|Home | About | Journals | Submit | Contact Us | Français|
Physiologically relevant concentrations of [Arg8]-vasopressin (AVP) induce repetitive action potential firing and Ca2+ spiking responses in the A7r5 rat aortic smooth muscle cell line. These responses may be triggered by suppression of KCNQ potassium currents and/or activation of non-selective cation currents. Here we examine the relative contributions of KCNQ5 channels and TRPC6 non-selective cation channels to AVP-stimulated Ca2+ spiking using patch clamp electrophysiology and fura-2 fluorescence measurements in A7r5 cells. KCNQ5 or TRPC6 channel expression levels were suppressed by short hairpin RNA constructs. KCNQ5 knockdown resulted in more positive resting membrane potentials and induced spontaneous action potential firing and Ca2+ spiking. However physiological concentrations of AVP induced additional depolarization and increased Ca2+ spike frequency in KCNQ5 knockdown cells. AVP-activated non-selective cation current was reduced by TRPC6 shRNA treatment or removal of external Na+. Neither resting membrane potential nor the AVP-induced depolarization was altered by knockdown of TRPC6 channel expression. However, both TRPC6 shRNA and removal of external Na+ delayed the onset of Ca2+ spiking induced by 25 pM AVP. These results suggest that suppression of KCNQ5 currents alone is sufficient to excite A7r5 cells, but AVP-induced activation of TRPC6 contributes to the stimulation of Ca2+ spiking.
[Arg8]-vasopressin (AVP) is a pituitary hormone that is released into the systemic circulation when blood pressure falls. Circulating concentrations of AVP may increase from a few picomolar to several hundred picomolar in response to a large drop in blood pressure . AVP binding to V1a vasopressin receptors on vascular smooth muscle cells (VSMCs) is known to induce Ca2+ release from the sarcoplasmic reticulum and Ca2+ influx via multiple Ca2+-permeable non-selective cation channels [2, 3]. The resulting elevation of cytosolic Ca2+ concentration ([Ca2+]i) activates the contractile apparatus of the vascular smooth muscle cells in the artery walls and leads to vasoconstriction. This mechanism is generally thought to be essential for the well characterized ability of AVP to restore blood pressure.
More recently, it has become apparent that AVP can induce Ca2+ responses in vascular smooth muscle cells via two distinct concentration-dependent pathways [4–6]. At physiological (picomolar) concentrations of AVP, inhibition of KCNQ family potassium currents results in membrane depolarization and activation of voltage-sensitive Ca2+ channels; the resulting influx of Ca2+ increases [Ca2+]i [4–6]. Nanomolar concentrations of AVP induce a robust Ca2+ release from the sarcoplasmic reticulum mediated through the classical receptor-activated phospholipase (PLC)/inositol triphosphate (IP3)-diacylglycerol (DAG) pathway [2, 3]. Although the clinical use of AVP  and V1a vasopressin receptor antagonists [8, 9] for a plethora of indications has increased significantly in recent years, the mechanisms of action of physiologically or therapeutically relevant concentrations of AVP still remain to be clarified.
The recently proposed hypothesis that KCNQ (Kv7) family of K+ channels might be the downstream target of physiological concentrations of AVP  has yet to be fully accepted. KCNQ channels were first discovered in neurons and recognized as the molecular correlate of M-currents . Five members have been recognized in the KCNQ channel family (Kv7.1–7.5) [11, 12]. Though expression was initially thought to be restricted to neurons and cardiac myocytes, KCNQ channels have now been shown to be expressed in many cell types, including VSMCs [4, 6, 12–14]. KCNQ channels conduct voltage-dependent non-inactivating outward K+ currents that tend to oppose membrane depolarization to the threshold for triggering action potential generation; hence they act as a brake for neuronal excitation . These properties of KCNQ channels may also enable them to serve as regulators of excitation of VSMCs. We previously provided evidence that physiological concentrations (10–100 pM) of AVP indeed produce membrane depolarization and Ca2+ influx in A7r5 rat aortic smooth muscle cells through a protein kinase C (PKC)-dependent suppression of KCNQ5 currents . We also showed that KCNQ currents in freshly isolated rat mesenteric artery smooth muscle cells are similarly regulated by AVP .
However, KCNQ channel inhibition may not be the only mechanism by which AVP regulates membrane excitability to increase [Ca2+]i in VSMCs. Signal transduction mediators implicated in KCNQ current suppression have also been implicated as regulators of other membrane-depolarizing effectors. For example, DAG, not only activates PKC and consequently suppresses KCNQ currents, but also activates non-selective cation channels independently of PKC activation . The DAG-sensitive TRPC6 channel has been proposed as a candidate cation channel that could increase [Ca2+]i following stimulation of G protein coupled receptors [17, 18]. TRPC6 is a member of the transient receptor potential (TRP) superfamily of cation-selective ion channels. An earlier attempt to implicate TRPC6 in AVP signaling used a supraphysiological concentration of AVP (100 nM)  at which AVP increases [Ca2+]i by stimulating both Ca2+ release and Ca2+ influx. In the present study, we have examined the contribution of TRPC6 at physiological concentrations of AVP (10–100 pM) that stimulate action potential-mediated Ca2+ spiking, a phenomenon that is independent of the release of intracellular Ca2+ stores . Our findings demonstrate that KCNQ5 channels are essential channels that conduct K+ at resting membrane potential and implicate both TRPC6 channel activation and KCNQ5 channel suppression in the stimulation of Ca2+ spiking by physiological concentrations of AVP in A7r5 cells.
For RNA interference, short hairpin RNAs (shRNAs) were designed against the rat TRPC6 sequence, 5′-GAAGGTCTTTATGCAATCGCGGTGGTTTT-3′ and the KCNQ5 sequence 5′-GGCTTCAAGTTGCCTCTTAATTCTGGAGT-3′. A control nucleotide sequence shRNA was made against the scrambled nonsense sequence 5′-TGCAGCTGTGGTCGCCCAGATTCGCCGAC-3′. Complementary nucleotides encoding the short hairpin sequences were cloned into RNAi-Ready pSIREN-DNR-DsRed-Express shuttle vector, and the recombinant shuttle vector was then used to construct the “AdenoX-TRPC6sh” or “AdenoX-KCNQ5sh” adenovirus using the Knockdown Adenoviral RNAi System 2 (BD Clontech). The AdenoX-TRPC6sh or KCNQ5sh adenovirus plasmid was digested with Pac I prior to transfection into HEK-293 cells using SuperFect Transfection Reagent (Qiagen). Within 7–10 days, when red-fluorescent plaques and cytopathic effects were apparent, lysates were prepared and used for adenovirus amplification and purification. Purified adenovirus was used directly on A7r5 cells in culture at a multiplicity of infection (MOI) of 50. Infected cells could be identified by red fluorescence within 24 hours.
A7r5 cells derived from embryonic rat aorta were cultured as described previously . Cells were sub-cultured in 6-well plates for fura-2 fluorescence measurements. Appropriate experimental groups were infected for 3–5 days with an adenoviral vector to co-express the fluorescent marker DsRed with TRPC6, KCNQ5, or scrambled control shRNAs.
Confluent monolayers of A7r5 cells cultured in 6-well plates were washed twice with control medium (135 mM NaCl, 5.9 mM KCl, 1.5 mM CaCl2, 1.2 mM MgCl2, 11.5 mM glucose, 11.6 mM HEPES, pH 7.3) and then incubated in the same medium with 1 μM fura-2-AM (Invitrogen/Molecular Probes), 0.1% bovine serum albumin and 0.02% Pluronic F127 detergent for 60 min at room temperature (22–25°C) in the dark. The cells were then washed twice and incubated in the dark in control medium for 1.5 h. Sodium-free medium contained 135mM N-methyl D-glucamine (NMDG) instead of sodium. Fura-2 fluorescence was measured using a Biotek Synergy™ HT plate reader (340 and 380 nm excitation; 510 nm emission). All experiments were performed at room temperature. Prior to loading the cells with fura-2-AM, cells were observed with a fluorescence microscope for DsRed fluorescence (535 nm excitation wavelength, 610 nm emission wavelength) to confirm the expression of TRPC6 or KCNQ5 shRNA. Background fluorescence at 340 and 380 nm wavelengths was also measured in the Biotek plate reader in the same wells before the start of fura-2 loading and subtracted from individual wavelength measurements before calculating 340/380 fluorescence ratio. The latency to onset of spiking was measured as the time from the start of treatment to the start of repetitive Ca2+ spiking (at least 2 spikes/min). Frequency of spiking was calculated by the number of spikes per min from the time of onset of repetitive Ca2+ spiking. Each ‘n’ is an average value of 3 wells tested. The frequency of spiking and the latency to start of repetitive Ca2+ spiking were compared among control and treatment groups using Student’s paired ‘t’ test or one way analysis of variance (ANOVA) as appropriate, with ‘p’ values < 0.05 considered statistically significant.
The whole cell perforated patch configuration was used to measure KCNQ5 currents under voltage-clamp conditions and membrane potentials under current-clamp conditions in single A7r5 cells trypsinized and plated on glass coverslips; the ruptured patch configuration was used to measure TRPC6 currents. All experiments were performed at room temperature with continuous perfusion of bath solution. The standard bath solution contained (in mM): 5 KCl, 135 NaCl, 10 HEPES, 2 CaCl2, 1.2 MgCl2, 5 glucose, pH 7.3. To isolate KCNQ currents, 100 μM GdCl3 (sufficient to block L- and T-type Ca2+ channels and non-selective cation channels) was added to the standard external solution. Internal (pipette) solution to record KCNQ currents contained (in mM): 110 K gluconate, 30 KCl, 5 HEPES, 1 K2EGTA, 1 MgCl2, pH 7.2. Osmolality was adjusted to 270 mOsm/l with D-glucose. 200 μg/ml Amphotericin B was added to the internal solution for membrane patch perforation. KCNQ currents were recorded in perforated patch configuration by application of 5 s voltage steps from a −74 mV holding potential to test potentials ranging from −94 mV to +46 mV. The series resistance in perforated patch recordings was below 30 MΩ. ICAT activated by 100 nM AVP was recorded in ruptured patch configuration with a 100 ms voltage ramp protocol (from +85 to −115 mV) from −15 mV holding potential every 5 sec. Internal solution to record ICAT contained (in mM): 110 Cs aspartate, 30 CsCl, 5 HEPES, 1 Cs2EGTA, 1 MgCl2, 2 Na2ATP, pH 7.2. Osmolality was adjusted to 270 mOsm/l with D-glucose. 10 μM verapamil and 100 μM spermine chloride were added to the external solution to block L-type Ca2+ channels and Mg2+-inhibitable currents, respectively. Sodium-free medium contained 135 mM NMDG instead of Na+. In knockdown experiments, current densities or membrane potentials were measured in control A7r5 cells and in brightly DsRed fluorescent cells. Current-clamp recording of membrane potential was performed in I = 0 mode. Data are presented as mean ± standard error of the mean (S.E.). For comparisons between groups, rank sum test, student’s paired ‘t’ test or one-way analysis of variance (ANOVA) was used, with ‘p’ values < 0.05 considered statistically significant.
A7r5 cells were infected with adenoviral vectors for TRPC6shRNA or KCNQ5shRNA for 5 days. The cells were then trypsinized, resuspended with an equal number of uninfected control A7r5 cells, and then re-plated on 12 mm round glass coverslips 1 day prior to fixing. The cells were fixed with 2% paraformaldehyde in phosphate-buffered saline. Coverslips with KCNQ5shRNA-expressing cells were incubated with rabbit polyclonal anti-KCNQ5 antibodies (Chemicon Inc) at a 1:2000 dilution at 4°C overnight. Coverslips with TRPC6shRNA-expressing cells were incubated with rabbit polyclonal anti-TRPC6 antibodies (Alomone Labs) at a 1:200 dilution at 4°C over night. Alexa Fluor® 488 goat anti-rabbit IgG at 1:400 dilution was used as secondary antibody. Cell images were acquired with an Olympus 1X71 inverted epifluorescence microscope and Hamamatsu Orca 12-bit digital camera. Grey scale images were captured at excitation wavelengths of 480, 535 and 360 nm, emission wavelength of 535, 610 and 460 nm for Alexa Fluor® 488 fluorescence, DsRed fluorescence and DAPI florescence, respectively. Colors were applied to grey scale images using Adobe Photoshop™ software to represent the different emission colors: green for Alexa Fluor® 488, red for DsRed, and blue for DAPI. Regions of interest were defined using Compix SimplePCI™ software by outlining nonfluorescent cells or cells expressing DsRed fluorescence. Coverslips incubated without primary antibody had no detectable fluorescence with 490 nm excitation at the gain and exposure settings used. The specificities of the antibodies were confirmed as shown in Supplemental Figure 1.
In our previous work, we established that 100 pM AVP induces a PKC-dependent suppression of KCNQ currents in A7r5 cells . To more clearly establish that this function is mediated specifically by KCNQ5, which was the only member of the KCNQ family detected by RT-PCR in A7r5 cells , we knocked down KCNQ5 channels by infecting A7r5 cells with an adenoviral vector for bicistronic expression of KCNQ5shRNA and DsRed fluorescent marker. Suppression of KCNQ5 channel protein was detected by immunofluorescence using rabbit polyclonal anti-KCNQ5 antibody. KCNQ5 immunofluorescence was invariably low in A7r5 cells expressing DsRed (Figure 1A), whereas control uninfected A7r5 cells often exhibited robust and in many cases punctuate immunofluorescence staining. The specificity of the KCNQ5 antibody was confirmed by the increased immunoflourescence in A7r5 cells overexpressing KCNQ5 and the absence of immunoreactivity when the primary antibody was omitted (Supplemental figure 1A). KCNQ currents measured by whole cell perforated patch clamp electrophysiology indicated that KCNQ5shRNA expression completely abolished KCNQ currents in A7r5 cells (Figure 1B and 1C).
To test our hypothesis that KCNQ current suppression regulates Ca2+ entry, we measured Ca2+ spiking responses in fura-2-loaded A7r5 cells expressing KCNQ5shRNA. In parallel cell cultures, spontaneous Ca2+ spikes were rarely observed in control cells (n=3) or scrambled shRNA-expressing cells (n=3), whereas expression of KCNQ5shRNA produced significantly greater repetitive spontaneous Ca2+ spiking activity (2.5 ± 0.5 spikes/min, n=6, P<0.005) (Figure 2A & 2B). The Ca2+ spiking frequency was increased significantly with the addition of 25pM AVP in KCNQ5shRNA-expressing A7r5 cells (compare KCNQ5shRNA bars in Figure 2 panels B and D; p<0.05). However the AVP-induced Ca2+ spiking frequency in KCNQ5shRNA-expressing A7r5 cells (5.0 ± 0.6 spikes/min) was not significantly different (p>0.1) from AVP-induced Ca2+ spiking in control cells (5.1 ± 0.4 spikes/min; n=10) or scrambled shRNA expressing cells (6.4 ± 0.7 spikes/min; n=3) (Figure 2D). A single Ca2+ spike generally occurred at the time of AVP injection, but this was likely a mechanically-induced transient because injection of control medium also induced an instantaneous spike, but did not produce repetitive Ca2+ spiking (results not shown). These observations indicate that K+ conductance through KCNQ5 channels regulates Ca2+ entry in A7r5 cells. These results also indicate that, apart from KCNQ5 current suppression, AVP regulates Ca2+ spiking by some additional mechanism.
To explore the contribution of TRPC6 non-selective cation channels to the AVP-induced Ca2+ spiking response, A7r5 cells were infected with an adenoviral vector for bicistronic expression of TRPC6shRNA and DsRed fluorescent marker. TRPC6 channel protein was detected in the same cells by immunostaining with a polyclonal anti-TRPC6 antibody. Robust TRPC6 immunofluorescence was observed in uninfected A7r5 cells, whereas those expressing TRPC6shRNA had little detectable immunofluorescence (Figure 3A). The specificity of the TRPC6 antibody was confirmed by the loss of immunofluorescence when the TRPC6 antibody was pre-incubated with a TRPC6-specific blocking peptide (Supplemental figure 1B). The function of TRPC6 channels in A7r5 cells was evaluated by electrophysiological measurement of AVP-stimulated non-selective cation currents (ICAT). ICAT activation by 100 nM AVP was suppressed in all TRPC6shRNA-expressing cells (Figure 3B, 3C), with ICAT activation reduced to undetectable levels in 7 of 13 such cells. Both outward and inward components of ICAT were significantly reduced (by ~70% on average) in TRPC6shRNA-expressing cells (n=13) compared to non-fluorescent control cells (n=11, Figure 3D). In contrast, KCNQ5shRNA expression did not affect ICAT current densities (Figure 3C, 3D).
Having previously proposed that PKC-dependent suppression of KCNQ channels contributes to AVP-induced Ca2+ spiking in A7r5 cells and AVP-induced mesenteric artery constriction [4, 6], we chose to explore whether specifically suppressing KCNQ5 activity alone might be sufficient to induce membrane depolarization. KCNQ5shRNA-expressing A7r5 cells were significantly depolarized (−16 ± 4.1 mV; n=9, p<0.001) compared to uninfected control cells (−51.6 ± 1.9 mV; n=6) (Figure 4A). Addition of 100pM AVP induced additional depolarization (by 5.9 ± 1.1 mV; n=9, p<0.001) in KCNQ5shRNA-expressing A7r5 cells (Figure 4A). However, this additional AVP-induced depolarization in KCNQ5shRNA-expressing cells was significantly less than that in control A7r5 cells (9.8 ± 1.2 mV; n=6, p<0.05). Approximately 25% of cells expressing KCNQ5shRNA were spontaneously electrically active (repetitive action potential firing observed in 4 out of 16 cells; figure 4B); the remaining 75% of the cells examined were depolarized to such an extent that these cells did not show any spontaneous electrical activity. Neither resting potential nor the level of AVP-induced depolarization was significantly altered by TRPC6shRNA expression (Figure 4A).
We next chose to see whether the reduction of AVP-induced ICAT in TRPC6shRNA-expressing cells (Figure 3D) would have any significance for the stimulation of Ca2+ spiking by physiological concentrations of AVP. Ca2+ spiking measurements in fura-2-loaded cells revealed that the latency to onset of Ca2+ spiking in TRPC6shRNA-expressing A7r5 cells was significantly greater than in control cells (13.2 ± 1.9 Vs. 7.8 ± 1.6 min; n=6 for each condition, p<0.005; Figure 5A and 5B). AVP-stimulated Ca2+ spiking was not observed during 25 min recordings in 4 out of the 18 wells plated with TRPC6shRNA-expressing A7r5 cells. In contrast, all the corresponding 18 wells plated with control A7r5 cells spiked normally. Interestingly, after initiation of Ca2+ spiking, the frequency of Ca2+ spiking in TRPC6shRNA-expressing cells was not different from control A7r5 cells (4.1 ± 0.6 Vs 4.5 ± 0.2 spikes/min; n=6 for each condition, p>0.1) (Figure 5C). In current-clamp recording, application of 100 pM AVP stimulated action potential (AP) firing in a substantial fraction of control A7r5 cells (observed in 9 out of 21 cells), whereas stimulation of AP firing by 100 pM AVP was not observed during similar recording periods in any cells expressing TRPC6shRNA (in 17 out of 17 cells, data not shown). These data further suggest that the mechanism of AVP-induced membrane depolarization is not limited to suppression of KCNQ channels alone, but also involves TRPC6 activation.
Diacylglycerol (DAG) produced via breakdown of membrane phospholipids may contribute to both the PKC-dependent suppression of KCNQ5 currents and the activation of TRPC6 currents by AVP. To evaluate the relative ability of DAG to effect these downstream responses, we tested the effects of a membrane-permeant DAG analog, 1-oleoyl-2-acetyl-sn-glycerol (OAG), on KCNQ5 currents, ICAT activation, and Ca2+ spiking. OAG (100 μM) significantly suppressed KCNQ5 currents at all voltages positive to −60 mV (p<0.05; Figure 6A and 6B) and stimulated Ca2+ spiking in A7r5 cells (Figure 6D). Both effects were abolished by a selective PKC inhibitor, Ro-31-8220 (2 μM) (Figure 6C and 6D). OAG also stimulated Ca2+ spiking in TRPC6shRNA-expressing A7r5 cells, but the latency to its onset was significantly increased when compared to control (19.4 ± 0.9 Vs. 17.7 ± 1.3 min; n=4 for each condition, P<0.05; Figure 6E). The frequency of OAG-induced Ca2+ spiking in TRPC6shRNA-expressing cells was not different from control A7r5 cells (6.3 ± 0.9 Vs 7.3 ± 0.6 spikes/min; n=4 for each condition, p>0.1; Figure 6F). These results suggest that despite its ability to directly activate TRPC6 channels (Supplemental figure 2), DAG contributes to the stimulation of Ca2+ spiking primarily via PKC-dependent suppression of KCNQ5 currents.
To determine whether Na+ influx through TRPC6 contributes to AVP-stimulated Ca2+ spiking, ICAT and Ca2+ spiking were measured in Na+-free conditions. Inward ICAT activated by 1 nM AVP was abolished when Na+ in the external solution was replaced with NMDG, an impermeant cation (Figure 7A). The latency to onset of Ca2+ spiking in A7r5 cells in Na+-free medium was also significantly increased from that in control medium (16.6 ± 3.6 Vs 4.7 ± 1.0 min; n=4 for each condition; p<0.05, Figure 7B and 7C). The frequency of 25pM AVP-induced Ca2+ spiking in A7r5 cells in Na+-free medium was significantly less than in cells maintained in Na+-containing medium (3.5 ± 0.4 Vs 6.1 ± 0.6 spikes/min; n=4 for each condition, p<0.01, Figure 7B and 7D).
Although the signal transduction of AVP in VSMCs has been studied for decades, there is still a paucity of information about the mechanisms of action of physiologically or therapeutically relevant concentrations of AVP. In the systemic circulation, concentrations of 100 pM or less are sufficient to produce a robust pressor response . Most studies exploring AVP signal transduction pathways have utilized supraphysiological (nanomolar to micromolar) concentrations of AVP. Based on such studies, the actions of AVP on vascular smooth muscles cells are often attributed to the release of intracellular Ca2+ stores. There is nonetheless general recognition that voltage-sensitive Ca2+ channels (VSCC) also contribute to vasoconstrictor responses, but with little agreement as to the mechanisms whereby those channels may become activated. We previously elucidated a potential mechanism activated by physiological (picomolar) concentrations of AVP that leads to VSCC-dependent action potential firing in A7r5 vascular smooth muscle cells. We found that this process involves PKC-mediated suppression of KCNQ K+ channel activity. KCNQ channel inhibition provides a depolarizing stimulus for activation of L-type VSCC. In the present study, we have investigated another mechanism by which AVP may depolarize vascular smooth muscle cells: activation of TRPC6 non-selective cation channels. Using a combination of molecular, electrophysiological, and cell physiological approaches, we provide evidence that specifically implicates both KCNQ5 and TRPC6 channels in the stimulation of Ca2+ spiking in VSMCs by physiological concentrations of AVP. Our results indicate that though suppression of KCNQ5 channels alone is sufficient to excite VSMCs, TRPC6 channel activation contributes to enhance excitation and Ca2+ entry in response to physiological concentrations of AVP.
Membrane potential of VSMCs controls Ca2+ influx through VSCC; membrane depolarization can lead to enhanced Ca2+ influx and vasoconstriction. Resting membrane potential is largely determined by K+ conductance (GK). This electrical conductance, which reflects the efflux of K+ through channels that are active in unstimulated cells, maintains the resting membrane potential at very negative voltages that prevent opening of VSCC.
Of the many types of K+ channels expressed in VSMCs, KCNQ channels are among the few that are thought to be appreciably active in unstimulated vascular myocytes [4, 6]. The membrane resistance (Rm) of arterial smooth muscle cells is high, in the order of 5–15GΩ . Hence even closure of a few channels that conduct K+ at resting membrane potential would produce significant membrane depolarization. We found that knockdown of KCNQ5 in A7r5 cells shifted the membrane potential by +35mV, revealing an important contribution of GK through KCNQ channels at resting membrane potentials. This membrane depolarization (Figure 4A), was associated with spontaneous Ca2+ spiking (Figure 2A) and action potential firing (Figure 4B). These results complement the previous findings from our laboratory in which pharmacological suppression of KCNQ currents with linopirdine produced membrane depolarization, Ca2+ spiking, and action potential firing [4, 6].
Expression of KCNQ channel subunits has been detected in a variety of arterial tissues or vascular myocytes: KCNQ1 in murine portal vein , KCNQ 1, 4 and 5 in mouse aorta, carotid artery, femoral artery and mesenteric artery , KCNQ5 in A7r5 rat aortic smooth muscle cells and KCNQ1 and 5 in rat aorta , KCNQ 1, 4 and 5 in rat mesenteric artery myocytes . The function of KCNQ channels as negative regulators of vasoconstriction has been demonstrated recently in murine portal vein , pulmonary artery  and mesenteric artery . It remains to be determined if KCNQ5 is the functionally predominant KCNQ subtype in all VSMCs. Further studies to elucidate the expression and function of KCNQ family of channels in various vascular beds will greatly enhance our understanding of their role in regulation of blood vessel diameter.
The L-type Ca2+ channel-dependent Ca2+ spiking observed at low concentrations of AVP  is different from the IP3-mediated Ca2+ release oscillations observed in A7r5 cells  or rat hepatocytes  treated with nanomolar concentrations of AVP. Supra-physiological (nanomolar to micromolar) concentrations of AVP activate PLC and subsequently induce IP3-mediated Ca2+ release from the sarcoplasmic reticulum [3, 27, 28]. In some cells, repetitive cycles of Ca2+ release and reuptake are exhibited as oscillations in [Ca2+]i [25, 26]. However, unlike the Ca2+ spiking responses, the Ca2+ release oscillations are resistant to L-type Ca2+ channel blockers. Furthermore, at physiological concentrations (≤100pM) of AVP the Ca2+ release responses in A7r5 cells are barely detectable [24, 29]. Suppression of KCNQ5 currents, leading to membrane depolarization and Ca2+ entry through voltage-sensitive Ca2+ channels, may provide a different Ca2+ signaling mechanism by which physiological [AVP] can exert its vasoconstrictor actions [4, 6].
Signal transduction mechanisms activated at picomolar concentrations of AVP have rarely been examined and therefore it is unclear whether suppression of KCNQ channels is the only downstream target of V1a vasopressin receptor activation in the physiological concentration range. In KCNQ5shRNA-expressing cells with no detectable KCNQ currents (Figure 1B), addition of a physiological concentration of AVP (100pM) produced a significant additional depolarization (Figure 4A) and increase in Ca2+ spiking frequency (Figure 2D), suggesting that additional signaling elements do contribute to these functional responses. The extent of depolarization is greater with KCNQ5 knockdown (compared to AVP-induced depolarization), but Ca2+ spiking is more robust with AVP treatment. With AVP treatment, suppression of KCNQ5 currents is combined with an increase in inward current (e.g. through activation of TRPC6 channels). This may be expected to increase the rate of depolarization and hence the frequency of firing compared with even more complete suppression of KCNQ5 current alone, because in the latter case only the resting inward current (leak current) provides the depolarizing stimulus.
While a 10–100 pM concentration of AVP produced a Ca2+ spiking response, these concentrations did not stimulate an appreciable release of intracellular Ca2+ stores  and hence it is unlikely that store-operated currents (Isoc) play a significant role in Ca2+ entry at these concentrations of AVP. Hence we chose to explore the contribution of the non-selective cation current (ICAT) to 25pM AVP-stimulated Ca2+ responses.
Members of the transient receptor potential (TRP) family of non-selective cation channels have been implicated as targets for vasoconstrictor signal transduction [30, 31]. Expression of TRPC1, 4, 6 and 7 of the classical TRP channel subfamily (TRPC) has been reported in A7r5 cells [17, 18, 32, 33]. TRPC1 is a candidate store-operated channel or part of the store operated channel in VSMCs, including A7r5 cells [32, 34, 35]. TRPC4 channels have also been recently reported to mediate store operated Ca2+ entry in VSMCs . TRPC6 and TRPC7 are members of the DAG-sensitive TRPC subgroup (TRPC3/6/7) [37, 38], of which TRPC6 channels have been proposed as candidate receptor-operated channels in A7r5 cells [17, 18]. TRPC6 channels, which have low selectivity for Ca2+ over Na+ [39, 40], can increase [Ca2+]i via at least 3 mechanisms: 1) the influx of cations through TRPC6 channels would shift the membrane potential to more positive values to activate VSCC; 2) influx of Na+ through TRPC6 channels may result in elevated cytosolic Na+ concentration and increase [Ca2+]i by reversing the Na+/Ca2+-exchanger (NCX) ; 3) influx of Ca2+ directly through TRPC6 channels may contribute to elevate [Ca2+]i. Previous efforts to evaluate the contributions of TRPC6 in AVP signaling have focused on effects induced by maximal [AVP] (100nM), which produces a robust Ca2+ release and influx of Ca2+ [17, 18], but may not be relevant for more physiological concentrations of AVP.
In our studies, knockdown of TRPC6 reduced AVP-stimulated ICAT by ~70% (Figure 3D), indicating that TRPC6 is the principal non-selective cation channel stimulated by AVP. These results also confirm that endogenous TRPC6 is the major mediator of ICAT in A7r5 cells [17, 18]. Examination of Ca2+ spiking responses to physiological concentrations of AVP in A7r5 cells, showed that knockdown of TRPC6 produced a shift in latency to onset of spiking but did not alter the frequency of spiking in response to 25pM AVP (Figure 5). Furthermore, knockdown of TRPC6 did not alter the resting membrane potential or depolarization in response to AVP (Figure 4A). These results suggest that ICAT through TRPC6 contributes only modestly to membrane potential when KCNQ channels are open. However when KCNQ channel activity is suppressed by physiological [AVP], membrane resistance (Rm) increases . According to Ohm’s law [Voltage (V) = Current (I) × Resistance (R)], when the resistance is high, even a small increase in ICAT through TRPC6 will be able to depolarize the VSMC more effectively. Hence the influx of cations through TRPC6 channels might be sufficient to more rapidly attain the threshold to activate VSCC and trigger firing of Ca2+-dependent action potentials (AP).
In a previous study, OAG-induced Ca2+ entry in A7r5 cells was abolished by nimodipine, a blocker of L-type Ca2+ channels, suggesting a membrane depolarizing effect of the DAG analog . The authors of that study concluded that OAG-induced Ca2+ entry was a result of membrane depolarization (via activation of TRPC6 and enhanced cation influx) and activation of L-type Ca2+ channels. However, in the same study, knockdown of TRPC6 in A7r5 cells did not prevent Ca2+ entry induced by application of OAG (100 μM) . We found that, although 100 μM OAG activated ICAT (Supplemental Figure 2) as observed in the previous study, it also significantly suppressed KCNQ5 currents in A7r5 cells (Figure 6A and 6B). TRPC6shRNA expression only increased the latency to onset of OAG-induced Ca2+ spiking, but, inhibition of PKC completely abolished OAG-induced Ca2+ spiking (Figure 6D). Inhibition of PKC had no effect on OAG-activated TRPC6 currents  and in general, PKC activation is thought to inhibit rather than activate TRPC channels . If Ca2+ spiking were caused primarily by TRPC6 activation, PKC inhibition would enhance rather than inhibit Ca2+ spiking. Hence AVP-stimulated Ca2+ spiking can more reasonably be attributed to DAG-induced PKC activation and suppression of KCNQ currents. It is also likely that the OAG-stimulated Ca2+ entry in the previous study  might have involved PKC-dependent suppression of KCNQ channel activity rather than activation of TRPC6 alone.
Knockdown of TRPC6 channels in A7r5 cells shifted the latency to onset of Ca2+ spiking, but did not alter the frequency of spiking in response to 25pM AVP (Figure 5). However in Na+-free conditions both the latency to onset of spiking and frequency of spiking were changed (Figure 7). This raises the possibility that there might be some contribution from other Na+-conducting pathways that influences spike frequency.
A functional coupling of TRPC6 and Na+/Ca2+ exchanger (NCX) has been proposed for rat aortic smooth muscle cells [41, 44, 45]. While purinergic receptor stimulation produced reversal of the NCX to mediate Ca2+ influx, it remains to be seen whether the same mechanism also contributes to AVP-induced Ca2+ influx. The contribution of Ca2+ influx directly though TRPC6 is likely to be small because the selectivity of TRPC6 channels for Ca2+ is very low compared to Na+ [39, 40, 46].
Studies to understand the physiological role of TRPC6 channels remain a significant challenge because selective TRPC6 channel blockers and activators are not yet available. The picture is complicated further because TRPC6 is known to interact with other members of the DAG-responsive TRPC subfamily, TRPC3 and TRPC7, to form tetrameric channels [33, 37]. In A7r5 cells, TRPC6 was found to interact with TRPC7. Hence it is highly likely that TRPC6 exists as a heteromultimer in vascular smooth muscle with TRPC3, TRPC7 or perhaps other members of the TRPC family. Nonetheless, we and others  have found that knocking down TRPC6 expression in A7r5 cells results in a dramatic reduction of AVP-stimulated non-selective cation currents. Our present findings further suggest that activation of TRPC6-mediated Na+ entry contributes to the Ca2+ responses of VSMCs to physiological concentrations of AVP.
We propose that at physiological concentrations, AVP signal transduction involves production of diacylglycerol (DAG), presumably by activation of phospholipases D and C (PLD and PLC) . DAG-induced PKC activation causes suppression of KCNQ5 channel-mediated K+ efflux and robust membrane depolarization that results in Ca2+ influx through voltage-sensitive Ca2+ channels. Increased concentrations of DAG, apart from activating PKC, also directly activate TRPC6 channels. Na+ and Ca2+ influx through TRPC6 channels contribute modestly to membrane depolarization, but this contribution is sufficient to enhance the membrane depolarization produced by suppression of KCNQ5 currents. This reduces the time required to reach the membrane potential threshold for activation of voltage-sensitive Ca2+ channels and Ca2+ influx. Ca2+ influx through TRPC6 and reversal of Na+/Ca2+ exchange following influx of Na+ may also modestly contribute to the robust increase in [Ca2+]i through the voltage-sensitive Ca2+ channels (Figure 8).
In summary, from the present study we conclude that (1) KCNQ5 channels conduct K+ at resting membrane potentials in A7r5 cells. The resulting outward K+ current opposes membrane depolarization and prevents Em from reaching the threshold for triggering action potential firing; (2) Suppression of KCNQ5 channel activity by physiological concentrations of AVP will depolarize the membrane and stimulate Ca2+ influx though voltage-sensitive Ca2+ channels; (3) Activation of TRPC6 non-selective cation currents modestly contributes to AVP-induced membrane depolarization and generation of Ca2+ spikes at physiological concentrations of AVP.
A. A7r5 cells were transfected with vector containing KCNQ5 and GFP genes (left, showing green fluorescence). The anti-KCNQ5 immunofluorescence (middle, showing red fluorescence with Alexa-594 secondary antibody) was increased in cells over expressing KCNQ5. Note that the untransfected cells (no green fluorescence in left panel) in the same field have less immunofluorescence than the KCNQ5 overexpressing cells. When the same procedure was done with the absence of anti-KCNQ5 primary antibody, no immunoflourescence (red fluorescence) was detected (right). B. A7r5 cells expressing TRPC6shRNA and DsRed, mixed and replated with uninfected control A7r5 cells. DAPI staining shows presence of 3 cells (left) in the microscopic field, but only one cell expresses TRPC6shRNA (and DsRed, middle). Pre-incubating the anti-TRPC6 antibody with a TRPC6-specific blocking peptide prevented the fluorescence with anti-TRPC6 antibody (green fluorescence) in both control and TRPC6shRNA expressing A7r5 cells (right).
A. OAG-activated ICAT recorded at −74 mV holding potential (100 μM OAG applied as indicated by black bar; C = 89.5 pF). The mean current density of OAG-activated ICAT was 1.17 ± 0.59 pA/pF, n= 5. OAG-activated current was blocked by 100 μM Gd3+ (white bar). B. Representative IV curve shows ICAT recorded before (a) and during OAG application (b).
This work was supported by the National Heart Lung & Blood Institute (R01 HL070670 to KLB)
The authors gratefully acknowledge the technical support of Chris Hill.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.