCCCTCGATGTTA GAAGCATTTCGAGCATCTCC ACTGAGCACCCCTGCTTCTA CATCTTCTCAAAATTCGAGTGACAA GAGGATACCACTCCCCAACAGACC GACCTCCACCAGCAGCTCCAA GACGTGGGAGATGTCCTCAAC CCACCTACATCCTCAGCTACATGA AGATCAACACCTGTGGTAACACC GACCATAGGGGTCTTGACCAA GGAAAATCCCGAACAAGGAG ATTACCTCCTTCCCGACACC CCAGTCTCTGACTGTGAGAGC CGGAGGAATGCACAGGCTCGAG

TGTAGTTTCATGGATGCCACAG CCAGGCTGAGTCATATCATCAATATC CCAGGCAAGGAGGACAGAGTT CGCAACGCAAGGATTTTATT CAAAAATTTGGCCTCCTCAA GAAGACGATTTCGCCAAGAG AGATTAGTCAGCCGGCAGGAA TGGGAGTAGACAAGGTACAACCC AAGTGCATCATCGTTGTTCATACA ACCCCCACCTGCTGCAT GGTACCTTTGCACCCTCCAGTA TGGGCCCTTGTTGTTCTTCT CTCTAGGGCCTCCACGAACA AGACTTGCTCTTTCTGAAAAGCC CCCAAAGCAAAGATGTCCAC CAAACTCCACCTCCTGATGC GCATCACTGTGCTGCTGGGAC CGATGTACTTCCCAGCACAGC

2.11. Statistical analysis Data represent the mean ± standard error of the mean (SEM). Statistical significance between individual comparisons was determined using Student’s t test. For multiple comparisons, one-way ANOVA test with Bonferroni’s correction was used. The calculations were performed with the statistical software SPSS version 11.0. Values of p < 0.05 indicate statistical significance. 3. Results 3.1. Identification of ovarian granulosa cells The primary granulosa cells and peritoneal macrophages were identified by examining the expression of follicle-stimulating hormone receptor (FSHR), a marker of granulosa cells (Camp et al., 1991), and F4/80, a marker of macrophages (Hume et al., 1984). Real-time qRT-PCR results showed that granulosa cells express

FSHR (Fig. 1A, left panel). By contrast, FSHR mRNA was not detected in peritoneal macrophages from the same mice. Moreover, FSHR protein was confirmed in the granulosa cells by immunofluorescence staining (Fig. 1A, middle panel), which was not detected in macrophages (Fig. 1A, right panel). Based on staining analysis, the purity of granulosa cells was over 95%. F4/80 mRNA in granulosa cell preparations was <5% of that in macrophages (Fig. 1B, left panel). Immunofluorescence staining using specific antibodies against F4/80 confirmed no macrophage contamination in the granulosa cell preparations (Fig. 1B, middle panel). Whereas, the purity of peritoneal macrophages was over 93% based on the immunofluorescence staining (Fig. 1B, right panel). 3.2. Expression of RIG-I, MDA5 and TLR3 in ovarian granulosa cells To investigate the antiviral response of ovarian granulosa cells, we examined the expression of major viral PRRs, including RIG-I, MDA5, TLR3, TLR7, TLR8 and TLR9, in granulosa cells in comparison

Fig. 1. Identification of granulosa cells (Gc) and macrophages (Mu). (A) Expression of follicle-stimulating hormone receptor (FSHR). Total RNA was extracted from primary mouse ovarian Gc and peritoneal Mu and analyzed for relative mRNA levels of FSHR using real-time quantitative RT-RCR (qRT-PCR) by normalizing to internal b-actin (left panel). Immunofluorescence staining using specific antibodies against mouse FSHR was performed in Gc (middle panel) and Mu (right panel). (B) Expression of F4/80. Relative mRNA levels of F4/80 in Gc and Mu were determined using real-time qRT-PCR (left panel). F4/80 protein was confirmed using immunofluorescence staining with anti-F4/80 antibodies. Cells were co-stained with 40 ,6-diamidino-2-phenylindol to show nuclei. The insets in the upper right corners of the images are negative controls. Data of qRT-PCR indicate the mean ± SEM of three experiments. Images are representatives of at least three experiments. Bar = 20 lm.

K. Yan et al. / Molecular and Cellular Endocrinology 372 (2013) 73–85

77

Fig. 2. Expressions of pattern recognition receptors (PRRs) that recognize viral nucleic acids. (A) mRNA levels of the PRRs. Total RNA was extracted from granulosa cells (Gc) and macrophages (Mu). Relative mRNA levels of the indicated PRRs were determined using real-time qRT-PCR by normalizing to b-actin. (B) Protein levels of PRRs. The cell lysates were analyzed by Western blot using specific antibodies to probe retinoic acid-inducible gene I (RIG-I), melanoma differentiation-associated gene 5 (MDA5), and Tolllike receptor 3 (TLR3). b-actin was used as a loading control. (C) Distribution of RIG-I, MDA5, and TLR3 proteins. Immunofluorescence staining was performed in primary Gc using specific antibodies. Cells were co-stained with 40 ,6-diamidino-2-phenylindol. (D) Localization of RIG-I, MDA5, and TLR3 in the ovary. Immunohistochemistry was performed in paraffin sections of the ovary using specific antibodies. Asterisk, arrows, and arrowheads indicate oocytes, Gc, and stroma cells, respectively. Insets in the upper right corners of the images represent negative controls for immunostaining. Images are representatives of at least three experiments. Data indicate the mean ± SEM of three experiments. Bar = 20 lm.

with peritoneal macrophages that are known to express all of these PRRs. Real-time qRT-PCR showed that granulosa cells express relatively high levels of RIG-I, MDA5 and TLR3, and low levels of TLR7, TLR8 and TLR9 compared with macrophages (Fig. 2A). We thus focused on the function of RIG-I, MDA5 and TLR3 in granulosa cells. High levels of the three viral sensors in granulosa cells were confirmed by Western blot (Fig. 2B). Immunofluorescence staining showed that RIG-I, MDA5 and TLR3 are localized in the cytoplasm (Fig. 2C). Immunohistochemistry confirmed that RIG-I and MDA5 are predominantly located in granulosa cells (arrows), whereas TLR3 is detected in granulosa cells and some stroma cells (arrowheads) in the ovary (Fig. 2D). RIG-I is also detected in oocytes (asterisk) (Fig. 2D, left panel). 3.3. Poly(I:C) induces activation of NF-jB and IRF3 in granulosa cells To determine whether or not TLR3, RIG-I and MDA5 are functional in ovarian granulosa cells, we examined the TLR3 and RIGI/MDA5 signaling pathways in cells after transfection with poly(I:C). TLR3 and RIG-I/MDA5 initiate different signaling pathways that converge at the activation of NF-jB and IRF3 (Swiecki

et al., 2011). NF-jB and IRF3 activation can be characterized by translocation of the molecules from the cytoplasm into the nucleus. Immunofluorescence staining showed that poly(I:C) efficiently induces the translocation of NF-jB and IRF3 into the nuclei of granulosa cells in a time-dependent manner (Fig. 3A). The nuclear translocations of NF-jB and IRF3 were observed at 2 h and peaked at 3 h post poly(I:C) transfection (Fig. 3B, left panel). The percentages of NF-jB- and IRF3-positive nuclei sharply decreased 4 h after transfection. Moreover, poly(I:C) induced the nuclear translocation of NF-jB and IRF3 in a dose-dependent manner (Fig. 3B, right panel). The plateaus of nuclear translocation were observed 3 h after transfection with 2 and 5 lg/ml poly(I:C). In the control, the transfection reagent alone did not induce nuclear translocation of NF-jB and IRF3. Phosphorylation of NF-jBp65 (p65) and IRF3 is required for the activation and subsequently nuclear translocation of the transcription factors. The time-dependent phosphorylation of p65 and IRF3 in granulosa cells after poly(I:C) transfection was detected by Western blot (Fig. 3C, left panel). Peak levels of phosphorylated p65 and IRF3 were observed 3 h after transfection (Fig. 3C, right panel), time-corresponding with the peak efficiencies of nuclear transloca-

78

K. Yan et al. / Molecular and Cellular Endocrinology 372 (2013) 73–85

Fig. 3. Activation of interferon regulatory factor 3 (IRF3) and nuclear factor kappa B (NF-jB). (A) Translocation of IRF3 and NF-jBp65 (p65) from the cytoplasm to nuclei. Granulosa cells (Gc) were transfected with 2 lg/ml poly(I:C) using lipofectamine RNAiMAX for the indicated durations. Cells that were treated with RNAiMAX alone for 3 h were used as a control. Distributions of IRF3 and p65 were determined by immunofluorescence staining using specific antibodies against IRF3 (upper panels) and p65 (lower panels). (B) Quantitative analysis of the nuclear translocation efficiency of IRF3 and p65. The percentages of positive nuclei were calculated based on immunofluorescence staining on the cells after transfection with poly(I:C) for the indicated durations (left panel) or with increasing doses of poly(I:C) for 3 h (right panel). (C) Phosphorylation of p65 and IRF3. Gc were transfected with poly(I:C) for the specific durations. Cells that were transfected with RNAiMAX alone for 3 h served as a control. Cell lysates were analyzed via Western blot using specific antibodies against phosphor-p65 (p-p65), p65, phosphor-IRF3, and IRF3. The specificities of the antibodies were assessed using macrophages (data not shown). b-actin was used as loading controls for Western blot analyses. Images are representatives of three separate experiments. Data represent the mean ± SEM of three experiments. Bar = 20 lm.

tion. In the control, the transfection reagent did not induce phosphorylation of p65 and IRF3 in granulosa cells. 3.4. Poly(I:C)-induced cytokine production Given that the PRR-mediated activation of NF-jB and IRF3 induces inflammatory cytokines and IFN-a/b, we examined the expression of TNF-a, IL-6 and IFN-a/b in granulosa cells. Real-time qRT-PCR results showed that poly(I:C) markedly induces the expressions of TNF-a and IL-6 in a time-dependent manner (Fig. 4A). Peak mRNA levels were detected 6 h after poly(I:C) treatment. Similarly, mRNA levels of IFN-a/b were significantly up-regulated in the granulosa cells by poly(I:C) in a time-dependent manner (Fig. 4B). Notably, the exogenously presence of poly(I:C) in culture media without transfection agent induces 6- to 8-fold less of cytokine mRNA levels in granulosa cells compared with

those in the cells transfected with poly(I:C). Therefore, we performed experiments with poly(I:C) transfection in this study. ELISA results confirmed that cytokine levels in the media of granulosa cells are increased in a time-dependent manner after poly(I:C) transfection (Fig. 4C). Significant levels of the cytokines were detected in the culture media 12 h after transfection. The peak levels of TNF-a and IL-6 were detected 18 h after transfection, whereas IFN-a/b levels peaked at 24 h. By contrast, the transfection reagent did not induce cytokine production by granulosa cells (data not shown). 3.5. Effects of poly(I:C) on the expression of antiviral proteins and P450 aromatase Given that a subset of IFN-inducible antiviral proteins, including 20 -50 -oligoadenylate synthetase (OAS1), Mx GTPase 1 (MX1), IFN-

K. Yan et al. / Molecular and Cellular Endocrinology 372 (2013) 73–85

79

transfection (Fig. 5B). By contrast, PKR protein was significantly expressed at basal condition in granulosa cells, and comparable before and after poly(I:C) transfection. Given that LPS/TLR4 signaling modulates endocrine function in bovine granulosa cells (Herath et al., 2007), we examined the impact of poly(I:C) on the expression of genes associated with endocrine function, including FSHR, LHR and P450 aromatase. We found that poly(I:C) markedly suppresses expression of P450 aromatase in a time-dependent manner (Fig. 5C). By contrast, the expression of FSHR and LHR was not significantly affected by poly(I:C) (data not shown).

3.6. Involvement of NF-jB and IRF3 activation in poly(I:C)-induced gene expression Signaling pathways that are responsible for poly(I:C)-induced gene expression in ovarian granulosa cells were then examined. Poly(I:C)-induced phosphorylation of p65 and IRF3 in granulosa cells was efficiently blocked by pre-incubation of the cells with the appropriate inhibitors (BAY11-7082 for NF-jB activation and BX795 for IRF3 activation) (Fig. 6A). The presence of inhibitors differentially decreased poly(I:C)-induced TNF-a, IL-6 and IFN-a/b secretion in the media (Fig. 6B). BAY11-7082 significantly decreased poly(I:C)-induced TNF-a and IL-6 levels; however, it did not affect IFN-a/b secretion. By contrast, BX795 markedly inhibited the secretion of IFN-a/b and did not affect TNF-a and IL-6. Poly(I:C)-induced up-regulation of OAS1, MX1 and ISG15 was significantly inhibited by BX795 but not by BAY11-7082 (Fig. 6C). In the control, the inhibitors did not affect cytokine production and antiviral protein expression at basal conditions. MTT assay results showed that the cell viability was not significantly affected by poly(I:C) and chemical inhibitors (Fig. 6D). These results indicate that poly(I:C) induces TNF-a and IL-6 expression through NF-jB activation and induces IFN-a/b and antiviral proteins through IRF3 activation.

3.7. Cooperation of TLR3 and RIG-I/MDA5 signaling in inducing antiviral responses

Fig. 4. Poly(I:C)-induced expression of cytokines. (A, B) Poly(I:C) was internalized by transfection of granulosa cells with 2 lg/ml poly(I:C) [Int poly(I:C)], or exogenously presented in culture medium without transfection agent [Exo Poly(I:C)] for the indicated durations. Total RNAs were extracted. The relative mRNA levels of TNF-a and IL-6 (A) and IFN-a/b (B) were analyzed using real-time qRT-PCR by normalizing to b-actin. (C) Cytokine secretion in media. Gc were cultured in 6-well plates at 5  105 cells/well and transfected with poly(I:C). The culture media were subjected to ELISA to determine cytokine levels at the indicated time points after transfection. Data indicate the mean ± SEM of three experiments.

stimulated gene 15 (ISG15) and dsRNA-activated protein kinase (PKR), play crucial roles in restricting virus replication within cells (Sadler and Williams, 2008), the effects of poly(I:C) on the expression of these antiviral proteins in granulosa cells were examined. Real-time qRT-PCR results showed that poly(I:C) significantly upregulates OAS1, MX1 and ISG15 at mRNA levels in a time-dependent manner (Fig. 5A). By contrast, PKR was slightly up-regulated by poly(I:C). The peak mRNA levels of the antiviral proteins were detected in cells 8 h after poly(I:C) transfection. Western blot analysis showed that the protein levels of OAS1, MX1 and ISG15 were significantly increased in granulosa cells 24 h after poly(I:C)

TLR3 and RIG-I/MDA5 trigger different signaling pathways to mediate antiviral responses (Thompson et al., 2011). To study the contribution of TLR3 and RIG-I/MDA5 signaling to poly(I:C)-induced innate antiviral responses in ovarian granulosa cells, we analyzed the effects of poly(I:C) on cytokine and antiviral protein expression in granulosa cells of TLR3–/– mice. Expression of MDA5 and RIG-I at both mRNA and protein levels was similar in TLR3–/– and wild-type (WT) granulosa cells (Fig. 7A). ELISA results showed that poly(I:C)-induced TNF-a and IL-6 secretion is significantly reduced in TLR3–/– granulosa cells compared with WT cells (Fig. 7B). BAY11-7082 significantly inhibited poly(I:C)-induced TNF-a and IL-6 secretion by both TLR3–/– and WT cells. Secretion of IFN-a/b by TLR3–/– cells was significantly decreased compared with WT cells 18 h after poly(I:C) transfection (Fig. 7C). BX795 significantly inhibited poly(I:C)-induced IFN-a/b secretion by WT cells and completely abolished its production by TLR3–/– cells. In consistent with cytokine production, poly(I:C)-induced mRNA levels of ISG15, OAS1 and MX1 in TLR3–/– granulosa cells were significantly low compared with WT cells (Fig. 7D). By contrast, comparable PKR mRNA levels were observed in WT and TLR3–/– ovarian granulosa cells. BX795 almost completely inhibited the poly(I:C)-induced expression of OAS1, MX1 and ISG15 in TLR3–/– granulosa cells (Fig. 7E). These results suggest that TLR3 and MDA5/RIG-I signaling pathways synergically mediate innate antiviral responses in mouse ovarian granulosa cells.

80

K. Yan et al. / Molecular and Cellular Endocrinology 372 (2013) 73–85

Fig. 5. Effects of poly(I:C) on expression of antiviral proteins and P450 aromatase. (A) Relative mRNA levels of antiviral proteins. Granulosa cells (Gc) were transfected with 2 lg/ml poly(I:C) using lipofectamine RNAiMAX for the specified durations. Real-time qRT-RCR was used to analyze relative mRNA levels of 20 -50 -oligoadenylate synthetase (OAS1), Mx GTPase 1 (MX1), IFN-stimulated gene 15 (ISG15), and dsRNA-activated protein kinase (PKR). b-actin was used as a control. (B) Western blot analyses of the antiviral proteins. Gc were transfected with 2 lg/ml poly(I:C) or treated with RANiMAX alone (control). At 24 h after transfection, cell lysates were subjected to Western blot using specific antibodies against the antiviral proteins (left panel). b-actin was used as a loading control. Poly(I:C)-induced antiviral proteins were quantitatively determined by analyzing band densities (right panel). (C) Inhibition of P450 aromatase expression. Gc were transfected with poly(I:C) for the indicated durations, and mRNA level of P450 aromatase was determined by real-time quantitative RT-PCR. Images are the representatives of at least three experiments. Data indicate the mean ± SEM of three experiments. p < 0.01.

3.8. Role of MDA5 and RIG-I in poly(I:C)-induced antiviral responses

4. Discussion

To assess the role of individual MDA5 and RIG-I, we examined the effects of poly(I:C) on TLR3–/– granulosa cells after silence of MDA5 or RIG-I using specific small interfering RNA (siRNA). We demonstrated that each siRNA reduced >70% target mRNA at 24 h post transfection (Fig. 8A). Dramatical reduction of target proteins by siRNA was confirmed (Fig. 8B). Therefore, the cells were stimulated with poly(I:C) at 24 h after siRNA transfection. Each siRNA targeting MDA5 or RIG-I significantly reduced the mRNA levels of cytokines (Fig. 8C). The transfection of the cells with siRNAs did not affect the cell viability (data not shown). These results indicate that MDA5 and RIG-I cooperatively mediate cytokine expression in granulosa cells in response to poly(I:C) stimulation.

Viral infections in the ovary may result in various pathological conditions, such as premature ovarian failure (Sinha and Kuruba, 2007), impairment of ovarian function (Fray et al., 2000; Zhao et al., 2011), and even transmission of pathogens to the fetus (Jin et al., 2011). Understanding the mechanisms underlying innate antiviral responses in the ovary may provide novel clues into the strategies for developing therapeutic and preventive approaches against viral infections. The current study demonstrates that TLR3, RIG-I and MDA5 are constitutively expressed in the mouse ovarian granulosa cells and cooperate in mediating innate antiviral responses, which could contribute to the protection of the ovary from viral infections.

K. Yan et al. / Molecular and Cellular Endocrinology 372 (2013) 73–85

81

Fig. 6. Involvement of nuclear factor kappa B (NF-jB) and interferon regulatory factor 3 (IRF3) activation in poly(I:C)-induced gene expression. (A) Inhibition of NF-jB and IRF3 activation by appropriate inhibitors. Granulosa cells (Gc) were transfected with 2 lg/ml poly(I:C) alone or after 2 h of pre-incubation with inhibitors [1 lg/ml BX795 for IRF3, 10 lM BAY11-7082 (BAY) for NF-jB]. At 3 h after transfection, cell lysates were subjected to Western blot using specific antibodies to probe the indicated proteins. (B) secretion of cytokines. Gc were cultured in 6-well plates at a density of 5  105 cells/well, and treated as in (A). At 18 h after transfection, cytokine concentrations in the culture media were measured by ELISA. ND, not detectable. (C) Expression of antiviral proteins. Gc were treated as in (A) for 8 h. Relative mRNA levels of OAS1, MX1, and ISG15 were determined using real-time qRT-PCR by normalizing to b-actin. (D) Cell viability. Gc were treated with the indicated conditions. After treatment with MTT assay agents, the cells were lyzed and the absorbance at 570 nm was recorded. Images are the representatives of at least three experiments. Data represent the mean ± SEM of three experiments. p < 0.01.

82

K. Yan et al. / Molecular and Cellular Endocrinology 372 (2013) 73–85

Fig. 7. Comparison of poly(I:C)-induced gene expressions in wild-type (WT) and TLR3 knockout (TLR3–/–) cells. (A) Expression of TLR3, MDA5 and RIG-I. Relative mRNA (left panel) and protein (right panel) levels in primary WT and TLR3–/– granulosa cells were analyzed using real-time qRT-PCR and Western blotting. (B) Secretion of TNF-a and IL6. Granulosa cells (Gc) from WT and TLR3–/– mice were cultured in 6-well plates at a density of 5  105 cells/well, and transfected with 2 lg/ml poly(I:C) alone or with the poly(I:C) 2 h after pre-incubation with 10 lM BAY11-7082 (BAY). At 18 h after transfection, the concentrations of TNF-a and IL-6 in the culture media were measured by ELISA. (C) Secretion of IFN-a and IFN-b. Cells were transfected with 2 lg/ml poly(I:C) alone or with the poly(I:C) 2 h after pre-incubation with 1 lg/ml BX795 for 18 h. IFN-a and IFN-b levels in the media were measured by ELISA. (D) Comparison of antiviral protein expression in WT and TLR3–/– cells. Cells were transfected with 2 lg/ml poly(I:C) for 6 h. Relative mRNA levels of OAS1, MX1, ISG15, and PKR were determined using real-time qRT-PCR. (E) Inhibition of antiviral protein gene expression in TLR3–/– cells by BX795. TLR3–/– Gc were transfected with 2 lg/ml poly(I:C) alone or 2 h after pre-incubation with 1 lg/ml BX795 for 8 h. Relative mRNA levels of OAS1, MX1, and ISG15 were analyzed using real-time qRT-PCR. b-actin was used as control for real-time qRT-PCR and Western blot. Data indicate the mean ± SEM of three independent experiments.  p < 0.05, p < 0.01.

PRR-triggered innate immune responses in immune and nonimmune cells to microbial pathogens play critical roles in building the first lines of defense against invading microbes (Akira et al., 2006). TLRs and RLRs represent two distinct families of PRRs (Kawai and Akira, 2008). TLRs are the best-characterized PRRs that recognize various PAMPs of microbes, including bacteria and viruses (Beutler, 2009). TLR3, TLR7, TLR8 and TLR9 are localized to the endosome membranes and recognize viral dsRNA, ssRNA and DNA (Kawai and Akira, 2006). The RLR family, which is composed of RIG-I, MDA5 and LGP2, represents cytoplasmic PRRs that

recognize viral dsRNA (Takeuchi and Akira, 2008). Since LGP2 lacks a domain that is responsible for triggering signaling pathways, it does not induce innate immune responses (Rothenfusser et al., 2005). In this study, we focused on the roles of TLR3, RIG-I and MDA5 in mediating ovarian innate antiviral responses because they are co-expressed in granulosa cells. Notably, we found that TLR3 is also expressed in some stroma cells and that RIG-I is abundantly expressed in oocytes. The roles of TLR3 and RIG-I in mediating innate antiviral responses in these ovarian cells should be worthy of investigation in the future.

K. Yan et al. / Molecular and Cellular Endocrinology 372 (2013) 73–85

83

Fig. 8. Role of MDA5 and RIG-I in mediating innate antiviral responses. (A, B) Reduction of MDA5 and RIG-I expression by specific small interfering RNA (siRNA). Granulosa cells from TLR3–/– mice were transfected with siRNA for MDA5 (siMDA5) or RIG-I (siRIG-I) or control (siCtrl). MDA5 and RIG-I mRNAs (A), and proteins (B) were assessed 24 h post transfection by real-time qRT-PCR and Western blotting. (C) Expression of cytokines. TLR3–/– granulosa cells were transfected with respective siRNAs. At 24 h post siRNA transfection, the cells were transfected with 2 lg/ml poly(I:C) for 6 h. Expression of the indicated cytokines was examined by real-time qRT-PCR. The images of Western blots are the representatives of at least three experiments. b-actin was used as control for real-time qRT-PCR and Western blot. Data represent the mean ± SEM of at least three seperate experiments. p < 0.05, p < 0.01.

Following the recognition of agonists, TLR3 and RIG-I/MDA5 initiate different signaling pathways that induce the expression of inflammatory cytokines, such as TNF-a, IL-6 and IFN-a/b (Thompson et al., 2011). IFN-a/b play critical roles in eliminating invading viruses through up-regulation of antiviral proteins in infected cells and promotion of adaptive immune responses against viruses (Theofilopoulos et al., 2005). We found that poly(I:C) efficiently activates both NF-jB and IRF3 in primary ovarian granulosa cells, thus resulting in induction of TNF-a, IL-6 and IFN-a/b. Notably, transfection of granulosa cells with poly(I:C) induced much stronger responses than the exogenous presence of poly(I:C) in cultures, agreeing with previous reports showing that internalization of poly(I:C) promotes innate immune responses in various cell types (Li et al., 2005; Silhol et al., 1986; Starace et al., 2008). Therefore, we transfected granulosa cells with poly(I:C) to induce strong innate antiviral responses. IFN-a/b facilitate innate cellular antiviral responses by inducing a set of antiviral proteins that inhibit viral replication at multiple levels (Sadler and Williams, 2008). The four best-characterized IFN-inducible antiviral proteins are ISG15, OAS1, PKR and MX1, which individually amplify antiviral signaling, block viral transcription, degrade viral RNA and inhibit translation of viral proteins (MacMicking, 2004). We found that PKR is constitutively expressed in granulosa cells and not evidently up-regulated by poly(I:C). By contrast, ISG15, OAS1 and MX1 were faintly expressed in granulosa cells at basal conditions and highly induced by poly(I:C). Given that all of these antiviral proteins belong to IFN-inducible genes (Sadler and Williams, 2008), it is reasonable to speculate that poly(I:C)-induced IFN-a/b may up-regulate antiviral proteins in granulosa cells. Notably, mRNA levels of ISG15, OAS1 and MX1 are evidently up-regulated at 4 and 6 h after poly(I:C) treatment (Fig. 5A), when IFN-a/b are not detectable in the medium (Fig. 4C). These data suggest that poly(I:C) may induce antiviral proteins in ovarian granulosa cells in an IFN-independent manner at early time after poly(I:C) treatment. Clarification of IFN-independent pathways

by which poly(I:C) induces antiviral protein expression in granulosa cells would be an interesting endeavor. TNF-a and IL-6 are potent regulators of inflammation. These cytokines are broadly involved in the regulation of some physiological events at low levels. IL-6 plays an important role in coordinating meiotic maturation and cumulus cell-oocyte complex expansion (Liu et al., 2009). TNF-a and IL-6 are associated with ovulation (Willis et al., 2003). However, high levels of pro-inflammatory cytokines may result in pathological conditions, such as ovarian cancer (Maccio and Madeddu, 2012). Substantial evidence shows that chronic production of high levels of TNF-a facilitates ovarian carcinogenesis (Charles et al., 2009). Moreover, increased TNF-a and IL-6 levels may interfere with steroidogenesis (Bornstein et al., 2004). In this study, we showed that poly(I:C) dramatically down-regulates expression of P450 aromatase, a key enzyme for estradiol synthesis. This observation corresponds to a previous study showing that TLR4-mediated innate immune responses suppress oestradiol synthesis associated with down-regulation of P450 aromatase in bovine granulosa cells (Herath et al., 2007). The effects of PRR-mediated TNF-a and IL-6 up-regulation on the pathophysiology of the ovary are thus worthy of consideration. Although TLR3, RIG-I and MDA5 recognize a common agonist dsRNA, they trigger different signaling pathways (Kawai and Akira, 2008). TLR3 is a membrane-bound receptor that is localized to the endosomes. Activation of TLR3 upon recognition of dsRNA initiates Toll/Il-1 receptor domain-containing adaptor inducing IFN-b dependent pathway. By contrast, both RIG-I and MDA5 are cytosolic viral sensors that initiate a common IFN-b promoter stimulator-1 pathway (Bruns and Horvath, 2012). TLR3 and RIG-I/MDA5 signaling pathways culminate in the phosphorylation of IRF3 through activation of TBK1 and IjB kinase, thus inducing IFN-a/b expression. To study the contributions of TLR3 and RIG-I/MDA5 signaling to antiviral responses in granulosa cells, we compared the effects of poly(I:C) on the initiation of innate antiviral responses in WT and TLR3–/– granulosa cells. We found that

84

K. Yan et al. / Molecular and Cellular Endocrinology 372 (2013) 73–85

poly(I:C)-induced antiviral responses in TLR3–/– granulosa cells are significantly lower than those in WT cells (Fig. 7). In particular, BX795 (a specific inhibitor of TBK1) significantly inhibited poly(I:C)-induced antiviral responses in WT granulosa cells and completely abolished responses in TLR3–/– cells. These results suggest that the TLR3 and RIG-I/MDA5 signaling pathways cooperate in mediating innate antiviral responses in mouse ovarian granulosa cells. Further, knockdown of individual MDA5 and RIG-I using specific siRNA significantly reduced poly(I:C)-induced cytokine expression in TLR3–/– cells, suggesting that both MDA5 and RIG-I are functional in mediating innate immune responses to poly(I:C). Notably, we found that the up-regulation of IFN-a/b and antiviral proteins depends on IRF3 activation, whereas TNF-a and IL-6 production is attributable to NF-jB activation. Manipulation of IRF3 and NF-jB activation may amplify antiviral responses and limit pro-inflammatory cytokine production. In conclusion, we provide evidence that mouse ovarian granulosa cells are equipped with innate antiviral system. TLR3, RIG-I and MDA5 are co-expressed in the granulosa cells, and cooperate in mediating innate antiviral responses. Data show a possible role of granulosa cells in the protection of ovarian follicles from viral infections. Notably, natural viral infections in the ovary were mostly reported in human, rather than in murine. It should be an interesting topic to clarify whether differential ovarian innate antiviral states exist in human and murine. A comprehensive analysis by means of in vivo viral infectious models will further clarify the innate antiviral state in the ovary and its impact on ovarian function. We are now investigating this issue in vivo. Disclosure summary The authors have nothing to disclose. Acknowledgement This work was supported by the National Natural Science Foundation of China (Grant Nos. 30971459, 31071270, 31171445). References Akira, S., Uematsu, S., Takeuchi, O., 2006. Pathogen recognition and innate immunity. Cell 124, 783–801. Beutler, B.A., 2009. TLRs and innate immunity. Blood 113, 1399–1407. Bornstein, S.R., Rutkowski, H., Vrezas, I., 2004. Cytokines and steroidogenesis. Mol. Cell Endocrinol. 215, 135–141. Bromfield, J.J., Sheldon, I.M., 2011. Lipopolysaccharide initiates inflammation in bovine granulosa cells via the TLR4 pathway and perturbs oocyte meiotic progression in vitro. Endocrinology 152, 5029–5040. Bruns, A.M., Horvath, C.M., 2012. Activation of RIG-I-like receptor signal transduction. Crit. Rev. Biochem. Mol. Biol. 47, 194–206. Camp, T.A., Rahal, J.O., Mayo, K.E., 1991. Cellular localization and hormonal regulation of follicle-stimulating hormone and luteinizing hormone receptor messenger RNAs in the rat ovary. Mol. Endocrinol. 5, 1405–1417. Charles, K.A., Kulbe, H., Soper, R., Escorcio-Correia, M., Lawrence, T., Schultheis, A., Chakravarty, P., Thompson, R.G., Kollias, G., Smyth, J.F., Balkwill, F.R., Hagemann, T., 2009. The tumor-promoting actions of TNF-alpha involve TNFR1 and IL-17 in ovarian cancer in mice and humans. J. Clin. Invest. 119, 3011–3023. Chong, M.M., Metcalf, D., Jamieson, E., Alexander, W.S., Kay, T.W., 2005. Suppressor of cytokine signaling-1 in T cells and macrophages is critical for preventing lethal inflammation. Blood 106, 1668–1675. Fray, M.D., Mann, G.E., Clarke, M.C., Charleston, B., 2000. Bovine viral diarrhoea virus: its effects on ovarian function in the cow. Vet. Microbiol. 77, 185–194. Herath, S., Williams, E.J., Lilly, S.T., Gilbert, R.O., Dobson, H., Bryant, C.E., Sheldon, I.M., 2007. Ovarian follicular cells have innate immune capabilities that modulate their endocrine function. Reproduction 134, 683–693. Hu, X.L., Zhou, X.P., Qian, Y.L., Wu, G.Y., Ye, Y.H., Zhu, Y.M., 2011. The presence and expression of the hepatitis B virus in human oocytes and embryos. Hum. Reprod. 26, 1860–1867. Hume, D.A., Halpin, D., Charlton, H., Gordon, S., 1984. The mononuclear phagocyte system of the mouse defined by immunohistochemical localization of antigen F4/80: macrophages of endocrine organs. Proc. Natl. Acad. Sci. USA 81, 4174– 4177.

Iwasaki, A., Medzhitov, R., 2010. Regulation of adaptive immunity by the innate immune system. Science 327, 291–295. Jin, Y., Ye, F., Shi, J., Qiu, H., Zhao, Y., Lin, S., Chen, T., Liu, M., He, Y., Zhang, S., 2011. Hepatitis B virus infection and replication in primary cultured human granulosa cells. Arch. Virol. 156, 1–7. Kawai, T., Akira, S., 2006. Innate immune recognition of viral infection. Nat. Immunol. 7, 131–137. Kawai, T., Akira, S., 2008. Toll-like receptor and RIG-I-like receptor signaling. Ann. NY Acad. Sci. 1143, 1–20. Kumar, H., Kawai, T., Akira, S., 2011. Pathogen recognition by the innate immune system. Int. Rev. Immunol. 30, 16–34. Li, K., Chen, Z., Kato, N., Gale Jr., M., Lemon, S.M., 2005. Distinct poly(I-C) and virusactivated signaling pathways leading to interferon-beta production in hepatocytes. J. Biol. Chem. 280, 16739–16747. Li, M., Zhou, Y., Feng, G., Su, S.B., 2009. The critical role of Toll-like receptor signaling pathways in the induction and progression of autoimmune diseases. Curr. Mol. Med. 9, 365–374. Liu, Z., Shimada, M., Richards, J.S., 2008. The involvement of the Toll-like receptor family in ovulation. J. Assist. Reprod. Genet. 25, 223–228. Liu, Z., de Matos, D.G., Fan, H.Y., Shimada, M., Palmer, S., Richards, J.S., 2009. Interleukin-6: an autocrine regulator of the mouse cumulus cell-oocyte complex expansion process. Endocrinology 150, 3360–3368. Maccio, A., Madeddu, C., 2012. Inflammation and ovarian cancer. Cytokine 58, 133– 147. MacMicking, J.D., 2004. IFN-inducible GTPases and immunity to intracellular pathogens. Trends Immunol. 25, 601–609. Martinon, F., Mayor, A., Tschopp, J., 2009. The inflammasomes: guardians of the body. Annu. Rev. Immunol. 27, 229–265. Matsumoto, M., Oshiumi, H., Seya, T., 2011. Antiviral responses induced by the TLR3 pathway. Rev. Med. Virol. 21, 67–77. Petrovska, M., Dimitrov, D.G., Michael, S.D., 1996. Quantitative changes in macrophage distribution in normal mouse ovary over the course of the estrous cycle examined with an image analysis system. Am. J. Reprod. Immunol. 36, 175–183. Price, J.C., Cronin, J., Sheldon, I.M., 2012. Toll-Like Receptor Expression and Function in the COV434 Granulosa Cell Line. Am. J. Reprod. Immunol. 68, 205–217. Rothenfusser, S., Goutagny, N., DiPerna, G., Gong, M., Monks, B.G., Schoenemeyer, A., Yamamoto, M., Akira, S., Fitzgerald, K.A., 2005. The RNA helicase Lgp2 inhibits TLR-independent sensing of viral replication by retinoic acid-inducible gene-I. J. Immunol. 175, 5260–5268. Sadler, A.J., Williams, B.R., 2008. Interferon-inducible antiviral effectors. Nat. Rev. Immunol. 8, 559–568. Samuel, C.E., 2001. Antiviral actions of interferons. Clin. Microbiol. Rev. 14, 778– 809. Serke, H., Vilser, C., Nowicki, M., Hmeidan., F.A., Blumenauer, V., Hummitzsch, K., Losche, A., Spanel-Borowski, K., 2009. Granulosa cell subtypes respond by autophagy or cell death to oxLDL-dependent activation of the oxidized lipoprotein receptor 1 and toll-like 4 receptor. Autophagy 5, 991–1003. Sheldon, I.M., Bromfield, J.J., 2011. Innate immunity in the human endometrium and ovary. Am. J. Reprod. Immunol. 66 (suppl 1), 63–71. Shimada, M., Hernandez-Gonzalez, I., Gonzalez-Robanya, I., Richards, J.S., 2006. Induced expression of pattern recognition receptors in cumulus oocyte complexes: novel evidence for innate immune-like functions during ovulation. Mol. Endocrinol. 20, 3228–3239. Shimada, M., Yanai, Y., Okazaki, T., Noma, N., Kawashima, I., Mori, T., Richards, J.S., 2008. Hyaluronan fragments generated by sperm-secreted hyaluronidase stimulate cytokine/chemokine production via the TLR2 and TLR4 pathway in cumulus cells of ovulated COCs, which may enhance fertilization. Development 135, 2001–2011. Silhol, M., Huez, G., Lebleu, B., 1986. An antiviral state induced in HeLa cells by microinjected poly(rI).poly(rC). J. Gen. Virol. 67 (9), 1867–1873. Sinha, P., Kuruba, N., 2007. Premature ovarian failure. J. Obstet. Gynaecol. 27, 16–19. Starace, D., Galli, R., Paone, A., De Cesaris, P., Filippini, A., Ziparo, E., Riccioli, A., 2008. Toll-like receptor 3 activation induces antiviral immune responses in mouse sertoli cells. Biol. Reprod. 79, 766–775. Swiecki, M., McCartney, S.A., Wang, Y., Colonna, M., 2011. TLR7/9 versus TLR3/ MDA5 signaling during virus infections and diabetes. J. Leukocyte Biol. 90, 691– 701. Takaya, R., Fukaya, T., Sasano, H., Suzuki, T., Tamura, M., Yajima, A., 1997. Macrophages in normal cycling human ovaries; immunohistochemical localization and characterization. Hum. Reprod. 12, 1508–1512. Takeuchi, O., Akira, S., 2008. MDA5/RIG-I and virus recognition. Curr. Opin. Immunol. 20, 17–22. Theofilopoulos, A.N., Baccala, R., Beutler, B., Kono, D.H., 2005. Type I interferons (alpha/beta) in immunity and autoimmunity. Annu. Rev. Immunol. 23, 307– 336. Thompson, M.R., Kaminski, J.J., Kurt-Jones, E.A., Fitzgerald, K.A., 2011. Pattern recognition receptors and the innate immune response to viral infection. Viruses 3, 920–940. Vanderhyden, B.C., Telfer, E.E., Eppig, J.J., 1992. Mouse oocytes promote proliferation of granulosa cells from preantral and antral follicles in vitro. Biol. Reprod. 46, 1196–1204. Wang, H., Chen, Y., Ge, Y., Ma, P., Ma, Q., Ma, J., Xue, S., Han, D., 2005. Immunoexpression of Tyro 3 family receptors – Tyro 3, Axl, and Mer – and their ligand Gas6 in postnatal developing mouse testis. J. Histochem. Cytochem. 53, 1355–1364.

K. Yan et al. / Molecular and Cellular Endocrinology 372 (2013) 73–85 Willis, C., Morris, J.M., Danis, V., Gallery, E.D., 2003. Cytokine production by peripheral blood monocytes during the normal human ovulatory menstrual cycle. Hum. Reprod. 18, 1173–1178. Woods, D.C., Schorey, J.S., Johnson, A.L., 2009. Toll-like receptor signaling in hen ovarian granulosa cells is dependent on stage of follicle maturation. Reproduction 137, 987–996. Wu, R., Van der Hoek, K.H., Ryan, N.K., Norman, R.J., Robker, R.L., 2004. Macrophage contributions to ovarian function. Hum. Reprod. Update 10, 119–133. Yoneyama, M., Fujita, T., 2007. Function of RIG-I-like receptors in antiviral innate immunity. J. Biol. Chem. 282, 15315–15318.

85

Yoneyama, M., Kikuchi, M., Matsumoto, K., Imaizumi, T., Miyagishi, M., Taira, K., Foy, E., Loo, Y.M., Gale Jr., M., Akira, S., Yonehara, S., Kato, A., Fujita, T., 2005. Shared and unique functions of the DExD/H-box helicases RIG-I, MDA5, and LGP2 in antiviral innate immunity. J. Immunol. 175, 2851–2858. Zhao, Y., Adams, Y.F., Croft, M., 2011. Preferential replication of vaccinia virus in the ovaries is independent of immune regulation through IL-10 and TGF-beta. Viral. Immunol. 24, 387–396. Zhou, M., McFarland-Mancini, M.M., Funk, H.M., Husseinzadeh, N., Mounajjed, T., Drew, A.F., 2009. Toll-like receptor expression in normal ovary and ovarian tumors. Cancer Immunol Immunother. 58, 1375–1385.