Blocking the BK_Ca channel induces NF-κB nuclear translocation by increasing nuclear calcium concentration

Abstract

Nuclear factor kappa B (NF-κB) transcriptionally regulates several genes involved in initiating uterine contractions. A key factor controlling NF-κB activity is its translocation to the nucleus. In myometrial smooth muscle cells (MSMCs), this translocation can be stimulated by the inflammatory molecule lipopolysaccharide (LPS) or by blocking the potassium calcium-activated channel subfamily M alpha 1 (KCNMA1 or BK_Ca) with paxilline (PAX). Here, we sought to determine the mechanism by which blocking BK_Ca causes NF-κB-p65 translocation to the nucleus in MSMCs. We show that LPS- and PAX-induced NF-κB-p65 translocation are similar in that neither depends on several mitogen-activated protein kinase pathways, but both require increased intracellular calcium (Ca²⁺). However, the nuclear transport inhibitor wheat germ agglutinin prevented NF-κB-p65 nuclear translocation in response to LPS but not in response to PAX. Blocking BK_Ca located on the plasma membrane resulted in a transient NF-κB-p65 nuclear translocation that was not sufficient to induce expression of its transcriptional target, suggesting a role for intracellular BK_Ca. We report that BK_Ca also localizes to the nucleus and that blocking nuclear BK_Ca results in an increase in nuclear Ca²⁺ in MSMCs. Together, these data suggest that BK_Ca localized on the nuclear membrane plays a key role in regulating nuclear Ca²⁺ and NF-κB-p65 nuclear translocation in MSMCs.

Blocking of BK channels by paxilline in myometrial smooth muscle cells increases nuclear calcium levels and translocation of NF-κB-p65.

Introduction

Throughout most of pregnancy, the uterus remains quiescent, generating only limited, weak, uncoordinated contractions. At term, the uterus becomes highly contractile to deliver the fetus [1, 2]. An important trigger for initiating and maintaining uterine excitability is intracellular calcium (Ca²⁺), which activates actomyosin contractility. A key ion channel involved in regulating Ca²⁺ dynamics in uterine (myometrial) smooth muscle cells (MSMCs) is the potassium calcium-activated channel subfamily M alpha 1 (KCNMA1, also known as BK_Ca), which is the predominant K⁺ channel expressed in the myometrium [3–6]. BK_Ca, which is activated by both depolarization of the plasma membrane and increases in intracellular Ca²⁺, maintains uterine quiescence during pregnancy by generating a potent repolarizing K⁺ current [7–9]. However, during labor, BK_Ca becomes less sensitive to Ca²⁺ and may thus allow contractility to increase [10]. Consistent with this idea, blocking BK_Ca results in membrane depolarization and the subsequent activation of L-type Ca²⁺ channels, which allow intracellular Ca²⁺ concentration to rise, resulting in increasing myometrial contractility [7].

Another important aspect of the myometrial transition from a quiescent state to a highly contractile state is intrauterine inflammation [11]. In the absence of an inflammatory trigger, the transcriptional regulator nuclear factor kappa B (NF-κB) is retained in the cytoplasm via binding to IκB [12–14]. In response to an inflammatory stimulus, such as bacterial lipopolysaccharide (LPS), NF-κB is released from IκB and translocates to the nucleus, where it regulates transcription of many target genes. For example, NF-κB increases expression of prostaglandin-endoperoxide synthase 2 (PTGS2, also known as COX2), a key enzyme in the biosynthesis of prostaglandins, which play a pivotal role in increasing myometrial contractility leading to parturition.

Given the importance of both inflammation and BK_Ca activity in regulating uterine contractility, we have been investigating the mechanisms by which these two processes intersect in the myometrium. We previously observed that blocking BK_Ca with paxilline (PAX) induced translocation of NF-κB into the nucleus, leading to contraction in both immortalized human MSMCs and primary human uterine myocytes [15]. Here, we sought to define the mechanism by which blocking BK_Ca activity induces NF-κB translocation and whether this mechanism mirrors that of LPS-induced NF-κB nuclear translocation. We report that these two mechanisms of NF-κB nuclear translocation both rely on increased intracellular Ca²⁺ but not on mitogen-activated protein kinase (MAPK) pathways. However, whereas LPS stimulates active transport of NF-κB into the nucleus, PAX may stimulate passive NF-κB translocation. Finally, we present evidence that BK_Ca present on the nuclear membrane likely plays a prominent role in PAX-induced NF-κB nuclear translocation and that PAX causes nuclear Ca²⁺ to increase. We conclude that LPS and BK_Ca inhibition lead to NF-κB nuclear translocation via independent mechanisms.

Materials and methods

Cell culture

Human myometrial smooth muscle cells immortalized with telomerase (hTERT) were maintained in DMEM/F12 medium supplemented with 10% fetal bovine serum (FBS) and 25 μg/ml gentamicin (all from Sigma, St. Louis, MO) [15, 16] at 37°C and 5% CO₂. To promote a contractile muscle-like phenotype, hTERT cells were transferred to DMEM/F12 medium supplemented with 0.5% FBS and 25 μg/ml gentamicin 12 h before treatments, unless noted otherwise [17].

Drug treatment experiments

To assess MAPK signaling, hTERT cells in 0.5% or 10% FBS were treated for 1 h with 10 μM U0126 (MEK1/2 inhibitor, CAS 109511-58-2, Sigma), 10 μM SB203580 (p38 MAPK inhibitor, CAS 152121-47-6, Sigma), or 25 μM SP600125 (JNK inhibitor, CAS 129-56-6, Sigma) and then treated with both the MAPK inhibitor and 10 ng/ml Escherichia coli lipopolysaccharide (LPS: 0111:B4, Sigma) or 20 μM Paxilline (PAX, CAS 57186-25-1, Tocris) for an additional 2 h. To assess requirement for Ca²⁺ signaling, hTERT cells in 0.5% FBS were treated with 5 μM BAPTA-AM (CAS 126150-97-8, Tocris, Minneapolis, MN) or 1 μM Thapsigargin (CAS 67526-95-8, Sigma) for 1 h and then treated with LPS or PAX as above. To assess NFκB-p65 nuclear translocation, hTERT cells in 0.5% FBS were treated with 10 μg/ml wheat germ agglutinin (WGA, L9640, Sigma) or 10 μM JSH-23 (CAS 749886-87-1, Sigma) for 1 h and then treated with LPS or PAX as above. To compare plasma membrane vs. intracellular BK_Ca, hTERT cells in 0.5% FBS were treated with 100 nM Iberiotoxin (IbTX, CAS 129203-60-7, Tocris) for 30–120 min. For all experiments, control cells were cultured in the same media as test groups with the appropriate vehicle control.

Immunofluorescence

hTERT cells were grown on chamber slides, fixed and permeabilized with 2% paraformaldehyde in phosphate-buffered saline (PBS) with 0.01% TritonX-100 for 30 min, and blocked with 10% heat-inactivated FBS and 1% heat-inactivated goat serum in PBS at 37°C for 30 min. Cells were then incubated with primary antibodies for 2 h at 37°C, washed, and incubated with Alexa Fluor 488-conjugated antimouse (1:1000 [Jackson Immuno Research, West Grove, PA]) for 1 h at 37°C. Nuclei were counterstained with TO-PRO-3 iodide or Hoechst 33342 (Thermo Fisher Scientific, Carlsbad, CA). Cells were imaged by confocal microscopy (Leica DMI4000 B, Wetzlar, Germany). Primary antibodies used were NFκB-p65 (AB 628017, 1:100, Santa Cruz Biotechnology, Dallas, TX) and Slo1/BK alpha (AB 2877291, 1:50, Neuro Mab, Davis, CA). For quantifying NFκB-p65 translocation, three independent experiments were performed, and 80–400 cells were counted per treatment group.

Protein isolation

Whole-cell lysates were obtained by homogenizing hTERT cells in ice-cold RIPA buffer (Sigma, R0278) and Complete Mini Protease Inhibitor Cocktail (EDTA-free, Roche Molecular Biochemicals, Indianapolis, IN). Cytoplasmic and nuclear lysates were isolated as previously described [2]. Briefly, hTERT cells were resuspended in NE1 buffer (10 mM HEPES, 10 mM MgCl₂, 5 mM KCl, 0.1% Triton X-100, 0.1 mM EDTA, pH 8.0) and homogenized by passage through a 23-gauge needle. Homogenates were centrifuged at 5000 × g for 10 min at 4°C to generate supernatant (cytoplasmic fraction) and pellets. Pellets were resuspended in NE2 buffer (25% glycerol, 20 mM HEPES, 500 mM NaCl, 1.5 mM MgCl₂, 0.2 mM EDTA, pH 8.0), incubated on ice for 30 min, and then centrifuged at 10 000 × g for 10 min at 4°C to generate supernatant (nuclear fraction).

Western blot analysis

Lysates (12.5 μg/well) were separated by SDS-PAGE, transferred to a nitrocellulose membrane, blocked with 5% nonfat dry milk in Tris buffered saline with 0.1% Tween 20 (TBST) for 1 h at room temperature, and probed with primary antibodies diluted in 5% BSA TBST overnight at 4°C. Membranes were washed with TBST and incubated with a horseradish peroxidase (HRP)-linked antirabbit secondary antibody (1:2000 in 5% BSA TBST [Cell signaling, Danvers, MA]) or HRP-linked antimouse secondary antibody (1:10 000 in 5% BSA TBST, Jackson ImmunoResearch) for 1 h at room temperature. The signal was developed with Clarity Western ECL substrate (Bio-Rad, Hercules, CA) or SuperSignal West Femto substrate (Thermo Fisher Scientific) and measured in a ChemiDoc MP imaging system (Bio-Rad). Primary antibodies used were p44/42 MAPK (AB 330744, 1:1000), Phospho-p44/42 MAPK (AB 331646, 1:1000), MAPKAPK-2 (AB 10694238, 1:1000), Phospho- MAPKAPK-2 (AB 2141311, 1:1000), JunB (AB 2130002, 1:1000), Phospho-JunB (AB 10950322, 1:1000 [Cell Signaling Technologies]), NFκB-p65 (AB 628017, 1:500, Santa Cruz), beta actin (ACTB, AB 476692, 1:10 000, Sigma Aldrich), and Gapdh (AB 2107445, 1:1000, Sigma Aldrich).

Quantitative reverse transcription-PCR (q-RT-PCR)

Total RNA from hTERT cells was extracted with an RNAeasy mini kit (Qiagen, Germantown, MD). The iScript reverse transcription supermix (Bio-Rad) was used to synthesize cDNA, which was amplified with iQ-SYBR Green Supermix (Bio-Rad) in a CFX96 Real-Time System (Bio-Rad). Primers spanning introns were used to measure cDNA abundance. Primer sequences (IDT, Newark, NJ) were human PTGS2, 5′ ACTTTCTGTACGCGGGTGG 3′ and 5′ TGTGCAACACTTGAGTGGCT 3′; and human TOP1, 5′ CCAGACGGAAGCTCGGAAAC 3′ and 5′ GTCCAAGGAGGCTCTATCTTGAA3′. Thermal cycling conditions were as follows: 95°C for 5 min, followed by 50 cycles of 95°C for 10 s and 60°C for 30 s. Each group consisted of three biological replicates run in duplicate. The delta–delta Ct method was used to calculate relative abundance of PTGS2 mRNA. PTGS2 expression was normalized to TOP1 mRNA abundance, which is commonly used as a control in myometrial cells [18].

Nuclei isolation

hTERT nuclei were isolated as previously reported [19]. Briefly, hTERT cells were collected and resuspended in a buffer containing 20 mM Tris, pH 7.6, 0.1 mM EDTA, 2 mM MgCl₂•6H₂O, 0.5 mM NaF, 0.5 mM Na₃VO₄, and protease inhibitors. After incubation on ice for 10 min, a final concentration of 1% Nonidet P-40 was added, and the cell membrane and cytoskeleton were disrupted by gently pipetting up and down three times with a 200 μl pipette tip. The nuclei were then pelleted by centrifugation at 500 g for 3 min at 4°C.

Nuclear Ca²⁺ imaging

Ca²⁺ imaging in isolated nuclei was performed as previously reported [2]. Briefly, the nuclei were incubated with 30 μg/ml Fluo-4 dextran (Thermo Fisher Scientific) for 30 min at 4°C or 20 μM Fluo-4-AM (Thermo Fisher Scientific) for 60 min at 4°C. The nuclei were then washed twice with intracellular buffer (125 mM KCl, 2 mM K₂HPO₄, 0.1 mM MgCl₂, 40 mM HEPES, pH 7.2, 100 nM Ca²⁺, with 10.2 mM EGTA and 1.65 mM CaCl₂) and then equilibrated in the same buffer supplemented with 1 μM of ATP and 300 nM Ca²⁺ for 10 min. Then, nuclei were washed twice with intracellular buffer without ATP and Ca²⁺. The Ca²⁺ imaging was performed at room temperature on an inverted iMIC digital microscope (Till Photonics, Gräfelfing, Germany) with a 20×/0.75 objective (Olympus, Tokyo, Japan) and a Polychrome V monochromator (Till Photonics). A CCD camera (Cooke Optics, Leicester, United Kingdom) was used to collect images at excitation wavelength 488 nm through a 510-nm emission filter at 1 s intervals. Nuclei with an increase in florescence above baseline were considered positive for Ca²⁺. After subtracting the matching background, the image intensities were plotted as a function of time for individual nuclei.

Statistical analysis

Categorical data were subjected to Fisher exact tests with Bonferroni’s correction for multiple tests. Continuous data were subjected to one-way ANOVA followed by Dunnett’s correction for multiple tests. GraphPad software (San Diego, CA) was used for all calculations. P ≤ 0.05 was considered significant. All data are presented as mean and standard error of the mean (SEM).

Results

Paxilline- and lipopolysaccharide-induced NF-κB-p65 nuclear translocation does not require MAPK signaling but does require increased intracellular Ca²⁺

In many cell types, lipopolysaccharide (LPS) binds to the transmembrane signaling receptor toll-like receptor 4, which leads to NF-κB nuclear translocation and activates mitogen-activated protein kinase (MAPK) signaling [20]. Once activated, some effectors of MAPK signaling can also influence NF-κB nuclear translocation. Thus, we first asked whether PAX- and LPS-induced NF-κB nuclear translocation in MSMCs required MAPK signaling. To answer this question, we incubated human MSMCs immortalized with telomerase (hTERT cells) in serum-starvation conditions and then treated them with small molecule inhibitors of the MEK1/2, p38, and JNK pathways. We first confirmed by western blotting that treatment with LPS, and to a lesser extent PAX, led to phosphorylation (activation) of components of these pathways and that the inhibitors prevented these phosphorylation events (Supplementary Figure S1A–C). However, immunofluorescence detection of RELA (also known as p65), a subunit of the NF-κB transcriptional complex, revealed that none of these inhibitors prevented LPS- or PAX-induced NF-κB-p65 nuclear accumulation (Supplementary Figure S1D and Supplementary Table S1). Because serum starvation can alter MAPK signaling, we repeated these experiments in serum-fed conditions and obtained similar results (Supplementary Figure S1A–C and Supplementary Table S2). Thus, all subsequent experiments were done in serum-starvation conditions.

In cerebellar granule neurons and lung epithelial cells [21, 22], increased intracellular Ca²⁺ activates the NF-κB pathway, so we wondered whether increased intracellular Ca²⁺ was required for LPS- or PAX-induced NF-κB-p65 nuclear translocation in hTERT cells. As previously shown [15], the majority of cells treated with LPS (63.24%) or PAX (68.04%) for 2 h showed nuclear accumulation of NF-κB-p65 (Figure 1). In contrast, when cells were pretreated with the cell permeant Ca²⁺ chelator BAPTA-AM for 1 h, only 13.94% of LPS-treated cells and 12.07% of PAX-treated cells had nuclear NF-κB-p65 (Figure 1). We then asked whether increased intracellular Ca²⁺ was sufficient to promote NF-κB-p65 nuclear translocation in hTERT cells. To answer this question, we treated the cells with thapsigargin, which blocks SERCA, thereby preventing Ca²⁺ movement from the cytosol to the sarcoplasmic reticulum and maintaining high cytosolic Ca²⁺ concentration. Compared to control cells, more cells treated with thapsigargin alone had NF-κB-p65 nuclear translocation (76.39% vs. 6.67%). LPS and PAX treatment did not further increase the percentage of cells with nuclear NF-κB-p65 (77.87% and 75.59%, respectively, Figure 1). We conclude that increased intracellular Ca²⁺ is necessary and sufficient for NF-κB-p65 nuclear translocation in hTERT cells.

Figure 4.

Membrane-impermeable BK channel blocker IbTX causes minimal, short-lived NF-kB-p65 translocation into the nucleus. Graph shows the percentage of hTERT myometrial cells with nuclear NF-κB-p65 after cells were treated with vehicle, 100 nM iberiotoxin (IbTX), 10 ng/ml LPS, or IbTX + LPS. At least 80 cells per group were counted in three independent experiments. Data are presented as mean ± SEM. Fisher exact tests with Bonferroni’s correction for multiple tests were performed on the total number of cells from all experiments. ^***P ≤ 0.001 LPS vs. Vehicle; ###P ≤ 0.001 IbTX vs. Vehicle; ⁺⁺⁺P ≤ 0.001 IbTX + LPS vs. Vehicle.

Figure 1.

Paxilline- and LPS-induced NF-κB-p65 nuclear translocation both require increased intracellular Ca²⁺ in human myometrial cells. The percentage of hTERT myometrial cells with nuclear NF-κB-p65 after cells was treated with vehicle (−), 5 μM BAPTA AM (BAPTA), or 1 μM thapsigargin (TG). After 1 h of treatment, 10 ng/ml lipopolysaccharide (LPS) or 20 μM paxilline (PAX) was added for an additional 2 h. Data are presented as mean ± SEM from 3 to 4 independent experiments, each with 105–315 cells counted per group. ^***P ≤ 0.001 by Fisher exact tests with Bonferroni’s correction for multiple tests on the total number of cells from all experiments.

LPS and PAX induce NF-κB-p65 nuclear translocation in hTERT cells by different mechanisms

The above results suggested that PAX and LPS stimulated NF-κB-p65 nuclear translocation through similar mechanisms, both requiring increased intracellular Ca²⁺ and not requiring MAPK signaling. To further test the similarity of the mechanisms, we treated hTERT cells with the small molecule JSH-23, which inhibits LPS-induced nuclear translocation of NF-κB-p65 through an unknown mechanism that is independent of IκB degradation [23]. As expected, significantly fewer cells treated with JSH-23 plus LPS had nuclear NF-κB-p65 than cells treated with LPS alone (8.14% vs. 63.24%, Figure 2A). In contrast, a similar percentage of cells treated with JSH-23 plus PAX vs. PAX alone had NF-κB-p65 nuclear translocation (62.28% vs. 68.04%, Figure 2A). These findings suggested that LPS and PAX induced NF-κB-p65 nuclear translocation by different mechanisms.

Figure 5.

Blocking nuclear BK_Ca with paxilline elevates nuclear Ca²⁺. (A) A representative example of Fluo-4-AM intensity measured in hTERT nuclei treated with vehicle, 20 μM PAX, or 10 ng/ml LPS. (B) The graph shows the percentage of nuclei with increased Fluo-4-AM intensity in four independent experiments, each with 24–32 nuclei analyzed per group. Data are presented as mean ± SEM. ^***P ≤ 0.001 by Fisher exact tests with Bonferroni’s correction for multiple tests performed on the total number of nuclei from all experiments.

Figure 2.

Nuclear pore blocker WGA blocks LPS-induced NF-kB-p65 but not paxilline-induced NF-kB-p65 nuclear translocation. (A) The graph shows the percentage of hTERT myometrial cells with nuclear NF-κB-p65 after cells was treated with vehicle (−), 10 μg/ml wheat germ agglutinin (WGA), or 10 μM JSH-23 (JSH). After 1 h of treatment, 20 μM PAX or 10 ng/ml LPS was added for an additional 2 h. Data are presented as the mean and SEM from 3–4 independent experiments, each with 100–315 cells counted per group. ^***P ≤ 0.001 by Fisher exact tests with Bonferroni’s correction for multiple tests performed on the total number of cells from all experiments. (B, C) Western blots showing NF-kB-p65 in the cytoplasmic (B) or nuclear (C) protein fraction of hTERT cells after the indicated treatments. β-ACTIN and GAPDH are included as loading controls. (D) Quantitation of NF-kB-p65 in western blots, normalized to GAPDH (nuclear) and β-actin (cytoplasmic) and expressed as the ratio between nuclear and cytoplasmic NF-kB-p65. Data are presented as mean ± SEM from three separate experiments. ^***P ≤ 0.001 by one-way ANOVA followed by Dunnett’s correction for multiple tests.

Figure 3.

BK_Ca localizes to the nucleus and plasma membrane in human myometrial cells. Confocal immunofluorescence images of hTERT myometrial cells showing BK_Ca (green) localization and nuclei (Hoechst 33342, blue). The boxed area indicates the magnified area shown below. Scale bars, 25 μm.

Proteins and other large molecules can enter the nucleus via either receptor-mediated active transport or passive transport through nuclear pores. We wondered which of these two mechanisms explained NF-κB-p65 nuclear translocation in PAX- and LPS-treated cells. To answer this question, we treated hTERT cells with wheat germ agglutinin (WGA), which blocks receptor-mediated transport but does not affect passive transport [24, 25]. WGA significantly reduced the percentage of LPS-treated cells that showed NF-κB-p65 nuclear translocation (4.81% vs. 63.24%) but had no effect on the percentage of PAX-treated cells that showed NF-κB-p65 nuclear translocation (60.60% vs. 68.04%) (Figure 2A). To confirm this result, we isolated the nuclear fraction from hTERT cells and immunoblotted for NF-κB-p65. Substantially more NF-κB-p65 protein was in the nuclear fractions in cells treated with LPS alone, PAX alone, or WGA plus PAX than in control cells or cells treated with WGA alone or WGA plus LPS (Figure 2B-D). These data indicated that NF-κB-p65 entered the nucleus via active transport in response to LPS but entered the nucleus via passive transport in response to PAX.

PAX induces long-term NF-κB-p65 nuclear translocation by inhibiting intracellular BK_Ca

Given that PAX-induced NF-κB-p65 nuclear translocation was not inhibited by WGA, and that BK_Ca localizes to the nuclear membrane in neurons [2], we wondered whether nuclear BK_Ca played a role in NF-κB-p65 nuclear translocation. To address this possibility, we first performed immunofluorescence on hTERT cells and noted that BK_Ca localized to both the plasma membrane and the nucleus (Figure 3). To determine the relative contributions of plasma membrane and intracellular BK_Ca on NF-κB-p65 nuclear translocation, we treated hTERT cells with the BK_Ca blocker iberiotoxin (IbTX) which, unlike PAX, is plasma membrane impermeant. Immunofluorescence revealed that approximately 20% of IbTX-treated cells had nuclear-localized NF-κB-p65 at 30–45 min, and the percentage decreased substantially by 60 and 120 min (Figure 4). In contrast, at 60 and 120 min, ~65% of cells treated with LPS or IbTX plus LPS had nuclear localized NF-κB-p65. Additionally, quantitative RT-PCR analysis showed that, unlike PAX, IbTX treatment did not lead to increased expression of the key NF-κB transcriptional target PTGS2 (Supplementary Figure S2). This indicated that inhibition of BK_Ca on the plasma membrane induced an initial wave of NF-κB-p65 nuclear translocation but was unable to promote long-lasting NF-κB-p65 nuclear localization and target gene expression.

Figure 6.

Proposed model of regulation of NF-kB in myometrial cells. Our data indicate that LPS initiates release of NF-kB from IkB by activating the TLR4 complex. NF-kB can then translocate to the nucleus in a WGA- and JSH-23-sensitive manner. In contrast, PAX allows NF-kB to translocate to the nucleus in a WGA- and JSH-23-insensitive manner. Although the mechanism by which PAX treatment inhibits IkB and allows transport though the nuclear pore is unknown, it likely involves Ca²⁺, as blocking BK_Ca in whole cells or nuclei increases intracellular and nuclear Ca²⁺ concentration, and NF-kB translocation into the nuclei is Ca²⁺ dependent.

Given the importance of intracellular BK_Ca and Ca²⁺ in PAX-induced nuclear NF-κB-p65 translocation, we hypothesized that PAX treatment altered nuclear Ca²⁺ concentration in hTERT cells. To address this possibility, we isolated hTERT nuclei, loaded them with the fluorescent Ca²⁺ indicator Fluo-4-AM, and treated them with LPS or PAX. LPS treatment had no effect on nuclear Ca²⁺, but PAX treatment increased nuclear Ca²⁺ concentration (Figure 5). These findings suggest that PAX promotes NF-κB-p65 nuclear translocation by inhibiting nuclear BK_Ca and increasing the nuclear Ca²⁺ concentration.

Discussion

Together, our data indicate that LPS and PAX induce NF-κB-p65 nuclear translocation by mechanisms that are similar in some respects but differ in others (Figure 6). We found that both LPS- and PAX-induced NF-κB-p65 nuclear translocation required an increase in cytoplasmic Ca²⁺, and neither depended on MAPK signaling. However, WGA and JSH-23 blocked NF-κB-p65 nuclear translocation in response to LPS but did not prevent PAX-induced NF-κB-p65 nuclear translocation. This suggests that NF-κB-p65 nuclear translocation occurs through receptor-mediated transport in response to LPS but through passive transport in response to PAX. Our observation that the cell membrane-impermeant BK_Ca blocker IbTX did not induce long-lasting NF-κB-p65 nuclear translocation suggested that intracellular BK_Ca contributes to the mechanism. Finally, we found that PAX treatment caused Ca²⁺ to increase in the nucleus, suggesting that the increased nuclear Ca²⁺ promotes passive NF-κB-p65 nuclear translocation.

Our findings that increased Ca²⁺ was both necessary and sufficient for NF-κB-p65 nuclear translocation is consistent with observations that Ca²⁺ increases promote NF-κB activity in other systems [7, 21, 26, 27]. Several pathways that are key to myometrial activity, such as calmodulin and calcium/calmodulin-dependent protein kinases and the phosphatidylinositol 3-kinase Akt pathway [21, 28–31], are involved in Ca²⁺-dependent NF-κB translocation in other cell types. Future work should address whether these pathways are also involved in NF-κB-p65 translocation in myometrial cells.

Our WGA experiments suggest that NF-κB-p65 enters the nucleus via different mechanisms in response to LPS and PAX. In other cell types, NF-κB nuclear import is mediated by the classical importin α and importin β system in which importin α binds to the NF-κB nuclear localization sequence, and importin β mediates transport through the nuclear pore [32]. Although we suggest that NF-κB-p65 enters the nucleus via passive transport in response to PAX, NF-κB translocation could be mediated by exportin and importin 8. These proteins bind p65 independent of its nuclear localization sequence and may not be inhibited by WGA [33]. Whether this mechanism of transport is involved in PAX-induced NF-κB-p65 nuclear translocation is unknown.

Our findings suggest that nuclear BK_Ca plays an important role in myometrial cells. First, we noted an increase in nuclear Ca²⁺ in response to PAX in myometrial cells. This is similar to a previous finding that blocking nuclear BK_Ca induced increased nuclear Ca²⁺ and gene transcription in rodent hippocampal neurons [2]. Second, our finding that thapsigargin treatment promoted NF-κB-p65 nuclear translocation is consistent with a role for nuclear BK_Ca, as thapsigargin increases cytoplasmic Ca²⁺ and influences the Ca²⁺ gradient across the nuclear envelope [34, 35]. Nuclear BK_Ca has similar electrophysiological properties as plasma membrane BK_Ca and could regulate the nuclear transmembrane potential, but whether this contributes to PAX-induced NF-κB-p65 translocation in myometrial cells is unknown.

Our findings do not rule out the possibility that BK_Ca present in other organelles contributes to PAX-induced NF-κB-p65 translocation. Several studies provide evidence that BK_Ca has functional roles on mitochondria, lysosomes, and endoplasmic reticulum of many different cell types [2, 36–39]. For example, BK_Ca on the membrane of lysosomes is involved in lysosomal membrane potential, Ca²⁺ release, and membrane trafficking in fibroblasts [39]. In cardiomyocytes, mitochondrial BK_Ca regulates the mitochondrial permeability transition pores by modulating Ca²⁺ concentration and reactive oxygen species production [40]. We are especially interested in the possibility that mitochondrial BK_Ca participates in PAX-induced NF-κB-p65 translocation given that we have detected BK_Ca in the mitochondria of hTERT cells (data not included). Moreover, myometrial mitochondria can modulate uterine contractions and intracellular Ca²⁺ dynamics [41]. Therefore, BK_Ca function in myometrial mitochondria warrants further study so as to fully delineate the mechanisms by which activity of BK_Ca contributes to regulating uterine quiescence before term and contractility during parturition.

Data availability

The data underlying this article are available in the article and in its online supplementary material.

Author contributions

LNK, YEL, MWP, and SKE designed experiments. LNK, YEL, MWP, MZN, and SGW conducted all experiments and analyzed data. LNK, YEL, MWP, and SKE wrote the manuscript.

Conflict of interest

The authors have no conflicts to declare.