MicroRNA-mediated downregulation of K⁺ channels in pulmonary arterial hypertension

Abstract

Downregulated expression of K⁺ channels and decreased K⁺ currents in pulmonary artery smooth muscle cells (PASMC) have been implicated in the development of sustained pulmonary vasoconstriction and vascular remodeling in patients with idiopathic pulmonary arterial hypertension (IPAH). However, it is unclear exactly how K⁺ channels are downregulated in IPAH-PASMC. MicroRNAs (miRNAs) are small non-coding RNAs that are capable of posttranscriptionally regulating gene expression by binding to the 3′-untranslated regions of their targeted mRNAs. Here, we report that specific miRNAs are responsible for the decreased K⁺ channel expression and function in IPAH-PASMC. We identified 3 miRNAs (miR-29b, miR-138, and miR-222) that were highly expressed in IPAH-PASMC in comparison to normal PASMC (>2.5-fold difference). Selectively upregulated miRNAs are correlated with the decreased expression and attenuated activity of K⁺ channels. Overexpression of miR-29b, miR-138, or miR-222 in normal PASMC significantly decreased whole cell K⁺ currents and downregulated voltage-gated K⁺ channel 1.5 (K_V1.5/KCNA5) in normal PASMC. Inhibition of miR-29b in IPAH-PASMC completely recovered K⁺ channel function and K_V1.5 expression, while miR-138 and miR-222 had a partial or no effect. Luciferase assays further revealed that K_V1.5 is a direct target of miR-29b. Additionally, overexpression of miR-29b in normal PASMC decreased large-conductance Ca²⁺-activated K⁺ (BK_Ca) channel currents and downregulated BK_Ca channel β1 subunit (BK_Caβ1 or KCNMB1) expression, while inhibition of miR-29b in IPAH-PASMC increased BK_Ca channel activity and BK_Caβ1 levels. These data indicate upregulated miR-29b contributes at least partially to the attenuated function and expression of K_V and BK_Ca channels in PASMC from patients with IPAH.

INTRODUCTION

Sustained pulmonary vasoconstriction is an important early cause of elevated pulmonary vascular resistance in patients with idiopathic pulmonary arterial hypertension (IPAH) (25, 61). Pulmonary arterial tone and vasoconstriction are controlled by the resting membrane potential (E_m) in pulmonary artery smooth muscle cells (PASMC) (30, 49, 73). A change in E_m plays a key role in excitation-contraction coupling in PASMC by regulating the cytosolic-free Ca²⁺ concentration ([Ca²⁺]_cyt). Membrane depolarization leads to an increase in [Ca²⁺]_cyt by opening voltage-dependent Ca²⁺ channels (VDCC) in PASMC (31, 32, 69). Increased [Ca²⁺]_cyt not only causes PASMC contraction and pulmonary vasoconstriction but also stimulates PASMC proliferation and migration, which are the major contributors to concentric pulmonary arterial wall remodeling (10, 15).

K⁺ channel activity in PASMC contributes significantly to E_m regulation (2, 41). Decreased K⁺ currents by downregulating channel expression (leading to a decreased number of K⁺ channels in the plasma membrane) and/or inhibiting channel activity would depolarize PASMC, open VDCC, and increase Ca²⁺ influx. Downregulated expression of K⁺ channels and decreased K⁺ currents in PASMC have been implicated in the development and progression of pulmonary hypertension; however, the underlying mechanisms are still unknown.

To date, more than eight different K⁺ channel families have been identified in the pulmonary vasculature; voltage-gated (K_V) and Ca²⁺-activated K⁺ (K_Ca) channels appear to participate in the regulation of E_m in PASMC. Each K_V channel comprises four pore-forming α subunits and four regulatory cytoplasmic β subunits that modulate channel activity by inactivating α subunits (47). The diversity of mammalian K_V channels is derived from genetic diversity, with over 40 genes encoding K_V channel α subunits from 12 subtypes (K_V1–12) (18, 42, 74). There are five families of Ca²⁺-activated K⁺ channels (K_Ca1–K_Ca5), including the large-conductance K_Ca channels [MaxiK or large-conductance Ca²⁺-activated K⁺ (BK_Ca)] (60). The BK_Ca channel consists of four α subunits (BK_Caα) that create a pore in the membrane associated with four ancillary β (BK_Caβ1–4) subunits. BK_Caβ2 (KCNMB2), BK_Caβ3 (KCNMB3), and BK_Caβ4 (KCNMB4) subunits are predominantly expressed in endocrine tissue, testis, and brain tissue, respectively; however, BK_Caβ1 (KCNMB1) appears to be solely expressed in smooth muscle tissue (5, 12). The function of BK_Ca channels in PASMC is finely tuned by its regulatory β subunits via enhancing α subunit sensitivity to intracellular Ca²⁺ and voltage (13). The presence of BK_Ca channels has been established in human and animal PASMC (3, 9, 20, 26). Downregulated K_V channel expression (e.g., K_V1.5/KCNA5) and decreased whole cell K_V currents in PASMC have been established in pulmonary hypertension (PH), including IPAH (8, 63, 75), whereas the role of BK_Ca channels in PH is not well established (1, 4, 19, 45).

MicroRNAs (miRNAs) are small non-coding regulatory RNAs that posttranscriptionally regulate gene expression by binding to the 3′-untranslated region (UTR) of their targeted mRNA, thereby preventing translation and/or decreasing stability of target mRNAs. miRNAs have been implicated in the development and progression of pulmonary hypertension by using lung specimen and isolated PASMC from patients with IPAH as well as genetically engineered mice, including miR-204, miR-21, miR-130/301, and the miR-17/92 cluster (11, 28, 46, 56, 76). It has been found that certain miRNAs target multiple K⁺ channels, including K_V7.5/KCNQ5 (34), K_V4.2/KCND2 (38), K_Ca2.3/KCNN3 (36), and K_IR2.1/KCNJ2 (40) channels in vascular smooth muscle cells and cardiac myocytes. The aim of this study is to investigate the role of miRNAs in the regulation of K_V channels and BK_Ca channels in PASMC from patients with IPAH.

MATERIALS AND METHODS

Cell culture.

The approval for using human cells was granted by the University Institutional Review Board. PASMC from six healthy (normal) subjects without pulmonary hypertension and six patients with IPAH were provided by the Pulmonary Hypertension Breakthrough Initiative (PHBI). Normal and IPAH-PASMC were cultured in 5% fetal bovine serum (FBS) smooth muscle growth media (LifeLine) and incubated in a humidified 5% CO₂ atmosphere at 37°C. Pulmonary arterial endothelial cells (PAEC) isolated from 4 normal subjects and 4 IPAH patients (PHBI) were cultured in 2% FBS growth media (LifeLine) and incubated in a humidified 5% CO₂ atmosphere at 37°C. Human PASMC and PAEC were isolated at PHBI facility as previously described (27, 67). PASMC authentication was carried out with fluorescence-activated cell sorting (FACS) and immunocytochemistry (ICC) with smooth muscle actin-α (SMA), smooth muscle 22-α (SM22α), and smooth muscle myosin heavy chain (SMMHC). The percentage of positive cells that exhibit SMA, SM22α, and SMMHC signal was over 95%. PAEC authentication was carried out with FACS and ICC using CD31 (FACS), von Willebrand factor (vWF), and vascular endothelial cadherin (VE-cadherin) (ICC). The percentage of positive cells that exhibit CD31, vWF, and VE-cadherin signal was over 93% in cultures. The cells were acquired at passages 2–3. Passages 4–6 were used for the experiments. The demographic information of human PASMC and PAEC is listed in Table 1.

Table 1.

Demographic information of human subjects from whom PASMC and PAEC were isolated for molecular biological and electrophysiological experiments

Subjects	Sex	Age, yr	Race	Type of Cells
Healthy subject-1	Female	36	White	PASMC
Healthy subject-2	Female	33	White	PASMC
Healthy subject-3	Male	34	White	PASMC
Healthy subject-4	Female	34	Asian	PASMC
Healthy subject-5	Male	46	Asian	PASMC
Healthy subject-6	Female	56	White	PASMC
IPAH patient-1	Female	56	White	PASMC
IPAH patient-2	Female	41	White	PASMC
IPAH patient-3	Male	27	White	PASMC
IPAH patient-4	Male	56	White	PASMC
IPAH patient-5	Female	33	White	PASMC
IPAH patient-6	Male	51	White	PASMC
Healthy subject-1	Female	49	White	PAEC
Healthy subject-2	Male	51	White	PAEC
Healthy subject-3	Female	55	White	PAEC
Healthy subject-4	Male	49	White	PAEC
IPAH patient-1	Male	51	White	PAEC
IPAH patient-2	Male	16	White	PAEC
IPAH patient-3	Female	16	White	PAEC
IPAH patient-4	Female	22	White	PAEC

IPAH, idiopathic pulmonary arterial hypertension; PAEC, pulmonary arterial endothelial cells; PASMC, pulmonary artery smooth muscle cells.

miRNA PCR array.

To determine the miRNA expression profile in PASMC, RNA was isolated from normal, and IPAH-PASMC and miRNA expression was explored using the Human Cancer Pathway Finder miRNA PCR Array from SABiosciences. The lung vasculopathy in IPAH patients was recently demonstrated to be similar to pathological chances in cancer, and miRNA dysregulation in affected lung vascular tissues from IPAH patients shares similarities with miRNA profile in cancer tissues (17, 52). The miRNA PCR array consisted of a 96-well plate, which contained distinct primers for a total of 89 specific miRNAs plus 4 housekeeping small nuclear RNAs for normalization. The miRNAs were reverse transcribed to generate cDNA using miScript Reverse Transcription Kit (Qiagen) and then added to each well of the miRNA PCR array containing distinct miRNA primers. miRNA expression levels were determined by using the CFX96 real-time PCR machine (Bio-Rad Laboratories). The results were analyzed via the ΔΔCt method, and miRNA expression was normalized to the four small housekeeping RNAs U6, SNORD44, SNORD47, and SNORD48 (Qiagen). Small housekeeping RNA U6 displayed the least variability between samples, making it a strong candidate as control. miRNA expression normalized to U6 in normal PASMC was defined as onefold, and miRNAs with a fold change of equal to or greater than 2.0 in IPAH-PASMC are shown in the heatmap.

In silico analysis.

We performed in silico analysis using the standard online software TargetScanHuman v7.2 (http://www.targetscan.org/) to predict the potential target mRNAs encoding K_V and BK_Ca channels for the specific miRNAs. The assigned context score to each 3′-UTR mRNA was used as a tool to select targets. The context score is the sum of the contributions of site-type, 3′ pairing, local AU, and position criteria. For each of these criteria, a more negative score indicates a more favorable site.

miRNA RT-PCR analysis.

The total RNAs were isolated from normal and IPAH-PASMC using a phenol/guanidine-based QIAzol reagent (Qiagen). To generate cDNA for sensitive and specific miRNA detection, the miScript II RT Kit (Qiagen) was used according to the manufacturer’s protocol. Real-time PCR was then performed by using the miScript SYBR Green PCR Kit (Qiagen) on the CFX384 real-time PCR machine (Bio-Rad Laboratories). miRNA specific primers (Qiagen) for human miR-29b, miR-138, and miR-222 were used to amplify miRNAs. The results were analyzed via the ΔΔCt method and miRNA expression was normalized to the small housekeeping RNA U6 since U6 displayed the least variability between samples. U6 is thus a good normalizing control compared with SNORD44, SNORD47, and SNORD48. To implement the ΔΔCt method, we compared our results with an average of the “healthy” cells. miRNA level was first normalized to the U6 level in all healthy subjects and the mean was calculated; then, the ratio of miRNA to U6 from each IPAH patient was normalized to the mean ratio from normal subjects. The mean ratio of miRNA expression level to U6 in normal PASMC was defined as onefold.

Patch-clamp experiments.

Whole cell and single-channel K⁺ currents were recorded with patch-clamp technique using an Axopatch-1D amplifier and a DigiData 1322 interface (Molecular Devices). The borosilicate patch pipettes (2–3 MΩ) were fabricated on a P-97 electrode puller (Sutter Instrument, Novato, CA) and polished with a Narishige MF-63 microforge for the current recording. The command voltage pulse protocols and the data acquisition were performed with pCLAMP 8.1 software. The currents, filtered at 1–2 kHz and digitized at 2–4 kHz, were elicited by depolarizing the cells from a holding potential of −80 mV to a series of test potentials (ranging from −100 to +100 mV, for 500 ms) in increments of 20 mV (every 15 s). The extracellular (bath) solution for recording K_V currents [I_K(V)] had an ionic composition of 140 mM NaCl, 4.7 mM KCl, 3 mM MgCl₂, 10 mM glucose, 10 mM HEPES, and 1 mM EGTA (at pH 7.4), while the intracellular (pipette) solution contained 140 mM KCl, 4 mM MgCl₂, 10 mM HEPES, 10 mM EGTA, and 5 mM Na₂ATP (at pH 7.2). To record whole cell K_Ca currents (I_K(Ca)), the bath (extracellular) solution contained 140 mM NaCl, 4.7 mM KCl, 1.8 mM CaCl₂, 1.2 mM MgCl₂, 10 mM glucose, and 10 mM HEPES (pH was adjusted to 7.4 by NaOH). The pipette (intracellular) solution contained 140 mM KCl, 4.4 mM CaCl₂, 4 mM MgCl₂, 5 mM Na₂ATP, 10 mM HEPES, and 5 mM EGTA (pH was adjusted to 7.2 by KOH); the pCa of the pipette solution under this condition was calculated to be ~6.0. The digital subtraction of the currents recorded in PASMC during application of iberiotoxin (IbTX, 100 nM) or Penitrem A (100 nM) from the currents recorded in PASMC before the application of blocker were defined as IbTX-sensitive or Penitrem A-sensitive BK_Ca currents. The single-channel BK_Ca currents [i_K(Ca)] in the cell-attached membrane patch were measured as we previously reported (50). We held the attached membrane patches at −70 mV before applying various test potentials. Open probability (P_open) and amplitude of the currents were measured with Fetchan and PStat software (Axon Instruments). High-K⁺ pipette (extracellular) solution was used in recording i_K(Ca) so that the K⁺ equilibrium potential (E_K) for the attached membrane was close to 0 mV (i.e., the [K⁺] in the pipette is close to the [K⁺] in the cytosol). The bath (extracellular) solution for measuring i_K(Ca) in the cell-attached membrane included 141 mM NaCl, 1.8 mM CaCl₂, 4.7 mM KCl, 1.2 mM MgCl₂, 10 mM glucose, and 10 mM HEPES (pH = 7.4). The pipette (extracellular) solution included 125 mM KCl, 10 mM NaCl, 4 mM MgCl₂, 5 mM ATP, and 10 mM HEPES (pH = 7.2). Results are presented as a function of the command potential (E_comm), which is the inverse of the applied (holding or test) potential. Because of a negative resting E_m (approximately −40 mV in cultured human PASMC), the actual transmembrane potential across the patched membrane is equal to the difference between the E_comm and resting E_m.

Gene (mRNA) expression RT-PCR analysis.

The total RNAs were isolated from human cells using QIAzol reagent (Qiagen). Real-time RT-PCR was performed as previously described (59). Human oligonucleotide primers (Integrated DNA Technologies) that were used in this study are listed as follows: KCNA1, CCCAGCAAGACGGACTTCTTCAAA (forward)/TCTGGTTTCCTTCCTGCTCAGCTA (reverse); KCNA2, GAACCGCCCTAGCTTTGATG (forward)/TCTCCCAGCTCATAAAACCGA (reverse); KCNA5, TTCTACCACCGGGAAACGGA (forward)/CTTCGGGCACTGTCTGCATT (reverse); KCNMB1, CTGTACCACACGGAGGACACT (forward)/GTAGAGGCGCTGGAATAGGAC (reverse); GAPDH, GCACCGTCAAGGCTGAGAAC (forward)/ATGGTGGTGAAGACGCCAGT (reverse). GAPDH, a housekeeping gene, was used to normalize the expression of the gene or transcript of interest.

Western blot analysis.

Cells were washed with PBS and resuspended into 1 × RIPA buffer (Millipore), which contained protease inhibitor cocktail tablets (Roche). Cells were incubated for 20 min on ice, followed by a 15-min centrifugation at 12,000 revolutions/min. The supernatant was collected and protein concentration was determined using a Nano-Drop spectrophotometer (Thermo Scientific). Samples were applied on 10% SDS-PAGE, and proteins were transferred onto nitrocellulose membranes. The membranes were blocked in 5% nonfat milk and incubated overnight at 4°C with primary antibodies for KCNA1 (cat. no. sc-11184), KCNA2 (cat. no. sc-292447), KCNA5 (cat. no. sc-25681), and KCNMB1 (cat. no. sc-377023) followed by 1-h incubation with secondary antibodies (either anti-rabbit, Cell Signaling, cat. no. 7074S, or anti-mouse, Cell Signaling, cat. no. 7076S). All samples were reprobed for β-actin (cat. no. sc-47778) as a loading control. Blots were developed using the SuperSignal West Pico Chemiluminescent Substrate (Pierce Biotechnology). To compare the levels of target proteins in PASMC between normal subjects and IPAH patients, we first normalized the level of target protein to β-actin level in each normal subject and calculated the mean ratio from all normal subjects as control. Then, the ratio of target protein to β-actin from each IPAH patient was normalized to the mean ratio from normal subjects. The mean ratio of target protein level to β-actin level in normal PASMC was defined as onefold.

Transfection experiments.

For the miRNA transfection, normal PASMC were transfected with 10 nM human miScript miRNA mimics for miR-29b, miR138, and miR-222 (Qiagen), while IPAH cells were transfected with 100 nM miR-29b, miR138, and miR-222 inhibitors (Qiagen) using Lipofectamine RNAiMAX Transfection Reagent (Invitrogen) according to the manufacturer’s protocol. AllStars Control siRNA (Qiagen, cat. no. 1027280) and miScript Inhibitor Negative Control (Qiagen, cat. no. 1027271) were utilized as a negative control (NC) for miRNA mimic and miRNA inhibitor transfection experiments, respectively (35, 39, 53, 65). The transfection was performed in Opti-MEM Reduced Serum Medium (Gibco) for 4–6 h. Then, the Opti-MEM was replaced with 5% FBS medium. After 48–72 h, the cells were collected for RNA or protein isolation and were used for electrophysiological recordings. The transfection efficiency was confirmed by real-time RT-PCR. For the gene transfection, human embryonic kidney (HEK) 293 cells were transiently transfected with 2 μg of KCNA5 (OriGene, plasmid no. RG219793) using X-tremeGENE 9 DNA Transfection Reagent (Roche) according to the manufacturer’s instructions. Experiments were performed 24–48 h after transfection.

Antibody validation.

All antibodies used in this study have undergone validation in the course of this study. Representative full-length blots and details of antibody validation are presented in Supplemental Fig. S1 (available at https://doi.org/10.6084/m9.figshare.9636893).

Luciferase assays.

gBlocks gene fragments containing the 3′-UTR of the human KCNA5, including the putative miR-29b binding site as well as enzyme restriction sites for PmeI and XbaI, were purchased from Integrated DNA Technologies. Cloning was performed according to the manufacturer’s protocol. The gBlocks and the pmirGLO Dual-Luciferase miRNA Target Expression Vector (Promega) were digested with PmeI and XbaI, followed by ligation and transformation. To mutate the miR-29b seed region, a site-directed mutagenesis kit (Invitrogen) with specific primers (Integrated DNA Technologies) was used according to the manufacturer’s protocol. Vector constructs (wild type, mutant-1, and mutant-2) were confirmed by DNA sequencing. Wild type, mutant-1, and mutant-2 3′-UTR vectors as well as an empty vector for a control were used in luciferase assays. To prepare for the luciferase assay, HEK293 cells were transfected in 6-well plates with 10 nM miR-29b mimic (Qiagen), 2 μg luciferase vector, and 40 ng Renilla luciferase vector (pRL-TK) using Lipofectamine RNAiMAX (Invitrogen). Forty-eight hours after transfection, luciferase assays were performed using the Dual-Luciferase Reporter Assay System (Promega), according to the manufacturer’s protocol. Cells were lysed using the passive lysis buffer. Twenty microliters of the resulting cell lysate was added to a white 96-well plate, which was then placed in a GloMax 96 Microplate Luminometer with Dual Injectors (Promega). The firefly luminescence from each experimental condition was normalized to the luminescence from empty control vectors.

Statistical analysis.

The data are expressed as means ± SE. The differences between groups were analyzed for statistical significance with Student’s t test (paired or unpaired as applicable) or ANOVA and post hoc tests (Student-Newman-Keuls) where appropriate. The differences were considered to be statistically significant when P < 0.05.

RESULTS

To discover novel miRNAs that might be involved in the development and progression of pulmonary hypertension, we first conducted an miRNA PCR array experiment to determine and compare miRNA expression levels in PASMC between normal subjects and patients with IPAH. To determine which miRNAs are potentially involved in downregulating K⁺ channel expression and decreasing whole cell K⁺ currents in PASMC, we focused on the miRNAs that were selectively upregulated in IPAH-PASMC compared with normal cells. Then, we performed a series of experiments to determine whether the upregulated miRNAs affected expression and activity of various K⁺ channels in PASMC from IPAH patients.

Decreased activity of K_V channels is associated with selectively upregulated miRNAs in IPAH-PASMC.

We first compared whole cell K_V currents in PASMC isolated from healthy control subjects and patients with IPAH. The amplitude (Fig. 1Aa and Ab) and current density (Fig. 1Ac) of whole cell K_V currents, elicited by depolarizing the cells from a holding potential of −70 mV to a series of test potentials ranging from −90 mV to +90 mV, were significantly (P < 0.01) lower in IPAH-PASMC than in normal (Nor) cells. To determine whether the dysfunction of K⁺ channels or the decreased current density of K_V channels in IPAH-PASMC was due to miRNA-mediated posttranscriptional regulation, we then performed a miRNA PCR microarray experiment to investigate the human miRNA expression profile in normal PASMC and IPAH-PASMC. Among 89 miRNAs screened, 5 miRNAs were significantly (twofold or greater) downregulated (blue) and 6 miRNAs were significantly (twofold or greater) upregulated (red) in IPAH-PASMC compared with normal PASMC (Fig. 1B). The real-time PCR experiments confirmed 3.3-fold, 3.3-fold, and 2.9-fold upregulation for miR-29b, miR-138, and miR-222, respectively, in IPAH-PASMC (Fig. 1C). We were unable to detect any significant differences in miR-15a, miR-29a, and miR-let-7e levels between normal and IPAH cells using the real-time PCR method. In addition to the upregulated miRNAs, miRNA PCR microarray showed there are 5 miRNAs (miR-18a, miR-183, miR-181b, miR-378, and miR-126) that are downregulated in IPAH-PASMC compared with normal PASMC. Therefore, miR-29b, miR-138, and miR-222 were selected for the downstream experiments to investigate whether they were involved in (or required for) the regulation of activity and/or expression of K⁺ channels in PASMC isolated from patients with IPAH.

Fig. 1.

Upregulated microRNAs (miRNAs) correlate with reduced voltage-gated K⁺ (K_V) currents in pulmonary artery smooth muscle cells (PASMC) from patients with idiopathic pulmonary arterial hypertension (IPAH). A: representative records (a) showing whole cell K_V currents elicited by depolarization from a holding potential of −70 mV to a series of test potentials ranging from −80 to +80 mV (in 20-mV increments) in PASMC isolated from normal subjects (Nor) and IPAH patients (IPAH). Summarized data (mean ± SE) showing the averaged current amplitude at +80 mV in Nor and IPAH PASMC (b). The current density-voltage (I-V) relationship curves (c) of whole cell K_V currents measured from normal (n = 25 cells from 4–5 different cell lines) and IPAH (n = 25 cells from 4–5 different cell lines) PASMC. I-V curves of normal and IPAH PASMC are significantly different (P < 0.01). B: a heatmap showing miRNAs that are upregulated (red) and downregulated (blue) in normal (n = 2 different cell lines) and IPAH (n = 2 different cell lines) PASMC. Six miRNAs (with greater than twofold upregulation) highlighted in red were further investigated by using real-time RT-PCR analysis. C: real-time RT-PCR analysis of miR-29b, miR-138, and miR-222 expression (normalized to U6) in PASMC from normal subjects (Nor, n = 6 different cell lines) and IPAH patients (IPAH, n = 6 different cell lines). The expression of miR-15a, miR-29a, and miR-let-7e in normal and IPAH PASMC is not significantly different (data not shown). Data are expressed as median (solid line)/mean (dashed line) ± confidence interval or as mean ± SE from 3–5 independent experiments. **P < 0.01, ***P < 0.001 vs. Nor. Statistical analysis was performed using unpaired Student’s t test.

Downregulation of specific K⁺ channels in PASMC from patients with IPAH.

In silico analysis revealed that miR-29b, miR-138, and miR-222 could target several K_V (i.e., K_V1.1/KCNA1, K_V1.2/KCNA2, K_V1.3/KCNA3, K_V1.5/KCNA5, K_V1.6/KCNA6, K_V1.7/KCNA7, K_Vβ1.1/KCNAB1, K_Vβ1.3/KCNAB3) and BK_Ca (BK_Caβ1/KCNMB1, BK_Caβ3/KCNMB3, BK_Caβ4/KCNMB4) channels. It has been established that K_V1.1/KCNA1, K_V1.2/KCNA2, K_V1.5/KCNA5, and BK_Caβ1/KCNMB1 are involved in the development of pulmonary hypertension (4, 9, 75). In this study, we aimed to determine whether miR-29b, miR-138, and miR-222 are responsible for the regulation of these channels in PASMC isolated from patients with IPAH. To test this possibility, we first compared the expression level of certain K⁺ channels (K_V1.1, K_V1.2, K_V1.5, and BK_Caβ1) in PASMC from normal subjects and patients with IPAH. Real-time RT-PCR experiments showed that the mRNA levels of K_V1.2 (KCNA2), K_V1.5 (KCNA5), and BK_Caβ1 (KCNMB1) were downregulated in PASMC isolated from patients with IPAH (Fig. 2A). However, there were no significant differences in K_V1.1 (KCNA1) mRNA expression between normal and IPAH-PASMC. Utilizing Western blot technique, we confirmed these results. The protein levels of KCNA2 (K_V1.2), KCNA5 (K_V1.5), and KCNMB1 (BK_Caβ1) were significantly reduced in IPAH-PASMC compared with control cells, while the protein level of KCNA1 did not change (Fig. 2, B and C). These data indicate that mRNA expression of certain K⁺ channels is significantly downregulated in PASMC from IPAH patients. Downregulated K⁺ channel expression is potentially responsible for the decreased amplitude and current density of whole cell K⁺ currents due to changes in K⁺ channel pore-forming subunits (i.e., KCNA2 and KCNA5) or from changes in K⁺ channel regulatory subunits (i.e., KCNMB1) in IPAH patients.

Fig. 2.

Downregulation of selective K⁺ channels in pulmonary artery smooth muscle cells (PASMC) isolated from patients with idiopathic pulmonary arterial hypertension (IPAH). A: real-time RT-PCR analysis of KCNA1, KCNA2, KCNA5, and KCNMB1 (normalized to GAPDH) in PASMC isolated from normal subjects (Nor, n = 6 different cell lines) and IPAH patients (IPAH, n = 6 different cell lines). *P < 0.05, ***P < 0.001 vs. Nor. B: representative Western blot images showing the protein expression level of KCNA1, KCNA2, KCNA5, and β-actin (a) as well as KCNMB1 and β-actin (b) in normal and IPAH-PASMC. C: summarized data showing protein level of KCNA1, KCNA2, KCNA5, and KCNMB1 (normalized to β-actin) in normal subjects (Nor, n = 6 different cell lines) and IPAH patients (IPAH, n = 6 different cell lines). *P < 0.05, **P < 0.01 vs. Nor. D: representative Western blot images (a) showing the protein expression level of KCNA2, KCNA5, KCNMB1, and β-actin and summarized data (b) in pulmonary arterial endothelial cells (PAEC) isolated from normal subjects (Nor, n = 4 different cell lines) and IPAH patients (IPAH, n = 4 different cell lines). *P < 0.05 vs. Nor. E: real-time RT-PCR analysis of microRNA (miR)-29b, miR-138, and miR-222 expression (normalized to U6) in PAEC from normal subjects (Nor, n = 4 different cell lines) and IPAH patients (IPAH, n = 4 different cell lines). Data are expressed as median/mean ± confidence interval from 3–5 independent experiments. **P < 0.01, ***P < 0.001 vs. Nor. Statistical analysis was performed using unpaired Student’s t test.

Moreover, we compared levels of K⁺ channels in pulmonary arterial endothelial cells (PAEC) between healthy subjects and IPAH patients to identify whether KCNA2, KCNA5, and KCNMB1 were decreased in endothelial cells. As shown in Fig. 2Da, we were able to detect KCNA2, KCNA5, and KCNMB1 channels in human PAEC but did not reveal significant changes in protein expression of KCNA2 and KCNA5 between normal and IPAH PAEC. KCNMB1 level was slightly increased in IPAH cells compared with controls (Fig. 2Db). We did not measure KCNA1 protein levels in normal and IPAH PAEC since there were no differences in KCNA1 expression between normal and IPAH-PASMC. These results demonstrate that the downregulation of KCNA2, KCNA5, and KCNMB1 are unique to PASMC in IPAH patients.

To further demonstrate that miR-29b, miR-138, and miR-222 specifically regulate K⁺ channels in PASMC, we also quantified the miRNAs in normal and IPAH PAEC. Real-time PCR experiments showed that the miR-138 level was significantly decreased, whereas the miR-222 level was markedly increased in PAEC isolated from IPAH patients compared with controls (Fig. 2E). We did not reveal any changes in miR-29b expression between normal and IPAH cells. These findings led us to speculate that unchanged K⁺ channel expression is related to miR-29b but not miR138 and miR-222 levels in IPAH PAEC.

The miR-29b, miR-138, and miR-222 mimics decrease the activity and downregulate the expression of K_V channels in normal human PASMC.

This set of experiments was designed to investigate whether miR-29b, miR-138, and miR-222 regulate the expression and/or activity of K⁺ channels in human PASMC. As shown in Supplemental Table S1, in silico analysis revealed that miR-29b, miR-138, and miR-222 target KCNA2, KCNA5, and KCNMB1 in human PASMC (Supplemental Material is available at https://doi.org/10.6084/m9.figshare.9636893). First, we aimed to determine whether the overexpression of miR-29b, miR-138, or miR-222 (which were detected at higher levels in IPAH-PASMC) in normal PASMC decreased K⁺ channel expression and/or activity. Normal PASMC transfected with miR-29b, miR-138, and miR-222 mimics had a dramatic increase of miR-29b, miR-138, and miR-222 compared with nontransfected (Mock) cells and cells transfected with AllStars Control siRNA (NC mimic) (Fig. 3A). There was no significant difference between mock and NC mimic groups. It is important to note that we should measure the level of functional (RNA induced silencing complex-associated) miRNA in a manner such that we can avoid detecting miRNA mimic trapped in nonfunctional locations. Because of the limitations of real-time RT-PCR technique, however, we were only able to detect the total, but not the functional, miRNA level (Fig. 3A). Then, we compared K_V channel activity in normal PASMC transfected with NC mimic or miRNA mimics (miR-29b/miR-138/miR-222). The amplitude and current density of whole cell K_V currents were significantly decreased by miR-29b (P < 0.001), miR-138 (P < 0.001), or miR-222 (P < 0.001) mimics (Fig. 3B); the miR-29b-induced inhibition (~80%) seemed to be greater than the miR-138/miR-222-mediated inhibition (at 60%–70% compared with NC mimic-transfected cells) (Fig. 3Bb). We then aimed to investigate which K_V channels were responsible for the miRNA-mediated decrease in K_V currents in normal PASMC. As shown earlier in this study (see Fig. 2), we found that both mRNA and protein expression levels of KCNA2 and KCNA5 were significantly downregulated in IPAH-PASMC compared with cells from healthy subjects. In the next set of experiments, we examined whether expression of KCNA2 and KCNA5 channels were regulated by miR-29b, miR-138, and miR-222 in normal PASMC. As shown in Fig. 3C, all three miRNAs (miR-29b/miR-138/miR-222) remarkably decreased the mRNA expression level of KCNA5 but not the mRNA expression of KCNA2 in normal PASMC. We were able to confirm the real-time PCR results in the Western blot experiments. However, we could not detect any changes in KCNA2 protein expression between the NC mimic-transfected cells and the miRNA-transfected cells, while the KCNA5 level was dramatically reduced in PASMC overexpressed miR-29b, miR-138, or miR-222 (Fig. 3D). These results indicate that the upregulation of miR-29b, miR-138, or miR-222 (observed in IPAH-PASMC) reduces whole cell K_V currents by, at least partially, downregulating KCNA5 channels in normal PASMC.

Fig. 3.

MicroRNA (miR)-29b, miR-138, and miR-222 reduce voltage-gated K⁺ (K_V) currents and downregulate KCNA5 expression in normal human pulmonary artery smooth muscle cells (PASMC). A: real-time RT-PCR analysis of miR-29b, miR-138, and miR-222 expression (normalized to U6) in nontransfected PASMC (mock) and PASMC transfected with AllStars Control siRNA [negative control (NC) mimic], miR-29b, miR-138, or miR-222 mimic (n = 3 different cell lines). Data are expressed as mean ± SE from 3–4 independent experiments. **P < 0.01, ***P < 0.001 vs. Mock; ^###P < 0.001 vs. NC mimic. B: representative whole cell K_V currents (a) elicited by depolarizing the cells from a holding potential of −70 mV to a series of test potentials ranging from −80 to +80 mV (in 20-mV increments) and summarized current-voltage (I-V) relationship curves (b) in normal human PASMC transfected with NC, miR-29b, miR-138, or miR-222 mimic. Summarized data showing that the current density of K⁺ currents induced by the test potentials of +40 mV to +80 mV in PASMC transfected with miR-29b, miR-138, or miR-222 mimic (n = 25 cells from 4–5 different cell lines) is significantly different from the NC mimic group (P < 0.001). C: real-time RT-PCR analysis (mean ± SE) of KCNA2 and KCNA5 (normalized to GAPDH) in normal PASMC transfected with NC, miR-29b, miR-138, or miR-222 mimic (n = 5 different cell lines). **P < 0.01, ***P < 0.001 vs. NC mimic. D: representative Western blot images (a) and summarized data (b) showing the protein expression levels of KCNA2, KCNA5, and β-actin in normal PASMC transfected with NC, miR-29b, miR-138, or miR-222 mimic (n = 5 different cell lines). Data are expressed as mean ± SE or mean/median ± confidence interval (b) from at least 3 independent experiments. *P < 0.05, **P < 0.01 vs. NC mimic. Statistical analysis was performed using one-way ANOVA and post hoc test (Student-Newman-Keuls).

Inhibition of miR-29b rescues the activity and expression of K_V channels in PASMC that were isolated from IPAH patients.

To investigate whether the inhibition of selected miRNAs (miR-29b, miR-138, or miR-222) in IPAH-PASMC results in recovering of K_V channel activity and/or expression, we transfected IPAH-PASMC with the miRNA inhibitors (or antagomiRs). As expected, transfection of the specific miRNA inhibitors, antagomiR-29b, antagomiR-138, or antagomiR-222 in IPAH-PASMC significantly decreased the levels of miR-29b, miR-138, and miR-222, respectively (Fig. 4A), whereas transfection of the miScript Inhibitor Negative Control (NC Inhibitor) had no effect on the miRNA levels (Fig. 4A) in comparison to the nontransfected cells (Mock). We then compared K_V channel activity in IPAH-PASMC transfected with the NC inhibitor or the miRNA inhibitors. As shown in Fig. 4Ba, antagomiR-29b completely rescued whole cell K_V currents in IPAH-PASMC to the normal level, while antagomiR-138 partially restored whole cell K_V currents. The amplitude of the whole cell K_V currents was significantly (P < 0.01) higher in IPAH-PASMC transfected with antagomiR-29b or antagomiR-138 than in NC inhibitor-transfected IPAH-PASMC (Fig. 4Bb). Interestingly, inhibition of miR-222 with antagomiR-222 had no effect on whole cell K_V currents in IPAH-PASMC (Fig. 4B).

Fig. 4.

Inhibition of microRNA (miR)-29b restores voltage-gated K⁺ (K_V) channel activity and KCNA5 expression in idiopathic pulmonary arterial hypertension (IPAH)-pulmonary artery smooth muscle cells (PASMC). A: real-time RT-PCR analysis (means ± SE) of miR-29b, miR-138, and miR-222 expression levels (normalized to U6) in nontransfected IPAH-PASMC (mock) and IPAH-PASMC transfected with miScript Inhibitor Negative Control (NC inhibitor), antagomiR-29b, antagomiR-138, or antagomiR-222 (n = 3 different cell lines). Data are expressed as mean ± SE from 3–4 independent experiments. *P < 0.05, **P < 0.01 vs. Mock; ^#P < 0.05, ^##P < 0.01 vs. NC inhibitor. B: representative whole cell K_V currents (a) elicited by depolarizing the cells from a holding potential of −70 mV to a series of test potentials ranging from −60 mV to +80 mV (in 20-mV increments) and summarized data (means ± SE) showing current-voltage (I-V) relationship curves (b) of the K_V currents in IPAH-PASMC transfected with NC inhibitor, antagomiR-29b, antagomiR-138, or antagomiR-222. Summarized data (b) showing that the current density of K⁺ currents induced by the test potentials of +40 mV to +80 mV in IPAH-PASMC transfected with antagomiR-29b and antagomiR-138 (n = 25 cells from 4–5 different cell lines) is significantly different from the NC inhibitor group (P < 0.01). C: real-time RT-PCR analysis of KCNA5 (normalized to GAPDH) in IPAH-PASMC transfected with NC inhibitor, antagomiR-29b, antagomiR-138, or antagomiR-222 (n = 5 different cell lines). D: representative Western blot images (a) and summarized data (b) showing the protein expression level of KCNA5 and β-actin in IPAH-PASMC transfected with NC inhibitor, antagomiR-29b, antagomiR-138, or antagomiR-222 (n = 5 different cell lines). Data are expressed as mean ± SE or mean/median ± confidence interval (b) from at least 3 independent experiments. ***P < 0.001 vs. NC inhibitor. Statistical analysis was performed using one-way ANOVA and post hoc test (Student-Newman-Keuls). miRNA, microRNA.

Next, we investigated whether inhibition of miR-29b, miR-138, or miR-222 resulted in increased KCNA5 expression in IPAH-PASMC. Out of the three miRNA inhibitors, only antagomiR-29b was able to significantly upregulate both mRNA and protein level of KCNA5 channels (Fig. 4, C and D). Inhibition of miR-138 or miR-222 by transfecting antagomiR-138 or antagomiR-222 had no effect on KCNA5 protein expression levels in IPAH-PASMC (Fig. 4, C and D). We did not quantify KCNA2 protein levels since KCNA2 mRNA was not changed in miRNA-transfected cells (see Fig. 3, C and D). These data demonstrate that only miR-29b inhibition is sufficient to recover or restore the whole cell K_V currents by increasing KCNA5 protein expression in PASMC isolated from patients with IPAH.

To further characterize the recovered or restored currents in antagomiR-29b-transfected IPAH-PASMC, we conducted pharmacological experiments using diphenyl phosphine oxide-1 (DPO-1), a specific K_V1.5/KCNA5 channel blocker (21, 23). To confirm that DPO-1 specifically blocks K_V1.5/KCNA5 currents, we used HEK293 cells transiently transfected with the KCNA5 gene (Fig. 5A) to determine whether DPO-1 decreases whole cell KCNA5 currents. Extracellular application of DPO-1 (0.3 µM) to the KCNA5-transfected cells significantly and reversibly decreased the amplitude and current density of whole cell KCNA5 currents (Fig. 5B). These results indicate that DPO-1 is a potent inhibitor of KCNA5 channels. Then, we examined the effect of DPO-1 on the recovered K⁺ currents in IPAH-PASMC transfected with antagomiR-29b. As shown in Fig. 5C, antagomiR-29b transfection resulted in an increase in KCNA5 protein expression level in IPAH-PASMC (see also Fig. 4, C and D). Extracellular application of 0.3 μM DPO-1 significantly and reversibly decreased the whole cell K_V currents in IPAH-PASMC transfected with antagomiR-29b (Fig. 5D). The amplitude and current density of K_V currents, elicited by a test potential of +80 mV from a holding potential of −70 mV, in antagomiR-29b-transfected IPAH-PASMC were reduced from 343.1 ± 49.8 pA and 22.9 ± 3.3 pA/pF to 128.6 ± 36.4 pA and 6.4 ± 1.8 pA/pF (P < 0.001) before (Control) and during (DPO-1) extracellular application of 0.3 μM DPO-1 (Fig. 5D, b and c). The DPO-1-mediated inhibitory effect on whole cell K_V currents was also reversible in some cells. These data led us to conclude that increased miR-29b and miR-29b-mediated KCNA5 downregulation are involved in decreasing whole cell K_V currents in IPAH-PASMC.

Fig. 5.

Whole cell K⁺ currents in idiopathic pulmonary arterial hypertension (IPAH)-pulmonary artery smooth muscle cells (PASMC) transfected with microRNA (miR)-29b inhibitor are sensitive to the specific voltage-gated K⁺ channel 1.5 (K_V1.5) channel blocker, diphenyl phosphine oxide-1 (DPO-1). A: Western blot analysis of KCNA5 and β-actin in control human embryonic kidney (HEK) cells (Control) and cells transfected with empty vector (Vector) or KCNA5. B: representative whole cell K⁺ currents (a), elicited by depolarization from a holding potential of −70 mV to a series of test potentials ranging from −80 mV to +80 mV (in 20-mV increments) in a KCNA5-transfected HEK cell before (Control), during (DPO-1), and after (Washout) extracellular application of 0.3 µM DPO-1. Summarized data (means ± SE) showing the amplitude of the KCNA5 currents at +80 mV (n = 10–15 cells from 3 different cell lines) (b) and the current-voltage (I-V) curves (n = 10–15 cells from 3 different cell lines) (c) in KCNA5-transfected HEK cells before (Control), during (DPO-1), and after (Washout) extracellular application of DPO-1. ***P < 0.001 vs. Control and Washout. The I-V curve of the KCNA5 currents in KCNA5-transfected cells superfused with DPO-1 is significantly different (P < 0.001) from the curves in the cells before (Control) and after (Washout) DPO-1 application. C: Western blot analysis of KCNA5 and β-actin in nontransfected control IPAH-PASMC (Mock) and IPAH-PASMC transfected with negative control (NC) inhibitor or antagomiR-29b. D: representative currents (a) elicited by depolarization from a holding potential of −70 mV to a series of test potentials ranging from −80 mV to +80 mV (in 20-mV increments) in an antagomir-29b-transfected IPAH-PASMC before (Control), during (DPO-1), and after (Washout) extracellular application of 0.3 µM DPO-1. Summarized data (means ± SE) showing the amplitude of the outward K⁺ currents at +80 mV (n = 10–15 cells from 3 different cell lines) (b) and the current-voltage (I-V) curves (n = 10–15 cells from 3 different cell lines) (c) in antagomir-29b-transfected IPAH-PASMC before (Control), during (DPO-1), and after (Washout) extracellular application of DPO-1. Data are expressed as mean ± SE from at least 3 independent experiments. ***P < 0.001 vs. Control and Washout. The I-V curve of the outward K⁺ currents in antagomir-29b-transfected IPAH-PASMC superfused with DPO-1 is significantly different (P < 0.001) from the curves in the cells before (Control) and after (Washout) DPO-1 application. Statistical analysis was performed using one-way ANOVA and post hoc test (Student-Newman-Keuls).

miR-29b directly targets 3′-UTR of human KCNA5.

There is a potential complementary binding site for miR-29b in the 3′-UTR of human KCNA5 mRNA (Fig. 6A). To determine if miR-29b directly binds KCNA5, we performed luciferase assays in which wild-type 3′-UTR of KCNA5 was cloned into a pmirGLO vector. To further confirm direct binding of miR-29b to the putative seed sequence in the 3′-UTR of KCNA5, we additionally designed 2 different luciferase constructs containing specific point mutations within the seed sequence (Fig. 6B). The firefly luminescence was significantly decreased in HEK293 cells transfected with wild-type KCNA5 3′-UTR vector and miR-29b mimic compared with cells transfected with wild-type KCNA5 3′-UTR vector and NC mimic (Fig. 6C). This decrease indicates that miR-29b binds to the 3′-UTR of KCNA5. As expected, the transfections with mutated plasmids (mutant-1 and mutant-2) did not significantly alter the luciferase activity between the NC inhibitor-transfected cells and the miR-29b-transfected cells (Fig. 6C). These results demonstrate that miR-29b directly binds to the 3′-UTR of KCNA5. Therefore, a selective miR-29b inhibitor would be an ideal therapeutic agent that may be able to rescue the attenuated K⁺ channel function and the decreased KCNA5 expression in PASMC from patients with IPAH.

Fig. 6.

MicroRNA (miR)-29b directly binds to the 3′-untranslated region (UTR) of KCNA5. A: the potential binding site of miR-29b on the 3′-UTR of human KCNA5. B: the sequences mutated in miR-29b binding site within the 3′-UTR of KCNA5. C: the summarized data showing luciferase activity in human embryonic kidney (HEK) 293 cells cotransfected with negative control (NC) or miR-29b mimic and pRL-TK vector containing the 3′-UTR region of KCNA5 (wild type or mutant) and the luciferase gene (n = 4). Data are expressed as mean ± SE from at least 3 independent experiments. *P < 0.05 vs. NC mimic. Statistical analysis was performed using unpaired Student’s t test.

Activity of BK_Ca channels is attenuated in PASMC isolated from IPAH patients.

Earlier in this study, we showed that KCNMB1 mRNA and protein level is lower in IPAH-PASMC than normal cells (see Fig. 2, Bb and C). In silico analysis revealed that miR-29b, miR-138, and miR-222 could also target the BK_Caβ₁ channel encoded by the KCNMB1 gene. Therefore, we aimed to measure the activity of BK_Ca channels in normal and IPAH-PASMC. In these experiments, whole cell BK_Ca currents were measured under conditions in which the intracellular [Ca²⁺] (in the pipette solution) was maintained at a high level (1 μM by using appropriate concentrations of EGTA and CaCl₂), and extracellular [Ca²⁺] (in the bath solution) was kept at 1.8 mM. Extracellular application of 100 nM iberiotoxin (IbTX), a selective blocker of BK_Ca channels, significantly reduced the whole cell BK_Ca currents in both normal and IPAH-PASMC (Fig. 7Aa). The subtraction of the currents recorded during the application of IbTX from the currents recorded before application of IbTX revealed IbTX-sensitive BK_Ca currents (Fig. 7Ab).

Fig. 7.

Large-conductance Ca²⁺-activated K⁺ (BK_Ca) channel currents are decreased in pulmonary artery smooth muscle cells (PASMC) from patients with idiopathic pulmonary arterial hypertension (IPAH). A: representative whole cell K⁺ currents (a) elicited by depolarizing cells from a holding potential of −70 mV to a series of test potentials ranging from −100 mV to +100 mV (in 20-mV increments, for 500 ms every 15 s) in PASMC from normal subjects (Nor) and PASMC from IPAH patients (IPAH) before (Control) and during [iberotoxin (IbTX)] extracellular application of 100 nM IbTX. IbTX-sensitive components of the currents (b) were obtained by subtracting the currents recorded during application of IbTX from the currents recorded under control condition. B: bar graphs summarizing the capacitance of PASMC isolated from normal subjects and IPAH patients (means ± SE, n = 6 cells, NS). C: summarized data (means ± SE) showing the current-voltage (I-V) relationship curves of the IbTX-sensitive K⁺ currents in normal PASMC and IPAH-PASMC (n = 25 cells from 4–5 different cell lines, P < 0.05 vs. Normal). Data are expressed as mean ± SE from at least 3 independent experiments. Statistical analysis was performed using unpaired Student’s t test.

In normal PASMC, the outward K⁺ currents elicited by depolarizing the cells from a holding potential of −80 mV to a series of test potentials ranging from −90 to +100 mV were mostly inhibited by 100 nM IbTX. On the other hand, the amplitude of whole cell outward K⁺ currents (including K_V and BK_Ca currents) in IPAH-PASMC was smaller than that in normal PASMC (Fig. 7A), and the sensitivity of the whole cell outward K⁺ currents to IbTX was also much less in IPAH-PASMC than in normal PASMC (Fig. 7A). It is important to note that the IbTX-insensitive currents, which are mainly generated by K⁺ efflux through K_V channels (I_K(V)), are very low in normal PASMC, indicating that high intracellular [Ca²⁺] reduces I_K(V) (55). There was no significant difference in cell capacitance between normal and IPAH-PASMC (Fig. 7B). The amplitude and current density of the IbTX-sensitive BK_Ca currents in IPAH-PASMC were significantly (P < 0.05) lower than in normal PASMC (Fig. 7C). These results indicate that the activity of BK_Ca channels is dramatically reduced in PASMC isolated from IPAH patients.

miR-29b regulates BK_Ca channel activity and KCNMB1 expression.

This set of experiments was designed to test whether the selected upregulated miRNAs (miR-29b/miR-138/miR-222) contributed to the attenuated BK_Ca channel activity and/or expression in PASMC from IPAH patients. At first, we compared KCNMB1 expression in normal PASMC transfected with miR-29b, miR-138, or miR-222 mimics. Western blot experiments showed that the protein level of KCNMB1 was decreased by miR-29b, miR-138, and miR-222 in normal PASMC (Fig. 8A). We were also able to confirm our results on channel activities in the single-channel cell-attached mode of patch-clamp experiments. The amplitude of single-channel current at holding potential of +40 mV indicates that this is the large-conductance BK_Ca channel current (Fig. 8Ba). The open probability (P_open) of the channels was significantly decreased in cells transfected with miR-29b mimic compared with NC mimic-transfected cells (Fig. 8B, a and b). Overexpression of miR-138 also had an inhibitory effect on BK_Ca channel activity in normal PASMC, but the P_open in miR-138-transfected cells was only partially decreased in comparison to cells transfected with NC mimic (Fig. 8C). We did not measure BK_Ca channel activity in miR-222-transfected cells because we showed earlier that miR-222 had no effect on the whole cell K⁺ currents in IPAH-PASMC (see Fig. 4B).

Fig. 8.

MicroRNA (miR)-29b, miR-138, and miR-222 downregulate KCNMB1 and reduce large-conductance Ca²⁺-activated K⁺ (BK_Ca) currents in normal pulmonary artery smooth muscle cells (PASMC). A: representative Western blot images (a) and summarized data (means ± SE) (b) showing the protein expression level of the BK_Ca channel β₁-subunit (KCNMB1) and β-actin in normal PASMC transfected with AllStars Control siRNA [negative control (NC) mimic], miR-29b, miR-138, or miR-222 mimic (n = 5 different cell lines). Data are expressed as mean/median ± confidence interval from at least three independent experiments. *P < 0.05 vs. NC. B: representative large-conductance single-channel K⁺ currents in cell-attached membrane patches held at +40 mV (a) in normal PASMC transfected with NC mimic or miR-29b. The symbol C and horizontal broken lines denote the current level when the channel is closed. Summarized data (means ± SE) (b) showing the steady-state open probability (P_open) of the currents recorded in normal PASMC transfected with NC or miR-29b mimic. C: representative single-channel BK_Ca current recordings in cell-attached membrane patches held at +40 mV (a) in normal PASMC transfected with NC and miR138 mimic. The symbol C and horizontal broken lines denote the current level when the channel is closed. Bar graph (b) summarizing the steady-state P_open of the single-channel currents in normal PASMC transfected with NC or miR-138 mimic. Statistical analysis was performed using one-way ANOVA and post hoc test (Student-Newman-Keuls) or unpaired Student’s t test.

Then, we aimed to determine whether antagomiR-29b, antagomiR-138, and antagomiR-222 recovered the expression of BK_Ca channels in PASMC isolated from patients with IPAH. As shown in Fig. 9A, only antagomiR-29b significantly restored KCNMB1 protein expression in IPAH-PASMC, whereas antagomiR-138 and antagomiR-222 had little effect on the protein level of KCNMB1 in IPAH-PASMC. These data suggest that miR-29b had the greatest ability to regulate BK_Ca channel expression and K_Ca currents in human PASMC; therefore, we excluded miR-138 and miR-222 from subsequent experiments. To further confirm miR-29b-mediated regulation of BK_Ca channels, we recorded the whole cell K⁺ current in the presence of a specific blocker of K_Ca channels. Extracellular application of Penitrem A (100 nM), a selective blocker of BK_Ca channels, was able to reduce the whole cell K⁺ currents in normal PASMC transfected with NC mimic but not in cells transfected with miR-29b mimic (Fig. 9Ba). In NC mimic-transfected cells, the outward K⁺ currents were significantly reduced by the BK_Ca channel blocker, Penitrem A. On the other hand, the whole cell outward K⁺ currents in miR-29b-transfected cells were smaller than that in NC mimic-transfected PASMC and insensitive to Penitrem A. The subtraction of the currents recorded during the application of Penitrem A from the currents recorded before application of the blocker revealed Penitrem A-sensitive BK_Ca currents (Fig. 9Bb). The amplitude of the Penitrem A-sensitive BK_Ca currents in PASMC transfected with miR-29b was significantly (P < 0.05) lower than that in NC mimic-transfected cells (Fig. 9Bc). These data indicate that miR-29b, in addition to its inhibitory effect on K_V channels, also decreases BK_Ca channel activity by, at least in part, reducing KCNMB1 mRNA and protein level in PASMC.

Fig. 9.

Inhibition of miR-29b restores KCNMB1 expression level and whole cell large-conductance Ca²⁺-activated K⁺ (BK_Ca) currents in pulmonary artery smooth muscle cells (PASMC) from patients with idiopathic pulmonary arterial hypertension (IPAH). A: representative Western blot images (a) and summarized data (mean ± SE or mean/median ± confidence interval) (b) showing the protein expression level of KCNMB1 and β-actin in IPAH-PASMC transfected with miScript Inhibitor Negative Control (NC inhibitor), antagomiR-29b, antagomiR-138, or antagomiR-222 (n = 5 different cell lines). *P < 0.05 vs. NC inhibitor. B: representative whole cell K⁺ currents, elicited by depolarization from a holding potential of −70 mV to a series of test potentials ranging from −80 mV to +80 mV (in 20-mV increments) in normal PASMC transfected with AllStars Control siRNA (NC mimic) or miR-29b mimic before (Control) and during extracellular application of 100 nM penitrem A (Penitrem A). Penitrem A-sensitive component of the currents (b) was obtained by subtracting the currents recorded during application of Penitrem A from the currents recorded under control condition. Summarized data (means ± SE) (c) showing the I-V curves of the Penitrem A-sensitive K⁺ currents in NC mimic- and miR-29b-transfected PASMC (n = 25 cells from 4 to 5 different cell lines, P < 0.05 vs. NC mimic). C: representative whole cell K⁺ currents, elicited by depolarizing an antagomir-29b-transfected IPAH-PASMC from a holding potential of −70 mV to a series of test potentials ranging from −80 mV to +80 mV (in 20-mV increments) before (Control), during (Penitrem A) and after (Washout) extracellular application of 100 nM penitrem A. Summarized data (means ± SE (b) showing the amplitude of K⁺ currents at +80 mV and the I-V curves of the currents (c) in IPAH-PASMC transfected with antagomiR-29b before (Control), during (Penitrem A) and after (Washout) extracellular application of Penitrem A (n = 10–15 cells from 3 different cell lines, ***P < 0.001 vs. Control). D: representative single-channel current recorded in cell-attached membrane patches held at +40 mV (a) in IPAH-PASMC transfected with NC inhibitor (top in red) and antagomir-29b (bottom in blue). Steady-state P_open in IPAH-PASMC transfected with NC inhibitor or antagomiR-29b (b). The symbolC and horizontal broken lines denote the current level when the channel is closed. Statistical analysis was performed using one-way ANOVA and post hoc test (Student-Newman-Keuls) or unpaired Student’s t test.

To further characterize the outward K⁺ currents recovered by transfecting the miR-29b inhibitor, we measured whole cell BK_Ca currents before, during, and after application of Penitrem A in IPAH-PASMC transfected with antagomiR-29b (Fig. 9C). The extracellular application of 100 nM Penitrem A significantly and reversibly reduced the whole cell outward K⁺ currents in antagomiR-29b-transfected IPAH-PASMC (Fig. 9C). The sensitivity of antagomiR-29b-recovered or restored whole cell K⁺ currents to Penitrem A indicates that BK_Ca currents are part of the K⁺ currents in IPAH-PASMC restored by antagomir-29b (Fig. 9C). These data also imply that increased miR-29b in IPAH-PASMC is involved in decreasing whole cell K⁺ currents by downregulating K_V channels (see Fig. 5D) and BK_Ca channels (see Fig. 9C) in PASMC.

Next, we performed single-channel patch-clamp experiments to further confirm that antagomiR-29b elevated BK_Ca currents in IPAH-PASMC. miR-29b inhibitor or antagomir-29b significantly increased the channel activity and P_open in IPAH-PASMC (Fig. 9D). The amplitude (~10 pA) of a single-channel current at a holding potential of +40 mV for the cell-attached membrane patch indicates that the antagomiR-29b-recovered K⁺ currents in IPAH-PASMC (Fig. 9D) are likely the large-conductance (~200 pS) BK_Ca channel currents. Based on these results, we conclude that miR-29b inhibition is sufficient to restore BK_Ca channel activity in PASMC isolated from patients with IPAH.

DISCUSSION

In this study, we demonstrate that miR-29b is specifically upregulated and negatively exerts posttranscriptional regulation onto K⁺ (K_V and BK_Ca) channels in PASMC isolated from IPAH patients in comparison to PASMC isolated from normal subjects. Additionally, we showed that inhibitors or antagomiRs specifically targeting miR-29b could be novel potential therapeutic agents for patients with IPAH. Our results indicate that 1) upregulated miR-29b, miR-138, and miR-222 are associated with reduced K⁺ channel activity and downregulated K_V1.2 (KCNA2), K_V1.5 (KCNA5), and BK_Caβ1 (KCNMB1) channel subunits in PASMC from IPAH patients; 2) miR-29b decreases K_V channel activity in IPAH-PASMC by directly targeting KCNA5; and 3) miR-29b also contributes to attenuated BK_Ca channel activity and BK_Caβ1 expression in IPAH-PASMC. The miR-29b-mediated posttranscriptional suppression of the K_V channel pore-forming α subunit, K_V1.5/KCNA5, and the large-conductance K_Ca channel regulatory β subunit, BK_Caβ1/KCNMB1, in PASMC is thus one of the important mechanisms responsible for attenuated whole cell K⁺ currents in PASMC from patients and animals with pulmonary hypertension. The decreased function and expression of K_V (e.g., K_V1.5/KCNA5) and BK_Ca (e.g., BK_Caβ1/KCNMB1) channel subunits by miR-29b in PASMC would contribute to causing pulmonary vasoconstriction (by membrane depolarization-mediated increase in [Ca²⁺]_cyt) and pulmonary vascular remodeling (by enhancing PASMC proliferation and inhibiting apoptosis).

We previously identified the expression of many different K⁺ channels in normal human PASMC (26). We and other investigators showed that mRNA and protein expression of K_V channels (e.g., K_V1.2/KCNA2, K_V1.5/KCNA5) was downregulated, and amplitude and current density of K_V currents were reduced in PASMC isolated from patients with IPAH compared with PASMC isolated from normal subjects (or patients without pulmonary hypertension) and in PASMC isolated from animals with experimental pulmonary hypertension (e.g., hypoxia-induced pulmonary hypertension) compared with the normoxic controls (8, 22, 24, 44, 63, 64, 75). Our current findings are consistent with the previous results: the whole cell K_V currents are reduced and K_V1.2/KCNA2 and K_V1.5/KCNA5 channels are downregulated in IPAH-PASMC in comparison to normal cells. However, we could not detect any significant changes in K_V1.1 (KCNA1) mRNA and protein levels between normal and IPAH-PASMC. Given that K_V1.5/KCNA5 channel is an important K⁺ channel involved in regulating the membrane potential and inducing membrane repolarization and hyperpolarization in PASMC (48), the downregulated K_V1.5/KCNA5 expression is responsible, at least partially, for the decreased amplitude and current density of whole cell I_K(V) and the membrane depolarization-associated increases in [Ca²⁺]_cyt in PASMC from IPAH patients under normoxia.

When we used intracellular pipette solution containing high [Ca²⁺]_cyt to record BK_Ca currents in human PASMC, we found that the IbTX-insensitive currents or the remaining IbTX-insensitive K_V currents in PASMC superfused with a high concentration of IbTX (100 nM) were very small in normal PASMC (see Fig. 7A). These results are consistent with early studies showing that intracellular Ca²⁺ may inhibit K_V channels in PASMC (51, 55). Post et al. (51) first reported in 1992 that acute hypoxia inhibited K_V channels in PASMC by an increase in [Ca²⁺]_cyt; the initial increase in [Ca²⁺]_cyt was likely due to hypoxia-mediated Ca²⁺ release (55) from IP₃-sensitive intracellular Ca²⁺ stores and/or Ca²⁺ influx through voltage-independent cation channels (57, 66). The resulting membrane depolarization induced by Ca²⁺-mediated K_V channel inhibition further results in Ca²⁺ influx through voltage-dependent Ca²⁺ channels, causing pulmonary vasoconstriction.

Unlike K_V channels, the expression and function of BK_Ca channels are not well understood in pulmonary hypertension (1, 4, 19, 45). In this study, we revealed that BK_Ca channel activity is attenuated and KCNMB1 expression is reduced in PASMC from IPAH patients in comparison to cells from healthy subjects. Our findings are consistent with the observations in animal models of pulmonary hypertension (7, 71, 77). BK_Ca channel activator NS1619 was able to reduce right ventricular systolic pressure, a surrogate of pulmonary arterial systolic pressure, in rats with monocrotaline-induced pulmonary hypertension (54). KCNMB1 is a regulatory subunit that modulates the electrophysiological property of the pore-forming subunit (e.g., KCNMA1) of the large-conductance Ca²⁺-activated and voltage-dependent K⁺ channels (BK_Ca channels) (13). Coassembly of the β subunit (KCNMB1) with the pore-forming α subunit (KCNMA1) significantly affects the biophysical (e.g., voltage- and Ca²⁺-sensitivity) and pharmacological (e.g., toxin selectivity and efficacy) properties of the BKCa channel (42). The modifications of Ca²⁺ sensitivity (which is regulated by β subunit/KCNMB1) are mainly responsible for the decreased activity of BK_Ca channels in PASMC isolated from rats with hypoxia-induced pulmonary hypertension (9), while kcnmb1^−/− mice develop worse pulmonary hypertension under hypoxic conditions (4). Interestingly, kcnmb1^−/− mice are also hypertensive owing to the inability of BK_Ca channels to be activated by Ca²⁺ sparks and the compensated pressure-induced vasoconstriction (13). Additionally, whole cell BK_Ca current density and Ca²⁺ sensitivity were also lower in vascular smooth muscle cells from patients with systemic hypertension due to KCNMB1 downregulation (72). The striking parallels to systemic hypertension suggest that impaired BK_Ca channel function may be an important contributor to sustained vasoconstriction in pulmonary hypertension.

In contrast to our results, another group revealed the upregulation of KCNMB1 mRNA and protein in lung tissue from donors and IPAH patients (45). Although KCNMB1 appeared to be restricted mainly to smooth muscle tissue, we at first detected its expression in human pulmonary arterial endothelial cells (PAEC). Using Western blot experiments, we demonstrated that KCNMB1 is decreased in PASMC but increased in PAEC from IPAH patients. We previously reported that the whole-lung tissue is mainly composed of lung vascular endothelial cells (62); therefore, the divergent changes in KCNMB1 expression in PASMC and PAEC could explain, at least partially, the different data on KCNMB1 level in PASMC and lung tissue-isolated patients with IPAH or animals with experimental pulmonary hypertension.

Various miRNAs have been implicated in the development and progression of pulmonary hypertension (11, 28, 46, 56, 76); however, miRNA-mediated regulation of K_V and BK_Ca channels is not well understood. Recently, Mondejar-Parreño et al. (43) revealed that KCNA5 is inhibited by the upregulated miR-1 in a hypoxia/SU5416-induced pulmonary hypertension model in rats. In contrast, Machado’s group reported that hypoxia attenuated miR-1 in in vitro and in vivo experiments (58). Additional studies are needed to further investigate these differing findings. Interestingly, it is reported that miR-1 is downregulated in PASMC isolated from IPAH patients (58), suggesting that downregulation of KCNA5 is not directly due to miR-1 in IPAH-PASMC. Our data showed for the first time that miR-29b directly inhibits KCNA5 and decreases KCNMB1, resulting in reduced whole cell K⁺ currents in IPAH-PASMC. The upregulation of miR-138 and miR-222 were also previously established in pulmonary hypertension (6, 29, 33, 37, 68), but in this study, we demonstrated that miR-29b has the most potential to regulate K⁺ channel function and expression in PASMC from patients with IPAH. Although miR-138 and miR-222 attenuate K⁺ channel activity and downregulate KCNA5 and KCNMB1, inhibition of miR-138 and miR-222 are not sufficient to restore K⁺ channel activity and expression in IPAH-PASMC. It is reasonable to speculate that miR-29b could be a potential interesting target in developing a novel therapeutic approach for different forms of pulmonary hypertension.

In human cancer, miR-29b is known to be a tumor suppressor; however, it may also act as a tumor promoter under certain conditions (70). Data about miR-29b in pulmonary hypertension is limited but suggest that miR-29b is involved in the development and progression of disease. In contrast, a recent study found that hypoxia decreases miR-29b in vivo and in vitro (16). Chen et al. showed that miR-29b suppresses proliferation and promotes apoptosis by targeting CCND2 and Mcl-1. Similar to cancer research miR-29b could have a different effect in hypoxia-induced pulmonary hypertension and idiopathic PAH. It is also unclear whether and how miR-29b affects PASMC contraction, proliferation, migration, and/or apoptosis in normal subjects and IPAH patients, this line of investigation will be pursued in future studies.

In conclusion, we identified three miRNAs (miR-29b, miR-138, miR-222) in this study that are upregulated in IPAH-PASMC compared with normal PASMC in addition to many miRNAs that have been implicated in the development and progression of pulmonary hypertension. Our results from this study indicate that downregulated KCNA5, a pore-forming α subunit for forming K_V channels, and KCNMB1, a regulatory β subunit for BKCa channels, and the reduced whole cell K_V and BK_Ca currents in IPAH-PASMC may be due partially to the upregulation of selected miRNAs (e.g., miR-29b, miR-138, miR-222). Overexpression of these miRNAs in normal PASMC makes normal cells phenotypically resemble IPAH-PASMC (e.g., with reduced whole cell K⁺ currents and downregulated KCNA5/KCNMB1), whereas inhibition of miR-29b only in IPAH-PASMC is sufficient to rescue the decreased K⁺ currents and KCNA5/KCNMB1 expression. AntagomiR-29b-recovered whole cell K⁺ currents in IPAH-PASMC contain both Penitrem A-insensitive K_V currents and Penitrem A-sensitive BK_Ca currents. These observations indicate that upregulated miR-29b in IPAH-PASMC is sufficient to decrease both K_V currents (by downregulating KCNA5) and BK_Ca currents (by downregulating KCNMB1). Significantly reduced whole cell K⁺ currents (e.g., by miR-29b, miR-138, miR-222) would subsequently cause membrane depolarization, enhance Ca²⁺ influx through VDCC, increase [Ca²⁺]_cyt and eventually cause pulmonary vasoconstriction and induce pulmonary vascular medial hypertrophy (by stimulating PASMC proliferation and migration). Furthermore, decreased K⁺ currents or K⁺ efflux through K_V and BK_Ca channels in IPAH-PASMC would also inhibit PASMC apoptosis by attenuating apoptotic volume decrease and inhibiting cytoplasmic caspase activity and further contribute to pulmonary vascular wall thickening (14). The impaired BK_Ca channel function or inhibited Ca²⁺- and voltage-mediated BK_Ca channel activation would also contribute to increasing myogenic vascular tone and further enhance sustained pulmonary vasoconstriction. These data imply that the specific inhibition of miR-29b in PASMC (by the use of chemically modified oligonucleotide antisense inhibitors or antagomirs) may be an important strategy to develop novel and therapeutically effective approaches to treat patients with IPAH.

GRANTS

This work was supported, in part, by grants from the National Heart, Lung, and Blood Institute of the National Institutes of Health (HL125208, HL135807, and HL126609).

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

A.B. and J.X.-J.Y. conceived and designed research; A.B., R.J.A., T.Z., J.F.E.V., N.M.P., A.Y., H.Y., and B.A.Q. performed experiments; A.B., R.J.A., T.Z., J.F.E.V., N.M.P., A.Y., H.Y., B.A.Q., S.R., F.B., A.H., R.R.V., A.M., and J.X.-J.Y. analyzed data; A.B., R.R.V., P.A.T., A.M., and J.X.-J.Y. interpreted results of experiments; A.B., R.J.A., T.Z., J.F.E.V., N.M.P., A.Y., H.Y., S.R., F.B., A.H., R.R.V., and J.X.-J.Y. prepared figures; A.B., N.M.P., and J.X.-J.Y. drafted manuscript; A.B., R.J.A., M.B., L.W., K.S.R., R.R.V., P.A.T., A.M., and J.X.-J.Y. edited and revised manuscript; A.B., R.J.A., T.Z., J.F.E.V., N.M.P., A.Y., H.Y., B.A.Q., M.B., L.W., K.S.R., S.R., F.B., A.H., R.R.V., P.A.T., A.M., and J.X.-J.Y. approved final version of manuscript.