NFATc3 regulates BK channel function in murine urinary bladder smooth muscle

Abstract

The nuclear factor of activated T-cells (NFAT) is a Ca²⁺-dependent transcription factor that has been reported to regulate the expression of smooth muscle contractile proteins and ion channels. Here we report that large conductance Ca²⁺-sensitive potassium (BK) channels and voltage-gated K⁺ (K_V) channels may be regulatory targets of NFATc3 in urinary bladder smooth muscle (UBSM). UBSM myocytes from NFATc3-null mice displayed a reduction in iberiotoxin (IBTX)-sensitive BK currents, a decrease in mRNA for the pore-forming α-subunit of the BK channel, and a reduction in BK channel density compared with myocytes from wild-type mice. Tetraethylammonium chloride-sensitive K_V currents were elevated in UBSM myocytes from NFATc3-null mice, as was mRNA for the Shab family member K_V2.1. Despite K_V current upregulation, bladder strips from NFATc3-null mice displayed an elevated contractile response to electrical field stimulation relative to strips from wild-type mice, but this difference was abrogated in the presence of the BK channel blocker IBTX. These results support a role for the transcription factor NFATc3 in regulating UBSM contractility, primarily through an NFATc3-dependent increase in BK channel activity.

stimulation of parasympathetic nerves in the urinary bladder induces the corelease of the neurotransmitters ATP and ACh, which act directly on the detrusor smooth muscle (SM) purinergic and cholinergic receptors, respectively, to generate the force necessary to expel urine. Electrical stimulation initiates excitatory junction potentials that trigger SM action potentials (18). Action potentials consist of a depolarizing upstroke followed by a repolarizing phase and an afterhyperpolarization (5, 17, 19). Ca²⁺ influx through L-type voltage-gated Ca²⁺ channels (L-VDCC) mediates the upstroke of the action potential (7, 17, 20). K⁺ efflux through two types of voltage-sensitive K⁺ channels [large-conductance, Ca²⁺-sensitive K⁺ (BK) channels and voltage-gated K⁺ (K_V) channels] contributes to the repolarization phase of the action potential (16, 17, 20, 45). In addition to their role in repolarization, K_V channels, as well as small conductance, Ca²⁺-sensitive K⁺ (SK) channels, likely play a role in the afterhyperpolarization (11, 17, 45).

Nerve-evoked contractions of urinary bladder smooth muscle (UBSM) are greatly enhanced by blocking BK channels or by targeted disruption of the gene for the pore-forming α-subunit of the BK channel (Slo or Kcnma1) (33, 43, 46). These effects likely reflect actions on UBSM BK channels because nerves in UBSM strips do not express BK channels (43, 46). BK channels in SM are composed of four pore-forming α-subunits and accessory β₁-subunits, which increase the apparent voltage and Ca²⁺ sensitivity of the α-subunits (13, 32, 44). Block of BK channels by the scorpion toxins iberiotoxin (IBTX) or charybdotoxin increases the duration and peak amplitude of UBSM action potentials (20, 30), resulting in an increase in the force of muscle contraction (16, 17, 22, 27). Mice lacking the pore-forming α-subunit exhibit significantly elevated bladder pressures, increased oscillations of intravesicular pressure, reduced bladder capacity, reduced volume per void, and urinary incontinence (33, 43), suggesting that a decrease in BK channel activity may contribute to the symptoms associated with urinary bladder dysfunction.

K_V channels are composed of four α-subunits arranged radially around a central pore (29, 48, 49). We (45) have previously evaluated the biophysical, pharmacological, and molecular properties of the murine UBSM K_V current. RT-PCR analysis of UBSM tissue, combined with biophysical comparisons of the murine UBSM K_V current to values reported in the literature for other K_V channel complexes, suggests that the UBSM K_V channel is most likely a heteromultimeric assemblage of K_V2.1 subunits and K_V5.1 and/or K_V6.1 subunits (45). This stands in contrast to vascular smooth muscle K_V currents, which are carried primarily by channels composed of members of the 4-aminopyridine (4-AP)-sensitive K_V1 subfamily (42, 50) and, as suggested by recent work, through K_V2.1 channels (2, 3).

Recent reports have shown that the expression of BK and K_V channels can be modulated by Ca²⁺-dependent transcription factors, providing a potential link between alterations in Ca²⁺ signaling patterns and changes in K⁺ channel activity. Specifically, the transcription factor nuclear factor of activated T cells (NFAT) has been reported to regulate the expression of K_V2.1 channels in arterial SM and in cardiac myocytes (2, 39) and to regulate the expression of the modulatory β₁-subunit of the BK channel in arterial SM (34).

Four members of the NFAT family (NFAT1/c2, NFAT2/c1, NFAT3/c4, and NFAT4/c3) are Ca²⁺ dependent. The transcriptional activity of these NFAT isoforms is controlled primarily through regulation of NFAT subcellular localization (9, 12, 52). A fifth isoform, NFAT5/TonEBP, is Ca²⁺ independent and shares limited homology with the other family members. Elevations in intracellular Ca²⁺ promote the nuclear localization of NFAT by activating the Ca²⁺- and calmodulin-dependent protein phosphatase calcineurin, which dephosphorylates NH₂-terminal NFAT serine residues, inducing a conformational change that exposes nuclear localization signals (4, 8, 10, 28, 36, 38). Once in the nucleus, NFAT acts as a transcriptional coregulator associating with cofactors at composite regulatory elements to modulate the transcription of target genes (10, 26, 38).

K⁺ channel expression and activity is a crucial factor in determining the contractile status of SM, including UBSM. The fact that K_V channels are putative regulatory targets of NFAT, in turn, implies that NFAT transcriptional activity might modulate UBSM contractility. There are currently no reports of NFAT activity in UBSM. However, in cultured UBSM myocytes, overexpression of a constitutively active form of calcineurin has been shown to induce a hypertrophic phenotype (35), suggesting a possible role for NFAT activity in the urinary bladder.

The goal of the present study was to examine the role of NFAT in regulating nerve-evoked UBSM contractility. We focused on NFATc3 since this isoform is prominently expressed in SM (14, 15, 25, 41) and has been previously reported to modulate the expression of proteins involved in SM contractility, including K⁺ channels and α-actin (2, 15, 34). We found that UBSM strips from NFATc3-null mice exhibited increased force generation in response to electrical field stimulation (EFS) compared with strips from wild-type mice. We attribute the increased nerve-evoked contractility of the UBSM from NFATc3-null mice to decreased expression of the BK channel pore-forming α-subunit and an associated decrease in overall BK channel activity. These results indicate that NFATc3 may play an important role in regulating UBSM excitability and contractility and further suggest that UBSM disorders, which are characterized by alterations in Ca²⁺ regulation, may reflect NFATc3-dependent alterations in BK channel expression and function.

METHODS

Isolation of myocytes.

Adult mice were euthanized with an overdose of pentobarbital sodium followed by exsanguinations, according to a protocol approved by the University of Vermont Office of Animal Care and Management. NFATc3-null mice, originally supplied by Dr. Laurie Glimcher, and wild-type background strain mice (Balb/C) were used in all experiments. Other researchers working with these NFATc3-null mice have reported that the loss of NFATc3 did not result in compensatory changes in the expression of remaining NFAT isoforms at the mRNA level (47).

The bladders were removed and transferred to an ice-cold Ca²⁺-free dissection solution (DS; in mM: 55 NaCl, 5.6 KCl, 2 MgCl₂, 80 sodium glutamate, and 10 glucose) buffered to pH 7.3 with 10 mM HEPES. After the external fatty and connective tissues were removed, the bladders were cut open and rinsed free of urine, and the urothelial layer was carefully peeled away. The detrusor was cut into small sections (∼2 mm × 2 mm) and digested for 20–30 min in DS containing 1 mg/ml papain (Worthington, Lakewood, NJ) and 1 mg/ml dithioerythritol (Sigma, St. Louis, MO), rinsed briefly with DS, and then further digested in DS containing 1 mg/ml collagenase type II (Sigma) and 100 μm CaCl₂ for 6–10 min. Both digestions were performed at 37°C. The digested tissue sections were gently triturated using a fire-polished glass Pasteur pipette to release individual UBSM myocytes.

Electrophysiology.

Whole cell currents were recorded using the perforated-patch configuration of the whole cell patch-clamp technique. Aliquots of dissociated myocytes were transferred to a 0.5-ml electrophysiology chamber and allowed to adhere. After 15–20 min, the cells were thoroughly rinsed in a bath solution consisting of (in mM) 134 NaCl, 6 KCl, 1 MgCl₂, 2 CaCl₂, 10 glucose, and 10 HEPES, pH 7.4. The pipette solution consisted of (in mM) 110 potassium aspartate, 10 NaCl, 30 KCl, 1 MgCl₂, 0.05 EGTA, and 10 HEPES, pH 7.2. Amphotericin, initially solubilized in DMSO, was added to the pipette solution at a final concentration of 200 μg/ml. Iberiotoxin (IBTX; 100 nM final concentration) and tetraethylammonium chloride (TEA; 5 mM final concentration) were both purchased from Sigma and solubilized in the bath solution. All recordings were made at 22°C.

Single-channel BK currents (I_BK) were recorded in excised inside-out patches using the patch-clamp technique. The patches were exposed to a 10 μM buffered Ca²⁺ solution. Single-channel currents were recorded for 2–5 min at potentials of −40 to + 40 mV at 22°C using pipettes fire polished to a final tip resistance of ∼4–6 MΩ. The bath and pipette solution consisted of (in mM) 140 KCl, 1.8 MgCl₂, 1 5-hydroxyethylene-diaminetriacetic acid, and 0.13 CaCl₂ (10 μM free), adjusted to pH 7.2 with NaOH. Data were analyzed using pCLAMP software (Axon Instruments). The number of channels per patch was determined at a holding potential of +60 mV. Channel open probability (P_o) was determined by dividing channel activity (NP_o) derived using the pCLAMP software by the number of channels (N) as in the respective patch.

UBSM contractility studies.

Bladders were harvested and rinsed, and the urothelium was removed, as detailed above. Detrusor strips (1 mm × 3 mm), cut longitudinally from the dome to the bladder base, were suspended in jacketed water baths (7 ml volume, 37°C) in an aerated bath solution (in mM: 119 NaCl, 4.7 KCl, 24 NaHCO₃, 1.2 KH₂PO₄, 2.5 CaCl₂, 1.2 MgSO₄, 0.023 EDTA, and 11 glucose, pH 7.3). Contractility was measured after strips were prestretched to 5 mN of tension for 45 min using a myograph (MyoMED, MED Associates). The EFS protocol consisted of sequential pulses from 0.5 to 50 Hz (20-V pulse amplitude, 0.2-ms pulse width, 2-s pulse duration) at 3-min intervals. All recording were made at 37°C.

Immunohistochemistry and nuclear localization assay.

Dissociated myocytes were placed into eight-well plates and the cells were allowed to adhere for 1 h at 37°C. The medium was then replaced with growth media consisting of Dulbecco's modified Eagle's medium (Invitrogen) with 10% fetal bovine serum (Invitrogen, Carlsbad, CA) and penicillin-streptomycin (10 U/ml, Invitrogen). The plates were maintained at 37°C, 5% CO₂ for 24 h, at which time the cells were ∼60% confluent. Stimulation with platelet-derived growth factor (PDGF; 20 ng/ml), endothelin (ET-1, 100 nM), or pretreatment with FK506 and cyclosporine A (CsA, 1 μM each) was performed as detailed in results. Immunostaining and quantitation of NFAT nuclear localization was performed as detailed previously (15). All NFAT antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA).

Semiquantitative RT-PCR.

Approximately 100-mg sections of mouse urinary bladder detrusor muscle, denuded of urothelium, were homogenized in TRIzol reagent (Sigma) using a motorized homogenizer, and the RNA was isolated as per the manufacturer's protocol. cDNA was generated using the Superscript First-Strand cDNA kit (Invitrogen) according to the manufacturer's protocol. Primers for K_V2.1 (forward: 5′-TGGACAT CGTGGTGGAGAA-3′, reverse: 5′-CAGATACTCTGATCCCGAG-3′) were described previously (45). Primers for the α- and β-subunits of the BK channel (GenBank accession numbers L16912 and AF020711, respectively) were α-BK forward: 5′-AGTCTGCATCTTTGGGGATG-3′, reverse: 5′-TGTCCATTCCAGGAGGTGT-3′; β-BK forward: 5′-TTCTGCACCTCAAGTCAACG-3′, reverse: 5′-GACAGCGTCATTGC-AGG TAA-3′. RT-PCR was performed using the Fail-Safe PCR kit (Epicenter) with Amplitaq Gold DNA polymerase. For semiquantitative analysis of K⁺ channel expression, PCR reactions were terminated at a point empirically determined to lie within the logarithmic phase (33 cycles for K_V2.1 and the BK α- and β-subunits) and expressed relative to the housekeeping gene β-actin (23 cycles). Densitometry of bands in ethidium bromide-stained agarose gels was determined using Quantity One imaging analysis software (Bio-Rad).

Statistical analyses.

The summary data are presented as means ± SE unless indicated otherwise. Statistical significance was determined using Student's t-test (unpaired or paired) with a P value <0.05 being considered statistically significant.

RESULTS

Expression and stimulation of NFAT nuclear localization in UBSM.

We have previously reported that NFATc3 and NFATc4 are expressed in cerebral vascular and ileal SM (14, 41). In primary cultured UBSM myocytes, immunohistochemistry revealed the expression of NFATc1, NFATc3, and NFATc4 but not NFATc2 (Fig. 1). Similarly, RT-PCR analysis of fresh UBSM detrusor tissue indicated that NFATc1, NFATc3, and NFATc4 were present (Fig. 1, inset). The lack of NFATc2 expression in UBSM is consistent with previously published reports on the absence of NFATc2 expression in other SM tissues (14, 41).

Fig. 1.

Nuclear factor of activated T cell (NFAT) expression in murine urinary bladder smooth muscle (UBSM) cultured cells and tissue. Immunostaining of primary cultures of UBSM myocytes indicates that murine UBSM expresses NFATc1, NFATc3, and NFATc4 but not NFATc2. NFAT expression is shown in red, nuclear staining in green, and regions of NFAT nuclear localization in white. Some nuclei exhibit NFATc3 nuclear localization under these basal conditions. Inset: RNA extracted from UBSM tissues was probed for the presence of mRNA for NFATc1-c4 using RT-PCR (H, heart; B, bladder; T, thymus).

The SM mitogens PDGF and ET-1 are capable of stimulating NFATc3 nuclear translocation in ileal and cerebral vascular SM tissues (14, 41) and have been suggested to play a role in the physiology and pathophysiology of the urinary bladder (1, 31, 40). To determine whether these growth factors activate NFATc3 in UBSM, we treated primary cultured UBSM myocytes with either ET-1 (100 nM) or PDGF (20 ng/ml) for 30 min and quantified NFATc3 nuclear accumulation using confocal microscopy. Under nonstimulated, basal conditions, NFATc3 was found primarily in the cytosol. Upon stimulation with ET-1 (Fig. 2A) or PDGF (not shown), NFATc3 displayed prominent nuclear localization. As summarized in Fig. 2B, NFATc3 nuclear localization was increased 3.3 ± 0.6- and 3.4 ± 0.5-fold by exposure to ET-1 and PDGF, respectively (P < 0.05). ET-1- and PDGF-dependent nuclear translocation was blocked by preincubation with the calcineurin inhibitors FK506 and CsA (1 μM each). Thus NFAT is expressed in intact and cultured UBSM tissue, and nuclear translocation of NFATc3 can be induced by physiologically relevant agents.

Fig. 2.

Stimulation of NFATc3 nuclear translocation in UBSM. A: example micrographs showing the subcellular distribution pattern of NFATc3 under basal conditions and after stimulation with endothelin-1 (ET-1). B: treatment with either ET-1 or platelet-derived growth factor (PDGF) led to a significant increase in NFATc3 nuclear localization compared with basal conditions (*P < 0.05; c = number of cells analyzed). Treatment with FK506 and cyclosporine A (CsA, 1 μM; 30 min) before the addition of the PDGF or ET-1 blocked NFATc3 nuclear translocation.

UBSM strips from NFATc3-null mice exhibit enhanced nerve-evoked contractility but reduced responses to the BK channel blocker IBTX.

We have previously reported that the NFATc3 isoform is expressed in vascular and ileal SM (14, 41) and regulates the expression of the contractile protein SM α-actin (15). NFATc3 has also been implicated in the regulation of both BK and K_V channels in vascular smooth muscle (2, 34), suggesting that the NFATc3 isoform may be important in the regulation of SM function.

To determine whether NFATc3 modulates UBSM contractility, we compared the force generated by UBSM strips taken from NFATc3-null and wild-type mice in response to EFS using wire myography. A representative recording is presented in Fig. 3A, and the averaged force-frequency plot is presented in Fig. 3B. UBSM strips from NFATc3-null mice generated more force than strips from wild-type mice at all stimulation frequencies >5 Hz (Fig. 3B; n = 16 strips from 3 animals each; P < 0.05). At 20 Hz, the mean force generated by NFATc3-null UBSM strips was ∼1.8-fold that of strips from wild-type animals. The contractile responses of UBSM strips from wild-type and NFATc3-null animals to direct membrane potential depolarization with 60 mM K⁺ were not significantly different (10.6 ± 2.9 mN for wild-type and 8.5 ± 1.5 mN for NFATc3-null mice; P > 0.05; n = 13), suggesting that the L-VDCCs, which mediate depolarization-induced Ca²⁺ influx, are unchanged in NFATc3-null mice. The enhanced contractile response to EFS of UBSM strips taken from NFATc3-null mice suggests that the NFATc3 isoform contributes to the regulation of detrusor SM contractility, independent of L-VDCCs.

Fig. 3.

UBSM strips from NFATc3-null mice are more contractile than strips from wild-type mice and display a reduced responsiveness to IBTX. A: representative myograph recordings obtained from UBSM strips from wild-type and NFATc3-null mice subjected to EFS from 1 to 50 Hz. B: UBSM tissues strips from NFATc3-null mice exhibited a greater contractile response than UBSM strips from wild-type mice at all frequencies of stimulation >5 Hz (*P < 0.05; n = 3 animals and 16 strips each). C: in the presence of iberiotoxin (IBTX), there was no significant difference (P > 0.05) in the amplitude of electrical field stimulation (EFS)-induced contractions of UBSM strips from wild-type and NFATc3-null mice. Note that the results presented in A and B above are from a separate series of experiments from those in C.

BK channels play a central role in the repolarization of the UBSM action potential (20) and figure prominently in nerve-evoked contractility of UBSM (33, 43, 46). Blocking BK channels with IBTX increases the contractility of UBSM strips by shifting the force-frequency relationship to the left and by increasing the maximum force generated in response to muscarinic or purinergic stimulation or to EFS (20, 23, 37, 43, 46). To determine whether a decrease in BK channel activity might account for the increased contractility of bladder strips from NFATc3-null mice, we evaluated the effects of IBTX on the EFS-induced contractions of UBSM strips from NFATc3 -null and wild-type mice. UBSM tissue strips were stimulated at frequencies from 1 to 50 Hz before and after treatment with IBTX. The averaged force-frequency plot in the presence of IBTX is presented in Fig. 3C. IBTX increased the amplitude of EFS-induced contractions in UBSM strips from both wild-type and NFATc3-null mice. However, IBTX treatment had a greater effect on the contractile response of UBSM strips taken from wild-type mice than on strips taken from NFATc3-null mice. For example, at 5 Hz the contraction amplitude in the wild-type UBSM strips was increased 2.2 ± 0.07-fold by treatment with IBTX but was increased only 1.5 ± 0.02-fold in the NFATc3-null mice (n = 6; P < 0.05). The decreased responsiveness of NFATc3 null UBSM strips to IBTX treatment suggests that BK channel activity is downregulated in these animals, which may account for the increase contractility in UBSM strips from NFATc3-null mice. In support of this, the force-frequency relationships of the NFATc3-null and wild-type mice were not significantly different when BK channels were blocked with IBTX (Fig. 3C).

SK channel activity has also been previously demonstrated to play a significant role in the contractility of UBSM tissues (11, 17, 21, 24). When we exposed UBSM strips to the SK channel blocker apamin (320 nM), the force generated by EFS was increased at all frequencies, but there was no significant difference in the magnitude of the increase between strips from wild-type and NFATc3-null mice (data not shown). At 5 Hz stimulation, for example, apamin treatment led to a 1.7 ± 0.2-fold increase in force in the wild-type UBSM strips and a 1.6 ± 0.1-fold increase in the strips from NFATc3-null mice (n = 6; P > 0.05). We conclude from these experiments that the role of SK channels in the regulation of the force response to EFS is similar in wild-type and NFATc3-null mice.

I_BK is reduced in UBSM myocytes from NFATc3-null mice.

UBSM strips from NFATc3-null mice displayed a reduced responsiveness to IBTX, which may reflect a decrease in the activity or expression of BK channels in detrusor SM cells. To explore this possibility, we compared whole cell currents elicited by 10-mV voltage steps from −70 to +70 mV from UBSM cells obtained from NFATc3 null and wild-type mice. Currents were recorded before and after the application of IBTX to selectively block BK channels. An evaluation of the whole cell K⁺ currents (I_K) recorded before the application of IBTX revealed no significant difference in either end-pulse or tail currents between the myocytes isolated from wild-type and NFATc3-null mice. Representative recordings are presented in Fig. 4A and averaged current-voltage plots for the end-pulse and tail currents are presented in Fig. 4B.

Fig. 4.

UBSM myocytes from wild-type and NFATc3-null mice have similar whole cell K⁺ currents (I_K). A: representative whole cell current recordings elicited by 10-mV membrane potential steps in freshly isolated UBSM myocytes from wild-type and NFATc3-null mice. B: analyses of the resulting end-pulse and tail currents revealed no significant difference between the myocytes isolated from wild-type (▪) and NFATc3-null mice (○) (P > 0.05).

Although there was no significant difference in I_K, the IBTX-sensitive current (I_BK) was significantly reduced in UBSM cells from NFATc3-null mice (Fig. 5, A and B). As shown in Fig. 5C, I_BK was significantly lower in UBSM myocytes from NFATc3-null animals at all membrane potentials more positive than −20 mV (P < 0.05). At +40 mV, I_BK was 4.1 ± 0.8 pA/pF (n = 9) in wild-type myocytes but was only 1.2 ± 0.6 pA/pF (n = 10) in NFATc3-null myocytes, a ∼70% reduction in I_BK. Thus despite the similarity in I_K, I_BK was reduced in myocytes isolated from NFATc3-null mice. The reduction of I_BK in UBSM myocytes from NFATc3-null mice is consistent with the contractility data and supports the hypothesis that the enhanced contractile response of UBSM strips from NFATc3-null mice is due to a reduction in BK channel activity.

Fig. 5.

UBSM myocytes from NFATc3-null mice have reduced IBTX-sensitive current (I_BK). A and B: representative whole cell current recordings elicited by 10-mV membrane potential steps in freshly isolated UBSM myocytes from wild-type (A) and NFATc3-null mice (B) before and after addition of IBTX to block BK channels. C: UBSM myocytes obtained from NFATc3-null mice (○) displayed reduced IBTX-sensitive I_BK compared with myocytes obtained from control mice (▪) (*P < 0.05).

BK channel expression is downregulated in NFATc3-null mice.

It is possible that expression of either or both pore-forming α- and modulatory β₁-subunits are altered in the NFATc3-null mice. To address this, we used semiquantitative RT-PCR to determine the expression levels of BKα- and β₁-subunit mRNA (Fig. 6A). There was no significant difference in the expression of mRNA for the β₁-subunit between wild-type and NFATc3-null mice (P > 0.05; Fig. 6B). However, the expression of mRNA for the pore-forming α-subunit was ∼44% lower in UBSM from NFATc3-null mice than in wild-type mice (α-subunit: β-actin ratio of 1.78 ± 0.28 for wild-type mice vs. 0.98 ± 0.03 in NFATc3-null mice P < 0.05; n = 5 wild-type and 4 for NFATc3-null mice; Fig. 6C). Thus the reduction in I_BK in UBSM myocytes from NFATc3-null mice is supported at the molecular level by a decrease in the mRNA for the pore-forming α-subunit of the BK channel.

Fig. 6.

Reduced expression of the large conductance K⁺ channel (BK) α-subunit in UBSM from NFATc3-null mice. A: representative agarose gel of semiquantitative RT-PCR analysis of RNA isolated from wild-type and NFATc3-null mice probed for β-actin, BK α-subunit, and BK β-subunits. B: there was no difference in the expression of mRNA for the β₁-subunit (relative to β-actin) in UBSM from wild-type and NFATc3-null mice (n = 3 animals each). C: UBSM myocytes isolated from NFATc3-null mice expressed significantly lower levels of the pore-forming BK α-subunit (relative to β-actin levels) (*P < 0.05; n = 5 wild-type and 3 NFATc3-null mice).

A reduction in the cell membrane expression of the pore-forming α-subunit of the BK channel should not affect the channel P_o but should lead to a reduction in the channel density; that is, there should be fewer channels per patch. In contrast, a loss of β₁-subunit expression should dramatically lower the single channel P_o (6, 37). To definitively rule out the possibility that altered β₁-subunit expression contributes to the reduction in BK channel current, we determined the P_o of BK channels in wild-type and NFATc3-null UBSM myocytes by evaluating single BK channel currents measured in excised inside-out patches (Fig. 7A). All recordings were made in symmetrical 140 mM K⁺ with 10 μM intracellular Ca²⁺. Under these conditions, we have previously shown that loss of the β₁-subunit in UBSM decreased BK channel P_o at −40 mV and +40 mV 40-fold and 1.6-fold, respectively (37). There was no significant difference in P_o between wild-type and NFATc3-null myocytes at negative (−40 mV) or positive (+40 mV) membrane potentials (Fig. 7B; P > 0.05; n = 9). However, the BK channel density was significantly lower (Fig. 7C; P < 0.05; n = 9) in excised patches isolated from NFATc3-null myocytes (1.3 ± 0.7 channels per patch) than in wild-type UBSM myocytes (2.6 ± 1.3 channels per patch). These data are in accord with the whole cell BK current noted previously (Fig. 5C) and indicate that the number of functional channels is significantly reduced in UBSM from NFATc3-null mice.

Fig. 7.

Single channel BK recordings reveal a reduced number of BK channels per patch in NFATc3-null UBSM myocytes. A: representative single channel recordings from excised patches held at −40 and +40 mV. Arrows indicate closed state. B: there was no significant difference in open probability (P_o) of excised BK channels isolated from wild-type or NFATc3-null UBSM myocytes. C: number of channels per patch was significantly reduced in UBSM myocytes from NFATc3-null mice (P < 0.05; n = 9; error bars represent means ± SD).

UBSM myocytes isolated from NFATc3-null mice have increased K_V currents.

Our observation that I_K in wild-type and NFATc3-null mice was similar (Fig. 4B), despite a significant decrease in I_BK in the NFATc3-null mice (Fig. 5C), was initially puzzling. However, I_K in UBSM is composed of currents through multiple families of K⁺ channels (7, 20, 24, 27), so it was conceivable that upregulation of another class of K⁺ channels might have offset the reduction in I_BK in NFATc3-null mice. Whereas our molecular, electrophysiological, and functional data indicate that BK expression is NFATc3 dependent, there are reports that NFATc3 may also regulate K_V channel expression (2, 39). Thus we investigated the possibility that upregulation of K_V channel activity in UBSM myocytes from NFATc3-null mice might account for the observed similarities in I_K.

The wild-type murine UBSM K_V current is insensitive to block by 4-AP but is sensitive to block by TEA with an IC₅₀ of 5.2 mM (45). Thus, to determine whether the K_V component (I_KV) of I_K was altered in UBSM myocytes from NFATc3-null mice, we evaluated whole cell currents elicited by 10-mV membrane potential steps before and after application of 5 mM TEA (Fig. 8, A and B, for wild-type and NFATc3-null myocytes, respectively). All recordings were made in the presence of IBTX to block BK channels. As shown in the Fig. 8C, I_KV, defined as the current blocked by the addition of 5 mM TEA in the presence of IBTX, is significantly larger in the NFATc3-null UBSM myocytes at all membrane potentials more positive than −20 mV (P < 0.05; Fig. 4B). At +40 mV, I_KV was 5.40 ± 1.31 pA/pF in myocytes from NFATc3-null mice and only 2.18 ± 0.32 pA/pF in wild-type myocytes (n = 7). These results indicate that the reduction in I_BK in NFATc3 null myocytes is offset by an increase in I_KV, resulting in an I_K that is essentially the same in both NFATc3-null and wild-type UBSM myocytes.

Fig. 8.

UBSM myocytes from NFATc3-null mice have elevated K_V ccurrent (I_KV). A and B: representative whole cell current recordings elicited by 10-mV membrane potential steps from freshly isolated UBSM myocytes from wild-type (A) and NFATc3-null mice (B) before and after addition of tetraethylammonium chloride (TEA) to block voltage-gated (K_V) channels. All cells were also pretreated with IBTX to block BK channels. C: analysis of the averaged difference currents from wild-type (▪) and NFATc3-null mice (○) after treatment with 5 mM TEA indicates that UBSM myocytes from NFATc3- null mice displayed significantly elevated I_KV (*P < 0.05).

Biophysical and molecular evidence for increased K_V2.1 expression in NFATc3-null mice.

The simplest explanation for the observed increase in I_KV in NFATc3-null mice would be an upregulation of the native K_V channel complexes, without any change in subunit composition. However, it is also possible that the changes in K_V activity may be due to alterations in the subunit composition of the channels. To distinguish between these two possibilities, we compared some of the biophysical and pharmacological properties of I_KV in isolated myocytes from wild-type and NFATc3-null mice. There was no significant difference in I_KV steady-state activation curves obtained in myocytes isolated from NFATc3-null or wild-type mice (Fig. 9A). The slope factors (k) of the activation curves for wild-type (n = 5) and NFATc3-null mice (n = 7) were 16.3 ± 3.3 and 17.6 ± 4.1, respectively (P > 0.05). Similarly, a comparison of the activation time constants (Fig. 9B) also revealed no significant difference between UBSM myocytes from NFATc39-null and wild-type mice (P < 0.05). In addition, neither the wild-type nor the NFATc39-null UBSM tissues strips responded to treatment with 4-AP (data not shown), indicating that expression of 4-AP9-sensitive K_V family members (K_V1 and K_V4) was not induced in NFATc39-null UBSM myocytes. Thus the biophysical similarities and the lack of responsiveness to 4-AP suggest that the upregulation of I_KV in the NFATc39-null myocytes is likely due to increased expression of the native K_V2.1-K_V5.1/K_V6.1 channel complex rather than the expression of an alternate channel assemblage.

Fig. 9.

Molecular and electrophysiological evidence for elevated K_V2.1 expression in NFATc3-null mice. There was no significant difference in the steady-state activation curves (A) or activation time constants (B) between UBSM myocytes obtained from wild-type and NFATc3-null mice (P < 0.05). C: UBSM tissue from wild-type and NFATc3-null mice was probed for K_V2.1 and β-actin mRNA expression. D: comparison of the averaged ratios of K_V2.1:β-actin in UBSM tissues indicates that the UBSM from NFATc3-null mice contained significantly higher levels of mRNA for K_V2.1 than did UBSM tissues from wild-type mice (D) (*P < 0.05; n = 3 animals each).

To determine whether evidence for I_KV upregulation was supported at the molecular level, we performed semiquantitative RT-PCR on UBSM from wild-type and NFATc3-null mice to compare the levels of mRNA for K_V2.1, which has previously been reported to be regulated by NFATc3 (2, 39) (Fig. 9C). Consistent with the increase in I_KV, UBSM from NFATc3-null mice exhibited a 40% increase in the expression of mRNA for K_V2.1 relative to that from wild-type mice (P < 0.05; n = 3 animals each; Fig. 9D). Thus these data suggest that I_K in NFATc3-null and wild-type mice are similar despite the downregulation of I_BK in the NFATc3-null mice because the expression of K_V2.1 is increased.

DISCUSSION

In this study, we have found that NFATc3-dependent changes in ion channel expression play a role in the regulation of UBSM contractility. UBSM tissue strips from NFATc3-null mice generated more force in response to EFS compared with strips from wild-type mice. We attribute this enhanced contractility to a reduction in BK channel activity. This interpretation is supported at the molecular level by a reduction in BK α-subunit expression in the NFATc3-null UBSM myocytes and at the electrophysiological level by a significant reduction in I_BK and BK channel density. However, I_K was similar in both NFATc3-null and wild-type mice, an observation that can be explained by an upregulation of I_KV in the NFATc3-null mice that is supported at the molecular level by an increase in the expression of mRNA for the K_V2.1 subunit, a major component of the mouse UBSM K_V channel complex.

Our observation that nerve-evoked contractility in NFATc3-null mice was increased despite the presence of an enhanced I_KV suggests that, at least in response to EFS, I_BK downregulation has a more significant impact than I_KV upregulation. This is presumably due to the prominent role that the BK channel plays in the repolarization phase of the action potential (20, 22, 24), where a small decrease in BK channel activity can lead to a relatively large increase in force. Furthermore, the observation that the force-frequency relationships in UBSM strips from NFATc3-null and wild-type mice were identical in the presence of the BK channel blocker IBTX suggests that modulation of other channel types (e.g., K_V) does not significantly contribute to NFATc3-dependent differences in EFS-induced contractions. We also found no significant difference in the amplitude of the contractile response to K⁺ (60 mM)-mediated depolarization between UBSM muscle strips isolated from NFATc3-null and wild-type mice, suggesting that L-VDCC expression was unchanged.

Several families of K⁺ channels are expressed in UBSM, including K_V, BK, SK, and ATP-dependent K⁺ (K_ATP) channels. In the absence of a channel opener, K_ATP channels make little or no contribution to UBSM contractility (27) and, in any case, would not contribute to voltage-dependent K⁺ current since their activity is independent of voltage. In contrast, blocking BK channels has been demonstrated to cause very significant increases in nerve-evoked contractions (21, 33, 43, 46). The absence of a significant difference in I_K between NFATc3-null and wild-type mice seems to be explained by reciprocal changes in BK and K_V channel expression and activity that we have described. Quantitatively, our data would support this hypothesis given that, at +40 mV, the I_BK was reduced by ∼3 pA/pF, whereas the I_KV was elevated by approximately the same amount.

Our observation that I_KV and K_V2.1 channel expression are upregulated in UBSM from NFATc3-null mice is consistent with previous reports. In rats, sustained administration of ANG II leads to a reduction in I_KV and K_V2.1 protein levels in cerebral arteries (2). This downregulation is blocked by the calcineurin-NFAT inhibitor CsA and is also abrogated in NFATc3-null mice. Similarly, Santana and colleagues (39) found that I_KV and K_V2.1 expression levels are reduced in mouse cardiac muscle after surgically induced myocardial infarction (MI). In these studies, administration of CsA prevented the post-MI reduction in I_KV, and there was no post-MI decrease in NFATc3-null mice. Conversely, the expression of a constitutively active form of NFATc3 recapitulated these changes in I_KV in the absence of MI. Thus these previous reports strongly suggest that NFATc3 mediates the downregulation of K_V2.1 channel expression and activity, similar to what we have reported in the current study. However, we found that EFS-induced contractions were the same in UBSM strips from wild-type and NFATc3-null mice in the presence of the BK channel blocker IBTX, suggesting that a decrease in BK channel activity can explain the increase in EFS-induced contractions in NFATc3- null mice. Although the current study does not provide evidence for a functional consequence of K_V channel upregulation, it is conceivable that modulation of this channel type may play a role under other circumstances.

We have found evidence for the downregulation of basal BK channel expression and activity in UBSM from NFATc3-null mice at multiple levels, including a reduction in BK α-subunit expression and channel density, a reduction in IBTX-sensitive I_BK, and a reduced responsiveness of UBSM muscle strips to IBTX treatment. Recently, Santana's laboratory (34) reported that activation of the calcineurin-NFATc3 pathway in arterial SM by ANG II stimulation leads to a decrease in β₁-subunit expression with no change in the expression of the α-subunit. In contrast to our current work, where we found that the α-subunit, but not the β₁-subunit, is constitutively downregulated in UBSM of NFATc3-null mice, this previous study found that neither α- nor β₁-subunit expression was changed in the arteries of NFATc3-null mice under basal conditions. Although it is tempting to speculate that these different outcomes reflect tissue-specific differences in NFATc3 regulatory patterns, an alternative explanation might be that engagement of the ANG II signaling pathway modulates NFAT target specificity, perhaps through effects on the relative abundance of potential NFAT cofactors. Additional studies will be required to resolve the molecular questions raised by these results, but collectively they indicate that, directly or indirectly, NFATc3 is capable of regulating the expression of both the α- and β₁-subunits of the BK channel, and thus, may be an important regulator of BK channel activity.

Our semiquantitative PCR data indicated that the mRNA for the α-subunit of the BK channel was reduced ∼44%, a value that is similar to the decrease in single BK channel density (∼50%) at +40 mV. However, at +40 mV, the ∼70% decrease in I_BK was much greater than would be expected based on the PCR and single-channel data. It should be noted that, at +70 mV, the reduction in I_BK was ∼50%, which is more consistent with the PCR and single-channel data. The possible discrepancy between the magnitude of the NFATc3-dependent decrease in BK-α message and the magnitude of the resulting decrease in the IBTX-sensitive current suggests that mechanisms in addition to decreased mRNA expression may contribute to the total reduction in functional channels. In this context, Toro's laboratory (51) found that a splice variant of the α-subunit of the BK channel (SV1), when coexpressed with the human BK α-subunit and β₁-subunits in HEK293 cells, reduced surface expression of the BK channel complex by ∼80%. Accordingly, it is possible that, in addition to inducing a decrease in BK α-subunit expression, the absence of NFATc3 might result in the expression of an alternative splice variant of the α-subunit capable of inhibiting surface expression of the native BK channel. Sequence analysis of putative 5′ upstream regulatory regions of the K_V2.1 gene from mouse, human, and rat genomes reveals 15 potential NFAT consensus binding sites (GGAAA) within 3600 base pairs of the transcriptional initiation site. Analysis of the promoter regions for the human and mouse BK channel α-subunit reveals eight potential NFAT binding sites within 2600 base pairs of the initiation site, two of which are conserved in both the human and mouse genome. In addition, it has been reported that the BK β-subunit, which is also a potential target of NFAT, contains putative NFAT recognition sites (34). Thus the K_V2.1, BK α-subunit and BK β-subunits all represent potential targets of NFAT regulation. In theory, NFATc3 might directly regulate the expression of both K_V2.1 and BK α-subunit genes. However, this possibility seems unlikely since the opposing influences of these two transcriptional events create an outcome that is physiologically contradictory. More likely, NFATc3 is directly involved in regulating either K_V2.1 or the BK α-subunit, but not both. In this scenario, the change in the non-NFATc3-regulated gene might constitute an example of developmental compensation that would have the effect of maintaining membrane potential within a normal physiological range. Although it has been previously shown that NFATc3 may directly downregulate K_V2.1 (2, 39), our data does not definitively rule out the possibility that the changes in K_V2.1 expression are an indirect consequence of NFATc3 action on an alternative target. Further studies will employ additional experimental approaches to deconvolute the underlying molecular events. These include the use of promoter-reporter constructs and chromatin binding assays to establish functional significance of putative NFAT binding sites, and temporally controlled disruptions of NFATc3 using small interfering RNA/antisense or dominant negative approaches to determine immediate and secondary consequences of NFAT inhibition. In vitro experiments employing cultured UBSM might also be helpful in this regard, for example, by demonstrating that NFAT-activating stimuli (e.g., endothelin, PDGF) induce the predicted changes in K_V2.1 and BK expression. However, UBSM undergoes significant functional changes when cultured for as little as 24 h, so results obtained using cultured cells would need to be considered in context with additional, more physiologically relevant, experimental approaches.

In conclusion, our data indicate that NFATc3 plays a role in regulating the contractility of UBSM by directly or indirectly regulating BK channel expression and function. NFATc3-dependent changes in K_V channel activity, which might be expected to modulate UBSM contractility, were also observed, although these changes had no significant effect on the functional parameters measured in this study. UBSM strips from NFATc3-null mice exhibited greater nerve-evoked contractions than strips from wild-type mice, and this enhanced contractility was attributable to a decrease in the expression and activity of BK channels in UBSM cells. I_K in NFATc3-null and wild-type mice were statistically indistinguishable, an observation that is largely explained by a concomitant upregulation in I_KV and K_V2.1 channel expression. At present, it is unclear whether BK and K_V2.1 channels are direct targets of NFATc3 or whether the expression of one is indirectly regulated by reciprocal change in the other channel in a kind of compensatory event. Notwithstanding these unresolved questions, our studies strongly support a role for NFATc3 in the regulation of nerve-evoked contractions through modulation of BK channel activity.

GRANTS

This research was funded by the Totman Trust for Medical Research, the National Heart, Lung, and Blood Institute Training Program in the Molecular Basis of Cardiovascular Disease Grants T32 HL-007944 and HL-007647-19, and National Institutes of Health Grants DK-53832, DK-65947, HL-44455, and HL-77378.