Zika virus infection induces expression of NRF2 and antioxidant systems in trophoblast cells

The nuclear factor erythroid 2-related factor 2 (NRF2) is a transcription factor that plays a critical role in the xenobiotic and stress responses. During viral infection, NRF2 can modulate the host metabolism and innate immunity; however, the most common activity of NRF2 in viral diseases is controlling reactive oxygen species (ROS). The Zika virus (ZIKV) is involved in a vertical infection in pregnancy, with reported fetal health consequences. However, the possibility that ZIKV regulates NRF2 expression in placental trophoblasts has not been investigated. In this report, we evaluated the upregulation of NRF2 and antioxidant enzymes in a trophoblast-like cell. These findings could help us understand the antioxidant mechanism underlying the ZIKV infection in the placenta during pregnancy.


Introduction
The ZIKV is a single-stranded positive-sense RNA virus of the Flaviviridae family that is principally transmitted by mosquitoes. However, ZIKV exhibits sexual transmission and vertical transmission [1]. Vertical transmission of pathogens could have severe consequences for fetal development. Additionally, ZIKV has a broader cell and tissue tropism Edited  and is detected in the brain, testis, uterus, vagina, and placenta [2,3]. In the viral replicative cycle, the host factors are recruited to promote virus production; this perturbs cellular homeostasis and triggers stress responses in the host cell [4]. For example ROS, which react directly with biomolecules and promote tissue damage, inflammatory responses, and finally cell death. One mechanism that regulates this stress is NRF2. It induces the transcription of cytoprotective genes, such as superoxide dismutase 1 (SOD) and catalase (CAT), to inhibit the accumulation of ROS. Viruses can regulate these pathways, modulating defense systems and making their infection cycle more efficient [5].
In the pregnancy, NRF2 preserves cellular homeostasis by enhancing the cell's innate antioxidant status to reduce oxidative stress and inflammatory damage in placental tissues, Indeed, NRF2 is actively involved in placental establishment, specifically in the differentiation of cytotrophoblasts to syncytiotrophoblasts by upregulating the expression of various cellular players, e.g., AhR, HMOX1, Aromatase, PPARy, C/EBPb or by downregulating the expression levels of Axin2 or GSK3b by miR-1246) [6,7]. The absence of this control system mediated by NRF2 affect the trophoblast functions, such as impaired differentiation and invasion [8]. ZIKV infection promotes ROS production and disrupts cellular antioxidant mechanisms, resulting in cell damage and death in hepatic and neural cells [9]. However, the regulation of NRF2 during ZIKV infection in trophoblast cultures as a placental model has not been investigated. The aim of this study was to examine whether ZIKV infection can upregulate NRF2 expression and anti-oxidant enzymes in a trophoblast cell model.

Results and discussion
To identify the behavior of NRF2 during ZIKV infection, we used a trophoblast cell model widely utilized in several in vitro studies (JEG-3 Cell ATCC HTB-36). JEG-3 cells were infected with the Puerto Rico strain (ATCC PRVABC59), which was propagated in the Vero cell line. Western blot assays were performed to identify changes in NRF2 expression during ZIKV infection at 16, 24, and 36 h, infecting JEG-3 cells at MOI 1; MOCK cells infected with a heat-inactivated virus were used as control. At the end of the infection time, protein was extracted and transferred to a nitrocellulose membrane. NRF2 proteins were then detected; ZIKV prM protein and GAPDH were used as infection control and load control, respectively. The following dilutions were used: anti-NRF2 (Abcam, 1:1000); anti-Zika prMprotein (Genetex, 1:500); and anti-GADPH (santacruz, 1:1000); The antibody secondary using an anti-mouse IgG-HRP (Genetex, 1:1000). The western blot analysis revealed an overexpression of the NRF2 protein in ZIKV-infected cells when compared to the control cell, where NRF2 levels were normal (Fig. 1a). Additionally, to compare the expression levels of NRF2 between the extracts under infection and control conditions, densitometric analysis was performed using the ImageJ Software (version 1.39, NIH, Bethesda, MD, USA) and normalizing according to the GADPH protein. Notably, changes in NRF2 expression were detected 16 h post-infection. However, a more substantial 24 h postinfection (Fig. 1b). These findings indicate that ZIKV infection stimulates NRF2 expression in the trophoblast lineage, most likely to regulate the expression of antioxidant systems.
To further investigate this phenomenon, we examined if antioxidant enzymes are stimulated during ZIKV infection as a result of NRF2 overexpression in trophoblast cells. Immunofluorescence assays were used to determine the cellular distribution of the enzymes SOD-1 and CAT. JEG-3 cells were seeded in coverslip-covered wells and infected under same conditions mentioned above. Subsequently, the cells were incubated for 24 h post-infection because this was when a more considerable change in NRF2 expression was observed. Following infection, the cells were incubated with specific antibodies Anti-SOD (Abcam, 1:100) and anti-Cat (Abcam, 1:100). To identify the infection, we used an anti-Flavivirus complex Envelope-protein (Millipore, 1:100). Secondary antibodies combined with fluorochromes were used to reveal immunofluorescence staining, which was then examined using confocal microscopy. The analysis of the stains demonstrated that under infection conditions, the subcellular distribution of the SOD and CAT enzymes displayed a pattern of aggregates or dots in the infected cells that were double positive for the viral antigen (Red) and the antioxidant enzymes (green). This phenomenon was not observed in uninfected cells, where the antioxidant enzymes were distributed more homogeneously in the cell cytoplasm (Fig. 1c). Furthermore, an analysis of the mean fluorescence intensity comparing the immunodetection signal of SOD and CAT under infection and non-infection conditions revealed that infected cells had a higher fluorescence intensity (Fig. 1d), indicating that these enzymes had a different distribution or accumulation during infection in the trophoblast, which could be related to a higher NRF2 expression.
Then we evaluated whether the increase in fluorescence levels achieved was associated with an overexpression of the SOD and CAT enzymes at the protein level during infection. For this, protein was extracted from JEG-3 cells infected under the conditions described above, and western blot assays were performed using specific antibodies for SOD (Abcam, 1:1000) and CAT (Abcam, 1:1000) and antizika prM-protein (Genetex, 1:500) and GADPH (santacruz, 1:1000) as infection marker and load control, respectively. The findings demonstrated that infected cells expressed more SOD and CAT than did control cells (Fig. 1e). Additionally, densitometric analysis was performed, and substantial differences in the protein levels of the enzymes between the two groups were observed, with a higher amount in the infected cells (Fig. 1f). As a result, these findings indicate that NRF2 is positively regulated by ZIKV infection of the trophoblast, leading to the expression of antioxidant enzymes.
Ultimately, we evaluated whether the increased antioxidant response observed in the trophoblast caused by ZIKV infection was related to infection-induced ROS production. For this, an indirect way of evaluating the presence of ROS during infection was to identify markers of damage by oxidative stress in lipids and proteins. For this, we determined two markers of this effect: malondialdehyde (MDA) and carbonylation levels. For this evaluation, cell extracts were obtained under the aforementioned conditions. Subsequently, the extracts were then incubated for 60 min with a mixture of 10 nM phenylindole (Sigma) and acetonitrile-methanol-HCl (Sigma) to detect MDA. At the end of the time, the absorbance was measured at 586 nm. In order to quantify the carbonylation levels, the samples were incubated with 2,4-dinitrophenylhydrazine (SIGMA) for 60 min; after incubation, the absorbance was measured at 370 nm. The results revealed that in ZIKV-infected cells, there was a more substantial detection of markers of oxidative damage to proteins and lipids, but this was not the case in the control cells (Fig. 1g and h), implying that the effect on NRF2 activation and the increase in the expression of antioxidant systems could be associated indirectly with the presence of markers damages by oxidative stress generated by ZIKV infection in the trophoblast.  16,24, and 36 h or mock-infected. Extracts proteins were obtained, and a western blot was performed. The protein expression level of NRF2 was identified using the corresponding antibody, anti-prM was used to detect viral protein, and anti-GADPH was used for loading control. The graphs represent NRF2 levels compared with GADPH. We measured the protein amount values from the control (mock-infected) and adjusted them to a value of 1. The values for the expression in infection were then expressed as a number relative to the control, and statistical significance was determined based on at least three independent experiments. *p < 0.05 for infected vs mockinfected. B Cellular distribution of SOD and CAT during ZIKV infection in JEG3 cells. The cells were infected under the conditions mentioned above. Twenty-four hours post-infection, the cells were fixed, and an immunofluorescence assay was performed using an antibody specific to SOD and CAT (green), and Envelope protein of ZIKV (Red); changes in the distribution cellular of enzymes were identified in cells infected (white arrows). C, D The graph represents the result expressed as mean fluorescence intensity (MFI) arbitrary units compared to infected and mock-infected cells. The error bars show the SD of three independent experiments ****p ≤ 0.0001, n = 30 per group. E JEG3 cells up-regulate the expression of antioxidant enzymes. The figure shows a representative western blot of total extracts from JEG3 cells non-infected (MOCK) or infected with ZIKV for 24 h. After blotting, the membranes were challenged with anti-SOD antibody, anti-CAT antibody, anti-prM protein, or anti-GAPDH as a protein load control. F, G Densitometry analysis shows results from three independent western blots. In all of them, enzyme expression was normalized concerning levels in non-infected conditions, and statistical analysis was performed as described previously. ZIKV induces ROS production and markers of oxidative stress damage in JEG3 cells. The cells were mock infected or infected with ZIKV 1 MOI, and 24 hpi cells were harvested to measure stress biomarkers carbonyl protein (H) and MDA lipoperoxidation (I). The results include data from three independent with *P < 0.01 indicate significant differences compared with control cells, Student's t test Vertical transmission of ZIKV during pregnancy is associated with fetal development complications, such as microcephaly. ZIKV infection of placental trophoblast, is a mechanism related to infection in-utero [10]. The placenta is a vital tissue for fetal development, where there is a balance between ROS production and antioxidant systems [8].
Maternal complications, such as preeclampsia, trigger the dysregulation of these axes, promote trophoblast apoptosis, and result in placental damage and inflammation [11]. Interestingly, lack of NRF2 may be related to the development of preeclampsia [7]. However, trophoblast have well-identified antioxidant mechanisms, including the NRF2 factor, an axis controlling oxidative stress in the placenta. Additionally, viral infections affect the functionality of NRF2; observing how it can be modulated depending on the cellular context. In DENV infection, NRF2 is upregulated in the mononuclear cells, enhancing tumor necrosis factor-alpha (TNF-α) production; however, the infection of dendritic cells with DENV is associated with NRF2 degradation mediated by the viral protease [12,13]. Our result demonstrated that there is an increase in the levels of NRF2 and the antioxidant enzymes SOD-1 and CAT in ZIKV-infected JEG3 cells, which was possibly induced by the accumulation of cytoplasmic ROS in JEG-3 cells infected with ZIKV. This cause-effect relation is also observed in the other members of the Flaviviridae family [5]. It is likely that ROS production is a defense mechanism during infection with flavivirus that can also regulate the production of pro-inflammatory cytokines [14,15]. Additionally, NRF2 can control the expression of certain immunoregulatory molecules such as AhR, PD-L1, or TNF-α, helping to modulate immune response [7]. However, the inflammatory response could also participate in the pathogenesis of the disease, as observed in clinical cases of dengue hemorrhagic fever [16]. Indeed, a failure of immune tolerance or an imbalance in the overall immune response can lead to adverse pregnancy outcomes, such as preeclampsia. Noting that NRF2 is important for the balance of the immune response as well as for antioxidant mechanisms in placenta. In addition, depending on the virus and the cell type, ROS may favor viral replication. The ROS production during Dengue and Sindbis virus infection upregulates the activity of NS5, contributing to genome replication [17]. These findings indicate the dual role of ROS in viral infections.
Notably, our findings are not in line with those obtained for the ZIKV infection of hepatic and neurons cells, where ZIKV promotes the downregulation of SOD-1 and CAT expression by inhibiting NRF2 activation [9]. However, it has recently been demonstrated that ZIKV infection in neuroblast cells models induced the activation of NRF2/ Glutathione (GSH) antioxidant response [18]. A separate mechanism on the ROS-NRF2 axis may be involved in ZIKV infection in the placenta. In a preeclampsia model evaluated in trophoblasts, tetraspanin-24 (CD151) increased the expression of antioxidant genes through ERK-NRF2 activation during an oxidative stress challenge [19]. The ZIKVinfected trophoblasts could have employed this strategy; future investigations are needed for a better understanding.
In conclusion, ZIKV infection induced an antioxidant response via NRF2 in a trophoblast cell line. Further research is required to understand the relationship between alternate pathways for inducing the expression of NRF2 and the progress of ZIKV infection, possibly using placental explants or animal models.