Heme oxygenase-1 promotes migration and β-epithelial Na⁺ channel expression in cytotrophoblasts and ischemic placentas

Abstract

Preeclampsia is thought to arise from inadequate cytotrophoblast migration and invasion of the maternal spiral arteries, resulting in placental ischemia and hypertension. Evidence suggests that altered expression of epithelial Na⁺ channel (ENaC) proteins may be a contributing mechanism for impaired cytotrophoblast migration. ENaC activity is required for normal cytotrophoblast migration. Moreover, β-ENaC, the most robustly expressed placental ENaC message, is reduced in placentas from preeclamptic women. We recently demonstrated that heme oxygenase-1 (HO-1) protects against hypertension in a rat model of placental ischemia; however, whether HO-1 regulation of β-ENaC contributes to the beneficial effects of HO-1 is unknown. The purpose of this study was to determine whether β-ENaC mediates cytotrophoblast migration and whether HO-1 enhances ENaC-mediated migration. We showed that placental ischemia, induced by reducing uterine perfusion suppressed, and HO-1 induction restored, β-ENaC expression in ischemic placentas. Using an in vitro model, we found that HO-1 induction, using cobalt protoporphyrin, stimulates cytotrophoblast β-ENaC expression by 1.5- and 1.8-fold (10 and 50 μM). We then showed that silencing of β-ENaC in cultured cytotrophoblasts (BeWo cells), by expression of dominant-negative constructs, reduced migration to 56 ± 13% (P < 0.05) of control. Importantly, HO-1 induction enhanced migration (43 ± 5% of control, P < 0.05), but the enhanced migratory response was entirely blocked by ENaC inhibition with amiloride (10 μM). Taken together, our results suggest that β-ENaC mediates cytotrophoblast migration and increasing β-ENaC expression by HO-1 induction enhances migration. HO-1 regulation of cytotrophoblast β-ENaC expression and migration may be a potential therapeutic target in preeclamptic patients.

preeclampsia is a pregnancy-specific condition characterized by hypertension, proteinuria, and maternal vascular dysfunction. It is the leading cause of maternal and perinatal mortality and morbidity. Currently, there are no effective preventive and treatment strategies except for the early delivery of the placenta and fetus. Because the symptoms of preeclampsia remit after delivery, the placenta has been implicated as the major contributor to the pathophysiology of the disorder. The prevailing hypothesis is that preeclampsia is initiated by abnormal cytotrophoblast migration and subsequent invasion of the maternal uterine spiral arteries, resulting in restricted blood flow to the developing uteroplacental unit and placental ischemia. Indeed, preeclamptic patients have smaller diameter, higher-resistance spiral arteries compared with normal pregnant women (22). However, the exact mechanisms underlying impaired migration and invasion of cytotrophoblasts are not fully established, and therapeutic targets remain unclear.

One potential mechanism underlying inadequate cytotrophoblast migration is a reduction in epithelial Na⁺ channel (ENaC) protein expression. There are three lines of evidence that suggest ENaC proteins might contribute to altered cytotrophoblast migration/invasion in preeclampsia. First, several ENaC proteins (α, β, γ) are expressed in cytotrophoblasts and normal placental tissues (5, 19, 21, 27). Second, ENaC inhibition using amiloride, its analog benzamil, or gene silencing suppresses chemotactic migration of several cell types, including glia, vascular smooth muscle cells, and cytotrophoblasts (6, 13, 18, 19). Third, the placentas of preeclamptic patients have reduced levels of β-ENaC (27). This latter finding, in conjunction with an early study by McDonald et al. (21) showing β-ENaC is the predominant ENaC message expressed in the placenta, suggests that β-ENaC may play a pivotal role in cytotrophoblast function. However, studies examining the specific importance of β-ENaC in cytotrophoblast migration are lacking. Additionally, studies assessing whether potential therapeutic targets for the treatment of preeclampsia, such as heme oxygenase (HO-1), can regulate β-ENaC expression and migration of cytotrophoblasts have not been conducted.

Recently, our laboratory reported that HO-1 induction using cobalt protoporphyrin (CoPP) protects against hypertension in a rat model of placental ischemia (10). HO-1 is an enzyme involved in the catabolism of heme to produce iron, carbon monoxide (CO), and biliverdin, which is then converted to bilirubin (25). HO-1 is the inducible form of the enzyme and is induced mainly in stressful conditions (24). Pharmacological induction of HO-1 has been shown to promote vessel relaxation (1) and reduce placental ischemia- and sFlt-1-induced hypertension in rats (9, 10) and attenuate inflammation-mediated cell damage in placental villus explants (1). Conversely, deficiency of HO-1 produces features of preeclampsia and intrauterine growth restriction (31) and is characterized by inadequate remodeling of the maternal spiral arteries (30). Additionally, pharmacological inhibition of HO-1 promotes hypertension and reactive oxygen species in pregnant rats (12). Interestingly, heme, the substrate of HO-1, and CO, a HO-1 metabolite, are powerful inducers of ENaC activity in renal epithelial cells (28). However, whether the beneficial effects of HO-1 induction in placental ischemia are mediated by β-ENaC is unknown. Furthermore, the effects of HO-1 on cytotrophoblast migration and β-ENaC expression are also unknown.

In this study, we demonstrate that, placental β-ENaC expression is reduced, and induction of HO-1 restores β-ENaC expression in a rat model of placental ischemia that mimics many characteristics of human preeclampsia. We then show that induction of HO-1 in cultured cytotrophoblasts promotes migration and β-ENaC expression.

METHODS

Animals.

Timed-pregnant Sprague-Dawley rats (Harlan) were received on gestational day 11, housed in the Lab Animal Facility at the University of Mississippi Medical Center, and maintained on a 12:12-h light-dark cycle. Food and water were provided ad libitum. All protocols were approved by the Institutional Animal Care and Use Committee and followed the National Institutes of Health's “Guidelines for the Care and Use of Laboratory Animals”.

HO-1 induction in placental ischemic rats.

On gestational day 14, rats were anesthetized using 3% isoflurane, and placental ischemia (reduced uterine perfusion pressure, or RUPP) was induced by placement of silver clips on the abdominal aorta and each of the ovarian artery bundles, as previously described (2). Following surgery, the HO-1 inducer, cobalt (III) protoporphyrin IX chloride (CoPP, Frontier Scientific), was administered via intraperitoneal injection (5 mg/kg). This dose has been shown to reduce mean arterial pressure and increase HO-1 activity and expression in pregnant rats (10). On gestational day 19, placentas were collected for determination of β-ENaC expression.

β-ENaC expression in ischemic placentas.

Placentas were collected from normal pregnant and placental ischemic rats treated or untreated with CoPP (n = 4 per group/treatment). Tissues were then homogenized and membrane proteins were extracted using the plasma membrane protein extraction kit (BioVision) following manufacturer's directions. An equal amount of protein was loaded in 4–20% graded gel (Bio-Rad), electrophoresed, and transferred to nitrocellulose membrane. Membrane was then incubated, after blocking with Odyssey blocking buffer (LI-COR) for 1 h, in 1:1,000 rabbit anti-β-ENaC and 1:5,000 mouse anti-β-actin antibody overnight. This antibody has been thoroughly characterized to demonstrate specificity by our laboratory (7, 8, 13–15). Membrane was then washed and incubated in 1:2,000 donkey anti-rabbit IRDye 700 and 1:5,000 donkey anti-mouse IRDye 800 secondary antibody for 1 h at room temperature. Membrane was then scanned and analyzed using Odyssey infrared imager and software. β-ENaC expression was quantified as a ratio of β-actin expression.

In vitro model of cultured cytotrophoblasts.

BeWo cells (human choriocarcinoma cells) were cultured in DMEM/Hams F-12 media supplemented with 1% penicillin-streptomycin, 1% amphotericin, and 1% FBS in a 37°C humidified incubator.

HO-1 induction in cytotrophoblasts.

To induce HO-1 expression, cells were treated with 0, 1, 10, and 50 μM CoPP for 48 h and harvested at 0, 24, or 72 h after removal of CoPP. HO-1 protein was determined in cell lysates by Western blot using standard procedures. Whole cell lysates were separated on 10% Tris·HCl gel (Criterion, Bio-Rad, Hercules, CA) and transferred to nitrocellulose. The membrane was probed using 1:1,000 rabbit anti HO-1 (Enzo Life Sciences, Plymouth Meeting, PA) and 1:5,000 mouse anti β-actin (Abcam, Cambridge, MA) followed by 1:5,000 Alexa Fluor 700 goat anti-rabbit (Invitrogen, Eugene, OR) and 1:5,000 donkey anti-mouse IRDye 800 (Rockland, Gilbertsville, PA). Membrane labeling was visualized using an Odyssey infrared scanner. For all additional experiments, cells were treated with CoPP for 48 h and then used immediately thereafter.

β-ENaC immunofluorescence.

β-ENaC labeling was assessed in cytotrophoblasts plated on 8-well glass slides exposed to CoPP (10 or 50 μg/ml) for 48 h. Cells were rinsed in PBS, fixed in 4% paraformaldehyde for 10 min, and rinsed. Cells were incubated in 5% normal donkey sera (NDS) for 1 h, followed by rabbit anti-β-ENaC COOH-terminal antibody for 1 h in 5% NDS. Samples were rinsed with PBS and then labeled with donkey anti-rabbit Alexa Fluor 488 (1:1,000, Molecular Probes) in 5% NDS for 1 h. Samples were then thoroughly rinsed and visualized on a Leica TCS-SP8 laser scanning confocal microscope. Expression of β-ENaC was quantified as the intensity of staining normalized to area and was represented as arbitrary units per square micrometer. Images were prepared in Adobe Photoshop.

Cytotrophoblast migration.

To determine the spontaneous migration capacity of cytotrophoblasts, cells were trypsinized, counted, and then plated at a density of 10,000 cells per 100 μl on 8-μm pore Costar inserts in 0.4% FBS and allowed to migrate for 4 h. Immediately after, inserts were rinsed with PBS, and the cells on the surface were gently removed with a cotton swab. Cups were rinsed and fixed with ice-cold methanol for 10 min. After rinsing, cells were stained with hematoxylin for 10 min and rinsed. Cell migration was quantified by counting the number of cells on the bottom of the migration membrane in six fields of view per cup using a 20× objective lens. Experiments were conducted in at least three separate trials.

β-ENaC silencing.

To suppress β-ENaC expression, we transiently transfected 100,000 cells with 15 μg cDNA encoding an NH₂-terminal truncation of β-ENaC fused to the COOH terminus of enhanced green fluorescent protein (EGFP) using electroporation. The construct has been used previously by our laboratory to silence endogenous β-ENaC expression (7, 13, 14) in hypoosmotic buffer at 790 mV for 40 μs using an Eppendorf electroporator, according to the manufacturer's instructions. Cells were plated after 10-min incubation in normal media and allowed 48 h to grow before migration assay was performed. EGFP expression was used as an indicator of transfection efficiency.

Statistics.

Data are represented as means ± SE. ANOVA with Student-Neuman-Keuls post hoc test or unpaired t-tests was used to assess differences between groups where appropriate. P < 0.05 was considered statistically significant. Graphs and statistical analyses were conducted using GraphPad Prism software.

RESULTS

Does placental ischemia suppress β-ENaC expression and can this be restored by induction of HO-1?

Studies have shown that β-ENaC expression is decreased in placentas from preeclamptic women (27) and that HO-1 induction reduces blood pressure in placental ischemic rats (10). Therefore, we determined whether β-ENaC protein is reduced in ischemic placentas. Placentas were collected from normal pregnant and placental ischemic rats (RUPP) that were treated or untreated with CoPP. Western blot analysis of placental β-ENaC expression revealed a 58% decrease in β-ENaC expression in placental ischemic rats compared with the normal pregnant rats (P = 0.036). HO-1 induction had no effect on β-ENaC expression in normal pregnant rats but normalized placental β-ENaC expression in placental ischemic rats (Fig. 1). These findings suggest that HO-1 induction restores placental β-ENaC expression associated with placental ischemia.

Fig. 1.

Heme oxygenase-1 (HO-1) induction normalizes β-ENaC expression in ischemic placentas. Representative Western blot and quantification of blots for β-epithelial sodium channel (β-ENaC) expression in rat placentas. The order of the bands on the Western blot corresponds to the order of the bar graph. Placental ischemia (RUPP) induces a reduction in β-ENaC expression. HO-1 induction increases β-ENaC expression in ischemic placentas to the level of the normal pregnant (NP) rats. Values presented in bars represent means ± SE; n = 4 per group/treatment. CoPP, cobalt protoporphyrin; RUPP, reduced uterine perfusion pressure.

Does CoPP stimulate HO-1 induction in cultured cytotrophoblasts?

To determine whether CoPP can induce HO-1 expression in a commonly used model of cultured cytotrophoblasts, we exposed BeWo cells to CoPP (1, 10, and 50 μM) for 48 h and then examined HO-1 expression by Western blot analysis. We found that while HO-1 was undetectable in untreated cells, CoPP concentration- and time-dependently increased the expression of HO-1 in BeWo cells, which was evident up to 72 h posttreatment (Fig. 2).

Fig. 2.

CoPP dose- and time-dependently induces HO-1 expression in cultured cytotrophoblasts. Western blot showing the expression of HO-1 and β-actin in cultured cytotrophoblasts treated with CoPP for 48 h. CoPP treatment induces HO-1 expression, which persists up to 72 h after removal of CoPP.

Does HO-1 induction stimulate β-ENaC expression in cytotrophoblasts?

To determine whether HO-1 induction enhances β-ENaC expression in vitro, we measured β-ENaC protein expression using semiquantitative immunofluorescence, an approach our laboratory has used successfully with our well-characterized β-ENaC antibodies (7, 13, 14, 26). For these experiments, cytotrophoblasts were cultured with or without CoPP (10 and 50 μM) for 48 h to induce HO-1 expression before β-ENaC labeling. As shown in Fig. 3, β-ENaC expression is low (44 ± 3 arbitrary units/μm²) in untreated cytotrophoblasts; however, treatment with CoPP (10 and 50 μM) robustly stimulated β-ENaC expression (66 ± 5 and 80 ± 5 arbitrary units/μm², respectively; P < 0.05). These findings demonstrate that HO-1 induction increases β-ENaC expression.

Fig. 3.

Induction of HO-1 using CoPP dose-dependently increases β-ENaC expression in cytotrophoblasts. Cytotrophoblasts were treated with CoPP for 48 h and then processed for immunocytochemistry. Representative images for each dose of CoPP are shown along with the quantified group data. Induction of HO-1 upregulates β-ENaC protein expression in a dose-dependent manner. Data are expressed as means ± SE; n = 14–30.

Does β-ENaC mediate cytotrophoblast migration?

Previous studies suggest that ENaC inhibition, using amiloride and its analogs, inhibits cytotrophoblast migration (5, 6). However, the importance of individual isoforms, such as β-ENaC, cannot be determined using these inhibitors. To specifically determine the role of β-ENaC, we used the expression of cDNA encoding for an NH₂-terminal truncation of β-ENaC. Previous studies from our laboratories demonstrate that expression of the NH₂-terminal truncated protein suppresses the endogenous subunit expression (7, 13, 14). As shown in Fig. 4, expression of the β-ENaC dominant-negative construct significantly decreased spontaneous migration in cultured cytotrophoblasts to 54 ± 12.6% of control (n = 6; P = 0.0325). These findings demonstrate that β-ENaC plays an important role in cytotrophoblast migration.

Fig. 4.

β-ENaC silencing inhibits cytotrophoblast migration. Silencing of β-ENaC using a dominant-negative β-ENaC NH₂-terminal truncated construct [enhanced green fluorescent protein (EGFP)-β-ENaC_I41X] reduces spontaneous migration of cultured cytotrophoblasts. Values in bars represent means ± SE.

Does enhanced β-ENaC expression in cytotrophoblasts stimulate migration?

To determine whether HO-1 induction can stimulate cytotrophoblast migration and whether the enhanced migration is mediated by ENaC, we used a standard Boyden chamber assay to assess spontaneous migration through 8-μm pores. As shown in Fig. 5, both CoPP and amiloride had a significant effect on cytotrophoblast migration (P = 0.002 and P < 0.001, respectively). There was no interaction between amiloride and CoPP on migration (P = 0.468). Both 1 and 10 μM CoPP pretreatment significantly increased migratory capacity above control (135 ± 12 and 143 ± 5%; P < 0.003 and 0.001, respectively). ENaC inhibition with 10 μM amiloride suppressed migration in the absence of CoPP (56 ± 7%) and abolished the enhanced migratory response to CoPP (66 ± 29, 89 ± 9% for 1 and 10 μM, respectively). These findings suggest that the enhanced cytotrophoblast migratory response to HO-1 induction is mediated by ENaC.

Fig. 5.

HO-1 induction enhanced cytotrophoblast migration in vitro is dependent on ENaC. Exposure to CoPP (1 and 10 μM, 48 h) enhances spontaneous cytotrophoblast migration using a modified Boyden chamber assay. ENaC inhibition with amiloride (10 μM) abolishes effect of CoPP on migration. Migration is represented as the percentage of control. Values in bars represent means ± SE; n = 4–11. *P < 0.05 compared with control, untreated cells.

DISCUSSION

The depth of cytotrophoblast migration and invasion in preeclamptic women is shallow and does not penetrate the myometrial boundary of the maternal spiral arteries compared with normal pregnant women (16, 17). Shallow invasion of the maternal spiral arteries by cytotrophoblasts results in poorly remodeled spiral arteries and reduced blood flow to the uteroplacental unit, creating a hypoxic/ischemic environment. The mechanisms underlying the insufficient migration/invasion are unclear. One potential mechanism may be altered expression and function of β-ENaC proteins. β-ENaC is robustly expressed in placental tissue and is an important mediator of glial and VSMC migration (13). The reduced cytotrophoblast β-ENaC expression in preeclamptic placentas may contribute to the impaired cytotrophoblast migration/invasion observed in preeclampsia. In the current study, we demonstrated that β-ENaC protein is reduced in ischemic placentas and that silencing β-ENaC in cultured cytotrophoblasts inhibited spontaneous migration by nearly 50%, a finding that suggests β-ENaC plays a critical role in cytotrophoblast migration and the etiology of preeclampsia.

Currently, there are no effective treatment options available for preeclampsia other than early delivery of the fetus. Therefore, developing therapeutic targets for the treatment of preeclampsia is desired. One potential therapeutic approach may be induction of HO-1. Two lines of evidence suggest HO-1 may be an ideal target. First, our laboratory demonstrated that induction of the enzyme HO-1 attenuates placental ischemia-induced hypertension in pregnant rats (9–11). Second, inhibition of HO activity (confirmed in liver and placentas) using tin mesoporphyrin in pregnant rats promotes the development of hypertension and oxidative stress (12). Third, inhibition of HO in primary human trophoblasts results in reduced invasion in vitro (20). Precisely how HO contributes to the etiology of preeclampsia is unclear; however, one possibility may be regulation of cytotrophoblast migration/invasion, the primary underlying cause of preeclampsia. In this study, we demonstrate that induction of HO-1 using cobalt protoporphyrin (CoPP) stimulates migration of cultured cytotrophoblasts.

Since HO and β-ENaC both contribute to cytotrophoblast migration, we considered the possibility that HO-1 might increase migration by enhancing β-ENaC expression. Using immunofluorescence, we found that HO-1 robustly stimulates cytotrophoblast β-ENaC protein expression. While there is limited evidence in the literature related to the role of HO-1 in regulating β-ENaC expression, a recent study by Wang et al. (28) found that heme (the substrate for HO) inhibits, while carbon monoxide (HO metabolite) stimulates, ENaC activity in renal epithelial cells. However, another study demonstrated that CO has the capacity to inhibit amiloride-sensitive channels in airway epithelial cells (3). These studies do not specify the ENaC subunit affected by the treatments; therefore, it is still unclear whether HO-1 has the capacity to increase expression and activity of all the ENaC subunits. Moreover, while our data show that HO-1 induction increases β-ENaC expression, we do not know whether activity is also increased. Future studies are needed to determine the exact mechanisms by which HO-1 induction stimulates β-ENaC expression. Additionally, whether the byproducts of heme metabolism contribute to the improvements in trophoblast migration and β-ENaC expression is an area of active investigation.

While we show that HO-1 induction plays an important role in cytotrophoblast function, HO-2 may also be involved. Indeed, placentas from guinea pigs demonstrate low staining for HO-2 protein (23) and have been reported to be expressed in extravillous trophoblasts (4). Importantly, HO-2 expression is decreased in endothelial cells from preeclamptic and fetal growth-restricted placentas (4). Lastly, HO-2 expression is decreased in synctiotrophoblasts and invasive trophoblasts from third-trimester preeclamptic patients (29). These seminal studies illustrate that HO-2, the constitutively expressed enzyme, may play a role in the pathophysiology of preeclampsia. Future studies will assess the role of HO-2 in our rat model of placental ischemia.

A key finding in this study is that HO-1 induction can positively regulate β-ENaC expression in ischemic placentas, as well as in cultured trophoblasts. Similar to the findings by Wang et al. (27), we show that knockdown of ENaC expression results in decreased migratory capacity of cultured cytotrophoblasts. Additionally, the expression of the three major subunits of ENaC is decreased in placentas from preeclamptic patients compared with normal pregnant patients (19, 27). Wang et al. (27) recently demonstrated that α and β-ENaC are temporally regulated in the placenta. αENaC is abundantly expressed during first trimester, while β-ENaC is more abundant at full term. The authors also demonstrated that knockdown of αENaC in first trimester extravillous trophoblasts reduced migration and invasion of these cells in vitro (27). On the basis of evidence that β-ENaC promotes and is necessary for the migration of vascular smooth muscle cells and syncytiotrophoblasts (6, 13), we hypothesize that HO-1 induction of β-ENaC (and potentially other ENaC subunits) restores the migratory capacity of the cytotrophoblasts, thereby improving placental function associated with placental ischemia. Whether the other major subunits of ENaC (α and γ) are decreased in our placental ischemic rat model and whether they contribute to increased migratory capacity are not known and is an area for future investigation.

Significance and Perspectives

Our data support the hypothesis that HO-1 improves placental ischemia-mediated hypertension through restoring and/or upregulating β-ENaC expression in ischemic placentas. Inducing HO-1 presents a potential therapy for improving preeclampsia symptoms in patients. Further studies are required to determine whether HO induction improves spiral artery remodeling and blood flow in placental ischemic conditions. Because the in vitro studies were conducted in human choriocarcinoma cell lines, it will be important to confirm these studies in primary trophoblast cultures.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Author contributions: J.P.W., K.C., C.S., P.A.H., E.M.G., and H.A.D. performed experiments; J.P.W. and H.A.D. analyzed data; J.P.W., P.A.H., and H.A.D. interpreted results of experiments; J.P.W. and H.A.D. prepared figures; J.P.W. and H.A.D. drafted manuscript; J.P.W., D.E.S., M.J.R., J.P.G., and H.A.D. edited and revised manuscript; J.P.W., D.E.S., M.J.R., J.P.G., and H.A.D. approved final version of manuscript; M.J.R., J.P.G., and H.A.D. conception and design of research.