Skip to main content
NCBI home page
The Journal of Clinical Endocrinology and Metabolism logoLink to The Journal of Clinical Endocrinology and Metabolism
. 2016 Jun 1;101(6):2612–2621. doi: 10.1210/jc.2016-1059

Nrf2 Activation Inhibits Effects of Thrombin in Human Amnion Cells and Thrombin-Induced Preterm Birth in Mice

Yoshitsugu Chigusa 1, Annavarapu Hari Kishore 1, Haruta Mogami 1, Ruth Ann Word 1,*
PMCID: PMC6287450  PMID: 27050800

Abstract

Context:

Nrf2 is a key transcription factor that modulates cell defense mechanisms against endogenous and exogenous stress. Previously, we reported that thrombin increased matrix metalloproteinases and prostaglandin synthesis in human amnion mesenchymal cells.

Objective:

We sought to determine whether activation of Nrf2 alters the effect of thrombin on prostaglandin synthesis, protease activation, and cytokine release in human amnion. Furthermore, we analyzed the effect of Nrf2 activation on thrombin-induced preterm labor in mice.

Design:

Primary human amnion mesenchymal cells and pregnant mice were employed to investigate the effect of Nrf2 on thrombin-induced inflammation and preterm birth.

Setting:

This was a laboratory-based study using cells and mice.

Results:

As expected, thrombin increased cyclooxygenase-2, IL-1β, IL-6, IL-8, and matrix metalloproteinase-1 in amnion mesenchymal cells. Preincubation with Nrf2 activators, diethyl maleate or 15-deoxy-Δ12, 14-prostaglandin J2 (15d-PGJ2), profoundly repressed thrombin-induced gene expression. In addition, Nrf2 activation inhibited thrombin-induced cyclooxygenase-2 protein levels and secretion of prostaglandin E2, IL-1β, IL-6, IL-8, TNFα, and granulocyte-macrophage colony-stimulating factor in the media. Whereas vehicle and 15d-PGJ2 did not alter gestational length, all pregnant mice treated with thrombin delivered preterm. 15d-PGJ2 delayed thrombin-induced preterm birth significantly.

Conclusions:

The results indicate that Nrf2 activation represents a key stress response in amnion mesenchyme cells and in pregnant mice to mitigate the adverse proinflammatory effects of thrombin on the fetal membranes. We suggest, therefore, that pharmacological activation of Nrf2 may prevent the increased risk of preterm premature rupture of the membranes associated with thrombin activation that accompanies subchorionic hemorrhage or bleeding during pregnancy.

Preterm birth, defined as the birth of an infant before 37 weeks of pregnancy, is the leading cause of perinatal morbidity and mortality (1, 2). In 2010, the estimated rate of preterm birth worldwide was 11.1% (3), and about 1.1 million neonates died from complications of preterm birth (4). Preterm premature rupture of the membranes occurs in 1 to 3% of all pregnancies and is associated with approximately 30% of preterm births (5, 6). The amnion is the load-bearing structure of the fetal membranes, and it is comprised of an avascular layer of epithelial cells, underlying mesenchymal cells, and extracellular matrix proteins (7). Previously, we found that thrombin activity was increased in human amnion from preterm deliveries and that thrombin increased expression of matrix metalloproteinase-1 (MMP-1), MMP-9, and prostaglandin-endoperoxide synthase 2 (cyclooxygenase-2 [COX-2]) in human amnion mesenchymal cells. Increasing MMP may cause collagen degradation and premature rupture of the membranes. Up-regulation of COX2 may lead to cervical ripening. Moreover, thrombin injection into pregnant mice uterus caused preterm birth (8).

Nuclear factor erythroid 2-related factor 2 (Nrf2) is one of the most important transcription factors that modulate cell defense mechanisms against endogenous and exogenous stress. Under normal conditions, Nrf2 is localized mainly in the cytoplasm and is ubiquitinated constitutively by Kelch-like ECH-associated protein 1 (Keap1), resulting in proteasomal degradation (9). Exposure to stress and reactive oxygen species leads to liberation of Nrf2 from Keap1, accumulation of Nrf2 in the nucleus, and induction of antioxidant and anti-inflammatory genes. For example, HO-1, glutamyl cysteine ligase modulatory (GCLM), glutamyl cysteine ligase catalytic (GCLC), and NAD(P)H:quinone oxidoreductase 1 (NQO1) are predominant target genes of Nrf2 and are commonly used to evaluate Nrf2 activation (10, 11). As one of the predominant Nrf2 target genes, heme oxygenase-1 (HO-1) is an inducible enzyme that catalyzes the degradation of heme, yielding iron, carbon monoxide, and biliverdin. HO-1 and its metabolites are known to have cytoprotective, antioxidative, and anti-inflammatory activities (12, 13). Thus far, the function of Nrf2 and HO-1 in the amnion in the context of preterm birth or premature rupture of the fetal membranes remains largely unknown.

Here, we hypothesized that Nrf2 activation may play a central role in preventing preterm birth. Based on our previous results indicating thrombin-induced activation of inflammation, protease activity, and prostaglandin synthesis in amnion and preterm birth in mice (8), we sought to determine whether Nrf2 activation mitigates the adverse proinflammatory effect of thrombin in human amnion mesenchymal cells and thrombin-induced preterm birth in mice.

Materials and Methods

Reagents

DMEM/F-12 (11320) and antibiotic-antimycotic solution (15240) were purchased from Invitrogen. Thrombin (T7009), 15-deoxy-Δ12, 14-prostaglandin J2 (15d-PGJ2) (D8440), and TRI reagent (T9424) were purchased from Sigma-Aldrich. Diethyl maleate (DEM) (sc202577) was purchased from Santa Cruz Biotechnology.

Isolation and culture of human amnion epithelial and mesenchymal cells

For human amnion cell culture, 30 human placentas were collected from normal pregnant subjects at Parkland Memorial Hospital (Dallas, Texas), as approved by the Institutional Review Board of the University of Texas Southwestern Medical Center. All pregnancies were uncomplicated and underwent elective cesarean section at 39–40 weeks of gestation without labor. Each placenta was processed and used individually for each experiment which, in most cases, was repeated at least three times. Preparation and isolation of human amnion epithelial and mesenchymal cells were performed as described previously (14). Briefly, human amnion tissue was separated by blunt dissection. The amnion tissue was minced, and cells were dispersed by enzymatic digestion. Isolated amnion cells were cultured in DMEM/F-12 supplemented with fetal bovine serum (10%, vol/vol) and antibiotic-antimycotic solution (1%, vol/vol), at 37ºC in a humidified atmosphere of 5% CO2 in air and allowed to replicate in a monolayer to confluence. Cell treatments were conducted in serum-free media. All experiments were conducted in at least three cell culture preparations obtained from different subjects.

Quantitative real-time PCR

Total RNA was extracted from cells and tissues using RNAqueous-4PCR total RNA Isolation Kit (Ambion) or TRI Reagent, respectively. Reverse transcription of 2 μg total RNA was performed using SuperScript III Reverse Transcriptase (Invitrogen). Primers for Hmox1, Gclc, Gclm, and Nqo1 were purchased from Thermo Scientific. Other primer sequences for amplifications are shown in Supplemental Table 1. Taqman probe with FAM dye label was used for COX2, Hmox1, Gclc, Gclm, and Nqo1, and SYBR Green was used for other genes. Quantitative RT-PCR was performed using 7900HT Fast Real-Time PCR System (Applied Biosystems). Gene expression was normalized to that of housekeeping gene, glyceraldehyde 3-phosphate dehydrogenase (GAPDH) for human amnion cells, and cyclophilin A for mouse tissue and cells.

Immunoblotting

Whole-cell lysates were extracted using radioimmunoprecipitation assay buffer (50 mm Tris-HCl, pH 8.0, 150 mm sodium chloride, 0.5% sodium deoxycholate, 0.1% 126 sodium dodecyl sulfate, and 1.0% Nonidet P-40 substitute) supplemented with cocktail protease inhibitor Complete Mini (Roche Diagnostics). Nuclear proteins were prepared as described elsewhere (15). Proteins were separated on 7.5% (for Nrf2) or 10% (for COX-2) SDS-polyacrylamide gels, respectively, and transferred onto nitrocellulose membranes. Membranes were probed with anti-Nrf2 (ab62352) (1:1000; Abcam), or anti-COX-2 (RM-9121-S1) (1:500; NeoMarkers) antibodies. Antibodies against β-actin (3700) (1:1000; Cell Signaling Technology) and TATA binding protein (ab818) (1:3000; Abcam) were used as loading controls. Blots were subsequently incubated with an appropriate secondary antibody (170-6515, 172-1011) (1:5000; BioRad). Signals were detected with SuperSignal West Pico Chemiluminescent Substrate (34080) (Thermo Scientific) and visualized using the LAS-3000 Imaging System (Fujifilm).

ELISA

Prostaglandin E2 (PGE2) concentration in the conditioned medium was assayed by Prostaglandin E2 Parameter Assay Kit (KGE004B) (R&D Systems) according to the manufacturer's instruction. Total protein concentration of conditioned medium was determined with a bicinchoninic acid protein assay kit (Thermo Scientific).

Cytokine measurement

Levels of five cytokines (IL-1β, IL-6, IL-8, TNFα, and granulocyte macrophage colony-stimulating factor [GM-CSF]) in the conditioned medium were quantified with Inflammatory Cytokine Human Magnetic 5-Plex Panel for Luminex Platform (LHC0003M) (Invitrogen) and MAGPIX (Luminex) according to the manufacturer's instruction. Values of the ELISA were adjusted to total protein concentration.

Mouse model of preterm birth

C57/Bl6 mice were maintained in a nonpathogen-free environment and exposed to a 12-hour light/12-hour dark cycle. To perform precise gestational age assessments, C57/Bl6 female mice were mated with male mice for 5 hours (from 10 am to 3 pm). On day 17 post coitum at 9 am, pregnant mice with six to 10 pups were anesthetized with tribromoethanol (T4840-100G; Sigma) (250 μg/g body weight, ip). After surgical preparation of the abdomen, a ventral incision was made, and both uterine horns were exteriorized.

Sample collection.

Four microliters of 15d-PGJ2 (5 μg) or vehicle (equal volume of dimethylsulfoxide in PBS) were injected in the interface between fetal membranes and uterine lining of each fetus opposite the placental site using a 33-gauge Hamilton syringe, taking care not to enter the amniotic cavity. Methylene blue was added to all of the stock solution, with a final concentration of 0.01%, and the injection was limited to six pups. The wound was closed in two layers with 5–0 vicryl. After 6 hours, mice were sacrificed, and fetal membranes from each pup were collected.

Monitoring delivery time.

To determine the lowest concentration of thrombin that initiates preterm birth, 4 μL of thrombin (2 U), thrombin (1 U), or vehicle was injected using the same methods in sample collection. Although injection of 1 U thrombin per pup did not cause preterm birth (delivery 56.5 ± 9.5 hours after injection), injection of 2 U per pup resulted in preterm birth similar to maximal doses of 4 U (24.8 ± 1.9 hours after injection). Based on these observations, 4 μL thrombin (2 U), 15d-PGJ2 (5 μg), thrombin+15d-PGJ2, or vehicle was injected into pregnant mice on day 17 as described above. Mice were monitored continuously and separately with closed-circuit television cameras and a digital video recorder.

All mice received buprenorphine (Reckitt Benckiser Pharmaceuticals Inc.) (0.1 μg/g body weight, ip). All animal experimental procedures were approved by the Institutional Animal Care and Use Committee of the University of Texas Southwestern Medical Center.

Isolation and culture of primary mouse amnion mesenchymal cells

Mouse fetal membrane cells were prepared according to the modified method of the isolation of human amnion epithelial and mesenchymal cells. Briefly, mouse fetal membranes were collected from pregnant wild-type C57/Bl6 mice on day 17. Eight to 10 fetal membranes from one mouse were minced together, and cells were dispersed by enzymatic digestion in 10 mL of Eagle's minimum essential medium containing 0.02 g of trypsin for 15 minutes at 37°C twice. Remaining tissue was collected by the cell strainer (70 μm) to enrich the cell preparation in mesenchymal cells. Thereafter, tissue fragments were incubated in 15 mL of Eagle's minimum essential medium with 11.4 mg of collagenase B and 1.14 mg of DNase I for 30 minutes at 37°C. Dispersed cells were filtered again and collected by centrifuge. Isolated amnion mesenchymal cells were cultured following the same methods of human amnion mesenchymal cell culture.

Statistical analysis

Data are presented as mean ± SD (in vitro experiments) or mean ± SEM (in vivo experiments). Statistical comparisons were performed with unpaired Student's t test or one-way ANOVA followed by Tukey's test, using Prism 6.0 (GraphPad Software). In the time course experiment, two-way ANOVA followed by Dunnett's test was used. The Mann-Whitney U test or Kruskal-Wallis test followed by Dunn's test was used as appropriate in the analysis of mice data. Statistical analysis of survival rates was conducted using the mean survival of each litter followed by ANOVA. P < .05 was deemed statistically significant.

Results

Nrf2 signaling in human amnion mesenchymal cells

To determine whether human amnion mesenchymal cells were responsive to Nrf2 activators, cells were treated with DEM (100 μm) or 15d-PGJ2 (10 μm) for 6 hours. Thereafter, nuclear proteins were extracted and subjected to Western blotting for Nrf2, and mRNA levels of Nrf2 target genes were assessed using quantitative PCR (qPCR). Both DEM and 15d-PGJ2 treatment resulted in accumulation of Nrf2 in the nucleus (Figure 1A) and increased mRNA expressions of GCLC, GCLM, and HO-1 in human amnion mesenchymal cells (Figure 1, B and C). GCLC and GCLM are rate-limiting enzymes in the biosynthesis of glutathione, also known as an essential endogenous antioxidant. DEM up-regulated another Nrf2 target gene, NQO1 mRNA, but 15d-PGJ2 did not. Although similar results were obtained in human amnion epithelial cells (Supplemental Figure 1), we focused this investigation on mesenchymal cells due to their response to thrombin and proinflammatory stimuli. Together, the results indicate that Nrf2 signaling pathways are intact in human amnion mesenchymal cells.

Figure 1.

Figure 1.

Open in a new tab

Effect of Nrf2 activators on human amnion mesenchymal cells.

A, Mesenchymal cells were treated with two different Nrf2 activators (DEM, 100 μm; 15d-PGJ2, 10 μm) for 6 hours. Thereafter, nuclear extracts (50 μg per lane) were examined by immunoblot analysis for Nrf2 protein levels. TATA binding protein (TBP) was used as a loading control. Blot represents consistent results from three different cell preps. B and C, Effect of DEM (B) or 15d-PGJ2 (C) on expression of Nrf2 target genes. Results represent the mean ± SD of three replicates. **, P < .01; ***, P < .001, compared with control.

Nrf2 activation represses thrombin-induced COX-2 and MMPs

To determine the effects of Nrf2 activation on thrombin-induced proinflammatory gene expression, human amnion mesenchymal cells were pretreated with the Nrf2 activator DEM (100 μm) for 1 hour before treatment with thrombin (2 U/mL/24 hours). Thrombin resulted in dramatic induction of COX2, IL-6, IL-8, MMP1, and MMP9. Interestingly, although Nrf2 activation did not alter basal expression, DEM decreased thrombin-induced up-regulation of these genes significantly (Figure 2A). Next, we used HO-1 as a marker for Nrf2 biological activity in amnion mesenchymal cells and found that HO-1 mRNA was increased dramatically by DEM within 4 hours, decreasing to basal levels by 48 hours (Figure 2B). Likewise, DEM increased expression of HO-1 17-fold within 4 hours, even in the presence of thrombin. Thus, gene induction by DEM preceded thrombin-induced activation of COX2, IL-6, IL-8, and MMP1, and activation of Nrf2 decreased thrombin-induced activation of proinflammatory genes at all time points from 8–24 hours (Figure 2B). Consistent with mRNA results, DEM also blocked thrombin-induced COX-2 protein levels (Figure 3A). Furthermore, thrombin-induced increases in PGE2 accumulation were reduced dramatically by Nrf2 activation (Figure 3B). To examine whether Nrf2 activation altered cytokine production by mesenchymal cells, levels of five cytokines (IL-1β, IL-6, IL-8, TNFα, and GM-CSF) were quantified in conditioned medium from mesenchymal cells treated with vehicle, thrombin (2 U/mL), DEM (100 μm), or DEM+thrombin for 24 hours (Figure 3C). Whereas thrombin treatment resulted in dramatic accumulation of these cytokines in the media after 24 hours, thrombin-induced increases were reduced significantly by the Nrf2 activator DEM (Figure 3C).

Figure 2.

Figure 2.

Open in a new tab

Effect of the Nrf2 activator, DEM, on thrombin-induced gene expression in human amnion mesenchymal cells.

A, Cells were pretreated with DEM (100 μm) for 1 hour before treatment with vehicle (control), thrombin (2 U/mL), DEM, or DEM+thrombin (DEM/Thro) for 24 hours. Results represent relative fold expression (mean ± SD; n = 3) using RT-qPCR and GAPDH as the normalizing housekeeping gene, which was invariant with treatment groups. B, Temporal activation of gene expression in cells treated with vehicle (●, black dotted line), thrombin (2 U/mL; ■, black solid line), DEM (▴, black dashed line), or DEM+thrombin (▾, gray solid line). Results represent mean ± SD of triplicates. *, P < .05; **, P < .01; ***, P < .001, thrombin vs DEM/Thro. §, P < .001 compared with control.

Figure 3.

Figure 3.

Open in a new tab

Effect of the Nrf2 activator, DEM, on thrombin-induced gene products and PGE2 accumulation in human amnion mesenchymal cells.

A, Immunoblot analysis of COX2 protein in cells treated with vehicle, thrombin (2 U/mL), DEM (100 μm), or DEM+thrombin for 24 hours. β-Actin served as the loading control (60 μg per lane). B, PGE2 was quantified in media from cells treated with vehicle (control), thrombin (2 U/mL), DEM (100 μm), or DEM+thrombin (DEM/Thro) for 24 hours. Results represent mean ± SD (n = 3). ***, P < .001. C, IL-1β, IL-6, IL-8, TNF-α, and GM-CSF were quantified in conditioned media of mesenchymal cells treated with vehicle, thrombin (2 U/mL), DEM (100 μm), or DEM+thrombin for 24 hours. ***, P < .001.

To investigate the effect of Nrf2 activator further, we employed another Nrf2 activator, 15d-PGJ2, which is physiologically relevant (16). The effects of Nrf2 activation by 15d-PGJ2 (5 μm) were similar to those of DEM. Specifically, treatment with 15d-PGJ2 for 1 hour resulted in profound abrogation of thrombin-induced COX2, IL-6, IL-8, MMP1, and MMP9 mRNA in mesenchymal cells for 24 hours (Figure 4A). Likewise, 15d-PGJ2 inhibited thrombin-induced COX-2 protein expression (Figure 4B), PGE2 accumulation (Figure 4C), and cytokine levels (Figure 4D) in the conditioned medium. Together, the results indicate that Nrf2 activation in mesenchymal cells mitigates thrombin-induced proinflammatory effects.

Figure 4.

Figure 4.

Open in a new tab

Effect of the Nrf2 activator, 15d-PGJ2, on thrombin-induced gene expression, gene products, and PGE2 accumulation in human amnion mesenchymal cells.

Cells were pretreated for 1 hour with 15d-PGJ2 (5 μm) before treatment with vehicle (control), thrombin (2 U/mL), 15d-PGJ2, or 15d-PGJ2+thrombin (PGJ2/Thro) for 24 hours. A, Results represent relative fold expression (mean ± SD; n = 3) using RT-qPCR and GAPDH as the normalizing housekeeping gene, which was invariant with treatment groups. ***, P < .001. B, Immunoblot analysis of COX2 protein in cells treated with vehicle, thrombin (2 U/mL), 15d-PGJ2 (5 μm), or PGJ2+thrombin for 24 hours. β-Actin served as the loading control (60 μg per lane). C, PGE2 was quantified in media from cells treated with vehicle (control), thrombin (2 U/mL), 15d-PGJ2 (5 μm), or 15d-PGJ2+thrombin (PGJ2/Thro) for 24 hours. Results represent mean ± SD (n = 3). **, P < .01. D, IL-1β, IL-6, IL-8, TNF-α, and GM-CSF were quantified in conditioned media of mesenchymal cells treated with vehicle, thrombin (2 U/mL), 15d-PGJ2 (5 μm), or 15d-PGJ2+thrombin for 24 hours. ***, P < .001.

Effect of 15d-PGJ2 on Nrf2 activation in vivo

Next, we determined whether 15d-PGJ2 could activate Nrf2 in fetal membranes of pregnant mice. Consistent with in vitro experiments using human amnion mesenchymal cells, 15d-PGJ2 increased Nrf2 target gene expression significantly in fetal membranes (Figure 5A). As expected, however, the magnitude of gene induction was not as robust in fetal membranes in vivo relative to in vitro conditions of cell culture. Nonetheless, the results indicated that intrauterine treatment of 15d-PGJ2 was sufficient to activate Nrf2 target genes in fetal membranes.

Figure 5.

Figure 5.

Open in a new tab

Effect of 15-PGJ2 on Nrf2 activation in vivo.

A, Either 15d-PGJ2 (5 μg/pup; n = 7 dams) or vehicle (control; n = 5 dams) was injected into the interface between the uterine wall and fetal membranes of each fetus on day 17. After 6 hours, fetal membranes were collected and analyzed for induction of Nrf2 target genes. Each fetal membrane tissue was from a different animal. Results represent relative fold expression (mean ± SEM) using RT-qPCR. B, Murine amnion mesenchymal cells were isolated and pretreated for 1 hour with either DEM (100 μm) or 15d-PGJ2 (5 μm) before treatment with vehicle (control), thrombin (4 U/mL), DEM+thrombin (DEM/Thro), or 15d-PGJ2+thrombin (PGJ2/Thro) for 24 hours. Results represent relative fold expression (mean ± SD; n = 3) using RT-qPCR. Cyclophilin A (Cphn) served as the normalizing housekeeping gene which was invariant with treatment groups. *, P < .05; **, P < .01; ***, P < .001. n.s., not significant.

Next, we determined whether 15d-PGJ2 altered thrombin-induced activation of COX-2, cytokines, or MMP expression in murine amnion mesenchymal cells. Cells were treated with thrombin (4 U/mL) with or without pretreatment with DEM (100 μm) or 15d-PGJ2 (5 μm). After 24 hours, Mmp8 and Mmp13 (the major interstitial collagenases in mice), Cox2, Mmp9, IL-1β, IL-6, and Tnfα gene expression was quantified by RT-qPCR (Figure 5B). Thrombin treatment resulted in significant, but relatively modest, increases in Cox2, Mmp8, and Mmp9 (1.8- to 5-fold). 15d-PGJ2 not only decreased basal expression of these genes, but also inhibited thrombin-induced increases in Cox2, Mmp8, and Mmp9 mRNA (Figure 5B). In contrast to Cox2, Mmp8, and Mmp9, thrombin treatment resulted in dramatic increases in Mmp13 (16-fold), IL-1β (>90-fold), IL-6 (25-fold), and Tnfα (35-fold). Surprisingly, despite these robust responses, DEM and 15d-PGJ2 completely blocked thrombin-induced increases in Mmp13 and Il1β and inhibited thrombin-induced increases in IL-6 and Tnfα significantly. The effects of 15d-PGJ2 were not as pronounced as inhibition by DEM (Figure 5B).

15d-PGJ2 inhibits thrombin-induced preterm birth

Here, our results demonstrated that Nrf2 could be activated in fetal membranes in pregnant mice and that Nrf2 activation mitigated thrombin-induced stimulation of proinflammatory and protease gene expression in murine amnion mesenchymal cells. Thus, we investigated the effect of intrauterine 15d-PGJ2 on thrombin-induced preterm birth in mice. Pregnant mice on gestation day 17 were injected with vehicle, thrombin (2 U/pup), 15d-PGJ2 (5 μg/pup), or thrombin+15d-PGJ2 (Figure 6A). Whereas vehicle and 15d-PGJ2 did not alter gestational length (69.3 ± 4.4 hours, 68.8 ± 7.5 hours after injection, respectively), 15d-PGJ2 delayed thrombin-induced preterm birth significantly (from 24.8 ± 1.9 to 58.2 ± 9.6 hours after injection; P = .013), and this was accompanied by a significant increase in fetal survival (Figure 6B).

Figure 6.

Figure 6.

Open in a new tab

Effect of 15d-PGJ2 on thrombin-induced preterm birth in mice.

Pregnant mice on gestation day 17 were injected with vehicle (n = 6), thrombin (2 U/pup; n = 7), 15d-PGJ2 (5 μg/pup; n = 6), or thrombin+15d-PGJ2 (n = 7). Mice were monitored continuously and separately with closed-circuit television cameras and digital video recorder. A, Delivery time (hours) after injection. B, Fetal survival rates (mean ± SEM). Pups alive 48 hours after delivery were considered survived. **, P < .01; ***, P < .001.

Discussion

In the present investigation, we found that Nrf2 activation in amnion mesenchymal cells strongly reduced: 1) thrombin-induced COX-2, PGE2 synthesis, cytokine production, and MMP9 and MMP1 mRNA levels in human amnion mesenchymal cells; and 2) cytokine and collagenase gene expression in mouse fetal membranes. Furthermore, a physiological activator of Nrf2, 15d-PGJ2, delayed thrombin-induced preterm birth in mice.

Nrf2 is a master regulator of the response to oxidative stress and toxic insults (17), thereby regulating antioxidant proteins (18), detoxification enzymes (19), and inflammation (20). The importance of this role in several physiological and pathological processes is widely recognized, including cellular adaptation to stress, redox homeostasis, inflammation, and aging. Due to its positive effect in several diseases, Nrf2 has become a major therapeutic target with novel natural synthetic and targeted small molecules currently under investigation to modulate the pathway in clinical trials. Despite its significance, the physiological role of Nrf2 during pregnancy and fetal responses to stress is poorly understood. Recently, Lim et al (21) reported that Nrf2 accumulation in amnion epithelial cells, chorionic cytotrophoblasts, and mesenchymal cells and in nuclear extracts of whole fetal membranes was significantly reduced after spontaneous term labor compared to those without labor, and it was suggested that low activation of Nrf2 may play a role in labor onset or the progress of labor and rupture of the fetal membranes. Herein, we show that although thrombin induced increases in proinflammatory genes and proteases, thrombin alone does not activate Nrf2 target genes. Activation of Nrf2 signaling, however, mitigated the adverse effects of thrombin on amnion cells and mouse fetal membranes with significant prolongation of thrombin-induced preterm birth in mice.

Previously, we found that COX-2, MMPs, and PGE2 were dramatically induced by thrombin in amnion mesenchymal, but not epithelial, cells (8). Thus, these cells were utilized to analyze the effects of Nrf2 activation on thrombin-induced activation of these proinflammatory pathways. We considered that amnion mesenchymal cells were deeply involved in the mechanism of preterm labor. On the other hand, we used two different reagents, DEM and 15d-PGJ2, both widely acknowledged as potent Nrf2 activators (20, 22, 23). In mice, however, 15d-PGJ2 was used because: 1) DEM was toxic to pregnant animals; and 2) 15d-PGJ2 was considered more physiological in that concentrations of 0.3–3 nm have been measured in human amniotic fluid samples (16).

Nrf2 activation in amnion mesenchymal cells resulted in inhibition of thrombin-induced COX-2 and PGE2 synthesis. COX-2, a rate-limiting enzyme in amnion prostaglandin synthesis, and PGE2 are involved in the initiation of labor through facilitation of uterine contractions and cervical ripening (24, 25). In addition, the rupture of fetal membranes is preceded by degradation of collagen that is mediated primarily by MMPs in amnion. Nrf2 activation mitigated thrombin-induced activation of major collagenases in human amnion cells and mouse fetal membranes. Collectively, our data raise the possibility that Nrf2 activation in mesenchymal cells may serve to prevent preterm birth or premature rupture of the membranes.

Several important cytokines were quantified in the conditioned medium of amnion mesenchymal cells, all of which were robustly reduced by Nrf2 activation. Intrauterine inflammation and high levels of cytokine are causally linked to preterm birth (26, 27) and to the development of cerebral palsy (28, 29), and importantly, the amnion is one of the major sources of cytokine production (30). Inhibition of cytokine production in amnion mesenchymal cells by Nrf2 activation, therefore, may have a distinct positive impact on perinatal and neonatal outcomes.

Regulation of Nrf2 is complex. Although most studies have focused on its repression by Keap1, recently it has become increasingly apparent that Nrf2 activity is also controlled by cross-talk with other signaling pathways including the glycogen synthase kinase-3 β-transducin repeat-containing protein axis (31), endoplasmic reticulum-associated degradation-associated E3 ubiquitin-protein ligase (Hrd1, also called synoviolin) (32), nuclear factor-kappa B (NF-κB) (33), Notch (34), and AMP kinase (35). Particularly relevant to the current study are findings that robust NF-κB and Nrf2 activities are essential to orchestrate cellular responses to resolve inflammatory pathways within many tissues (36–41).

Hence, although Nrf2 is believed to act predominantly through induction of cell defense mechanisms, its promiscuous interactions with NF-κB are also intricately linked to cell defense. Previously, we demonstrated that most thrombin-induced effects on proinflammatory genes in amnion were mediated by Toll-like receptor 4, not proteinase-activated receptor 1 (8). Thus, the mitigating effects of Nrf2 on thrombin-induced activation cytokines and proteases may be due to amelioration of NFκB gene activation. Consistent with our findings, Lee and Chau found that lipopolysaccharide (LPS)-induced COX-2 was negatively regulated by HO-1 induction, whereas inhibition of HO-1 increased LPS-induced COX-2 expression (42). Pirianov et al (43) showed that 15d-PGJ2 delayed LPS-induced preterm delivery in mice via inhibition of myometrial NF-κB but Nrf2 activation was not investigated.

In summary, these results indicate that Nrf2 activation represents a key stress response in amnion mesenchyme cells to mitigate the adverse proinflammatory effects of thrombin on the fetal membranes. We suggest, therefore, that pharmacological activation of Nrf2 may prevent the increased risk of preterm premature rupture of the membranes associated with subchorionic hemorrhage or bleeding during pregnancy.

Supplementary Material

jc-16-1059
Click here for additional data file. (140.8KB, pdf)

Acknowledgments

We thank Ms. Sylvia Wright, the hospital staff, and Parkland Hospital for assistance in obtaining fetal membranes for this study.

This work was supported by National Institutes of Health Grant P01 HD11149 and by March of Dimes Foundation Research Grant 21-FY13-35.

Disclosure Summary: The authors have nothing to disclose.

Abbreviations

DEM

diethyl maleate

15d-PGJ2

15-deoxy-Δ12, 14-prostaglandin J2

GAPDH

glyceraldehyde 3-phosphate dehydrogenase

GCLC

glutamyl cysteine ligase catalytic

GCLM

glutamyl cysteine ligase modulatory

GM-CSF

granulocyte-macrophage colony-stimulating factor

HO-1

heme oxygenase-1

Keap1

Kelch-like ECH-associated protein 1

LPS

lipopolysaccharide

MMP

matrix metalloproteinase

NF-κB

nuclear factor-κ B

NQO1

NAD(P)H:quinone oxidoreductase 1

Nrf2

nuclear factor erythroid 2-related factor 2

PGE2

prostaglandin E2

qPCR

quantitative PCR.

References

  • 1. Chang HH, Larson J, Blencowe H, et al. Preventing preterm births: analysis of trends and potential reductions with interventions in 39 countries with very high human development index. Lancet. 2013;381:223–234. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Hermans FJ, Kazemier BM, Mol BW. Prevention of preterm parturition. N Engl J Med. 2014;370:1860–1861. [DOI] [PubMed] [Google Scholar]
  • 3. Blencowe H, Cousens S, Oestergaard MZ, et al. National, regional, and worldwide estimates of preterm birth rates in the year 2010 with time trends since 1990 for selected countries: a systematic analysis and implications. Lancet. 2012;379:2162–2172. [DOI] [PubMed] [Google Scholar]
  • 4. Liu L, Johnson HL, Cousens S, et al. Global, regional, and national causes of child mortality: an updated systematic analysis for 2010 with time trends since 2000. Lancet. 2012;379:2151–2161. [DOI] [PubMed] [Google Scholar]
  • 5. Goldenberg RL, Culhane JF, Iams JD, Romero R. Epidemiology and causes of preterm birth. Lancet. 2008;371:75–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Mercer BM. Preterm premature rupture of the membranes: current approaches to evaluation and management. Obstet Gynecol Clin North Am. 2005;32:411–428. [DOI] [PubMed] [Google Scholar]
  • 7. Parry S, Strauss JF 3rd. Premature rupture of the fetal membranes. N Engl J Med. 1998;338:663–670. [DOI] [PubMed] [Google Scholar]
  • 8. Mogami H, Keller PW, Shi H, Word RA. Effect of thrombin on human amnion mesenchymal cells, mouse fetal membranes, and preterm birth. J Biol Chem. 2014;289:13295–13307. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Itoh K, Wakabayashi N, Katoh Y, et al. Keap1 represses nuclear activation of antioxidant responsive elements by Nrf2 through binding to the amino-terminal Neh2 domain. Genes Dev. 1999;13:76–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Suh JH, Shenvi SV, Dixon BM, et al. Decline in transcriptional activity of Nrf2 causes age-related loss of glutathione synthesis, which is reversible with lipoic acid. Proc Natl Acad Sci USA. 2004;101:3381–3386. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Komatsu M, Kurokawa H, Waguri S, et al. The selective autophagy substrate p62 activates the stress responsive transcription factor Nrf2 through inactivation of Keap1. Nat Cell Biol. 2010;12:213–223. [DOI] [PubMed] [Google Scholar]
  • 12. Choi AM, Alam J. Heme oxygenase-1: function, regulation, and implication of a novel stress-inducible protein in oxidant-induced lung injury. Am J Respir Cell Mol Biol. 1996;15:9–19. [DOI] [PubMed] [Google Scholar]
  • 13. Paine A, Eiz-Vesper B, Blasczyk R, Immenschuh S. Signaling to heme oxygenase-1 and its anti-inflammatory therapeutic potential. Biochem Pharmacol. 2010;80:1895–1903. [DOI] [PubMed] [Google Scholar]
  • 14. Casey ML, MacDonald PC. Interstitial collagen synthesis and processing in human amnion: a property of the mesenchymal cells. Biol Reprod. 1996;55:1253–1260. [DOI] [PubMed] [Google Scholar]
  • 15. Hari Kishore A, Li XH, Word RA. Hypoxia and PGE(2) regulate MiTF-CX during cervical ripening. Mol Endocrinol. 2012;26:2031–2045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Helliwell RJ, Keelan JA, Marvin KW, et al. Gestational age-dependent up-regulation of prostaglandin D synthase (PGDS) and production of PGDS-derived antiinflammatory prostaglandins in human placenta. J Clin Endocrinol Metab. 2006;91:597–606. [DOI] [PubMed] [Google Scholar]
  • 17. Ma Q. Role of nrf2 in oxidative stress and toxicity. Annu Rev Pharmacol Toxicol. 2013;53:401–426. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Wild AC, Moinova HR, Mulcahy RT. Regulation of γ-glutamylcysteine synthetase subunit gene expression by the transcription factor Nrf2. J Biol Chem. 1999;274:33627–33636. [DOI] [PubMed] [Google Scholar]
  • 19. Itoh K, Chiba T, Takahashi S, et al. An Nrf2/small Maf heterodimer mediates the induction of phase II detoxifying enzyme genes through antioxidant response elements. Biochem Biophys Res Commun. 1997;236:313–322. [DOI] [PubMed] [Google Scholar]
  • 20. Itoh K, Mochizuki M, Ishii Y, et al. Transcription factor Nrf2 regulates inflammation by mediating the effect of 15-deoxy-Delta(12,14)-prostaglandin j(2). Mol Cell Biol. 2004;24:36–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Lim R, Barker G, Lappas M. The transcription factor Nrf2 is decreased after spontaneous term labour in human fetal membranes where it exerts anti-inflammatory properties. Placenta. 2015;36:7–17. [DOI] [PubMed] [Google Scholar]
  • 22. Katsumata Y, Shinmura K, Sugiura Y, et al. Endogenous prostaglandin D2 and its metabolites protect the heart against ischemia-reperfusion injury by activating Nrf2. Hypertension. 2014;63:80–87. [DOI] [PubMed] [Google Scholar]
  • 23. Maruyama A, Mimura J, Itoh K. Non-coding RNA derived from the region adjacent to the human HO-1 E2 enhancer selectively regulates HO-1 gene induction by modulating Pol II binding. Nucleic Acids Res. 2014;42:13599–13614. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Challis JR, Sloboda DM, Alfaidy N, et al. Prostaglandins and mechanisms of preterm birth. Reproduction. 2002;124:1–17. [DOI] [PubMed] [Google Scholar]
  • 25. Kishore AH, Owens D, Word RA. Prostaglandin E2 regulates its own inactivating enzyme, 15-PGDH, by EP2 receptor-mediated cervical cell-specific mechanisms. J Clin Endocrinol Metab. 2014;99:1006–1018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Goldenberg RL, Hauth JC, Andrews WW. Intrauterine infection and preterm delivery. N Engl J Med. 2000;342:1500–1507. [DOI] [PubMed] [Google Scholar]
  • 27. Romero R, Grivel JC, Tarca AL, et al. Evidence of perturbations of the cytokine network in preterm labor. Am J Obstet Gynecol. 2015;213:836.e1–836.e18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Yoon BH, Jun JK, Romero R, et al. Amniotic fluid inflammatory cytokines (interleukin-6, interleukin-1β, and tumor necrosis factor-α), neonatal brain white matter lesions, and cerebral palsy. Am J Obstet Gynecol. 1997;177:19–26. [DOI] [PubMed] [Google Scholar]
  • 29. Yoon BH, Romero R, Park JS, et al. Fetal exposure to an intra-amniotic inflammation and the development of cerebral palsy at the age of three years. Am J Obstet Gynecol. 2000;182:675–681. [DOI] [PubMed] [Google Scholar]
  • 30. Keelan JA, Blumenstein M, Helliwell RJ, Sato TA, Marvin KW, Mitchell MD. Cytokines, prostaglandins and parturition–a review. Placenta. 2003;24(suppl A):S33–S46. [DOI] [PubMed] [Google Scholar]
  • 31. Hayes JD, Chowdhry S, Dinkova-Kostova AT, Sutherland C. Dual regulation of transcription factor Nrf2 by Keap1 and by the combined actions of β-TrCP and GSK-3. Biochem Soc Trans. 2015;43:611–620. [DOI] [PubMed] [Google Scholar]
  • 32. Kobayashi A, Kang MI, Okawa H, et al. Oxidative stress sensor Keap1 functions as an adaptor for Cul3-based E3 ligase to regulate proteasomal degradation of Nrf2. Mol Cell Biol. 2004;24:7130–7139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Wardyn JD, Ponsford AH, Sanderson CM. Dissecting molecular cross-talk between Nrf2 and NF-κB response pathways. Biochem Soc Trans. 2015;43:621–626. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Wakabayashi N, Skoko JJ, Chartoumpekis DV, et al. Notch-Nrf2 axis: regulation of Nrf2 gene expression and cytoprotection by notch signaling. Mol Cell Biol. 2014;34:653–663. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Mo C, Wang L, Zhang J, et al. The crosstalk between Nrf2 and AMPK signal pathways is important for the anti-inflammatory effect of berberine in LPS-stimulated macrophages and endotoxin-shocked mice. Antioxid Redox Signal. 2014;20:574–588. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Kim JE, You DJ, Lee C, Ahn C, Seong JY, Hwang JI. Suppression of NF-κB signaling by KEAP1 regulation of IKKβ activity through autophagic degradation and inhibition of phosphorylation. Cell Signal. 2010;22:1645–1654. [DOI] [PubMed] [Google Scholar]
  • 37. Liu GH, Qu J, Shen X. NF-κB/p65 antagonizes Nrf2-ARE pathway by depriving CBP from Nrf2 and facilitating recruitment of HDAC3 to MafK. Biochim Biophys Acta. 2008;1783:713–727. [DOI] [PubMed] [Google Scholar]
  • 38. Jeong WS, Kim IW, Hu R, Kong AN. Modulatory properties of various natural chemopreventive agents on the activation of NF-κB signaling pathway. Pharm Res. 2004;21:661–670. [DOI] [PubMed] [Google Scholar]
  • 39. Xu C, Shen G, Chen C, Gélinas C, Kong AN. Suppression of NF-κB and NF-κB-regulated gene expression by sulforaphane and PEITC through IκBα, IKK pathway in human prostate cancer PC-3 cells. Oncogene. 2005;24:4486–4495. [DOI] [PubMed] [Google Scholar]
  • 40. Sun CC, Li SJ, Yang CL, et al. Sulforaphane attenuates muscle inflammation in dystrophin-deficient mdx mice via NF-E2-related factor 2 (Nrf2)-mediated inhibition of NF-κB signaling pathway. J Biol Chem. 2015;290:17784–17795. [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]
  • 41. Mao L, Wang H, Qiao L, Wang X. Disruption of Nrf2 enhances the upregulation of nuclear factor-κB activity, tumor necrosis factor-α, and matrix metalloproteinase-9 after spinal cord injury in mice. Mediators Inflamm. 2010;2010:238321. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Lee TS, Chau LY. Heme oxygenase-1 mediates the anti-inflammatory effect of interleukin-10 in mice. Nat Med. 2002;8:240–246. [DOI] [PubMed] [Google Scholar]
  • 43. Pirianov G, Waddington SN, Lindström TM, Terzidou V, Mehmet H, Bennett PR. The cyclopentenone 15-deoxy-delta 12,14-prostaglandin J(2) delays lipopolysaccharide-induced preterm delivery and reduces mortality in the newborn mouse. Endocrinology. 2009;150:699–706. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

jc-16-1059
Click here for additional data file. (140.8KB, pdf)

Articles from The Journal of Clinical Endocrinology and Metabolism are provided here courtesy of The Endocrine Society

RESOURCES