Diphosphate Regulation of Adenosine Triphosphate Sensitive Potassium Channel in Human Bladder Smooth Muscle Cells

Diphosphate Regulation of Adenosine Triphosphate Sensitive Potassium Channel in Human Bladder Smooth Muscle Cells Shunichi Kajioka, Nouval Shahab, Haruhiko Asano, Hiromitsu Morita, Megumi Sugihara, Fumi Takahashi-Yanaga, Tatsuya Yoshihara, Shinsuke Nakayama,* Narihito Seki and Seiji Naito From the Departments of Urology (SK, NS, NS, SN) and Clinical Pharmacology (FTY, TY), Graduate School of Medical Sciences, Kyushu University and Special Patient Oral Care Unit, Kyushu University Hospital (HM, MS), Fukuoka, Department of Cell Science, Nagoya University Graduate School of Medicine (SN), Nagoya and Department of Biomedical Sciences, College of Life and Health Sciences, Chubu University (HA), Kasugai, Japan

Abbreviations and Acronyms ATP ⫽ adenosine triphosphate cADPR ⫽ cyclic adenosine diphosphate ribose GAPDH ⫽ glyceraldehyde-3phosphate dehydrogenase GDP ⫽ guanosine diphosphate KATP ⫽ ATP sensitive K⫹ channel NAADP ⫽ nicotinic acid adenine dinucleotide phosphate NAD ⫽ nicotinamide adenine dinucleotide PCR ⫽ polymerase chain reaction PSS ⫽ physiological saline solution RT ⫽ reverse transcriptase UDP ⫽ uridine diphosphate Submitted for publication October 11, 2010. Study received Kyushu University Hospital ethical committee approval. Supported by Grant-in-Aid for Special Purposes 20599012 from the Japan Society for the Promotion of Science and the Suzuki Foundation for Urological Medicine. Supplementary material for this article can be obtained at www.med.kyushu-u.ac.jp/uro/topics/ kajioka.html. * Correspondence: Department of Cell Physiology, Nagoya University Graduate School of Medicine, Nagoya 466-8550, Japan (telephone: ⫹81 92 744 2045; FAX: ⫹81 92 744 2048; e-mail: [email protected]).

736

www.jurology.com

Purpose: To clarify the properties of adenosine triphosphate sensitive K⫹ channel in human detrusor smooth muscle we examined the effect of the representative nicotinic acid derivatives ␤-nicotinamide adenine dinucleotide, cyclic adenosine diphosphate ribose and nicotinic acid adenine dinucleotide phosphate (Sigma-Aldrich®) on human detrusor adenosine triphosphate sensitive K⫹ channels. Materials and Methods: Patch clamp procedures were done in human detrusor cells. Reverse transcriptase and real-time polymerase chain reaction were performed to clarify the subunit components of adenosine triphosphate sensitive K⫹ channels. Results: The K⫹ channel opener levcromakalim induced a long lasting outward current that was inhibited by glibenclamide (Sigma-Aldrich) under the whole cell configuration. The single channel study revealed that the unitary conductance of the adenosine triphosphate sensitive K⫹ channel in the human detrusor was 11 pS and nucleotide diphosphates increased its open probability. Applying ␤-nicotinamide adenine dinucleotide also activated the adenosine triphosphate sensitive K⫹ channel but applying cyclic adenosine diphosphate ribose or nicotinic acid adenine dinucleotide phosphate had little effect on channel activation. Molecular studies indicated that Kir6.1 and SUR2B were the predominant components of the adenosine triphosphate sensitive K⫹ channel in the human detrusor. Conclusions: To our knowledge we report the first single channel study of the adenosine triphosphate sensitive K⫹ channel in the human detrusor. The properties of this channel, ie unitary conductance, adenosine triphosphate sensitivity and diphosphate activation, were consistent with those of other smooth muscle organs. ␤-Nicotinamide adenine dinucleotide has the potency to activate adenosine triphosphate sensitive K⫹ channels in the human detrusor. This channel likely has some role during ischemic conditions as well as physiological muscle motion leading to the activation of cell metabolism. Key Words: urinary bladder; muscle, smooth; potassium channels; adenosine triphosphate; nicotinamide arabinoside adenine dinucleotide STUDIES in the last 2 decades revealed that KATP channels are ubiquitous in many organs. Although many physio-

logical and pharmacological examinations have been done, it is still controversial whether KATP channels make

0022-5347/11/1862-0736/0 THE JOURNAL OF UROLOGY® © 2011 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION

Vol. 186, 736-744, August 2011 Printed in U.S.A. DOI:10.1016/j.juro.2011.03.153

AND

RESEARCH, INC.

DIPHOSPHATE REGULATION OF POTASSIUM CHANNEL IN BLADDER SMOOTH MUSCLE

a major contribution to resting membrane potential.1,2 However, in detrusor smooth muscle low KATP channel activation (less than 1%) decreased detrusor excitability and spontaneous phasic contractions, which are specific to detrusor and important for regulating urinary function.3 The KATP channel works as a critical metabolic sensor in processes such as hyper/hypoglycemia, ischemia and hypoxia.4,5 In the cardiovascular system the KATP channel has a protective role in metabolic stress and has become a possible therapeutic target.6 In nonvascular smooth muscle cells of the lower urinary tract the properties of the KATP channel have also been extensively investigated, eg in the guinea pig bladder,3,7,8 and in the pig urethra9 and bladder.10 This is because K⫹ channel openers, which definitely target the KATP channel, have great potential for overactive bladder treatment.11 To focus on the metabolic sensor, we used a nicotinic acid derivative such as ␤-NAD, which has a central role in the electron transfer system, has a diphosphate component in its structure. Some reports demonstrated the effect of nicotinic acids on the KATP channels of reconstituted pancreatic ␤ cells or cardiac cells.12–14 Thus, we were interested in investigating the effects of these nicotinic acids on bladder smooth muscle KATP channel activity. The current patch clamp study to investigate KATP channels in the human detrusor was done with 2 major objectives, including 1) to determine and clarify the properties of this channel, and 2) investigate the effect of nucleotide diphosphate derivatives, which are known to regulate cell metabolism and/or Ca2⫹ mobilization.

MATERIALS AND METHODS Tissue Specimens, Single Cell Preparation and Cell Culture Human tissue specimens were obtained from 5 males and 5 females with a mean ⫾ SD age of 70.0 ⫾ 7.4 years who underwent cystectomy due to bladder cancer. The average International Prostate Symptom Score of these patients was 7.5 ⫾ 2.4 (range 0 to 21). All patients provided informed written consent, as approved by the Kyushu University Hospital ethical committee. Tissue specimens were obtained from tumor-free parts of the bladder. Using a binocular microscope the muscle bundles were cut into pieces approximately 0.5 ⫻ 1 mm2 after the urothelial mucosa and connective tissue were removed. The pieces were digested in nominally Ca2⫹-free PSS containing 0.1% collagenase (type 3, Worthington) for 30 minutes at 37C. Cells were isolated by mechanical agitation using a fine bore pipette. Dispersed single cells were stored at 5C in PSS containing 0.5 mM Ca2⫹. These cells were also used to establish the cell culture protocol for human detrusor smooth muscle.

737

Patch Clamp Study Patch clamp studies were done at room temperature (21C to 23C), as previously described.15,16 All input voltage steps were generated by a patch clamp system using Instrutech ITC16 (HEKA, Bellmore, New York) and Macintosh® IIci or pClamp, version 6.04 (Axon Instruments®) with a TL-1 A/D-D/A board and IBM® compatible computer applied to the clamped cell through an EPC 7 patch clamp amplifier (List Electric, Darmstadt, Germany) or an Axopatch 1D patch clamp amplifier (Axon Instruments).

RT-PCR Examination Procedures were essentially the same as those previously used to detect KATP channel subunits in the pig detrusor.10 Human detrusor and cardiac total RNA (Clontech, Mountain View, California) was used. RNA was also extracted from cultured human detrusor smooth muscle cells. Using 1 ␮g RNA we analyzed the expression of Kir6.1, Kir6.2, SUR1, SUR2A, SUR2B, GAPDH and MYL6 mRNA by RT-PCR. Subsequently diluted cDNA transcripted from commercially purchased, cultured human detrusor RNA was subjected to TaqMan® real-time PCR using an ABI® 7500 Fast Real-Time PCR System according to manufacturer instructions. TaqMan probes used for the reactions were Hs01093761_m1 for SUR1, Hs01072316_m1 for SUR2A, Hs01074088_m1 for SUR2B and Hs02758991_g1 for GAPDH (ABI).

Solutions For whole cell recording the ionic composition of PSS used in the bath was 138 mM NaCl, 6 mM KCl, 2.4 mM CaCl2, 12 mM glucose and 5 mM HEPES. High K⫹ solution in the pipette was composed of 140 mM KCl, 0.1 mM MgCl2, 0.3 mM ethyleneglycoltetraacetic acid and 5 mM HEPES. Nominally Ca2⫹-free PSS was composed of 140.4 mM NaCl, 6 mM KCl, 12 mM glucose and 5 mM HEPES. The pH of these solutions was adjusted to 7.3 (25C) with tris base. For single channel recordings the pipette and bath solution were nominally Ca2⫹-free PSS with 100 ␮M levcromakalim and 100 nM charybdotoxin (Sigma-Aldrich), and high K⫹ solution, respectively.

Statistical Methods Numerical data are shown as the mean ⫾ SD with the number of cells or samples and the number of patients per experiment. Difference between means was evaluated using ANOVA. When a significant difference was identified by ANOVA in the same category of groups, it was evaluated by the Bonferroni-Dunn test with p ⬍0.05 considered statistically significant.

RESULTS Membrane Current in Human Detrusor To investigate levcromakalim induced current we used whole cell voltage clamp methods in isolated human detrusor smooth muscle. Ten ␮M levcromakalim induced a long lasting outward current to a mean of 58.7 ⫾ 20.2 pA at the holding potential of ⫺40 mV. This outward current was completely inhibited by 1 ␮M glibenclamide (fig. 1, A). Instantaneous currentvoltage relationships achieved using a ramp pulse pro-

738

DIPHOSPHATE REGULATION OF POTASSIUM CHANNEL IN BLADDER SMOOTH MUSCLE

Figure 1. Effect of levcromakalim and glibenclamide on membrane current of whole cell and patch configuration from human detrusor. A, upper trace indicates continuous recording of whole cell currents in response to 10 ␮M levcromakalim and application of 1 ␮M glibenclamide at ⫺40 mV holding potential. Continuous recordings show instantaneous current-voltage relationships obtained with ramp pulses of ⫺100 to 40 mV per indicated condition. Pipette and bath solutions were high K⫹ solution and PSS, respectively. B, in cell attached patch configuration single channel currents were induced by 100 ␮M levcromakalim and inhibited by 10 ␮M glibenclamide at 0 mV holding potential. C, 2 minutes after noting decrease in channel opening application of 1 mM GDP recovered channel opening. In following single channel experiments nominally Ca2⫹-free PSS with 100 ␮M levcromakalim and 100 nM charybdotoxin was used in pipette, and high K⫹ solution was used in bath at 0 mV holding potential unless otherwise noted. In all traces openings of maxi K⫹ channel are not indicated at full size. Arrows indicate time of inside-out patch configuration done from cell attached patch configuration. s, seconds.

tocol before and after applying 10 ␮M levcromakalim intersected at approximately ⫺80 mV (fig. 1, A). Human Detrusor Smooth Muscle Cells Single channel analysis. In the cell attached patch configuration 100 nM charybdotoxin was included in the patch pipette to decrease the open probability of the maxi K⫹ channel.17 Including 100 ␮M levcromakalim in the patch pipette gradually increased the opening of the 11 pS K⫹ channel with an amplitude of approximately 1 pA, as described, after the cell attached configuration was performed (fig. 1, B). Without levcromakalim in the pipette channel opening was hardly observed. Under the cell attached conditions described the mean open probability derived from the all points amplitude histogram was 0.24 ⫾ 0.13 in 5 preparations and 5 patients (fig. 1, B). Opening the 11 pS K⫹ channel was completely inhibited by 10 ␮M glibenclamide (fig. 1, B). GDP reactivated 11 pS Kⴙ channel. Burst-like openings of the 11 pS K⫹ channel were observed for

a mean of 44.1 ⫾ 7.7 seconds after the patch membrane was isolated. The mean open probability of this burst-like openings was 0.35 ⫾ 0.06 in 5 preparations and 5 patients. Opening of this channel gradually ceased even while 100 ␮M levcromakalim were present in the pipette. However, bath (intracellular) application of 1 mM GDP (Sigma-Aldrich) reactivated this channel (fig. 1, C). The mean open probability of the 11 pS K⫹ channel in the presence of 100 ␮M levcromakalim with 1 mM GDP was 0.38 ⫾ 0.05 in 5 preparations and 3 patients. After the patch membrane was isolated applying GDP alone hardly caused the 11 pS K⫹ channel to open without inclusion of levcromakalim. Thus, levcromakalim was included in the pipette for all single channel recordings. Unitary Conductance in Human Detrusor Figure 2, A shows representative traces obtained at various holding potentials in the inside-out patch configuration. Figure 2, A also shows the current-

DIPHOSPHATE REGULATION OF POTASSIUM CHANNEL IN BLADDER SMOOTH MUSCLE

739

Figure 2. Single channel conductance and ATP sensitivity of levcromakalim and GDP activated channel in inside-out patch configuration. A, single channel currents were recorded from inside-out patch membrane in presence of 1 mM GDP. s, seconds. Note representative traces at ⫺20, 0 and 20 mV holding potential (a) and current-voltage relationships of levcromakalim and GDP activated channel (b). Symbols indicate mean amplitude of channel openings with SD of 13 patch membranes and lines drawn by least squares method. B, inhibitory action of intracellular ATP on GDP activated channel at 0 mV holding potential. ATP concentrations 1, 0.3 and 0.1 mM were applied when GDP activated channel opening was noted in same patch membrane (a). Graph shows ATP inhibitory effect on open probability of GDP activated channel (b). Graphs were obtained from 2-minute recordings. Same analysis was done in results of following figures. Columns represent mean ⫾ SD of open probability in 3 to 5 preparations and 3 patients. Asterisk indicates Bonferroni-Dunn test p ⬍0.05.

voltage relationship of the GDP activated channel. Unitary conductance was 11.3 pS in the inside-out patch configuration in 13 preparations and 6 patients but 11.1 pS in the cell attached patch configuration in 6 preparations and 4 patients (data not shown). 11 pS Kⴙ Channel ATP Sensitivity Figure 2, B shows representative traces obtained from the same inside-out patch configuration. After activation the 11 pS K⫹ channel was observed after applying 1 mM GDP. Additional application of 1 mM ATP (Sigma-Aldrich) (free ATP4⫺ was estimated to be 0.93 mM in the presence of 0.1 mM Mg2⫹) greatly decreased 11 pS K⫹ channel opening. Decreasing the ATP concentration to 0.3 and 0.1 mM (free ATP4⫺ was estimated to be 241 ␮M and 77.1 ␮M, respectively, in the presence of 0.1 mM Mg2⫹) obviously increased 11 pS K⫹ channel opening in a concentration dependent manner. After removing ATP channel openings were recovered in the presence of

1 mM GDP. Figure 2, B also shows the significant inhibitory effect of ATP on GDP induced openings of the 11 pS K⫹ channel. The mean open probability was 0.010 ⫾ 0.004 for control, 0.38 ⫾ 0.05 for 1 mM GDP, 0.18 ⫾ 0.04 for 0.1 mM ATP with 1 mM GDP and 0.10 ⫾ 0.03 for 0.3 mM ATP with 1 mM GDP in 3 to 5 preparations and 3 patients. Diphosphate Molecules and KATP Channel It is generally accepted that the ribonucleotides adenosine diphosphate, UDP (Sigma-Aldrich) and GDP, which have a diphosphate structure, are essential to maintain KATP channel activation, especially in vascular smooth muscle.17,18 Nicotinic acid derivatives, which have a main role in cell metabolism, also have diphosphate moieties in their structure. Thus, we investigated the effects of representative derivatives on channel activity. ␤-NAD, cADPR and NAADP were selected since ␤-NAD has a central role in cell metabolism, and its derivatives

740

DIPHOSPHATE REGULATION OF POTASSIUM CHANNEL IN BLADDER SMOOTH MUSCLE

cADPR and NAADP are widespread, potent calcium mobilizing messengers.19,20 Figure 3, A shows representative traces of 1 mM UDP, ␤-NAD, cADPR and NAADP applied to the bath solution in the inside-out patch configuration. One mM ␤-NAD activated the KATP channel in human detrusor smooth muscle cells (fig. 3, A). However, neither 1 mM cADPR nor 1 mM NAADP induced KATP channel opening (fig. 3, A). Figure 3, B shows the effect of the various diphosphate structures on the open probability of the KATP channel. The order of potency to activate the KATP channel was UDP ⬎GDP ⬎␤-NAD ⬎⬎cADPR ⫽ NAADP. The mean open probability induced by 1 mM UDP, GDP and ␤-NAD was 0.35 ⫾ 0.13, 0.28 ⫾ 0.05 and 0.18 ⫾ 0.05, respectively, in 4 to 7 preparations and 2 to 5 patients. On the other hand, the mean open probability induced by 1 mM cADPR and NAADP was significantly subtle at 0.023 ⫾ 0.007 and 0.024 ⫾ 0.016, respectively, in 4 preparations and 3 patients. We also investigated the effects of cADPR and NAADP on KATP channel opening. After we identified KATP channel activation by 1 mM GDP,

cADPR and NAADP did not detectably modulate channel opening in 3 preparations and 3 patients (fig. 4, A). To confirm that the ␤-NAD activated channel was the KATP channel in the human detrusor we also investigated the effect of ATP on ␤-NAD activated channel opening. Intracellular application of ATP decreased the open probability of ␤-NAD activated channels in a concentration dependent manner (fig. 4, B). The mean open probability of the ␤-NAD activated channels was 0.26 ⫾ 0.10 in the absence of ATP, and 0.15 ⫾ 0.03 and 0.003 ⫾ 0.001 in the presence of 0.1 and 1 mM ATP, respectively, in 3 preparations and 3 patients (fig. 4, B). Molecular Study To evaluate differences in the expression level of KATP channel subunits in the human detrusor we performed RT-PCR and real-time PCR. Specific primers for Kir6.1, Kir6.2, SUR1#1, SUR1#2, SUR2A and SUR2B for RT-PCR were designed to produce amplicons of 427, 286, 357, 174, 375 and 285 bp, respectively (see table). The reaction condition was 3 minutes at 95C and then 35 cycles at 95C for 35

Figure 3. Effects of nicotinic acids on KATP channel activity, that is GDP activated channels, in human detrusor. A, single channel currents were recorded from inside-out patch configuration at 0 mV holding potential. Representative traces were recorded under various conditions. One mM UDP (a) and 1 mM ␤-NAD (b) were applied to bath solution. KATP channels in each patch membrane and no channel run down phenomenon were confirmed before and after 1 mM GDP application. One mM cADPR (c) and 1 mM NAADP (d) were applied in absence of GDP. Horizontal lines indicate nucleotide application periods. B, effects of 1 mM UDP, GDP, ␤-NAD, NAADP and cADPR on open probability (NPo) of each 2-minute recording, defined as number of channels ⫻ mean open probability of single channels. Bars represent mean ⫾ SD open probability in 4 to 7 preparations and 2 to 5 patients. Asterisk indicates Bonferroni-Dunn test p ⬍0.05.

DIPHOSPHATE REGULATION OF POTASSIUM CHANNEL IN BLADDER SMOOTH MUSCLE

741

Figure 4. No inhibitory effect of nicotinic acids on the KATP channel in human detrusor. A, single channel currents were recorded from inside-out patch configuration at 0 mV holding potential. Effects of cADPR (a) and NAADP (b) on KATP channel opening elicited by 1 mM GDP. s, seconds. B, effects of intracellular ATP on KATP channel, that is ␤-NAD activated channel (a). To same patch membrane 0.1 (b) and 1 (c) mM ATP were cumulatively applied. Note 1 mM ␤–NAD wash (d). Graph shows inhibitory effect of ATP on ␤-NAD activated channel open probability (e). Columns represent mean ⫾ SD in 3 preparations and 3 patients. Asterisk indicates BonferroniDunn test p ⬍0.05.

seconds, 58C for 35 seconds and 72C for 40 seconds for Kir6.1; 95C for 3 minutes and then 35 cycles at 95C for 35 seconds, 61C for 35 seconds and 72C for 40 seconds for Kir6.2; 95C for 3 minutes and then 3 cycles at 51C for 40 seconds and 72C for 40 seconds, followed by 33 cycles at 95C for

35 seconds, 56C for 35 seconds and 72C for 40 seconds for SUR2A; and 95C for 3 minutes, and then 3 cycles at 49C for 40 seconds and 72C for 40 seconds, followed by 33 cycles at 95C for 35 seconds, 54C for 35 seconds and 72C for 40 seconds for SUR2B.

PCR primers for human detrusor sample were designed using conserved sequences between humans and mice Clone SUR1: Sense Antisense SUR2A: Sense Antisense SUR2B: Sense Antisense Kir6.1: Sense Antisense Kir6.2: Sense Antisense

Primer Sequence

Primer Site

Product Length (bp)

Accession No.

174 5=- ACCTGGTGATTGTCCTGAAG - 3= 5=- AGTTAGAAAACCCAGGCAGG - 3=

Pig Pig

(p)CD572197 (p)CD572197 376

5=- CAGCTGAAGAATATGGTCAAATC - 3= 5=- CTACTTGTTGGTCATCACCAAAG - 3=

Human ⫽ mouse Human ⫽ mouse

5=- CAGCTGAAGAATATGGTCAAATC - 3= 5=- TTCATCACAATAACCAGGTCTGC - 3=

Human ⫽ mouse Human ⫽ mouse

(h)NM_005691 (m)D86037 285 (h)NM_020297 (m)D86038 403

5=- CCTGTTGATAACCCGCTTGAG - 3= 5=- GCTTCCTCAGAGAGTTCTGGT - 3=

Pig Pig

(p)CF181244 (p)CF181244 386

5=- GTGTTCACCACGCTGGTGGAC - 3= 5=- GCATGCTTGCTGAAGATGAGGGT - 3=

Human ⫽ mouse Human ⫽ mouse

(h)NM_000525 (m)NM_010602

DIPHOSPHATE REGULATION OF POTASSIUM CHANNEL IN BLADDER SMOOTH MUSCLE

Figure 5, A shows results suggesting that Kir6.1 and SUR2B were the predominant components. Two pairs of primers did not detect SUR1 under any PCR conditions in commercially purchased human detrusor RNA. To reinforce SUR2B expression we performed real-time PCR using cDNA samples prepared from commercially purchased and cultured human detrusor RNA. Expression levels of SUR1, SUR2A and SUR2B were determined shown relative to those of a commercially available human cardiac sample since all SUR subunits are expressed in the heart.21 Only SUR2B was detected in each detrusor samples (fig. 5, B). To verify the cell type we examined GAPDH and the smooth muscle positive marker MYL6 (fig. 5, C).

channel have been investigated in the pig and the guinea pig, the human detrusor is comparable to the pig detrusor. 1) Conductance of the human detrusor KATP channel was 11 pS (5 mMout/140 mMin K⫹), virtually identical to that of the pig detrusor at 12 pS (5 mMout/140 mMin K⫹).10 2) In terms of intracellular signaling of the KATP channel the dependence on diphosphates and sensitivity to ATP (50% inhibitory concentration approximately 100 ␮M) were similar to those of the pig detrusor (fig. 2) while GDP did not reactivate the KATP channel and a higher concentration of ATP was required to inhibit KATP channel openings in the guinea pig detrusor.7,10 Generally KATP channels consist of a K⫹ channel pore, Kir 6.X with sulfonylurea receptors, SUR X.22 The previous immunohistochemical study demonstrated that SUR1 might be expressed as well as SUR2B in the human detrusor.23 However, our RTPCR and real-time PCR studies did not detect SUR1 at the mRNA level. Thus, we conclude that Kir6.1 and SUR2B are the predominant components in human detrusor smooth muscle cells, well in line with previous reports of smooth muscle cells (fig. 5).24

DISCUSSION KATP Channels In lower urinary tract. To our knowledge we report the first study of KATP channel in freshly isolated human smooth muscle cells. While the physiological and pharmacological properties of the detrusor KATP

SUR1 #2

SUR2A

Kir6.1

Kir6.2

SUR2B

commercial human detrusor

A

SUR1 #1

742

cultured

commercial

100 bp marker

cultured human detrusor

cultured

commercial human detrusor

commercial

C

B

100 bp marker

700 500 400 300 200 100

0.15 0.1

500

SUR1

N.D.

N.D.

N.D.

0

trace

200

0.05

SUR2A

SUR2B

MYL6

GAPDH

Figure 5. KATP channel subunit mRNA expression. RNA samples commercially purchased and extracted from cultured human detrusor cells were reverse transcribed to cDNA using Anchored-oligo[dt]18 primer. A, amplified PCR fragments were run on 2% agarose gels using these samples. B, SUR1, SUR2A and SUR2B expression was determined by quantitative real-time PCR with values normalized to human cardiac RNA expression as 1.0. C, GAPDH and smooth muscle specific marker MYL6 were also examined in each sample.

DIPHOSPHATE REGULATION OF POTASSIUM CHANNEL IN BLADDER SMOOTH MUSCLE

Effect of pyridine nucleotides on human detrusor. We investigated 3 representative nicotinic acid derivatives, including ␤-NAD, cADPR and NAADP, on KATP channels in the human detrusor, as mentioned. Earlier studies showed that NAD blocks native KATP channels above 500 ␮M but stimulates channel activity at lower concentrations in pancreatic ␤ cells.13,14 The plausible pancreatic ␤-cell type of reconstructive KATP channels (Kir6.2/SUR1) is inhibited by NAD.12 In rat ventricular myocytes ADP ribose but not cyclic ADP ribose inhibits the native cardiac type of KATP channels. There may be much variety among the effects of nicotinic acid derivatives on KATP channel activity since ADP activates the KATP channel in smooth muscle cells9,18 but is a potent inhibitor of this channel in pancreatic ␤ cells.14,25 The fact that ␤-NAD but neither cADPR nor NAADP activated KATP channels in the human detrusor is favorable to assist Ca2⫹ mobilization. Since the estimation of the total cellular NAD is commonly below the mM range, NAD potentially affects KATP channel activation in the physiological range.12,26 Role and Therapeutic Potential Our finding of NAD induced KATP channel activation suggests another novel possibility of KATP channel function since NAD biosynthetic pathways are essential in mammalian cell systems. It is generally accepted that the KATP channel is the target of K⫹ channel openers, which were first tried as novel antihypertensive agents. These agents also have potential to treat other disorders due to the ubiquity of KATP channels. Above all overactive bladder is a disorder that could be plausibly treated with K⫹ channel openers. Low levels of KATP channel activation (less than 1%) decrease guinea pig bladder excitability and contractility.3 Accordingly the devel-

743

opment of bladder sensitive K⫹ channel openers has focused on newer agents such as ZD094722 and A-251179,27 which are more selective for the bladder and have been under investigation. Another K⫹ channel opener, nicorandil, was effective in human therapeutic trials and in in vivo studies.28,29 Thus, the specificity of a KATP channel opener for the bladder should be strengthened for overactive bladder treatment. Notably all of our specimens were obtained from geriatric detrusor with an average donor patient age of 70 years. Development and aging might modulate KATP channel distribution and characteristics. However, as a safer example, our study provides better understanding of this channel from the clinical standpoint that urinary dysfunction develops with aging.

CONCLUSIONS In the human detrusor KATP channels with a conductance of 11pS consisted of predominantly Kir6.1 and SUR2B. These KATP channels were activated by levcromakalim and diphosphates, and inhibited by ATP and glibenclamide. We found ␤-NAD activated KATP channels in the human detrusor, suggesting that the channels act as a metabolic sensor due to ATP production and nicotinic acid mobilization. To our knowledge we report the first single channel analysis of metabolic modulation of KATP channel activity in the human detrusor.

ACKNOWLEDGMENTS Prof. Arthur Weston provided levcromakalim. Dr. Tom Cunnane and Prof. Alison F. Brading, University of Oxford, provided discussion. Noriko Hakota and Seiko Kamori assisted with experiments.

REFERENCES 1. Fry CH, Wu C and Sui GP: Electrophysiological properties of the bladder. Int Urogynecol J Pelvic Floor Dysfunct 1998; 73: C110. 2. Quayle JM, Nelson MT and Standen NB: ATPsensitive and inwardly rectifying potassium channels in smooth muscle. Physiol Rev 1997; 77: 1165. 3. Petkov GV, Heppner TJ, Bonev AD et al: Low levels of K(ATP) channel activation decrease excitability and contractility of urinary bladder. Am J Physiol Regul Integr Comp Physiol 2001; 280: R1427. 4. Minami K, Miki T, Kadowaki T et al: Roles of ATP-sensitive K⫹ channels as metabolic sensors: studies of Kir6.x null mice. Diabetes 2004; 53: S176.

5. Quayle JM, Turner MR, Burrell HE et al: Effects of hypoxia, anoxia, and metabolic inhibitors on KATP channels in rat femoral artery myocytes. Am J Physiol Heart Circ Physiol 2006; 291: H71. 6. Fujita A and Kurachi Y: Molecular aspects of ATP-sensitive K⫹ channels in the cardiovascular system and K⫹ channel openers. Pharmacol Ther 2000; 85: 39. 7. Bonev AD and Nelson MT: ATP-sensitive potassium channels in smooth muscle cells from guinea pig urinary bladder. Am J Physiol 1993; 264: C1190. 8. Gopalakrishnan M, Whiteaker KL, Molinari EJ et al: Characterization of the ATP-sensitive potassium channels (KATP) expressed in guinea pig bladder smooth muscle cells. J Pharmacol Exp Ther 1999; 289: 551.

9. Teramoto N, McMurray G and Brading AF: Effects of levcromakalim and nucleoside diphosphates on glibenclamide-sensitive K⫹ channels in pig urethral myocytes. Br J Pharmacol 1997; 120: 1229. 10. Kajioka S, Nakayama S, Asano H et al: Levcromakalim and MgGDP activate small conductance ATP-sensitive K⫹ channels of K⫹ channel pore 6.1/sulfonylurea receptor 2A in pig detrusor smooth muscle cells: uncoupling of cAMP signal pathways. J Pharmacol Exp Ther 2008; 327: 114. 11. Andersson KE: Clinical pharmacology of potassium channel openers. Pharmacol Toxicol 1992; 70: 244. 12. Dabrowski M, Trapp S and Ashcroft FM: Pyridine nucleotide regulation of the KATP channel Kir6.2/

744

DIPHOSPHATE REGULATION OF POTASSIUM CHANNEL IN BLADDER SMOOTH MUSCLE

SUR1 expressed in Xenopus oocytes. J Physiol 2003; 550: 357.

inhibition by glibenclamide in vascular smooth muscle cells. Br J Pharmacol 1993; 110: 573.

13. Harding EA, Kane C, James RF et al: Modulation of three types of potassium selective channels by NAD and other pyridine nucleotides in human pancreatic beta-cells. NAD and K⫹ channels in human beta-cells. Adv Exp Med Biol 1997; 426: 43.

19. Galione A, Lee HC and Busa WB: Ca2⫹-induced Ca2⫹ release in sea urchin egg homogenates: modulation by cyclic ADP-ribose. Science 1991; 253: 1143.

14. Dunne MJ, Findlay I and Petersen OH: Effects of pyridine nucleotides on the gating of ATP-sensitive potassium channels in insulin-secreting cells. J Membr Biol 1988; 102: 205. 15. Kajioka S, Nakayama S, McMurray G et al: Ca2⫹ channel properties in smooth muscle cells of the urinary bladder from pig and human. Eur J Pharmacol 2002; 443: 19. 16. Morita H, Honda A, Inoue R et al: Membrane stretch-induced activation of a TRPM4-like nonselective cation channel in cerebral artery myocytes. J Pharmacol Sci 2007; 103: 417. 17. Kajioka S, Kitamura K and Kuriyama H: Guanosine diphosphte activates an adenosine 5=-triphosphate-sensitive K⫹ channel in the rabbit portal vein. J Physiol 1991; 444: 397. 18. Beech DJ, Zhang H, Nakao K et al: K channel activation by nucleotide diphosphates and its

20. Cancela JM, Churchill GC and Galione A: Coordination of agonist-induced Ca2⫹-signalling patterns by NAADP in pancreatic acinar cells. Nature 1999; 398: 74. 21. Kuniyasu A, Kaneko K and Kawahara K: Molecular assembly and subcellular distribution of ATPsensitive potassium channel proteins in rat hearts. FEBS Lett 2003; 552: 259. 22. Isomoto S, Kondo C, Yamada M et al: A novel sulfonylurea receptor forms with BIR (Kir6.2) a smooth muscle type ATP-sensitive K⫹ channel. J Biol Chem 1996; 271: 24321. 23. Aishima M, Tomoda T, Yunoki T et al: Actions of ZD0947, a novel ATP-sensitive K⫹ channel opener, on membrane currents in human detrusor myocytes. Br J Pharmacol 2006; 149: 542. 24. Inagaki N, Gonoi T, Clement JP 4th et al: Reconstitution of IKATP: an inward rectifier subunit plus the sulfonylurea receptor. Science 1995; 270: 1166.

25. Tung RT and Kurachi Y: On the mechanism of nucleotide diphosphate activation of the ATPsensitive K⫹ channel in ventricular cell of guineapig. J Physiol 1991; 437: 239. 26. Pollak N, Dölle C and Ziegler M: The power to reduce: pyridine nucleotides—small molecules with a multitude of functions. Biochem J 2007; 402: 205. 27. Shieh CC, Brune ME, Bucker SA et al: Characteriztion of a novel ATP-sensitive K⫹ channel opener, A-251179, on urinary bladder relaxation and cystometric parameters Br J Pharmacol 2007; 151: 467. 28. Kamiyama Y, Muto S, Nishio K et al: Nicorandil, a KATP channel opener and nitric oxide donor, was effective in the treatment of urinary frequency—a randomised clinical trial. Presented at the 36th Annual Meeting of the International Continence Society, Christchurch, New Zealand, November 27–December 1, 2006. 29. Kamiyama Y, Muto S, Masuda H et al: Inhibitory effects of nicorandil, a KATP channel opener and a nitric oxide donor, on overactive bladder in animal models. BJU Int 2008; 101: 360.