Poly(dA:dT) Suppresses HSV-2 Infection of Human Cervical Epithelial Cells Through RIG-I Activation

Epithelial cells of the female reproductive tract (FRT) participate in the initial innate immunity against viral infections. Poly(dA:dT) is a synthetic analog of B form double-stranded (ds) DNA which can activate the interferon (IFN) signaling pathway-mediated antiviral immunity through DNA-dependent RNA Polymerase III. Here we investigated whether poly(dA:dT) could inhibit herpes simplex virus type 2 (HSV-2) infection of human cervical epithelial cells (End1/E6E7). We demonstrated that poly(dA:dT) treatment of End1/E6E7 cells could significantly inhibit HSV-2 infection. Mechanistically, poly(dA:dT) treatment of the cells induced the expression of the intracellular IFNs and the multiple antiviral IFN-stimulated genes (ISGs), including IFN-stimulated gene 15 (ISG15), IFN-stimulated gene 56 (ISG56), 2’-5’-oligoadenylate synthetase 1 (OAS1), 2’-5’-oligoadenylate synthetase 2 (OAS2), myxovirus resistance protein A (MxA), myxovirus resistance protein B (MxB), virus inhibitory protein, endoplasmic reticulum-associated, IFN-inducible (Viperin), and guanylate binding protein 5 (GBP5). Further investigation showed that the activation of RIG-I was largely responsible for poly(dA:dT)-mediated HSV-2 inhibition and IFN/ISGs induction in the cervical epithelial cells, as RIG-I knockout abolished the poly(dA:dT) actions. These observations demonstrate the importance for design and development of AT-rich dsDNA-based intervention strategies to control HSV-2 mucosal transmission in FRT.


INTRODUCTION
Herpes simplex virus type 2 (HSV-2) is the leading cause of genital herpes and the most commonly sexual transmitted virus. It is estimated that 417 million people aged 15-49 (11%) worldwide are infected with HSV-2 infection in 2012 (1). Importantly, the epidemiological and biological studies have shown a strong association between human immunodeficiency virus 1 (HIV-1) and HSV-2 infection. HSV-2 infection increases the risk of HIV-1 acquisition by approximately three-fold (2) and facilitates transmissibility of HIV-1 up to five-fold through genital ulcers (3). In turn, HIV-1   infection enhances HSV-2 shedding frequency and quantity (4,5), people with HIV-1-related immunosuppression can have severe HSV-2 diseases (6, 7). Epithelial cells in the female reproductive tract (FRT) are the first barrier to pathogen invasion. At cellular level, epithelial cells constitute a unique microenvironment and participate in FRT innate immunity against viral infections, including HSV-2 (8,9). HSV-2 primarily infects genital epithelium and replicates within the vaginal keratinocytes (10). Human cervical epithelial cells have been extensively used to study FRT-mediated immunity against viral infections (11)(12)(13)(14). Studies have shown that these cells could be immunologically activated and produced the multiple antiviral factors against HSV-2 (12) and HIV-1 (15). As the outmost layer cells in FRT, human cervical epithelial cells are the first to contact with invading microbes. Thus, understanding the processes and mechanisms of these cellsmediated innate immunity against viral infections is of importance and significance.    (20). Type III IFN, also known as IFN lambda (IFN-l), can induce expression of ISGs and exert antiviral properties similar to type I IFNs (21). Studies have shown that activation of IFN-dependent innate immune defense through RIG-I signaling pathway is vital in antiviral response of epithelial cells (22)(23)(24). However, we know little about whether poly(dA:dT) has the ability to activate the intracellular antiviral immunity of human cervical epithelial cells, an essential component of the mucosal defense mechanisms in the FRT. In this study, we examined whether poly(dA:dT) has the ability to induce the intracellular antiviral factors against HSV-2 infection of human cervical epithelial cells. We also explored the cellular and molecular mechanisms underlying poly(dA:dT)-mediated IFN/ISG induction and HSV-2 inhibition in these cells.

Cell Lines and Virus
End1/E6E7 cell line was established from normal human endocervical epithelia immortalized by expression of human papillomavirus 16/E6E7. End1/E6E7 cells have exactly the same cytokeratin and involucrin patterns as primary End1 cells and the morphological and immunocytochemical characteristics of the End1/E6E7 cells closely resembled those of origin and primary cultures. Therefore, End1/E6E7 cells provide the basis for valid reproducible in vitro models for studies on The results were measured as HSV-2 gene levels relative (%) to control (without treatment, which is defined as 100%). Data are shown as mean ± SD of three independent experiments. Asterisks indicate statistically significant differences (*P < 0.05, **P < 0.01). cervicovaginal physiology and infections and for testing pharmacological agents for intravaginal application (25). In addition, End1/E6E7 cell line has been broadly used as an in vitro model of human female reproductive tract (13,26,27). End1/E6E7 cells were cultured in keratinocyte growth medium (Gibco, USA) supplemented with the provided recombinant epidermal growth factor (0.1ng/ml) and bovine pituitary extract (50µg/ml). African green monkey kidney epithelial cells (Vero) and 293T cells were cultured in Dulbecco's modified Eagle's culture medium (DMEM, Gibco) supplemented with 10% fetal bovine serum (FBS, Gibco) at 37°C in a humidified atmosphere of 5% CO 2 . HSV-2 G strain was provided by Dr. Qinxue Hu (State Key Laboratory of Virology, Wuhan Institute of Virology, Chinese Academy of Sciences, China). The HSV-2 G strain was propagated at a low multiplicity of infection (MOI = 0.001) in Vero cells.

Cell Viability Assay
The cytotoxic effect of poly(dA:dT) was evaluated by the MTT assay based on the manufacturer's instruction. End1/E6E7 cells were seeded in 96-well plate (1×10 4 cells/well) treated with different concentrations of poly(dA:dT) for 72h. Cells were then incubated with MTT working solution (0.5mg/ml) for 4h at 37°C in darkness. The formation of soluble formazan from MTT was measured by spectrophotometric determination of absorption at 490nm using a 96-well plate reader (SpectraMax i3, Molecular Devices, Sunnyvale, CA, USA).

Flow Cytometry Analysis of Apoptosis
We used Annexin V/7-AAD assay to measure the apoptosis effect of poly(dA:dT) on End1/E6E7 cells. Cells were seeded in 24-well plate (2×10 5 cells/well) and treated with different concentrations of poly(dA:dT) for 72h. Cells were washed twice with cold phosphate-buffered saline (PBS) and then resuspended in 1×binding buffer at a concentration of 1×10 5 cells/ml. Annexin V-PE (2.5µl) and 7-AAD (5µl) were added and then incubated for 15min at room temperature without light, finally analyzed by FCM (FACScan, Becton Dickinson, San Jose, CA).

In Vitro Antiviral Assay
End1/E6E7 cells were pretreated with poly(dA:dT) for 24h and infected with HSV-2 (MOI = 0.001) for 2h. The cells were then washed to remove unattached viruses and subsequently cultured for 48h. HSV-2 genome DNAs from HSV-2-infected cells and culture supernatant were extracted with DNA lysis buffer as previously described (12) and subjected to the real-time PCR. HSV-2 gD standards with known copy numbers were used to quantify HSV-2 gD copies in the culture supernatant. Total proteins were extracted from End1/E6E7 cells and subjected to the Western blot. In addition, the antiviral effect of poly(dA:dT) under different treatment conditions (before, simultaneously and after HSV-2 infection) was evaluated. Briefly, End1/E6E7 cells were pretreated with poly(dA:dT) (0.5µg/ml) for 24h, then infected with HSV-2 (before); End1/E6E7 cells were simultaneously (simul) treated with poly(dA:dT) and infected with HSV-2; End1/E6E7 cells were first infected with HSV-2 for 2h, then washed and treated with poly(dA:dT) (after). At 48h post infection, both HSV-2 genomic DNA and total proteins were extracted from End1/E6E7 cells and subjected to the realtime PCR or Western blot assay.

CRISPR Cas9
We used the CRISPR Cas9 system to abrogate RIG-I expression in End1/E6E7 cells. Briefly, we designed gRNA (5'-GGGTCTTCCGGGATATAATCC-3') targeting the conserved sites in human RIG-I genomic sequences based on CCTop (https://crispr.cos.uni-heidelberg.de/). The gRNA was then subcloned into the lentiCRISPRv2 plasmid to obtain the lentiCRISPRv2-gRNA clone that expressed both Cas9 and gRNA according to the publications' instruction (28,29). Then lentiCRISPRv2-gRNA plasmid and two packaging plasmids, psPAX2 and pMD2.G, were co-transfected into 293T cells to obtain lentivirus. LentiCRISPRv2 was used as an empty vector control. Three days after the lentivirus infection, End1/E6E7 cells were cultured with puromycin (0.5µg/ml) containing medium for 14 days. Total cellular proteins were then collected and subjected to Western blot for RIG-I protein expression.

Western Blot
Total cell lysates were prepared with the cell extraction buffer (Invitrogen, Shanghai, China) with 1% protease inhibitor cocktail (Sigma, MO) and 1% phosphatase inhibitor mixture (Applygen, Beijing, China). The total proteins were quantified by a BCA protein assay kit (Beyotime Institute of Biotechnology, Shanghai, China), and equal amount of proteins were separated on SDS-PAGE. After being transferred to a PVDF membrane (Millipore, Germany), non-

Statistical Analysis
Data were presented as mean ± SD from at least three independent experiments, and statistical significance was analyzed by Student's t-test using GraphPad Prism for Windows version 5.0 (GraphPad Software Inc., San Diego, CA). Statistical significance was defined as P < 0.05 or P < 0.01.

Poly(dA:dT) Has Little Cytotoxicity Effect on End1/E6E7 Cells
We first examined the effect of poly(dA:dT) on the viability and apoptosis of End1/E6E7 cells. As shown in Figures 1B, C, little cytotoxic and apoptosis effect was observed in End1/E6E7 cells treated with poly(dA:dT) at the dose as high as 10µg/ml.

Poly(dA:dT) Inhibits HSV-2 Infection of End1/E6E7 Cells
To determine the anti-HSV-2 effect of poly(dA:dT), End1/E6E7 cells were pretreated with poly(dA:dT) for 24h prior to HSV-2 infection. As shown in Figures 2A, C  the three treatment conditions, poly(dA:dT) were able to significantly suppress HSV-2 infection at both DNA and protein levels. Pretreatment of the cells with poly(dA:dT) was the most effective in HSV-2 inhibition ( Figures 2B, D, F). In addition, we examined the effect of poly(dA:dT) on several key HSV-2 genes, including two immediate early genes (IE: ICP0 and ICP27), two early genes (E: ICP8 and DNA polymerase) and two late genes (L: HSV-2 gC and gD). As shown in Figure 3, poly(dA:dT) could significantly inhibit the expression of HSV-2 IE (Figures 3A, B), E ( Figures 3C, D) and L ( Figures 3E, F) genes in the infected cells.

Poly(dA:dT) Activates the JAK/STAT Signaling Pathway
To determine the mechanisms by which poly(dA:dT) inhibits HSV-2 infection of End1/E6E7 cells (Figures 2 and 3), we examined whether poly(dA:dT) could activate IFN-based were measured as HSV-2 gene levels relative (%) to control (without treatment, which is defined as 100%). Data are shown as mean ± SD for three independent experiments. Asterisks indicate that the differences between the indicated groups are statistically significant (*P < 0.05, **P < 0.01). immunity in End1/E6E7 cells. As shown in Figures 4A, B, poly (dA:dT) induced the expression of IFN-b and IFN-l at both mRNA ( Figure 4A) and protein ( Figure 4B) levels. These effects of poly(dA:dT) on IFN-b and IFN-l induction were timedependent. We next studied whether IRF3 and IRF7, the key positive regulators of the IFN signaling pathway, were involved in the IFN-b and IFN-l induction by poly(dA:dT) in End1/E6E7 cells. As shown in Figure 4D, poly(dA:dT) facilitated the phosphorylation of both IRF3 and IRF7 (p-IRF3 and p-IRF7), which were positively associated with the treatment time of poly (dA:dT) in End1/E6E7 cells.

RIG-I Knockout Significantly Compromises Poly(dA:dT)-Mediated HSV-2 Inhibition
To determine whether RIG-I plays a role in poly(dA:dT)-mediated HSV-2 inhibition, we constructed RIG-I knockout End1/E6E7 cells (RIG-I −/− ) and End1/E6E7 V2 control cells (V2 control) through CRISPR Cas9 system ( Figure 1A). As shown in Figures  8A-C, the levels of both HSV-2 DNA (Figures 8A, B) and protein ( Figure 8C) in RIG-I −/− cells were significantly higher than those in V2 control cells. In addition, we examined the impact of RIG-I knockout on poly(dA:dT)-mediated HSV-2 gene inhibition and observed that RIG-I −/− cells has higher levels of HSV-2 IE, E and L genes expression than V2 control cells (Figure 9). The expression level of RIG-I in HSV-2 infected cells was lower than those in uninfected cells. Especially, the expression level of RIG-I in HSV-2 infected cells which were pretreated with poly(dA:dT) was significantly lower than those in uninfected cells.
RIG-I Knockout Diminishes the Effect of Poly(dA:dT) on the Activation of JAK/STAT Signaling Pathway To determine whether RIG-I is vital for poly(dA:dT)-mediated IFN induction, we examined the levels of IFN-b and IFN-l in RIG-I knockout cells (RIG-I −/− ) and control cells (V2 control) transfected with poly(dA:dT). As shown in Figures 10A, B, IFNb and IFN-l levels in RIG-I −/− cells were significantly lower than those in V2 control cells. In addition, we examined the impact of RIG-I knockout on the expression of IRF3 and IRF7, the key regulators of IFNs. We observed that there was a significant decrease in the levels of p-IRF3 and p-IRF7 in RIG-I −/− cells as compared to those in V2 control cells ( Figure 10D). To determine whether the suppression of IFN-b, IFN-l, and IRFs in RIG-I −/− cells is associated with the inhibition of JAK/STAT signaling pathway, we analyzed the effect of poly(dA:dT) on STATs expression. Figure 11 demonstrated that poly(dA:dT)mediated induction of p-STAT1, p-STAT2, and ISGF-3gp48 diminished in RIG-I −/− cells as compared to V2 control cells.

DISCUSSION
As the first line of defense in FRT, human cervical epithelial cells are in direct contact with the virus and thus play an important role in preventing viral infections, including HSV-2. HSV-2 is a dsDNA virus which can be recognized by the cytosolic DNA sensors. There are several cytosolic dsDNA sensors involved in the host cells innate immunity against viral infections, including cGAS, whose secondary messenger cyclic GMP-AMP (cGAMP) can activate downstream sensor protein stimulator of IFN genes (STING). In addition to cGAS (33)(34)(35), AIM2, IFI16, and DAI are the cytosolic DNA receptors, activation of which could induce innate immune response to virus infections (36)(37)(38)(39).
Studies have shown that classical B form dsDNA is a potent immune stimulator when presents in the cytosol (40,41). In the present study, we examined whether poly(dA:dT) has the ability to activate the intracellular innate immunity against HSV-2 infection of human cervical epithelial cells. We found that poly (dA:dT) could significantly induce the expression of IFNs/the antiviral ISGs (Figures 4 and 6) and inhibit HSV-2 replication ( Figure 2). The investigation on the mechanisms for the induction of IFNs showed ( Figures 4C, D) that poly(dA:dT) treatment of the cells enhanced the phosphorylation of IRF3 and IRF7, the key and positive regulators of IFNs during viral infections (42). IRF3 and IRF7 phosphorylation is a crucial step in activating the IFNs-mediated antiviral immunity (43). During viral infections, IRF3 is important in the early phase of inducing the transcription of IFN-a and IFN-b, which results in the IRF7 activation. Both IFN-b, IFN-l1 gene expression is regulated by virus-activated IRF3 and IRF7, whereas IFN-l2/3 gene expression is mainly controlled by IRF7 (44). In addition to the positive impact on the IRFs, poly(dA:dT) could also activate the JAK/STAT signaling pathway that is vital for IFN-mediated innate immune response. We observed that poly(dA:dT) not only enhanced the expression of STAT1, STAT2, and STAT3 ( Figure 5A) but also facilitate the phosphorylation of STAT1, STAT2, STAT3, and ISGF-3gp48 ( Figure 5B). Several studies (45)(46)(47) have shown that comparing with STATs 1 and 2, STAT3 is an acute phase response factor with a transient activation. Therefore, it is likely that activation duration of STAT3 is shorter than that of STAT1 and STAT2.  (50,51). Therefore, the induction of these anti-HSV-2 ISGs provides a sound mechanism for poly(dA:dT)-mediated HSV-2 inhibition.
To determine the initial recognition molecule(s) for the poly (dA:dT) action on HSV-2 inhibition and IFN/ISGs induction, we first examined the effect of poly(dA:dT) on the expression of the DNA sensors. We found that poly(dA:dT) had little effect on the expression of the DNA sensors at both mRNA and protein levels ( Figures 7A, B). In addition, we failed to identify a specific DNA sensor that plays a major role in poly(dA:dT)-mediated IFN/ISG induction (data not shown). In contrast, poly(dA:dT) could significantly induce the RIG-I expression in the cells ( Figures  7C, D). Importantly, the vital role of RIG-I in the poly(dA:dT) actions was shown in the experiments using RIG-I −/− and the parental control cells ( Figure 1A). We observed that while poly (dA:dT) could significantly inhibit HSV-2 replication in the parental control cells, it had little effect on HSV-2 infectivity in RIG-I −/− cells (Figures 8 and 9). In addition, RIG-I −/− cells A B C FIGURE 12 | Effect of RIG-I knockout on poly(dA:dT)-induced ISGs expression. (A, B) End1/E6E7 control cells (V2 control) and RIG-I knockout End1/E6E7 cells (RIG-I −/− ) were transfected with or without poly(dA:dT) (0.5mg/ml) for 12h. Total cellular RNAs were extracted and subjected to the real-time PCR for ISG15, ISG56, OAS1, OAS2, MxA, MxB, viperin, and GBP5 expression. (C) V2 control and RIG-I −/− cells were transfected with or without poly(dA:dT) (0.5mg/ml) for 24h. Cellular proteins were collected and subjected to Western blot with the antibodies against ISG15, ISG56, OAS1, OAS2, MxA, MxB, viperin, GBP5, and GAPDH, respectively. The results are the mean ± SD of three independent experiments. Asterisks indicate that the differences between the indicated groups are statistically significant (*P < 0.05, **P < 0.01). lacked the effective response to poly(dA:dT) stimulation and expressed significantly lower levels of IFN-b and IFN-l, p-IRF3 and p-IRF7 than the control cells ( Figure 10). Similarly, the expression of p-STAT1, p-STAT2, ISGF-3gp48, and the antiviral ISGs was lower in RIG-I −/− cells as compared with that in the control cells (Figures 11 and 12). Finally, the important role of RIG-I in the intracellular innate immunity against HSV-2 was also supported by the observation that RIG-I levels in HSV-2infected cells were lower than those in uninfected cells, particularly in the cells infected and transfected with poly(dA: dT). The decreased expression of RIG-I in the infected cells is likely due to the negative effects of HSV-2 on the intracellular antiviral innate immunity, which is one of the strategies for the virus to escape the host cell immune response. These combined observations indicate that RIG-I is indeed a necessary and key sensor for poly(dA:dT)-mediated activation of IFN/STAT signaling pathways and HSV-2 inhibition. Therefore, it is possible that the interaction of poly(dA:dT) with RIG-I through RNA polymerase (Pol) III is crucial in initiating intracellular anti-HSV-2 innate immunity ( Figure 13). Chiu et al. (17) reported that poly(dA-dT) serves as a template for the de novo synthesis of poly(A-U) RNA by DNA-dependent RNA Pol III. They showed that Pol-III is responsible for the synthesis of the RNA from poly(dA-dT), which binds directly to RIG-I. Cheng et al. also demonstrated that RIG-I is essential for B form (AT-rich) dsDNA signaling pathway as poly(dA-dT) is converted by RNA Pol III of the host cell into a 5' triphosphate RNA intermediate, which can be recognized by RIG-I (52). Collectively, our study for the first time has demonstrated that the activation of the RIG-I by poly(dA:dT) could effectively inhibit HSV-2 infection of human cervical epithelial cells. While the precise cellular and molecular mechanisms for poly (dA:dT)-mediated HSV-2 inhibition remain to be determined, the induction of IFNs and the multiple antiviral ISGs should be largely responsible for much of poly(dA:dT)-mediated anti-HSV-2 activity. These findings are clinically significant as they indicate that activating the intracellular innate immunity by poly (dA:dT) has potential for the prevention and treatment of HSV-2 infection. However, future ex vivo and in vivo investigations in animal models and clinical studies are necessary in order to determine whether poly(dA:dT) is effective in activating FRT innate immunity and beneficial for protecting human cervical epithelial cells from HSV-2 infection. These future studies will be crucial for the design and development of AT-rich dsDNA-based intervention strategies to control HSV-2 mucosal transmission in FRT.