Altered Redox State Modulates Endothelial K Ca 2.3 and K Ca 3.1 Levels in Normal Pregnancy and Preeclampsia

Abstract

Aims:

Altered redox state has been related to the development of normal pregnancy (NP) and preeclampsia (PE). Endothelial K_Ca2.3 and K_Ca3.1 (K_Cas) play an important role in vasodilation, and K_Cas levels are affected by oxidative stress. We investigated the mechanisms of oxidative stress-mediated K_Cas expression modulation during NP and PE.

Results:

Human uterine microvascular endothelial cells were incubated in serum from normal nonpregnant women (n = 13) and women with NP (n = 24) or PE (n = 15), or in vascular endothelial growth factor (VEGF), oxidized low-density lipoprotein (ox-LDL), progesterone, or estradiol-17β (E₂)-containing medium for 24 h. NP serum elevated H₂O₂ levels via reducing catalase and glutathione peroxidase 1 levels, thereby enhancing K_Cas levels via a H₂O₂/fyn/extracellular signal-regulated kinase (ERK)-mediated pathway. VEGF enhanced H₂O₂ and K_Cas levels and K_Ca3.1 currents. K_Cas were upregulated and K_Cas activation-induced endothelium-dependent relaxation (EDR) was augmented in vessels from pregnant mice and rats. Whereas PE serum, ox-LDL, progesterone, or soluble fms-like tyrosine kinase 1 (sFlt-1) elevated superoxide levels via elevating NADPH oxidase 2 (NOX2) and NOX4 levels and reducing superoxide dismutase (SOD) 1 levels, thereby downregulating K_Cas. sFlt-1 inhibited EDR. PE serum- or progesterone-induced alterations in levels of K_Cas were reversed by polyethylene glycol-SOD, NOX inhibition, or E₂.

Innovation and Conclusions:

This is the first study of how endothelial K_Cas levels are modulated during NP and PE. K_Cas were upregulated by soluble serum factors such as VEGF via H₂O₂ generation in NP, and were downregulated by serum factors such as progesterone and ox-LDL via superoxide generation in PE, which may contribute to hemodynamic adaptations in NP or to the development of PE.

Innovation

Endothelial K_Ca2.3 and K_Ca3.1 (K_Cas) induce vasodilation. K_Cas upregulation may inhibit blood pressure increases, with K_Cas downregulation having the opposite effect. Oxidative stress, which is increased in pregnancy, might affect K_Cas levels. We thus examined mechanisms for oxidative stress-induced K_Cas modulation in normal pregnancy (NP) and preeclampsia (PE). In NP, K_Cas were upregulated via downregulating H₂O₂-degrading catalase and glutathione peroxidase 1 and activating a H₂O₂/fyn/extracellular signal-regulated kinase (ERK) pathway. Whereas K_Cas were downregulated via upregulating NADPH oxidase 2 (NOX2) and NOX4 and downregulating superoxide dismutase (SOD) 1 in PE. Vascular endothelial growth factor (VEGF), estrogen, progesterone, and oxidized low-density lipoprotein in serum act as modulators of K_Cas levels.

Introduction

Oxidative stress is increased in preeclampsia (PE) compared with normal pregnancy (NP) (22, 25, 29). Superoxide production has been suggested to be greater in PE than in NP, whereas blood levels of antioxidants have been reported to be decreased in women with PE. Circulating plasma substances, such as placental debris, oxidized low-density lipoprotein (ox-LDL), and the antiangiogenetic factor soluble fms-like tyrosine kinase 1 (sFlt-1), stimulate reactive oxygen species (ROS) production (32), especially from endothelial cells (ECs), and contribute to the development of PE (21, 23, 37).

Increased oxidative stress leads to endothelial dysfunction, thereby causing increased blood pressure in PE. On the contrary, the proangiogenesis factor vascular endothelial growth factor (VEGF), which induces placental development during pregnancy, also stimulates ROS generation via activation of NADPH oxidases (NOXs) (37). In addition, pregnancy hormones (estrogen, progesterone), which control vascular contractility (9), stimulate ROS generation (5, 26, 38). These results suggest that redox state plays an important role in regulating blood pressure during pregnancy.

Ca²⁺-activated K⁺ channels (K_Ca2.3 and K_Ca3.1) play important roles in endothelial control of vasorelaxation. K_Ca2.3 and K_Ca3.1 evoke endothelium-dependent hyperpolarization and stimulate nitric oxide release by promoting Ca²⁺ influx through Ca²⁺ entry channels. Thus, K_Ca2.3 and K_Ca3.1 downregulation or deficiency enhances vascular contractility and elevates blood pressure (7, 28), whereas upregulation of these K⁺ channels promotes endothelium-dependent relaxation (EDR) (10).

Among these K⁺ channels, K_Ca3.1 levels and currents are enhanced by H₂O₂ and reduced by superoxide (11, 12). In addition, K_Ca3.1 levels are affected by various circulating plasma substances, such as platelet-derived growth factors and ox-LDL. Messenger RNA (mRNA) or protein levels of K_Ca3.1 are enhanced by platelet-derived growth factors and downregulated by ox-LDL, respectively (6, 11). Platelet-derived growth factors and ox-LDL stimulate ROS generation (18, 31). These results suggest that these plasma substances may regulate K_Ca3.1 levels via ROS generation. In addition, estrogen and progesterone may control K_Ca3.1 levels via ROS generation during pregnancy. However, little is known about whether these pregnancy hormones modulate K_Ca3.1 levels. Furthermore, although K_Ca2.3 plays an important role in endothelial functions, little is known about K_Ca2.3 level regulation in NP and PE.

Since redox state is controlled by the superoxide-producing membrane-bound enzymes NOXs and endogenous antioxidant enzymes such as superoxide dismutase (SOD), catalase, and glutathione peroxidase (GPX), altering levels of NOXs and/or antioxidant enzymes might change redox state. Therefore, we hypothesized that an altered redox state caused by altered levels of NOXs and antioxidant enzymes modulates K_Ca2.3 and K_Ca3.1 levels in NP and PE. We examined this hypothesis and the roles of plasma substances (VEGF, sFlt-1, ox-LDL, estrogen, and progesterone) in the modulation of these K⁺ channel levels via altering NOXs and antioxidant enzymes. The results showed that altered levels of these enzymes upregulate K_Ca2.3 and K_Ca3.1 by generating H₂O₂ in NP, whereas K_Ca2.3 and K_Ca3.1 are downregulated by superoxide generation in PE.

Results

NP serum upregulates endothelial K_Ca2.3 and K_Ca3.1 via VEGF receptor activation

We compared the effects of soluble serum factors in serum from normal nonpregnancy (NNP) and NP women on K_Ca2.3 and K_Ca3.1 expression by incubating human uterine microvascular ECs (HUtMECs) in serum from NNP and NP women for 24 h. K_Ca2.3 (Fig. 1A) and K_Ca3.1 levels (Fig. 1B) were markedly higher in HUtMECs treated with NP serum than in HUtMECs treated with NNP serum. The NP serum-mediated increase in these K⁺ channels was reduced by blocking VEGF receptors (VEGFRs) with an anti-VEGFR1 antibody (Ab) or anti-VEGFR2 Ab (Fig. 1A, B), indicating that NP serum increased K_Ca2.3 and K_Ca3.1 levels via VEGFR activation. We thus examined the effects of VEGF (10 ng/mL) on K_Ca2.3 and K_Ca3.1 levels by incubating HUtMECs in VEGF-containing culture medium for 24 h. VEGF treatment increased protein levels of K_Ca2.3 and K_Ca3.1 in a concentration-dependent manner (Fig. 1C, D), and K_Ca3.1 currents were enhanced from 23.3 ± 2.7 pA/pF of control to 48.3 ± 3.1 pA/pF of VEGF-treated ECs (Fig. 1E). We then examined the effects of VEGF (10 ng/mL) on NP serum- or PE serum-treated ECs. VEGF further enhanced K_Ca2.3 and K_Ca3.1 levels (Fig. 1F, G) and K_Ca3.1 currents from 46.2 ± 5.0 pA/pF of NP serum-treated ECs to 77.8 ± 6.9 pA/pF of (NP serum+VEGF)-treated ECs (Fig. 1H) in NP serum-treated cells. In contrast, VEGF did not affect K_Ca2.3 (Fig. 1F) and K_Ca3.1 levels (Fig. 1G) in PE serum-treated cells, indicating the presence of VEGF antagonists in PE serum. These data suggest that K_Ca2.3 and K_Ca3.1 are upregulated by soluble serum factors, such as VEGF, during pregnancy.

FIG. 1.

Endothelial K_Ca2.3 and K_Ca3.1 are upregulated during NP. (A, B) Western blot showing protein levels of K_Ca2.3 (A) and K_Ca3.1 (B) (n = 3 or 5 per group). HUtMECs were exposed to NNP or NP serum for 24 h. Increases in K_Ca2.3 and K_Ca3.1 levels by treatment with NP serum were inhibited by anti-VEGFR1 Ab or anti-VEGFR2 Ab. (C, D) Western blot showing protein levels of K_Ca2.3 (C) and K_Ca3.1 (D) in ECs treated with VEGF for 24 h (n = 3 or 5 per group). (E) HUtMECs were exposed to VEGF for 24 h, and functional K_Ca3.1 expression was determined by whole-cell patch clamp. (E, H) K_Ca3.1 currents were activated by loading 1 μM Ca²⁺ via a patch pipette in whole-cell clamped HUtMECs and adding the K_Ca2.3 and K_Ca3.1 activator 1-EBIO (100 μM) to the external solution (n = 6 per group). K_Ca3.1 current was normalized to cell capacitance and TRAM-34-sensitive current was measured as the K_Ca3.1 current. Summary of K_Ca3.1 current densities at +50 mV (lower panel). (F, G) Western blot showing protein levels of K_Ca2.3 (F) and K_Ca3.1 (G) in ECs treated with NP serum or (NP serum+VEGF) and in ECs treated with PE serum and (PE serum+VEGF) for 24 h (n = 3 or 5 per group). (H) HUtMECs were exposed to NP serum or (NP serum+VEGF) for 24 h, and K_Ca3.1 currents were recorded (n = 8 per group). Data are mean ± SEM. *p < 0.05, **p < 0.01. For original immunoblot files, see Supplementary Fig. S5. 1-EBIO, 1-ethyl-2-benzimidazolinone; Ab, antibody; ECs, endothelial cells; HUtMECs, human uterine microvascular ECs; NNP, normal nonpregnancy; NP, normal pregnancy; PE, preeclampsia; SEM, standard error of the mean; VEGF, vascular endothelial growth factor; VEGFR, VEGF receptor.

K_Ca2.3 and K_Ca3.1 upregulation enhances EDR during pregnancy

K_Ca2.3 and K_Ca3.1 upregulation might promote EDR. We thus examined whether these K⁺ channels were upregulated to enhance EDR in maternal blood vessels during pregnancy by studying nonpregnant and pregnant mice and rats. Protein levels of K_Ca2.3 (Fig. 2A) and K_Ca3.1 (Fig. 2B) were substantially increased in aortic tissues from pregnant mice compared with levels in nonpregnant mice. We then compared the magnitude of EDR of aortic rings from these mice and rings of the first branch and trunk of rat superior mesenteric artery (SMA). The EDR response to acetylcholine (ACh) was significantly increased in aortic rings from pregnant mice (Fig. 2C) and in rings of SMA branch from pregnant rats (Fig. 2D) compared with the response in rings from nonpregnant mice and in SMA rings from nonpregnant rats, respectively. We then inhibited nitric oxide production by pretreatment with N ^ω-nitro-l-arginine (l-NOARG), and examined K_Ca2.3 and K_Ca3.1 activator-induced EDR of endothelium-intact aortic rings. The K_Ca2.3 and K_Ca3.1 activator 1-ethyl-2-benzimidazolinone (1-EBIO) relaxed precontracted endothelium-intact aortic rings (Fig. 2E), but did not relax precontracted endothelium-denuded aortic rings (data not shown), indicating that 1-EBIO-induced relaxation was endothelium dependent. Compared with relaxation in nonpregnant mice, EDR to 1-EBIO was significantly increased in pregnant mice (Fig. 2E), indicating that the contribution of K_Ca2.3 and K_Ca3.1 to EDR was markedly increased in pregnant mice. In addition, we examined ACh-induced EDR in rings of the first branch and trunk of SMA from pregnant and nonpregnant rats in which nitric oxide production was inhibited by pretreatment with l-NOARG. l-NOARG-resistant EDR to ACh was significantly increased in the first branch of SMA (Fig. 2F) and SMA trunk (Fig. 2G) from pregnant rats compared with relaxation in the first branch of SMA and SMA trunk from nonpregnant rats, respectively. These results suggest that increased expression of K_Ca2.3 and K_Ca3.1 enhances endothelial functions such as vasorelaxation in maternal blood vessels during pregnancy.

FIG. 2.

K_Ca2.3 and K_Ca3.1 upregulation enhanced EDR during pregnancy. (A, B) Western blot showing protein levels of K_Ca2.3 (A) and K_Ca3.1 (B) in aortas from 25-week-old nonpregnant and pregnant mice. K_Ca2.3 and K_Ca3.1 are not expressed in vascular smooth muscle cells but are expressed in ECs. Thus, the K⁺ channel levels measured in aortic tissues represent the levels in aortic ECs. One aortic ring was obtained from each mouse, and graphs were computed with pooled data from nonpregnant and pregnant mice (n = 3 per group). (C–G) EDR was induced by ACh (C, D, F, G) or by the K_Ca2.3 and K_Ca3.1 activator 1-EBIO (100 μM) in aortic rings from 25-week-old nonpregnant or pregnant mice (C, E) and in rings of SMA branch (D, F) and trunk (G) from 25-week-old nonpregnant or pregnant rats. In each experiment, one arterial ring was obtained from each mouse or rat, and graphs were computed with pooled data from nonpregnant or pregnant mice or rats (n = 8 per group). The magnitude of maximal relaxation during each treatment was expressed as a percentage of initial prostaglandin F_2α- or norepinephrine-induced contraction. EDR in arterial rings without l-NOARG pretreatment (C, D) or in arterial rings in which nitric oxide production was inhibited by l-NOARG pretreatment (E–G). Data are mean ± SEM. *p < 0.05, **p < 0.01. For original immunoblot files, see Supplementary Fig. S5. ACh, acetylcholine; EDR, endothelium-dependent relaxation; l-NOARG, N ^ω-nitro-l-arginine; SMA, superior mesenteric artery. To see this illustration in color, the reader is referred to the web version of this article at www.liebertpub.com/ars

Pregnancy-associated K_Ca2.3 and K_Ca3.1 upregulation is attenuated in PE

We then examined K_Ca2.3 expressions in placental tissues from NP and PE using an immunohistochemical technique targeted at the K_Ca2.3 proteins. Figure 3A shows a cross-sectioned chorionic villus containing blood vessels in placental tissues. K_Ca2.3 expressions were substantially decreased in the ECs of placental tissues from PE compared with placental tissues from NP (Fig. 3A). The decrease in K_Ca2.3 expressions in placental tissues from women with PE was confirmed by Western blot analysis. Protein levels of K_Ca2.3 were markedly decreased in PE placental tissues compared with levels in NP placental tissues (Fig. 3B). In addition, K_Ca2.3 levels in ECs exposed to PE serum were significantly reduced compared with those in ECs exposed to NP serum (Fig. 3C). Then, HUtMECs were exposed to ox-LDL or lysophosphatidylcholine (LPC). Treatment of HUtMECs with ox-LDL or LPC reduced K_Ca2.3 levels (Fig. 3C). K_Ca3.1 levels were also markedly decreased in placental tissues from PE compared with levels in placental tissues from NP (Supplementary Fig. S1; Supplementary Data are available online at www.liebertpub.com/ars). Previously, we reported that endothelial K_Ca3.1 is downregulated in umbilical vessels from PE subjects and that PE plasma, ox-LDL, or LPC reduced K_Ca3.1 levels (11). These results suggest that both K_Ca3.1 and K_Ca2.3 are downregulated in ECs from PE.

FIG. 3.

Pregnancy- associated K_Ca2.3 upregulation is attenuated in PE. (A, B) K_Ca2.3 protein expression was compared in placental tissues from women with NP and PE by immunohistochemistry (A) and Western blotting (B). (A) Brown color (black arrowheads) denotes K_Ca2.3 proteins. Higher magnification images of the boxed area in each panel are displayed within the panel. Scale bar: 10 μm. (B) K_Ca2.3 protein level was quantified for four subjects per experimental group. (C) Western blot showing K_Ca2.3 protein levels in ECs treated with NP or PE serum (n = 4 per group), or treated with ox-LDL (n = 5 per group) and LPC (n = 3 per group). Data are mean ± SEM. **p < 0.01. For original immunoblot files, see Supplementary Fig. S5. LPC, lysophosphatidylcholine; ox-LDL, oxidized low-density lipoprotein. To see this illustration in color, the reader is referred to the web version of this article at www.liebertpub.com/ars

NOX4 and antioxidant enzyme levels in ECs treated with NP and PE serum

We previously reported that endothelial NOX2 is upregulated in umbilical vessels from PE subjects (11). However, levels of NOX1, NOX3, and NOX5 are not altered in vessels from NP and PE (11). We thus compared NOX4 levels, which are highly expressed under normal physiological conditions in ECs and might be constitutively activated (3), between ECs treated with NNP and NP serum, and between ECs treated with NP and PE serum. NOX4 levels were significantly decreased in ECs treated with NP serum versus ECs treated with NNP serum, whereas NOX4 levels were markedly enhanced in ECs treated with PE serum versus ECs treated with NP serum (Fig. 4A). SOD1 levels were significantly increased in ECs treated with NP serum versus ECs treated with NNP serum, whereas SOD1 levels were markedly reduced in ECs treated with PE serum versus ECs treated with NP serum (Fig. 4B). We then compared levels of catalase and GPX1 between ECs treated with NNP and NP serum, and between ECs treated with NP and PE serum (Fig. 4C, D). Catalase and GPX1 levels were significantly reduced in ECs treated with NP serum versus ECs treated with NNP serum, whereas catalase and GPX1 levels were markedly enhanced in ECs treated with PE serum versus ECs treated with NP serum. These results suggest that redox state is altered via alterations in NOXs and antioxidant enzyme levels in NP and PE.

FIG. 4.

Levels of NOX4 and antioxidant enzymes in ECs treated with NP and PE serum. HUtMECs were incubated in NNP, NP, or PE serum for 24 h, and protein levels of NOX4 or antioxidant enzymes were measured using Western blotting. (A, B) NOX4 (A) and SOD1 levels (B) were compared between ECs treated with NNP and NP serum, and between ECs treated with NP and PE serum (n = 4 or 6 per group). NOX upregulation and SOD downregulation may elevate superoxide levels. (C, D) Levels of catalase (C) and GPX1 (D) were compared between ECs treated with NNP and NP serum, and between ECs treated with NP and PE serum (n = 3 or 5 per group). Data are mean ± SEM. *p < 0.05, **p < 0.01. For original immunoblot files, see Supplementary Fig. S5. GPX, glutathione peroxidase; NOX, NADPH oxidase; SOD, superoxide dismutase.

Altered redox state modulates K_Ca2.3 and K_Ca3.1 levels in NP and PE

Since catalase and GPX degrade H₂O₂, catalase and GPX1 downregulation might elevate H₂O₂ levels. We thus examined whether NP serum-mediated catalase and GPX1 downregulation elevated H₂O₂ levels by using peroxy-orange 1 in ECs treated with NNP and NP serum. H₂O₂ levels were markedly increased in ECs treated with NP serum versus ECs treated with NNP serum (Fig. 5A; 156% ± 15% of control). In addition, treating HUtMECs with VEGF (10 ng/mL) elevated H₂O₂ levels (Fig. 5A; 5 ng/mL: 112% ± 4% of control; 10 ng/mL: 119 ± 6% of control). In contrast, NP serum and VEGF treatment did not elevate superoxide levels in HUtMECs (Supplementary Fig. S2A, B). We then examined whether H₂O₂ was involved in NP serum-mediated K_Ca2.3 and K_Ca3.1 upregulation. NP serum-mediated K_Ca2.3 and K_Ca3.1 upregulation was reversed by the membrane-permeable catalase, polyethylene glycol (PEG) catalase (Fig. 5B, C). H₂O₂ activates the Src family kinase fyn (24). We therefore compared phosphorylated fyn (p-fyn) levels in ECs treated with NNP and NP serum. Treatment with NP serum increased p-fyn levels compared with treatment with NNP serum, and this NP serum-mediated p-fyn increase was reversed by treatment with PEG catalase (Fig. 5D). As H₂O₂ and fyn play critical roles in extracellular signal-regulated kinase (ERK) activation and K_Ca3.1 protein is produced via a H₂O₂/fyn/ERK-dependent pathway (10, 12, 35), we examined whether NP serum elevated phosphorylated ERK (p-ERK) levels in HUtMECs. Treatment with NP serum increased p-ERK levels compared with treatment with NNP serum, and this NP serum-mediated p-ERK increase was reversed by treatment with PEG catalase (Fig. 5E). Since K_Ca3.1 upregulation by NP serum was reversed by treatment with PEG catalase (Fig. 5C), we examined whether treatment with PEG catalase reduced K_Ca3.1 currents in ECs treated with VEGF (10 ng/mL). PEG catalase reduced K_Ca3.1 currents from 46.4 ± 3.8 pA/pF to 30.6 ± 1.3 pA/pF in ECs treated with NP serum (Fig. 5F). These results were consistent with our previous results that K_Ca2.3 and K_Ca3.1 are upregulated by catalase and GPX1 knockdown using catalase and GPX1 double knockout mice (10), and suggest that catalase and GPX1 downregulation elevates H₂O₂ levels, thereby enhancing K_Ca2.3 and K_Ca3.1 levels in NP.

FIG. 5.

K_Ca2.3 and K_Ca3.1 are upregulated via a H₂O₂/fyn/ERK-mediated pathway in NP. (A) H₂O₂ levels were measured in ECs treated with NNP and NP serum and in ECs treated with VEGF by using the H₂O₂-sensitive dye peroxy-orange 1 (n = 6 per group). (B, C) Western blots (upper panels) and their quantification (lower panels) showing NP serum-mediated enhancement of K_Ca2.3 (B) and K_Ca3.1 (C) levels and the reversal by treatment with PEG catalase (20 units; n = 3 or 4 per group). (D, E) Western blot showing protein levels of p-fyn (D) and p-ERK (E) (n = 4 per group). NP serum enhanced levels of p-fyn and p-ERK, and these enhancements were reversed by PEG catalase. (F) Functional K_Ca3.1 expression was determined in ECs treated with VEGF or with VEGF+PEG catalase for 24 h (upper panels) by whole-cell patch clamp (n = 6 per group). Summary of K_Ca3.1 current densities at +50 mV. Data are mean ± SEM. *p < 0.05, **p < 0.01. For original immunoblot files, see Supplementary Fig. S6. ERK, extracellular signal-regulated kinase; p-ERK, phosphorylated ERK; PEG, polyethylene glycol.

Next, we examined whether NOX4 upregulation or SOD downregulation affected endothelial K_Ca2.3 and K_Ca3.1 levels via superoxide generation in PE. PE serum markedly elevated superoxide levels in ECs (Supplementary Fig. S2A, B), and K_Ca3.1 levels were markedly decreased in ECs treated with PE serum versus levels in ECs treated with NP serum (Supplementary Fig. S3A). We then inhibited PE serum-mediated superoxide generation by treatment with small interfering RNA (siRNA) against NOX4 or PEG-SOD. The PE serum-mediated decrease in K_Ca3.1 levels was recovered by inhibiting NOX4 with siRNA against NOX4 (Supplementary Fig. S3A). In addition, PE serum-mediated K_Ca2.3 downregulation was reversed by treatment with the membrane-permeable SOD, PEG-SOD (Supplementary Fig. S3B). Previously, we showed that PE plasma reduces endothelial K_Ca3.1 levels via superoxide generation (11). These results suggest that NOX4 upregulation and SOD downregulation enhance superoxide levels, thereby reducing K_Ca2.3 and K_Ca3.1 levels in PE.

Progesterone modulates K_Ca2.3 and K_Ca3.1 levels

Progesterone promotes superoxide release via NOX activation and SOD downregulation in blood vessels from progesterone-treated animals (38). Plasma levels of progesterone are increased during pregnancy, and progesterone levels are higher in PE compared with the levels in NP (33). We therefore exposed HUtMECs to progesterone. Progesterone elevated superoxide levels (Supplementary Fig. S2A, B) and reduced H₂O₂ levels (Supplementary Fig. S2C). The effects of progesterone on K_Ca3.1 levels varied according to the progesterone concentration. Progesterone increased protein levels of K_Ca3.1 at low concentrations (≤50 ng/mL), but decreased the K⁺ channel levels at high concentrations (≥250 ng/mL) (Fig. 6A). In addition, high progesterone concentration (500 ng/mL) reduced protein levels of K_Ca2.3 (Fig. 6B). As the plasma levels of progesterone range from 8 to 48 ng/mL and from 99 to 342 ng/mL in the first and third trimester, respectively (1), the concentrations of progesterone needed to affect K_Ca2.3 and K_Ca3.1 expression are within the ranges observed in NP. We then examined the effects of progesterone on levels of NOX2, NOX4, SOD1, catalase, and GPX1. Treatment of HUtMECs with progesterone increased NOX2 and NOX4 levels (Fig. 6C) and decreased SOD1 levels (Fig. 6D). GPX1 and catalase levels were enhanced by progesterone in a concentration-dependent manner (Fig. 6E). We then examined whether progesterone reduces K_Ca3.1 levels via superoxide generation. Progesterone (500 pg/mL) decreased K_Ca3.1 levels, and the progesterone-induced decrease in K_Ca3.1 levels was prevented by inhibiting superoxide generation using PEG-SOD, the pan-NOX inhibitor VAS2870, or the NOX4 inhibitor GKT137831 (Fig. 6F). These results suggested that progesterone reduced K_Ca2.3 and K_Ca3.1 levels by generating superoxide.

FIG. 6.

Progesterone downregulates K_Ca2.3 and K_Ca3.1. Western blots (upper panel) and their quantification (lower panel) showing protein levels of K⁺ channels, NOX4, and antioxidant enzymes in ECs treated with progesterone for 24 h. (A, B) Levels of K_Ca3.1 (A) or K_Ca2.3 (B) were measured in ECs treated with progesterone (n = 3 per group). (C, D) Treatment of HUtMECs with progesterone elevated NOX2 and NOX4 levels (C) and reduced SOD1 levels (D) (n = 3–5 per group). (E) Protein levels of catalase and GPX1 were elevated in ECs treated with progesterone (n = 4 per group). (F) The pan-NOX inhibitor VAS2870, the NOX4 inhibitor GKT137831, or PEG-SOD (20 units) enhanced K_Ca3.1 levels in progesterone-treated ECs (n = 3 or 4 per group). Data are mean ± SEM. *p < 0.05, **p < 0.01. For original immunoblot files, see Supplementary Figure S6.

Estrogen inhibits progesterone-induced K_Ca3.1 downregulation

Estrogen inhibits superoxide generation via nitric oxide generation, thereby preventing endothelial dysfunction (2). We thus examined whether estrogen inhibited progesterone-induced changes in the levels of NOX4, antioxidant enzymes, or K_Ca3.1. HUtMECs were exposed to progesterone or progesterone+estradiol-17β (E₂). Progesterone induced NOX4 upregulation and SOD1 downregulation, and these effects were prevented by E₂ (Fig. 7A, B). However, catalase upregulation by progesterone was not significantly prevented by E₂ (Fig. 7C). In contrast, GPX1 upregulation by progesterone was prevented by E₂ (Fig. 7D), and K_Ca3.1 downregulation by progesterone was prevented by E₂ (Fig. 7E). Meanwhile, E₂ alone did not increase protein levels of K_Ca3.1 at low concentrations (≤20 ng/mL), but significantly increased K_Ca3.1 (Fig. 7F) and K_Ca2.3 (Supplementary Fig. S4) levels at high concentrations (≥50 ng/mL). In addition, E₂ alone did not increase superoxide levels in ECs (Supplementary Fig. S2A, B). Since the plasma levels of E₂ range from 188 to 7192 pg/mL during pregnancy (1), the concentrations of E₂ needed to significantly increase protein levels of K_Ca3.1 were above the ranges observed in NP. However, the concentrations of E₂ needed to inhibit progesterone were very close to the ranges observed in NP.

FIG. 7.

Altered levels of NOX4, antioxidant enzymes, and K_Ca3.1 by progesterone were recovered by E₂. HUtMECs were exposed to progesterone, progesterone+E₂, or E₂ for 24 h. (A–E) Western blots (upper panel) and their quantification (lower panel) showing that E₂ reversed progesterone-induced changes in protein levels of NOX4, SOD1, catalase, GPX1, and K_Ca3.1 (n = 3 or 4 per group). Levels of NOX4 (A), SOD1 (B), catalase (C), GPX1 (D), and K_Ca3.1 (E) were compared between ECs treated with progesterone and with progesterone+E₂. (F) Western blots (upper panel) and their quantification (lower panel) showing protein levels of K_Ca3.1 in ECs treated with E₂ (n = 3 per group). E₂ enhanced protein levels of K_Ca3.1 at high concentrations. Data are mean ± SEM. *p < 0.05, **p < 0.01. For original immunoblot files, see Supplementary Figure S6. E₂, estradiol-17β.

sFlt-1 inhibits EDR by elevating superoxide levels

sFlt-1, which acts as an antagonist of VEGF, might contribute to the development of PE (21, 32). We thus examined the effects of sFlt-1 on NOX2, NOX4, superoxide, and K_Ca3.1 levels. Treatment with sFlt-1 for 24 h markedly elevated NOX2, NOX4 (Fig. 8A), and superoxide levels (Supplementary Fig. S2A, B) and reduced protein levels of K_Ca3.1 (Fig. 8B) in HUtMECs. Reduced protein levels of K_Ca3.1 were reversed by PEG-SOD. Then, the effects of sFlt-1 on mRNA levels of K_Ca3.1 and EDR were examined ex vivo. Rat aortas and SMA trunks were incubated in sFlt-1- or sFlt-1+PEG-SOD-containing culture medium for 24 h. sFlt-1 reduced mRNA levels of K_Ca3.1 in rat aortas, and reduced mRNA levels of K_Ca3.1 were reversed by PEG-SOD (Fig. 8C). Treatment with sFlt-1 attenuated EDR to ACh or 1-EBIO in SMA trunks from rats (Fig. 8D, E). In addition, l-NOARG-resistant EDR to 1-EBIO was reduced by the treatment with sFlt-1 (Fig. 8F). Reduced EDR to ACh or 1-EBIO was reversed by the treatment with PEG-SOD (Fig. 8D–F). These results suggest that sFlt-1 downregulates K_Ca3.1 by elevating superoxide levels, thereby attenuating EDR.

FIG. 8.

sFlt-1 induces K_Ca3.1 downregulation and attenuates EDR. (A, B) HUtMECs were exposed to sFlt-1 or (sFlt-1+PEG-SOD) for 24 h. Western blots (upper panel) and their quantification (lower panel) showing that sFlt-1 elevated NOX2 and NOX4 levels (A) and reduced K_Ca3.1 levels (B) (n = 3–7 per group). PEG-SOD reversed sFlt-1-induced changes in protein levels of K_Ca3.1 (n = 4 per group). (C–F) Aortas and SMA from nonpregnant rats were exposed to sFlt-1 or (sFlt-1+PEG-SOD) for 24 h. (C) K_Ca3.1 mRNA expression was compared between these aortic samples using real-time PCR (n = 5 per group). (D–F) EDR was induced by ACh (D) or by 1-EBIO (100 μM) or (1-EBIO+ACh) (E, F) in rings of SMA trunk (n = 6 per group). The magnitude of maximal relaxation during each treatment was expressed as a percentage of initial norepinephrine-induced contraction. EDR in arterial rings without l-NOARG pretreatment (D, E) or in arterial rings in which nitric oxide production was inhibited by l-NOARG pretreatment (F). Data are mean ± SEM. *p < 0.05, **p < 0.01. For original immunoblot files, see Supplementary Figure S6. mRNA, messenger RNA; sFlt-1, soluble fms-like tyrosine kinase 1. To see this illustration in color, the reader is referred to the web version of this article at www.liebertpub.com/ars

Discussion

In this study, we observed for the first time that endothelial K_Ca2.3 and K_Ca3.1 levels were modulated by altered redox state during pregnancy. During NP, catalase and GPX1 levels were reduced and SOD levels were enhanced, which may increase H₂O₂ levels and thereby upregulate K_Ca2.3 and K_Ca3.1 via a H₂O₂/fyn/ERK-mediated pathway. Whereas NOX2 and NOX4 levels were enhanced and SOD levels were reduced in PE, which might increase superoxide levels and thereby downregulate K_Ca2.3 and K_Ca3.1. Endothelial K_Ca2.3 and K_Ca3.1 play important roles in vasodilation. Therefore, K_Ca2.3 and K_Ca3.1 upregulation might inhibit blood pressure elevation in NP. Whereas K_Ca2.3 and K_Ca3.1 downregulation might contribute to the development of high blood pressure in PE (Fig. 9).

FIG. 9.

A schematic model for altered redox state-induced modulation of endothelial K_Ca2.3 and K_Ca3.1 in NP and PE. Soluble serum factors, such as VEGF, estrogen, progesterone, and ox-LDL, alter levels of NOXs and antioxidant enzymes in NP and PE, thereby modulating redox state. Compared to NNP (A), downregulation of catalase and GPX1 and SOD upregulation may increase H₂O₂ levels, thereby enhancing K_Ca2.3 and K_Ca3.1 levels in NP (B, left panel). In PE (B, right panel), SOD downregulation and NOX upregulation might enhance superoxide generation, thereby reducing K_Ca2.3 and K_Ca3.1 levels. K_Ca2.3 and K_Ca3.1 upregulation during NP might augment endothelium-dependent vasorelaxation by potentiating endothelium-dependent hyperpolarization and NO release, thereby compensating for effect of increased blood volume on BP. In contrast, K_Ca2.3 and K_Ca3.1 downregulation during PE may impair endothelium-dependent vasorelaxation. To see this illustration in color, the reader is referred to the web version of this article at www.liebertpub.com/ars

H₂O₂-induced K_Ca2.3 and K_Ca3.1 upregulation in NP and superoxide-induced K_Ca2.3 and K_Ca3.1 downregulation in PE suggest that endothelial functions can be modulated by altering redox state during pregnancy. Since SOD upregulation promotes the degradation of superoxide into H₂O₂, and catalase and GPX downregulation reduces H₂O₂ degradation, such changes in antioxidant enzymes on treatment with NP serum may increase H₂O₂ levels and decrease superoxide levels, thereby upregulating K_Ca2.3 and K_Ca3.1. Since NOX2 and NOX4 upregulation and SOD downregulation enhance superoxide levels and catalase and GPX upregulation reduces H₂O₂ levels, such changes in NOXs and antioxidant enzymes by treatment with PE serum may increase superoxide levels and decrease H₂O₂ levels, thereby downregulating K_Ca2.3 and K_Ca3.1. During pregnancy, soluble serum factors such as VEGF, progesterone, estrogen, and ox-LDL might act as modulators of NOXs and antioxidant enzyme levels, thereby regulating redox state. Thus, these factors might affect endothelial functions via modulating K_Ca2.3 and K_Ca3.1 levels.

Levels of estrogen and progesterone are elevated during pregnancy (1), and these steroids stimulate VEGF production (17, 27). VEGF, estrogen, and progesterone activate ERK (8, 14, 19, 20, 30). Since K_Ca3.1 protein synthesis is ERK dependent (34), VEGF and sex steroids might increase protein levels of K_Ca3.1 by stimulating K_Ca3.1 protein synthesis, consistent with the finding that VEGF upregulates K_Ca3.1 mRNA (16). In addition, H₂O₂, levels of which were increased in NP, activates fyn and ERK (12, 35). Thus, K_Ca3.1 upregulation that occurs during NP and on treatment with low concentrations of progesterone might be caused by increased production of the K⁺ channel proteins via activating ERK-dependent pathways. However, little is known about how K_Ca2.3 proteins are produced in ECs, and further studies are required to clarify regulatory mechanisms behind K_Ca2.3 protein production.

In contrast, high concentrations (≥100 or 250 pg/mL) of progesterone reduced K_Ca2.3 and K_Ca3.1 levels. Since progesterone stimulated superoxide generation via NOX2 and NOX4 upregulation and SOD downregulation, superoxide generated by progesterone might be involved in K_Ca2.3 and K_Ca3.1 downregulation by progesterone, which is supported by the finding that progesterone-induced K_Ca2.3 downregulation was reversed by PEG-SOD. In addition, we previously reported that superoxide induces K_Ca3.1 downregulation in PE (11). Enhanced NOX4 and GPX1 levels and reduced SOD1 levels by progesterone treatment were reversed by treatment with E₂, suggesting that progesterone-induced superoxide production was inhibited by E₂. In addition, estrogen has been shown to act as an antioxidant and to inhibit superoxide generation via nitric oxide generation (2, 4, 39). Thereby, estrogen might reverse K_Ca2.3 and K_Ca3.1 downregulation by progesterone and recover progesterone-induced endothelial dysfunction.

Progesterone levels in serum are increased during PE compared with NP, and are strongly associated with PE (33). Endothelial K_Ca2.3 and K_Ca3.1 downregulation by progesterone at high concentrations might contribute to the development of endothelial dysfunction in PE. Since estrogen inhibited progesterone-induced K_Ca3.1 downregulation, the progesterone/estrogen ratio might play an important role in controlling K_Ca2.3 and K_Ca3.1 levels and endothelial functions as pregnancy progresses. The progesterone/estradiol ratio might correlate positively with mean arterial pressure during pregnancy, and an increased ratio might contribute to the development of PE, as shown in women at high altitude (40).

Other soluble serum factors, such as sFlt-1 and ox-LDL, might also be responsible for the reduced pregnancy-associated K_Ca2.3 and K_Ca3.1 upregulation in PE. sFlt-1 elevates superoxide levels and reduces free-circulating levels of proangiogenic factors VEGF and placental growth factor (21), thereby reducing K_Ca2.3 and K_Ca3.1 levels. In addition, as reported in our previous studies, PE plasma-induced NOX2 upregulation and K_Ca3.1 downregulation were inhibited by anti-lectin-like ox-LDL receptor Ab (11), and the major component of ox-LDL, LPC, induces superoxide overload in ECs by reducing SOD1 levels and increasing catalase levels (13). These results suggest that ox-LDL, present in high levels in the sera of women with PE, induces superoxide production, thereby reducing K_Ca2.3 and K_Ca3.1 levels.

Endothelial K_Ca2.3 and K_Ca3.1 upregulation augments EDR of vascular smooth muscle (10). Increased expression of K_Ca2.3 and K_Ca3.1 protein compensated for diminished nitric oxide-induced EDR by enhancing K_Ca2.3 and K_Ca3.1 activation-induced relaxation in aged mice (10). The mechanism for K_Ca3.1 upregulation in aged mice was similar to the mechanism in NP, since downregulation of catalase and GPX1 contributed to K_Ca3.1 upregulation via H₂O₂ generation. Estrogen receptor agonists and E₂ induce EDR via activating endothelium-dependent hyperpolarization and stimulate EC proliferation (15, 36). Thus, estrogen-induced EDR might be, at least in part, caused by K_Ca2.3 and K_Ca3.1 upregulation.

The present study showed that catalase and GPX1 downregulation increased H₂O₂ levels, thereby enhancing K_Ca2.3 and K_Ca3.1 levels in NP. Whereas NOX2 and NOX4 upregulation and SOD downregulation might increase superoxide levels, thereby reducing K_Ca2.3 and K_Ca3.1 levels in PE. VEGF, sex steroids, and ox-LDL acted as modulators of these enzymes. Progesterone was involved in the K_Ca2.3 and K_Ca3.1 downregulation in PE, and estrogen inhibited progesterone-induced K_Ca2.3 and K_Ca3.1 downregulation. Since endothelial K_Ca2.3 and K_Ca3.1 play important roles in vasodilation, altered redox state-induced K_Ca2.3 and K_Ca3.1 levels may play important roles in controlling blood pressure during pregnancy.

Materials and Methods

Studies involving human subjects were approved by the local ethics committee, the Institutional Review Board of the Ewha Womans University Mokdong Hospital, and Korea University Guro Hospital, and were conducted in accordance with the Declaration of Helsinki. All patients gave their written informed consent before inclusion in the study. Experiments with mice were approved by the local ethics committee, the Institutional Review Board of the Ewha Womans University Mokdong Hospital, and were conducted in accordance with the Declaration of Helsinki, the Animal Care Guidelines of the Ewha Womans University, Medical School, and the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Human subjects

The study population consisted of Asian women who were not pregnant or had either an NP or PE (Table 1). Pregnancies were considered normal when patients did not have medical and obstetric complications of pregnancy and delivered a newborn at a gestational age of 37–42 weeks. PE was defined by a systolic blood pressure over 140 mmHg and a diastolic blood pressure over 90 mmHg after 20 weeks of gestation in a previously normotensive woman, and the onset of proteinuria exceeding 300 mg of protein during 24 h of urine collection. Nonpregnant women were healthy premenopausal volunteers taking no medications. Preeclamptic patients and normal pregnant women were matched for age (±3 years) and gestational age (±2 weeks), and nonpregnant healthy female volunteers were matched for age (±3 years). Blood samples were obtained from subjects during the third trimester of pregnancy. The study population was monitored at the department of obstetrics and gynecology from the first trimester until their pregnancy was completed without complications. Exclusion criteria included the following: altered renal function, diabetes or chronic diseases, twin pregnancies, recurrent miscarriages, fetal growth retardation, and abruptio placenta. All smokers and women with a history of essential hypertension were also excluded from this study. Gestational age was defined as the interval between the first day of the mother's last menstrual period and the date of delivery.

Table 1.

Blood Pressure and Protein Urea Levels of Study Groups

Group	Normal nonpregnancy (n = 13)	Normal pregnancy (n = 24)	Preeclampsia (n = 15)
Systolic blood pressure (mmHg)	118.3 ± 2.5	117.5 ± 2.7	151.1 ± 3.1
Diastolic blood pressure (mmHg)	75.1 ± 2.3	76.0 ± 2.2	97.9 ± 3.0
Protein urea (mg/24 h)	N/S	N/S	>300

Values shown are mean ± SEM and exclusively composed of serum donors.

N/S, not specified; SEM, standard error of the mean.

Animals and tissue collection

We studied young C57BL/6 wild-type mice (about 20 weeks old; n = 24) and Sprague-Dawley rats (about 20 weeks old; n = 36). To produce pregnancy, a single male and two to three female mice or rats in proestrous or estrous stage were housed together for about 48 h. Female mice or rats were checked for vaginal plugs each morning and evening. When a vaginal plug was observed, the female mice or rats were separated from the male mice or rats. Gestational day 0 was defined as the day on which a vaginal plug was observed. Pregnant (on gestational days 17—18; n = 12) or age-matched nonpregnant female mice (n = 12) or rats (n = 12) were anesthetized by an intraperitoneal injection of pentobarbital sodium (50 mg/kg body weight) and sacrificed by cervical dislocation. The thoracic aortas were then dissected from mice and rats, and the SMAs (trunk and the first branches) were then dissected from rats.

Cell culture and treatment of cells with serum or reagents

HUtMECs, which were purchased from PromoCell GmbH (Heidelberg, Germany), were maintained in EC growth medium MV2 (PromoCell GmbH). For serum treatment, HUtMECs were plated in six-well plates for 24 h. The concentration of FBS in culture medium was then gradually decreased from 10% to 5%, 2%, and 0% over 30 min, and HUtMECs were incubated in serum-free medium for 30 min. Next, culture medium was substituted with serum from NNP women or women with NP or PE. Cells were incubated in serum for 24 h. For VEGF (751-VE; R&D Systems, Minneapolis, MN), ox-LDL (Intracel, Inc., Frederick, MD), LPC, progesterone (P9776; Sigma-Aldrich, St Louis, MO), E₂ (E7879; Sigma-Aldrich), or (E₂+progesterone) treatment, cells were incubated in one of these reagents or (E₂+progesterone)-containing serum-free medium for 24 h. A pan-NOX inhibitor VAS2870 (SML0273; Sigma-Aldrich), an NOX4 inhibitor GKT137831 (17764; Cayman Chemicals, Ann Arbor, MI), PEG catalase (C4963; Sigma-Aldrich), and PEG-SOD (C4936; Sigma-Aldrich) were pretreated with cells 1 h before reagent application.

Measurement of intracellular ROS

HUtMECs plated on 96-well microplates (for direct quantification) were incubated in the superoxide-sensitive dye dihydroethidine (DHE; 10 μM) or the H₂O₂-sensitive dye peroxy-orange1 (5 μM) for 20 min. Samples were read directly using a microplate fluorescence reader (model SpectraMax; Molecular Devices, Sunnyvale, CA) or detected by confocal laser microscopy (model LSM 510; Carl Zeiss). Excitation/emission wavelengths used for detecting respective fluorescent dyes in 96-well microplate reader were DHE ex 518 nm/em 605 nm and peroxy-orange 1 ex 540 nm/em 585 nm. To further confirm the superoxide production for each sample tested, Diogenes enhanced superoxide detection kit (National Diagnostics, Atlanta, GA) was used. Cells seeded on 96-well microplates were incubated in serum (from NNP, NP, or PE) or sFlt-1 (AG-40T-0049; AdipoGen, San Diego, CA)-, VEGF-, progesterone- or E₂-containing medium for 24 h, and luminescence was measured by using a GloMax 96 luminometer (Promega Corporation, Madison, WI).

Electrophysiology

The patch-clamp technique was used in whole-cell configurations at 20–22°C. Whole-cell currents were measured using ruptured patches and monitored in voltage-clamp modes with an EPC-9 (HEKA Elektronik, Lambrecht, Germany). The holding potential was 0 mV, and currents were monitored by the repetitive application of voltage ramps from −100 to +100 mV with a 10-s interval (sampling interval 0.5, 650 ms duration). The standard external solution contained (in mM) the following: 150 NaCl, 6 KCl, 1.5 CaCl_2, 1 MgCl₂, 10 HEPES, 10 glucose, pH adjusted to 7.4 with NaOH. The pipette solution for whole-cell recording contained (in mM) the following: 40 KCl, 100 K-aspartate, 2 MgCl₂, 0.1 EGTA, 4 Na₂ATP, 10 HEPES, pH adjusted to 7.2 with KOH. To buffer free Ca²⁺, the appropriate amount of Ca²⁺ (calculated using CaBuf software; G. Droogmans, Leuven, Belgium) was added in the presence of 5 mM EGTA.

K_Ca3.1 currents were activated by loading 1 μM Ca²⁺ via a patch pipette in whole-cell clamped HUtMECs and adding the K_Ca2.3 and K_Ca3.1 activator 1-EBIO (100 μM) to the external solution. K_Ca3.1 current was normalized to cell capacitance and the selective K_Ca3.1 blocker TRAM-34-sensitive current was measured as the K_Ca3.1 current.

Contraction measurement on isolated aortic rings

Thoracic aorta samples from mice and SMA samples from rats were cut into 1.0–2.0 mm rings. A custom myograph was used to record mechanical responses from the ring segments at 37°C. Each aortic or SMA (trunk and the first branch) ring was threaded with two strands of tungsten wire (120 μm diameter) or stainless wire (100 μm diameter for trunk and 40 μm diameter for branches), respectively: one anchored in the organ bath chamber (1 mL) and the other connected to a mechanotransducer (FT03; Natus Neurology, Inc., Middleton, WI). The chamber was perfused at a flow rate of 2.5 mL/min with oxygenated (95% O₂ to 5% CO₂) Krebs–Ringer bicarbonate solution using a peristaltic pump. The composition (in mM) of the Krebs buffer was 118.3 NaCl, 4.7 KCl, 1.2 MgCl₂, 1.22 KH₂PO₄, 2.5 CaCl₂, 25.0 NaHCO₃, 11.1 glucose, pH 7.4. Optimal resting tension (0.5–1 g) was applied.

Arterial rings were precontracted with prostaglandin F_2α (1 μM) or norepinephrine (1 μM) and EDR was induced by ACh. l-NOARG (30 μM for 1 h) was used to inhibit nitric oxide production. K_Ca3.1 activation-induced EDR was induced by 1-EBIO (100 μM). In all experiments where EDR was measured, indomethacin (10 μM) was present in the Krebs buffer to inhibit prostacyclin production. Since prostacyclin production was inhibited by indomethacin, the EDR response to ACh might be evoked by nitric oxide released from ECs and by endothelium-dependent hyperpolarization via endothelial K_Ca2.3 and K_Ca3.1 activation.

Immunoblotting

For immunoblotting, cell lysates were prepared for each type of sample; total protein was measured using a bicinchoninic acid assay (Pierce Biotechnology, Rockford, IL). The same amounts of total protein were separated by sodium dodecyl sulfate/polyacrylamide gel electrophoresis on 7.5–12% gels and transferred to nitrocellulose membranes (Invitrogen, Eugene, OR). Membranes were blocked for 1 h in 5% bovine serum albumin in Tris-buffered saline with 0.1% Tween-20 and incubated overnight at 4°C with primary Abs diluted in blocking buffer. After that, membranes were washed three times with Tris-buffered saline with 0.1% Tween-20 and incubated for 1 h with horseradish peroxidase-conjugated secondary Abs diluted in blocking buffer. The immunoblots were visualized by chemiluminescence reagents obtained from GE Healthcare (Piscataway, NJ). Data processing was performed using a luminescent image analyzer LAS-3000 (Fujifilm, Tokyo, Japan) and IMAGE GAUSE software.

Abs against K_Ca2.3 (sc-28621), K_Ca3.1 (sc-32949), p-ERK (sc7383), NOX4 (sc3014), GAPDH (sc-25778), and β-actin (sc-130656) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-ERK Ab (4695S) was purchased from Cell Signaling (Danvers, MA). Abs specific to NOX2 (ab31092), GPX1 (ab22604), and catalase (ab16731) were obtained from Abcam (Cambridge, MA). Anti-fyn (clone EPR5500) and p-fyn (07-909) antibodies were obtained from Millipore (Temecula, CA; MABT208).

Tissue preservation and immunohistochemistry

Placental tissues from NP and PE subjects were fixed by immersion in a periodate-lysine-2% paraformaldehyde solution overnight at 4°C. Tissues were cut transversely into 1–2-mm-thick slices and processed for immunohistochemical studies using a horseradish peroxidase technique. Tissue slices were embedded in paraffin. Four-micrometer sections were deparaffinized with xylene and hydrated in a graded series of ethanol. After rinsing in tap water, sections were incubated with 3% H₂O₂ for 30 min to eliminate endogenous peroxidase activity. The sections were treated with blocking serum for 30 min and incubated overnight at 4°C in primary Abs (K_Ca2.3 and K_Ca3.1). After being washed in phosphate-buffered saline (PBS), the sections were incubated for 2 h with the peroxidase-conjugated donkey anti-rabbit IgG Fab fragment (Jackson ImmunoResearch Laboratories, West Grove, PA) diluted 1:100 in PBS. After being rinsed with Tris-HCl buffer, the sections were exposed to a mixture of 0.05% 3,3′-diaminobenzidine and 0.01% H₂O₂ for 5 min at room temperature. The sections were dehydrated with graded ethanol and xylene, mounted in Permount, and examined by light microscopy.

Reagents

Reagents were obtained from Sigma-Aldrich and dissolved in autoclaved distilled water unless indicated otherwise. Cells were treated with progesterone (dissolved in ethanol), E₂ (dissolved in ethanol), ox-LDL, or LPC (dissolved in chloroform:methanol, 2:1) for 24 h. To neutralize cell surface proteins, cells were pretreated with Abs specific to VEGFR1 (ab32152; Abcam), VEGFR2 (No. 2479; Cell Signaling), or anti-isotype control (rabbit IgG monoclonal) Ab (ab172730; Abcam) for 1 h. 1-EBIO (Tocris Bioscience, Bristol, United Kingdom) was used to activate K_Ca2.3 and K_Ca3.1 currents.

The final concentration of dimethyl sulfoxide (DMSO), chloroform, methanol, or ethanol in media was <0.1%, and these solvents did not have any effect on the activities tested in this study (data not shown).

Statistics

Data are mean ± standard error of the mean. To prove the statistical significance between groups, two-tailed Student's t-test or one-way analysis of variance was used. A p value of 0.05 or lower was considered statistically significant. Calculations were performed with SPSS 14.0 for Windows (SPSS, Chicago, IL).

Footnotes

Acknowledgments

This research was supported by the Basic Science Research Program through the Nation Research Foundation of Korea funded by the Ministry of Education, Science and Technology (NRF-2013R1A1A2010851, NRF-2013R1A1A2064543, 2016R1D1A1A09919073, and 2016R1D1A1A09918769) and intramural research promotion grants from Ewha Womans University, School of Medicine.

Author Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Abbassi-Ghanavati

, Greer

, and Cunningham

. Pregnancy and laboratory studies: a reference table for clinicians. Obstet Gynecol, 114: 1326–1331, 2009.

Arnal

, Fontaine

, Billon-Gales

, Favre

, Laurell

, Lenfant

, and Gourdy

. Estrogen receptors and endothelium. Arterioscler Thromb Vasc Biol, 30: 1506–1512, 2010.

Bedard

and Krause

. The NOX family of ROS-generating NADPH oxidases: physiology and pathophysiology. Physiol Rev, 87: 245–313, 2007.

Behl

, Skutella

, Lezoualc'h

, Post

, Widmann

, Newton

, and Holsboer

. Neuroprotection against oxidative stress by estrogens: structure-activity relationship. Mol Pharmacol, 51: 535–541, 1997.

Bellanti

, Matteo

, Rollo

, De Rosario

, Greco

, Vendemiale

, and Serviddio

. Sex hormones modulate circulating antioxidant enzymes: impact of estrogen therapy. Redox Biol, 1: 340–346, 2013.

, Toyama

, Lemaitre

, Takai

, Fan

, Jenkins

, Wulff

, Gutterman

, Park

, and Miura

. The intermediate conductance calcium-activated potassium channel KCa3.1 regulates vascular smooth muscle cell proliferation via controlling calcium-dependent signaling. J Biol Chem, 288: 15843–15853, 2013.

Brahler

, Kaistha

, Schmidt

, Wolfle

, Busch

, Kaistha

, Kacik

, Hasenau

, Grgic

, Si

, Bond

, Adelman

, Wulff

, de Wit

, Hoyer

, and Kohler

. Genetic deficit of SK3 and IK1 channels disrupts the endothelium-derived hyperpolarizing factor vasodilator pathway and causes hypertension. Circulation, 119: 2323–2332, 2009.

Breslin

, Pappas

, Cerveira

, Hobson

2nd , and Duran

. VEGF increases endothelial permeability by separate signaling pathways involving ERK-1/2 and nitric oxide. Am J Physiol Heart Circ Physiol, 284: H92–H100, 2003.

Chang

and Lubo

. Review article: steroid hormones and uterine vascular adaptation to pregnancy. Reprod Sci, 15: 336–348, 2008.

10.

Choi

, Kim

, Li

, Shin

, Oh

, Lee

, Oh

, Pewzner-Jung

, Futerman

, and Suh

. KCa 3.1 upregulation preserves endothelium-dependent vasorelaxation during aging and oxidative stress. Aging Cell, 15: 801–810, 2016.

11.

Choi

, Kim

, Na

, Kim

, Park

, Han

, Kim

, and Suh

. NADPH oxidase 2-derived superoxide downregulates endothelial KCa3.1 in preeclampsia. Free Radic Biol Med, 57: 10–21, 2013.

12.

Choi

, Na

, Kim

, Cho

, and Suh

. Contradictory effects of superoxide and hydrogen peroxide on KCa3.1 in human endothelial cells. Korean J Physiol Pharmacol, 17: 181–187, 2013.

13.

Choi

, Park

, Liang

, Kim

, and Suh

. Superoxide generated by lysophosphatidylcholine induces endothelial nitric oxide synthase downregulation in human endothelial cells. Cell Physiol Biochem, 25: 233–240, 2010.

14.

Filardo

, Quinn

, Bland

, and Frackelton

Jr.

Estrogen-induced activation of Erk-1 and Erk-2 requires the G protein-coupled receptor homolog, GPR30, and occurs via trans-activation of the epidermal growth factor receptor through release of HB-EGF. Mol Endocrinol, 14: 1649–1660, 2000.

15.

Geraldes

, Sirois

, Bernatchez

, and Tanguay

. Estrogen regulation of endothelial and smooth muscle cell migration and proliferation: role of p38 and p42/44 mitogen-activated protein kinase. Arterioscler Thromb Vasc Biol, 22: 1585–1590, 2002.

16.

Grgic

, Eichler

, Heinau

, Si

, Brakemeier

, Hoyer

, and Köhler

. Selective blockade of the intermediate-conductance Ca²⁺-activated K⁺ channel suppresses proliferation of microvascular and macrovascular endothelial cells and angiogenesis in vivo. Arterioscler Thromb Vasc Biol, 25: 704–709, 2005.

17.

Hyder

and Stancel

. Regulation of angiogenic growth factors in the female reproductive tract by estrogens and progestins. Mol Endocrinol, 13: 806–811, 1999.

18.

Kong

, Lee

, and Kim

. Inhibition of rac1 reduces PDGF-induced reactive oxygen species and proliferation in vascular smooth muscle cells. J Korean Med Sci, 16: 712–718, 2001.

19.

Miao

, Dong

, Fang

, Shang

, Liu

, Zhang

, Li

, and Chen

. VEGF increases paracellular permeability in brain endothelial cells via upregulation of EphA2. Anat Rec (Hoboken), 297: 964–972, 2014.

20.

Migliaccio

, Piccolo

, Castoria

, Di Domenico

, Bilancio

, Lombardi

, Gong

, Beato

, and Auricchio

. Activation of the Src/p21ras/Erk pathway by progesterone receptor via cross-talk with estrogen receptor. EMBO J, 17: 2008–2018, 1998.

21.

Powe

, Levine

, and Karumanchi

. Preeclampsia, a disease of the maternal endothelium: the role of antiangiogenic factors and implications for later cardiovascular disease. Circulation, 123: 2856–2869, 2011.

22.

Raijmakers

, Dechend

, and Poston

. Oxidative stress and preeclampsia: rationale for antioxidant clinical trials. Hypertension, 44: 374–380, 2004.

23.

Ray

, Murdoch

, Wang

, Santos

, Zhang

, Alom-Ruiz

, Anilkumar

, Ouattara

, Cave

, Walker

, Grieve

, Charles

, Eaton

, Brewer

, and Shah

. Endothelial Nox4 NADPH oxidase enhances vasodilatation and reduces blood pressure in vivo. Arterioscler Thromb Vasc Biol, 31: 1368–1376, 2011.

24.

Saksena

, Gill

, Tyagi

, Alrefai

, Ramaswamy

, and Dudeja

. Role of Fyn and PI3K in H₂O₂-induced inhibition of apical Cl⁻/OH⁻ exchange activity in human intestinal epithelial cells. Biochem J, 416: 99–108, 2008.

25.

Shaker

and Sadik

. Pathogenesis of preeclampsia: implications of apoptotic markers and oxidative stress. Hum Exp Toxicol, 32: 1170–1178, 2013.

26.

Sheng-Huang

, Chieh-Hsin

, Mu-Chun

, Wen-Tung

, Chia-Ying

, Ya-Ting

, Wan-Ling

, Jiuan-Jen

, Chih-Yang

, and Jer-Yuh

. Effects of estrogen on glutathione and catalase levels in human erythrocyte during menstrual cycle. Biomed Rep, 3: 266–268, 2015.

27.

Shifren

, Tseng

, Zaloudek

, Ryan

, Meng

, Ferrara

, Jaffe

, and Taylor

. Ovarian steroid regulation of vascular endothelial growth factor in the human endometrium: implications for angiogenesis during the menstrual cycle and in the pathogenesis of endometriosis. J Clin Endocrinol Metab, 81: 3112–3118, 1996.

28.

, Heyken

W-T

, Wölfle

, Tysiac

, Schubert

, Grgic

, Vilianovich

, Giebing

, Maier

, Gross

, Bader

, de Wit

, Hoyer

, and Köhler

. Impaired endothelium-derived hyperpolarizing factor-mediated dilations and increased blood pressure in mice deficient of the intermediate-conductance Ca²⁺-activated K⁺ channel. Circ Res, 99: 537–544, 2006.

29.

Siddiqui

, Jaleel

, Tamimi

, and Al Kadri

. Role of oxidative stress in the pathogenesis of preeclampsia. Arch Gynecol Obstet, 282: 469–474, 2010.

30.

Singh

. Ovarian hormones elicit phosphorylation of Akt and extracellular-signal regulated kinase in explants of the cerebral cortex. Endocrine, 14: 407–415, 2001.

31.

Sundaresan

, Yu

, Ferrans

, Sulciner

, Gutkind

, Irani

, Goldschmidt-Clermont

, and Finkel

. Regulation of reactive-oxygen-species generation in fibroblasts by Rac1. Biochem J, 318 (Pt 2): 379–382, 1996.

32.

Tam Tam

, Lamarca

, Arany

, Cockrell

, Fournier

, Murphy

, Martin

Jr. , and Granger

. Role of reactive oxygen species during hypertension in response to chronic antiangiogenic factor (sFlt-1) excess in pregnant rats. Am J Hypertens, 24: 110–113, 2011.

33.

Tamimi

, Lagiou

, Vatten

, Mucci

, Trichopoulos

, Hellerstein

, Ekbom

, Adami

, and Hsieh

. Pregnancy hormones, pre-eclampsia, and implications for breast cancer risk in the offspring. Cancer Epidemiol Biomarkers Prev, 12: 647–650, 2003.

34.

Tharp

and Bowles

. The intermediate-conductance Ca²⁺-activated K⁺ channel (KCa3.1) in vascular disease. Cardiovasc Hematol Agents Med Chem, 7: 1–11, 2009.

35.

Toni

, Spisni

, Griffoni

, Santi

, Riccio

, Lenaz

, and Tomasi

. Cellular prion protein and caveolin-1 interaction in a neuronal cell line precedes Fyn/Erk 1/2 signal transduction. J Biomed Biotechnol, 2006: 69469, 2006.

36.

Traupe

, Stettler

, Li

, Haas

, Bhattacharya

, Minotti

, and Barton

. Distinct roles of estrogen receptors alpha and beta mediating acute vasodilation of epicardial coronary arteries. Hypertension, 49: 1364–1370, 2007.

37.

Ushio-Fukai

. VEGF signaling through NADPH oxidase-derived ROS. Antioxid Redox Signal, 9: 731–739, 2007.

38.

Wassmann

, Wassmann

, and Nickenig

. Progesterone antagonizes the vasoprotective effect of estrogen on antioxidant enzyme expression and function. Circ Res, 97: 1046–1054, 2005.

39.

Yen

, Hsieh

, Chou

, and Lau

. 17Beta-estradiol inhibits oxidized low density lipoprotein-induced generation of reactive oxygen species in endothelial cells. Life Sci, 70: 403–413, 2001.

40.

Zamudio

, Leslie

, White

, Hagerman

, and Moore

. Low serum estradiol and high serum progesterone concentrations characterize hypertensive pregnancies at high altitude. J Soc Gynecol Investig, 1: 197–205, 1994.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.07 MB

0.08 MB

0.05 MB

0.03 MB

0.13 MB

0.14 MB

0.03 MB

Altered Redox State Modulates Endothelial K Ca 2.3 and K Ca 3.1 Levels in Normal Pregnancy and Preeclampsia

Abstract

Aims:

Results:

Innovation and Conclusions:

Introduction

Results

NP serum upregulates endothelial KCa2.3 and KCa3.1 via VEGF receptor activation

KCa2.3 and KCa3.1 upregulation enhances EDR during pregnancy

Pregnancy-associated KCa2.3 and KCa3.1 upregulation is attenuated in PE

NOX4 and antioxidant enzyme levels in ECs treated with NP and PE serum

Altered redox state modulates KCa2.3 and KCa3.1 levels in NP and PE

Progesterone modulates KCa2.3 and KCa3.1 levels

Estrogen inhibits progesterone-induced KCa3.1 downregulation

sFlt-1 inhibits EDR by elevating superoxide levels

Discussion

Materials and Methods

Human subjects

Animals and tissue collection

Cell culture and treatment of cells with serum or reagents

Measurement of intracellular ROS

Electrophysiology

Contraction measurement on isolated aortic rings

Immunoblotting

Tissue preservation and immunohistochemistry

Reagents

Statistics

Footnotes

Acknowledgments

Author Disclosure Statement

Abbreviations Used

References

Supplementary Material

NP serum upregulates endothelial K_Ca2.3 and K_Ca3.1 via VEGF receptor activation

K_Ca2.3 and K_Ca3.1 upregulation enhances EDR during pregnancy

Pregnancy-associated K_Ca2.3 and K_Ca3.1 upregulation is attenuated in PE

Altered redox state modulates K_Ca2.3 and K_Ca3.1 levels in NP and PE

Progesterone modulates K_Ca2.3 and K_Ca3.1 levels

Estrogen inhibits progesterone-induced K_Ca3.1 downregulation