cGMP does not activate two-pore domain K⁺ channels in cerebrovascular smooth muscle

Abstract

Two-pore domain K⁺ (K_2P) channels are a new channel family. The goal of this study was to determine if K_2P channels are activated by the nitric oxide (NO)/cGMP/PKG pathway in vascular smooth muscle. Relative levels of message for K_2P channels were assessed in rat middle cerebral arteries (MCAs) using quantitative RT-PCR, and K⁺ currents were measured in freshly dispersed vascular smooth muscle cells of the MCA. The rat MCA expresses a number of K_2P channels. Message for TREK-1 was the most abundant K_2P channel, followed by TASK-1 and TWIK-2, which were expressed at ∼10% of the level of TREK-1. Message for other K_2P channels was 1% or less than that of TREK-1. A number of K_2P channels, including TREK-1, TWIK-2, and TASK-1, have putative PKG phosphorylation sites in the intracellular domains. The NO donor sodium nitroprusside (100 μM) or the membrane permeable analog of cGMP 8-bromo-cGMP (10 μM) elicited transient increases in whole cell current of vascular smooth muscle from the rat MCA. However, after large-conductance Ca²⁺-activated K⁺ channels had been blocked with 10 mM tetraethylammonium (TEA), no increase in whole cell current was observed. Since K_2P channels are resistant to the blocking effects of TEA, we conclude that K_2P channels in vascular smooth muscle were not activated by the NO/cGMP/PKG pathway. Although K_2P channels are highly expressed, K_2P currents are not activated via the NO/cGMP pathway in rat MCA smooth muscle, despite the presence of numerous putative PKG phosphorylation sites.

relaxation of vascular smooth muscle by nitric oxide (NO) involves the activation of soluble guanylyl cyclase, increases in cGMP, and activation of PKG. PKG, a serine/threonine-specific kinase, relaxes smooth muscle by two parallel mechanisms: 1) activation of K⁺ channels and 2) inhibition of Ca²⁺-dependent enzymes that promote smooth muscle contraction (8, 36). The activation of K⁺ channels decreases intracellular Ca²⁺_, the second messenger for contraction, by hyperpolarizing the plasma membrane and closing voltage-dependent L-type Ca²⁺ channels. In vascular smooth muscle, PKG can activate at least three types of K⁺ channels, including large-conductance Ca²⁺-activated K⁺ (BK_Ca) channels, voltage-dependent K⁺ (K_v) channels, and ATP-sensitive K⁺ (K_ATP) channels (8, 17, 21, 22, 27, 28, 28). Although the activation of K⁺ channels by NO through mechanisms other than PKG have been reported (2, 38), they appear to be less important in the relaxation of vascular smooth muscle.

In the mid-1990s, a new family of K⁺ channels was discovered while searching for genes that code for the highly conserved pore domain, the region where K⁺ passes through the channel (18, 25, 34). The family has been given the name two-pore domain K⁺ (K_2P) channels based on the number of pore domains for each α-subunit of the protein. Presently, there are 15 genes coding for members of this family (13). K_2P channels are abundantly expressed in arteries and have a role in the regulation of the contractile state of vascular smooth muscle (1, 3, 4, 11, 12, 14–16, 29).

Several members of the K_2P family, including TREK-1, TWIK-2, and possibly TALK-1 and TALK-2, are activated by the NO/cGMP/PKG system in heterologous expression systems (5, 6, 19, 30). In gastrointestinal smooth muscle, the NO/cGMP/PKG pathway activates a native channel that strongly resembles TREK-1 (20, 35). It is not presently known if the NO/cGMP/PKG system can activate K_2P channels in vascular smooth muscle. Given that a number of K_2P channels, including TREK-1 and TWIK-2, which are activated by the NO/cGMP/PKG system, are expressed in arteries of the brain and periphery (1, 3, 4, 11, 15, 16, 29), we tested the hypothesis that K_2P channels in smooth muscle cells from the rat middle cerebral artery (MCA) can be activated by the NO/cGMP/PKG system. First, we determined the relative expression of K_2P channels, as determined by quantitative PCR, in the rat MCA. Second, we determined which members of the K_2P family contain putative PKG phosphorylation sites. Finally, we determined if NO or cGMP activated K⁺ currents after blockade of classical K⁺ channels. In this study, we present evidence that members of the K_2P family of K⁺ channels, while expressed in rat MCA smooth muscle and containing multiple PKG consensus sequences, are not sensitive to NO/cGMP stimulation and, therefore, are not likely to play a role in NO-mediated regulation of the diameter of the rat MCA.

METHODS

All experiments were approved by the Animal Protocol Review Committee of the Baylor College of Medicine. Male Long-Evans rats were anesthetized with 3% isoflurane and decapitated. Brains were removed from the skull and placed in chilled Krebs solution. MCAs were carefully taken from the brain surface and used for real time RT-PCR or electrophysiological experiments.

Gene expression.

RNA was isolated from MCAs pooled from three rats, treated with DNAse (Invitrogen), and assessed for purity by spectrophotometry with aborbance at 260/280 nm > 1.9. Total RNA (0.2 μg) was reverse transcribed using random hexamers (Invitrogen). Primers for 10 of the K_2P channels were designed using Primer Express 2.0 software and rat K_2P channel sequences. The primers were as follows: TWIK-1 (KNCK1), forward 5′-AACCTTCTGTGAGCTCCACGA-3′ and reverse 5′-AGGACAGTTGGTCATGCTCCA-3′; TREK-1 (KCNK2), forward 5′-TTGCCAAAGTGGAGGACACAT-3′ and reverse 5′-CGAAGAGGACACAGCCAAACA-3′; TASK-1 (KNCK3), forward 5′-TGTCCATGGCCAACATGGT-3′ and reverse 5′-GGAAGAAAGTCCAGCGCTCAT-3′; TRAAK (KCNK4), forward 5′-TTATGTACCAGGCGATGGCAC-3′ and reverse 5′-TGGTGAACACTGAGGCGAAGT-3′; TWIK-2 (KCNK6), forward 5′-AACAGGCAAGGAACTGACCCA-3′ and reverse 5′-CACGGCTATCAATGCCCAGTA-3′; TASK-3 (KCNK9), forward 5′-CCTTGATCGCCTGTACCTTCA-3′ and reverse 5′-GGCGGACCTCTTCTGCTTTAA-3′; TREK-2 (KCNK10), forward 5′-CTGATCCAGCACGCACTTGAT-3′ and reverse 5′-CCCAGCAAAGAAGAAGGCACT-3′; THIK-2 (KCNK12), forward 5′-TCTGCTTCGTCACCTTCAGCAC-3′ and reverse 5′-GGATGAAAAGGAAGTTGCCCAG-3′; THIK-1 (KCNK13), forward 5′-AACCTTCTGTGAGCTCCACGA-3′ and reverse 5′-AGGACAGTTGGTCATGCTCCA-3′; and TASK-5 (KCNK15), forward 5′-GGATTCTCATCACTTTCCCGG-3′ and reverse 5′-TGCATCAGGAAACTCTGCCAC-3′.

The efficiency of each primer set was assessed using high-quality cDNA from the rat brain and was determined to be >95%. Real-time PCR was performed with SYBR green (Invitrogen) using an Applied Biosystems 7000 Sequence Detection System. The relative quantitation of each K_2P channels was performed using the Pfaffl method (31) and normalized to GAPDH (10).

Protein motif analysis.

Each primary amino acid sequence was individually analyzed for the presence of high-stringency putative PKG phosphorylation sites: (R/K)_2-3X_1-2(S/T), where (R/K)_2-3 is either arginine (R) or lysine (K) with two or three repeats followed by any amino acid (X) with one or two repeats followed by serine (S) or threonine (T). Either serine or threonine is the phosphorylated amino acid. Optimal PKG phosphorylation sites include only one X (37). Each consensus site was cross reference under hydrophatic plot analysis for surface accessibility and validated using Scansite 2.0.

Cloning of TWIK-2 and the COOH-terminus of TREK-1 from the rat MCA.

One male Long-Evans rat was anesthetized with 3% isoflurane and decapitated. The brain was immediately removed from the skull and placed in chilled Ringer solution. Both the left and right MCAs were carefully dissected, snap frozen in liquid nitrogen, and homogenized. Total RNA was extracted using a Qiagen Micro RNA Isolation Kit according to the manufacturer's instructions. Strands of RNA containing a poly-A sequence were purified and separated from the total RNA pool using an Oligotex mRNA Miniprep Kit (Qiagen). mRNA was reverse transcribed using oligo-dT primers (Super Script III Invitrogen).

For TWIK-2, cDNA was amplified by PCR using Pfu DNA polymerase (Invitrogen) and primers that spanned the coding region. The primers were as follows: forward 5′-ATGCGGCGGGGCGCGCTCCTGGCT-3′ and reverse 5′-GATCCTACCTGGGGATGGAGGCGTAATT-3′.

For the COOH-terminus of TREK-1, cDNA was amplified with platinum Taq polymerase (Invitrogen) with the following primers: forward 5′-ATACAAGCTTGGACTTCTACAAGCCCGTTG-3′ and reverse 5′-AATGGAATTCGCAGCACAGTGTGGTGTCAGA-3′.

PCR products were separated by gel electrophoresis and showed products at the predicted sizes for the TWIK-2 coding region (942 bp) and COOH-terminus of TREK-1 (401 bp, amino acids 283-415). For TWIK-2, cDNA was cloned into the pGEM-T Easy Vector (Promega). The clone was sequenced and confirmed to be TWIK-2.

For TREK-1, PCR products were digested with restriction enzymes HindIII and EcoRI and subcloned into multiple cloning sites of pSPT18. The clone was sequenced and validated to be completely homologous with the COOH-terminus of rat TREK-1 [KCNK2, National Center for Biotechnology Information (NCBI) Accession No. NM_172041].

Isolation of vascular smooth muscle cells from the rat MCA.

Single vascular smooth muscle cells (VSMCs) were enzymatically isolated as previously described (4). Rat MCAs were placed in a digestion buffer consisting of 135 mM NaCl, 5 mM KCl, 1.5 mM MgCl₂, 0.42 mM Na₂HPO₄, 0.44 mM NaH₂PO₄, 4.2 mM NaHCO₂, 10 mM HEPES, and 1 mg/ml BSA (pH adjusted to 7.25 with NaOH). The MCA was cut into small pieces and digested with 18 U/ml papain and 1 mg/ml dithioerythritol in digestion buffer for 35 min at 37°C. After several washes in the digestion buffer, MCA segments were further digested with 1.2 mg/ml collagenase type II, 0.8 mg/ml soybean trypsin inhibitor, and 60 U/ml elastase for 10 min at 37°C. Enzyme-treated MCA segments were washed several times and triturated with a pipette that had been coated with BSA by wash with digestion buffer. Cells were placed on ice and used within 6 h.

Electrophysiology.

VSMCs were placed in a chamber on the stage of an inverted microscope (Olympus IX 71) and continually superfused with buffer. Whole cell currents in individual VSMCs were measured using an integrating patch-clamp amplifier (Axopatch 200B) and pCLAMP 9.2 software (Axon Instruments, Union City, CA) after access had been obtained to the inside of the cell through perforation with amphotericin B. Amphotericin B (1.0 mg) was dissolved in 25 μl DMSO and used within 9 h of preparation. Patch pipettes (5–7 MΩ) were pulled from glass tubing (catalog no. 64-0819, Warner Instruments) in two stages by a pipette puller (model PP-830, Narishige) and polished with a Narishige microforge (model MF-830). Data were filtered at 1 kHz with a four-pole Bessel filter, digitized at 5 kHz, and stored on a hard disk. There was no compensation for cellular capacitance, series resistance, or leak current. The liquid junction potential was calculated using pCLAMP and corrected. The pipette buffer consisted of (in mM) 100 K⁺-gluconate, 43 KCl, 1 MgCl₂, 0.1 EGTA, and 10 HEPES (pH adjusted to 7.1 with NaOH). The bath buffer contained (in mM) 140 NaCl, 4.2 KCl, 3 NaHCO₃, 1.2 KH₂PO₄, 2 MgCl₂, 0.1 CaCl₂, 10 glucose, and 10 HEPES (pH adjusted to 7.4 with NaOH). VSMCs were held at −50 mV before the initiation of any protocol. Current-voltage (I-V) plots were obtained by changing the membrane potential to −130 mV for 30 ms and then ramping the voltage to +100 mV over 1 s.

Materials.

BSA, papain, dithioerythritol, soybean trypsin inhibitor, amphotericin B, tetraethylammonium (TEA) chloride, sodium nitroprusside (SNP), and arachidonic acid (Na⁺ salt) were obtained from Sigma. Elastase was obtained from Calbiochem, collagenase type II was obtained from Worthington, and 8-bromo-cGMP (8-Br-cGMP) was obtained from EMD Biosciences. Amphotericin B was dissolved in DMSO; arachidonic acid was dissolved in ethanol. All other reagents were dissolved in H₂O.

Data analysis and statistics.

Data are expressed as means ± SE or SE of the least-squares mean. Data were analyzed using two-way repeated-measures ANOVA followed by Student-Newman-Keuls post hoc analysis where appropriate.

RESULTS

Figure 1 shows the expression of message (on a log scale) for 10 K_2P channels relative to TREK-1, the K_2P channel most abundantly expressed. The second and third most abundant channels are TASK-1 and TWIK-2, which showed expression of message at a level ∼10% of TREK-1. All other K_2P channels showing expression in the rat MCA were ∼1% or less of the level for TREK-1. We did not find any message for TASK-5. We did not determine the relative levels of message for several of the K_2P channels. At the time of the analysis, there was some uncertainty of the base sequences, whether the gene was expressed in rats, or if the gene expresses functional channels.

Fig. 1.

Relative expression of two-pore domain K⁺ (K_2P) channels in the rat middle cerebral artery using real-time PCR. Values are expressed as a percentage of the most abundant K_2P channel, TREK-1, using the Pfaffl calculation method. ΔC_T is defined as the number of PCR cycles for threshold detection minus the number of cycles for threshold detection for the control gene, GADPH (n = 3), using the Pfaffl calculation method.

Table 1 shows the high-stringency consensus sequences for PKG phosphorylation sites in K_2P channels, the location of the sequence, and the NCBI Accession Number used for the analysis. Each consensus site shown in Table 1 has an intracellular location and lies outside a transmembrane-spanning domain. A number of the K_2P channels expressing message in the rat MCA have consensus sequences for PKG phosphorylation. TREK-1, which is the most abundantly expressed (Fig. 1), has two potential phosphorylation sites. TASK-1 and TWIK-2, the second and third most abundant channels, have two and one potential PKG phosphorylation sites, respectively. Other K_2P channels expressed were THIK-1 and THIK-2, which have two potential sites each, TREK-2, TRAAK, and TASK-3, which have one potential site each, and TWIK-1, which does not have any potential PKG phosphorylation sites.

Table 1.

Potential PKG phosphorylation sites in rat two-pore domain K⁺ channels

NCBI Accession No.	Gene Designate	Member	PKG Consensus Sequence	Optimum PKG Phosphorylation Sites
NM_021688	KCNK1	TWIK-1	None	None
NM_172041	KCNK2	TREK-1	RRLS	S348
			RKLS	S366
			RRTLS	S385
NM_033376	KCNK3	TASK-1	KRRS	S409
			RRSS	S410
NM_053804	KCNK4	TRAAK	KKPS	S382
NM_001039516	KCNK5	TASK-2	None	None
NM_053806	KCNK6	TWIK-2	RRVS	S260
NM_053405	KCNK9	TASK-3	RRKS	S394
NM_023096	KCNK10	TREK-2	KKQVS	S227
			RRLS	S359
NM_022292	KCNK12	THIK-2	RRGS	S198
			RRLS	S365
NM_022293	KCNK13	THIK-1	RRGT	T170
			RKLDS	S296
			RRLS	S343
NM_130813	KCNK15	TASK-5	RKQS	S5
			RRAS	S262
			RRKS	S317
NM_001003820	KCNK18	TRESK-2	None	None

NCBI, National Center for Biotechnology Information.

TWIK-2 and the COOH-terminus of TREK-1 were cloned and sequenced to determine if the putative phosphorylation sites of the native channels in the rat MCA conformed to the phosphorylation sites reported in the NCBI database. The cloned cDNA corresponding to the COOH-terminus of TREK-1 (amino acids 283-415) demonstrated putative phosphorylation sties at S348, S366, and S385. These phosphorylation sites are identical to those reported in the NCBI database (Table 1). Similarly, the cloned cDNA corresponding to the message for TWIK-2 demonstrated a putative phosphorylation site at S260, as provided by the NCBI database. Thus, native TREK-1 and TWIK-2 in the rat MCA contain putative phosphorylation sites for PKG.

The NO/cGMP/PKG system is known to activate BK_Ca channels (28, 32, 33). In the first experiment, we wanted to demonstrate that SNP, a NO donor, or 8-Br-cGMP, a membrane-permeable analog of cGMP, stimulated PKG in our hands by showing that they activate BK_Ca channels. Figure 2 shows I-V plots from isolated VSMCs at baseline (A and B) and after the the addition of 100 μM SNP (C) or 10 μM 8-Br-cGMP (D). Note that all graphs (Fig. 2, A–D) have five individual VSMC plots in each of the graphs. The baseline I-V plots were relatively flat until approximately +10 mV, at which point the trace showed an increase in slope. Note that the baseline traces in Fig. 2, A and B, are superimposed on one another. Figure 2, C and D, shows that after the addition of SNP or 8-Br-cGMP, transient increases in currents occurred in all cells. These transient currents resemble spontaneously transient outward currents (STOCs), which are due to the localized activation of multiple BK_Ca channels by the release of Ca²⁺ from ryanodine-sensitive Ca²⁺-release channels in the sarcoplasmic reticulum (28). The baselines (Fig. 2, A and B) did not have STOCs; however, after the addition of SNP or 8-Br-cGMP, all VSMCs had at least one STOC, and most had multiple STOCs.

Fig. 2.

Whole cell currents [current-voltage (I-V) plots] from five vascular smooth muscle cells (VSMCs) before (A) and after the addition of 100 μM sodium nitroprusside (SNP), a nitric oxide (NO) donor (B), as well as before (C) and after the addition of 10 μM 8-bromo-cGMP (8-Br-cGMP), a cell-permeable cGMP analog (D). Some of the baseline traces are obscured because they are superimposed on one another. V_m, membrane voltage.

Figure 3 shows summary data for the I-V plots presented in Fig. 2. For the summary data, the currents for each VSMC were averaged over a voltage range of 20 mV and presented in the form of a bar graph. Figure 3A shows mean currents (±SE) for baseline (n = 5) and after the addition of 100 μM SNP (n = 5). Individual data points are also shown with each bar. A similar graph is shown for 8-Br-cGMP in Fig. 3B. Currents after the addition of SNP or 8-Br-cGMP were statistically different from baseline (P = 0.011 and 0.012, respectively).

Fig. 3.

Summary data for the I-V plots presented in Fig. 2. Currents for each VSMC were averaged over a voltage range of 20 mV. A: mean currents (±SE) for baseline (n = 5) and after the addition 100 μM SNP (n = 5, P = 0.011). B: mean currents (±SE) for baseline (n = 5) and after the addition 10 μM 8-Br-cGMP (n = 5, P = 0.012). Individual data points are also shown with each bar in A and B. *P < 0.05 compared with the corresponding baseline using Student-Newman-Keuls post hoc analysis.

K_2P channels are resistant to TEA, a blocker commonly used to block BK_Ca channels (24, 26). Therefore, we conducted similar experiments as shown in Figs. 2 and 3 but in the presence of 10 mM TEA. At this concentration, TEA blocks not only the BK_Ca channels but also blocks some K_v channels. If K_2P channels in VSMCs of the MCA can be activated by either SNP or 8-Br-cGMP, then we should see increased K⁺ currents without the interference of the STOCs produced by BK_Ca channels. Figure 4, A and B, shows raw traces for two individual cells treated with 100 μM SNP in the presence of TEA. The black trace in Fig. 4 is baseline, and the red dotted line in Fig. 4 is after SNP. Similar traces for two cells given 10 μM 8-Br-cGMP are shown in Fig. 4, C and D. Note that when BK_Ca channels were blocked with TEA, neither SNP nor 8-Br-cGMP elicited STOCs as when TEA was absent (compare Fig. 4 with Fig. 2). The summary data for SNP and 8-Br-cGMP in the presence of TEA are shown in Fig. 5 (n = 5 cells/group). In the presence of TEA, the currents slightly but significantly decreased after treatment with SNP (P = 0.025). There was no effect on the currents when 8-Br-cGMP was administered in the presence of TEA.

Fig. 4.

Individual whole cell currents (I-V plots) from VSMCs in the presence of 10 mM tetraethylammonium (TEA; black traces in A–D) and in the presence of TEA plus 100 μM SNP, a NO donor (red traces in A and B), or in the presence of TEA plus 10 μM 8-Br-cGMP, a cell-permeable cGMP analog (red traces in C and D).

Fig. 5.

Summary data for I-V plots in the presence of TEA and in the presence of TEA plus SNP or TEA plus 8-Br-cGMP. Currents for each VSMC were averaged over a voltage range of 20 mV. A: mean currents (±SE) in the presence of 10 mM TEA (n = 5) and in the presence of 10 mM TEA plus 100 μM SNP (n = 5, P = 0.025). B: mean currents (±SE) in the presence of 10 mM TEA (n = 5) and in the presence of 10 mM TEA plus 10 μM 8-Br-cGMP (n = 5, P = 0.806). Individual data points are also shown with each bar in A and B. *P < 0.05 compared with the corresponding control using Student-Newman-Keuls post hoc analysis.

The addition of 10 mM TEA alone significantly decreased currents. For example, compare the mean currents in the baseline state as shown in Fig. 3 with those after TEA as shown in Fig. 5. Baseline currents in the absence of TEA were 200–300 pA at 80–100 mV (baseline in Fig. 3, A and B) and 60–80 pA at 80–100 mV in the presence of TEA (baseline in Fig. 5, A and B). In presence of 10 mM TEA, 20 μM arachidonic acid elicited large K⁺ currents (data not shown), as previously reported (4). The purpose of the administration of arachidonic acid was to demonstrate that K⁺ currents could be elicited in the presence of 10 mM TEA.

DISCUSSION

We report three new findings in this present study. First, the level of message for TREK-1 is greater than the level of message for other K_2P channels expressed in the rat MCA. Messages for TASK-1 and TWIK-2 were ∼10% of that for TREK-1. Other K_2P channels expressed were THIK-1, THIK-2, TREK-2, TWIK-1, TRAAK, and TASK-3; however, these latter channels were only 1% or less than the level of message for TREK-1. Second, a number of K_2P channels have putative phosphorylation sites for PKG. This includes TREK-1, TWIK-2, and TASK-1, the three most abundantly expressed channels. Third, the activation of PKG does not activate K_2P channels in VSMCs of the rat MCA.

A number of K_2P channels have been reportedly expressed in the vascular system (1, 3, 4, 11, 12, 14–16, 29). Although there appear to be some variations from artery to artery as to what is expressed (message and/or protein), TASK-1 (KCNK3), TASK-2 (KCNK5), TWIK-2 (KCNK6), and THIK-1 (KCNK13) appear to be expressed in all arteries studied to date (for a review, see Ref. 15). TREK-1 has been reported to be expressed in cerebral and mesenteric arteries but not expressed in pulmonary or carotid arteries (1, 4, 11, 12). While these previous studies have qualitatively demonstrated the presence or absence of message, no previous study has determined the relative expression of K_2P channels using quantitative RT-PCR. Our results using quantitative RT-PCR now provide relative levels of message for 10 K_2P channels studied in the rat MCA. Given that a number of K_2P channels are expressed in the vascular system, the order in which scientists investigate their role in vascular regulation could be a daunting issue. We suggest that in the MCA, at least, the three K_2Ps to study first are TREK-1, TASK-1, and TWIK-2. While a correlation between the levels of expression and importance of function may not hold true, it does give us a place to begin. Of interest, TASK-1 and TWIK-2, the second and third most abundantly expressed K_2P channels in the MCA, appear to be expressed by arteries in general (15). Furthermore, TREK-1, the most abundant channel in the rat MCA, appears to have a major role in the regulation of vascular tone in the brain and other arteries (1, 12).

Our analysis of putative phosphorylation sites of K_2P channels by PKG reveals potential sites for a number of K_2P channels (Table 1). Of particular importance to this study are TREK-1, TASK-1, and TWIK-2, the three most abundantly expressed K_2P channels in the rat MCA. It must be emphasized that the sites identified in our analysis are only putative at this time. While they may have an ideal protein sequence for PKG phosphorylation, PKG may not be accessible and thus cannot phosphorylate. TREK-1 and TWIK-2, when expressed in a heterologous expression system, are activated by NO or cGMP (5, 19, 30). In gastrointestinal smooth muscle, NO/cGMP/PKG activates a native channel that strongly resembles TREK-1 (20, 35). We are not aware of other reports where the effects of NO or cGMP on other K_2P channels, including TASK-1, have been studied. The fact that TREK-1 or TWIK-2 can be activated by the NO/cGMP/PKG system in heterologous expression systems and possibly in gastrointestinal smooth muscle does not necessarily mean that TREK-1, TWIK-2, or other K_2P channels can be activated by this same system in cerebrovascular smooth muscle.

In preliminary experiments, we demonstrated that the activation of the NO/cGMP/PKG system elicited STOCs, which are characteristic of BK_Ca channels (Fig. 2). The activation of BK_Ca channels demonstrates that we were successful in activating PKG by the addition of SNP or 8-Br-cGMP.

Given the abundance of K_2P channels in the vascular system and the potential PKG phosphorylation sites for a number of K_2P channels, it would seem reasonable that the K_2P channel would be responsible for at least part of the relaxation to NO in vascular smooth muscle.

With a few exceptions, K_2P channels are resistant to the blocking effects of TEA (26). Those that are resistant include TREK-1, TASK-1, and TWIK-2, the most abundant K_2P channels in the rat MCA (7, 9, 23, 26, 30). We used TEA to block BK_Ca (and some K_v) channels to determine if SNP or 8-Br-cGMP could elicit an increase in current without the contaminating effects of BK_Ca channel activation. Not only did TEA reduce the baseline currents, but it also blocked the STOCs produced by BK_Ca channels (compare Figs. 2 and 4). Furthermore, whole cell current was not increased by the addition of either SNP or 8-Br-cGMP with TEA present. Thus, any K_2P channels present, including TREK-1, TWIK-2, or TASK-1, were not activated by the NO/cGMP/PKG pathway. Although the NO/cGMP/PKG system has been shown to activate some K_2P channels in heterologous expression systems (5, 19, 30), K_2P channels are not activated in native cerebrovascular smooth muscle cells with a NO donor or an analog of cGMP. Therefore, the activation of K_2P channels by NO does not appear to be a mechanism of smooth muscle relaxation in rat cerebral arteries.

We did see a slight but significant decrease in current with SNP at more positive potentials (+60–80 and +80–100 mV; Fig. 5). It might be interpreted that SNP acting through PKG was inhibiting a channel in VSMCs. However, this conclusion is not consistent with the effects of 8-Br-cGMP, which also activates PKG and did not inhibit currents. Although we do not have an explanation for this effect of SNP, the data support the conclusion that NO does not relax vascular smooth muscle in rat cerebral arteries by the activation of K_2P channels.

In summary, we have demonstrated that rat cerebral arteries express a number of K_2P channels, with TREK-1, TASK-1, and TWIK-2 being the most abundant. The amount of message for TASK-1 and TWIK-2 was ∼10% of the amount for TREK-1. Furthermore, dilations elicited by NO in rat cerebral arteries are not likely through the activation of K_2P channels in vascular smooth muscle.

GRANTS

This work was supported by National Institutes of Health Grants PO1-NS-38660 (to R. M. Bryan, Jr.), RO1-NS-46666 (to R. M. Bryan, Jr.), and RO1-HL-088435 (to S. P. Marrelli).