Intravascular pressure enhances the abundance of functional K_v1.5 channels at the surface of arterial smooth muscle cells

Abstract

Voltage-dependent potassium (K_v) channels are present in various cell types, including smooth muscle cells (myocytes) of resistance-sized arteries that control systemic blood pressure and regional organ blood flow. Intravascular pressure depolarizes arterial myocytes, stimulating calcium (Ca²⁺) influx through voltage-dependent Ca²⁺ (Ca_v) channels that results in vasoconstriction and also K⁺ efflux through K_v channels that oppose vasoconstriction. We hypothesized that pressure-induced depolarization may not only increase the open probability of plasma membrane–resident K_v channels but also increase the abundance of these channels at the surface of arterial myocytes to limit vasoconstriction. We found that K_v1.5 and K_v2.1 proteins were abundant in the myocytes of resistance-sized mesenteric arteries. K_v1.5, but not K_v2.1, continuously recycled between the intracellular compartment and the plasma membrane in contractile arterial myocytes. Using ex vivo preparations of intact arteries, we showed that physiological intravascular pressure through membrane depolarization or membrane depolarization in the absence of pressure inhibited the degradation of internalized K_v1.5 and increased recycling of K_v1.5 to the plasma membrane. Accordingly, by stimulating the activity of Ca_v1.2, membrane depolarization increased whole-cell K_v1.5 current density in myocytes and K_v1.5 channel activity in pressurized arteries. In contrast, the total amount and cell surface abundance of K_v2.1 were independent of intravascular pressure or membrane potential. Thus, our data indicate that intra-vascular pressure–induced membrane depolarization selectively increased K_v1.5 surface abundance to increase K_v currents in arterial myocytes, which would limit vasoconstriction.

INTRODUCTION

Intravascular pressure stimulates membrane depolarization in small, resistance-sized arteries, leading to the activation of smooth muscle cell (myocyte) voltage-dependent calcium (Ca²⁺) (Ca_v) channels, an increase in intracellular Ca²⁺ concentration ([Ca²⁺]_i), and vasoconstriction (1, 2). Myocyte membrane depolarization also activates voltage-dependent potassium (K⁺) (K_v) channels, inducing a negative feedback mechanism that partially opposes depolarization and vasoconstriction (3–9). This “myogenic response” is an autoregulatory mechanism that modulates regional blood flow, systemic blood pressure, and sets a resting arterial diameter from which other stimuli can induce vasoconstriction or vasodilation.

K_v channels comprise a large family of ~40 proteins that are subdivided into 12 classes (K_v1 to K_v12) (10). The genes encoding several K_v channel family members are expressed, and the encoded proteins are functional in arterial myocytes, including K_v1.2, K_v1.5, K_v2.1, K_v2.2, K_v7.1, and K_v7.4 (5, 11–13). Functional channels may be composed of homomeric subunits or heteromultimeric K_v subunits, both of which are also present in arterial myocytes (7, 14). Membrane potential directly controls the open probability (P_O) of surface K_v channels to modulate arterial contractility. However, cellular K_v current amplitude (I) is dependent not only on single-channel P_O but also on the number of channels present in the plasma membrane (N) and single-channel amplitude (i), such that I = NP_Oi. Whether membrane potential regulates K_v channel surface abundance is unclear, and the role of multiple K_v channel subtypes in arterial myocytes is poorly understood. Conceivably, specific K_v channel subtypes may respond to distinct stimuli or perform specific physiological functions in arterial myocytes.

Here, we investigated the hypothesis that intravascular pressure and membrane potential modulate the abundance of functional K_v channels at the surface of myocytes of resistance-sized mesenteric arteries. Our data showed that pressure-induced membrane depolarization inhibited K_v1.5 channel degradation, stimulated K_v1.5 surface abundance, and enhanced K_v1.5 currents in arterial myocytes. In contrast, the amount of K_v2.1 at the myocyte surface was unaffected by intravascular pressure in these arteries. We also show that the pressure-induced increase of K_v1.5 channel surface abundance reduced arterial contractility. Thus, our data demonstrated that membrane depolarization activated K_v1.5 currents through two distinct pathways in arterial myocytes: (i) direct channel activation and (ii) subtype-specific enhancement of K_v channel abundance at the cell surface.

RESULTS

The gene encoding K_v1.5 is highly expressed relative to other K_v channel members in mesenteric artery myocytes

We performed reverse transcription polymerase chain reaction (PCR) on freshly isolated mesenteric artery myocytes and quantified the relative expression of six K_v channel family members that are expressed in arterial myocytes (11–13). Analysis of transcripts encoding myosin heavy chain 11 (a myocyte marker), platelet endothelial cell adhesion molecule-1 (an endothelial cell marker), and fatty acid–binding protein 4 (adipocyte marker) served as controls for the purity of the myocyte preparations. We only detected the transcript encoding myosin heavy chain 11 in the preparations, indicating that complementary DNA (cDNA) was from pure myocytes (fig. S1A). Quantitative PCR analysis indicated that of the transcripts encoding the six K_v channel subtypes measured (fig. S1B), mRNAs encoding K_v1.5 accounted for ~60%, K_v2.1 and K_v2.2 each accounted for ~15%, and K_v1.2, K_v7.1, and K_v7.4 together represented ~10% of the K_v transcripts (Fig. 1A). Because the genes encoding K_v1.5 and K_v2.1 were most highly expressed, we focused on the regulation of these channels to pressure and membrane potential in arterial myocytes.

Fig. 1. Arterial isolation leads to a decrease in total and surface K_v1.5.

(A) Quantitative real-time PCR mean data for transcripts of six K_v family members in pure arterial myocytes. n = 3 for each. Data are expressed as the percent of total K_v transcripts. (B) Representative Western blot images of arterial biotinylation samples showing the nonbiotinylated [intracellular (Intracell.)] and biotinylated (Surface) abundance of K_v1.5 and K_v2.1 in mesenteric arteries immediately after (0 hour) and 3 hours after arterial isolation. (C) Mean data showing the percent of total K_v1.5 and K_v2.1 at the cell surface for arteries immediately and 3 hours after isolation. n = 6 for each. (D) Mean data showing percent of K_v1.5 and K_v2.1 remaining at the cell surface after 3 hours relative to the 0-hour samples. n = 6 for each. *P < 0.05 versus 0 hour.

Arterial isolation leads to loss of surface K_v1.5, but not K_v2.1, channels

To measure the amount of K_v1.5 and K_v2.1 channels at the plasma membrane in arterial myocytes, we used surface biotinylation after confirming the specificity of the antibodies for these proteins when expressed in human embryonic kidney (HEK) 293 cells (fig. S2, A and B). The K_v1.5- and K_v2.1-specific antibodies did not detect any proteins in mock-transfected HEK293 cells (fig. S2, A and B). Each antibody detected proteins in arterial lysates, and the most intense bands were at similar molecular weights to their recombinant counterparts (fig. S2C). The lower molecular weight bands may be subunit cleavage products.

Freshly isolated (assayed immediately after isolation, 0 hour) mesenteric arteries had ~48% of total K_v1.5 and ~81% of total K_v2.1 channels at the cell surface (Fig. 1, B and C). We examined the time-dependent effects of the loss of intravascular pressure due to arterial isolation on the relative amounts of K_v1.5 and K_v2.1 at the myocyte surface. Three hours after arterial isolation and maintenance at 37°C, the amount of K_v1.5 at the cell surface was ~37% of that in freshly isolated (0 hour) arteries (Fig. 1, B and D). The total amount of K_v1.5 also decreased during 3 hours without pressure, but the relative amount of K_v1.5 at the cell surface compared to the total amount of K_v1.5 was unchanged (Fig. 1, B and C). In contrast, both the amount of K_v2.1 at the surface and the relative cellular distribution of K_v2.1 were unchanged after 3 hours without pressure (Fig. 1, B to D). These data indicated that arterial isolation reduced the amount of K_v1.5 at the cell surface but did not alter K_v2.1 surface protein or the relative cellular distribution of K_v1.5 or K_v2.1 channels in myocytes.

Arterial isolation leads to time-dependent decrease in total K_v1.5 but not K_v2.1

After arterial isolation, we found that surface K_v1.5 decreased, but the abundance of the protein in the intracellular fraction did not increase, suggesting that the internalized K_v1.5 was degraded. We monitored the abundance of total K_v1.5 and K_v2.1 from isolation through 18 hours to investigate the differences in stability of the proteins in depressurized vessels. One hour after arterial isolation, total K_v1.5 was similar to that in freshly isolated arteries (Fig. 2, A and B). In contrast, 3 hours after arterial isolation, total K_v1.5 was ~33% of that in the freshly isolated arteries, and this reduction was maintained for 18 hours, the longest time point measured (Fig. 2, A and B). In contrast, total K_v2.1 did not change up to 18 hours after arterial isolation (Fig. 2, A and B). Quantitative analysis of transcript abundance showed that 3 hours after arterial isolation, K_v1.5- and K_v2.1-encoding mRNAs were similar to those in freshly isolated arteries, indicating that the reduction in K_v1.5 protein was not due to a decrease in mRNA for K_v1.5 (Fig. 2C). These data indicated that, in the absence of pressure, isolated artery K_v1.5, but not K_v2.1, is internalized and degraded within 3 hours.

Fig. 2. Bafilomycin and MG132 prevent K_v1.5 loss in mesenteric arteries.

(A) Representative Western blot images of total K_v1.5 and K_v2.1at 0, 1, and 3 hours of isolation. (B) Quantification of total K_v1.5 and K_v2.1 at the indicated times after isolation. n = 5 for the 1-hour group, n = 6 for the remaining groups. *P < 0.05 versus 0 hour. (C) Quantitative PCR data expressed as the percent of transcripts for each channel remaining after 3 hours with the amount of transcript at 0 hour set at 100%. n = 3 for each. (D) Representative Western blots and quantification of total K_v1.5 and K_v2.1 0 and 3 hours after arterial isolation, and the effect of bafilomycin (Baf; 50 nM) or MG132 (10 μM). n = 6 for each. *P < 0.05 versus 0 hour. (E) Representative Western blot images of arterial biotinylation samples showing the abundance of surface and intracellular K_v1.5 and K_v2.1 at 0 and after 3 hours of arterial isolation in the presence of bafilomycin. (F) Mean data showing the percent of total K_v1.5 and K_v2.1 at the cell surface for arteries immediately and 3 hours after isolation. n = 5 for each. (G) Mean data showing percent of K_v1.5 and K_v2.1 remaining at the cell surface after 3 hours in the presence of bafilomycin relative to the 0-hour samples. n = 5 for each.

Internalized proteins are either degraded in lysosomes, by the proteasome, or both degradation pathways may contribute. Arteries were placed in physiological saline solution (PSS) containing bafilomycin, an inhibitor of lysosomal degradation, or MG132, a blocker of proteasomal degradation, immediately after isolation to determine which pathways contributed to K_v1.5 instability in the unpressurized arteries. Either inhibitor prevented the reduction in total K_v1.5, with bafilomycin more effective than MG132 (Fig. 2D). Neither inhibitor affected the amount of K_v2.1 (Fig. 2D) or the relative surface/intracellular distribution of either K_v1.5 or K_v2.1 (Fig. 2, E and F). Bafilomycin completely prevented the loss of K_v1.5 from the cell surface 3 hours after isolation and depressurized conditions (Fig. 2G). These data indicated that, after arterial isolation, myocyte K_v1.5 is degraded by both the lysosomal and proteasomal pathways. Blocking either of these intracellular degradation pathways maintained K_v1.5 at the myocyte surface.

Intravascular pressure–induced membrane depolarization prevents arterial myocyte K_v1.5 loss and degradation

We hypothesized that intravascular pressure regulates K_v1.5 degradation and surface abundance. Pressurizing mesenteric arteries to 80 mmHg for 3 hours inhibited the reduction in total K_v1.5 (Fig. 3A) and limited the loss of K_v1.5 from the cell surface (Fig. 3, B and C), when compared to 0-hour controls. In contrast, the cell surface and total amounts of K_v1.5 were reduced in arteries maintained at 10 mmHg for 3 hours compared with that in 0-hour controls (Fig. 3, A to C). Intravascular pressure did not alter total or surface K_v2.1 (Fig. 3, A to C) or the relative surface/intracellular distribution of K_v1.5 or K_v2.1 (Fig. 3, B and D). These data indicated that intravascular pressure controls K_v1.5 degradation.

Fig. 3. Intravascular pressure inhibits K_v1.5 degradation and maintains surface expression.

(A) Representative Western blots and quantitative data of total K_v1.5 and K_v2.1 in control arteries (0 hour) or arteries exposed to 10 or 80 mmHg intravascular pressure for 3 hours. (B) Representative Western blot of arterial biotinylation samples showing the abundance of surface and intracellular K_v1.5 and K_v2.1. (C) Mean data showing the percent of K_v1.5 and K_v2.1 at the cell surface compared to the amount at isolation (0 hour) after 3 hours at 10 and 80 mmHg intravascular pressure. (D) Mean data showing the percent of K_v1.5 and K_v1.2 at the cell surface as a percent of the total of each channel protein in arteries exposed to the indicated pressure for 3 hours. For all quantitative data, n = 5 for each group and *P < 0.05 versus 0 hour.

Intravascular pressure stimulates membrane depolarization (1). To investigate regulation of K_v channel degradation by membrane potential, we compared the effects of an increase in K⁺ in the PSS from 6 to 30 mM K⁺ to depolarize arteries from ~−60 to ~−40 mV. This voltage shift is similar to that which occurs in arteries when increasing pressure from 0 mmHg to physiological intravascular pressure (2, 15). Whereas the 6 mM K⁺ solution was ineffective at stabilizing K_v1.5, the 30 mM K⁺ solution prevented the isolation-induced reduction in total and surface K_v1.5 (Fig. 4, A to C), as determined by Western blotting and cell surface biotinylation. In contrast, 30 mM K⁺ did not alter total or surface amounts of K_v2.1 (Fig. 4, A to C) or the relative cellular distribution of K_v1.5 or K_v2.1 channels (Fig. 4, B and D). To test the hypothesis that K_v1.5 is trafficked between the surface and internal pools in arterial myocytes, we also analyzed the amount of K_v1.5 at the cell surface by immunofluorescence colocalization with wheat germ agglutinin, a plasma membrane marker (Fig. 4E). The 30 mM K⁺ pre- vented the isolation-induced reduction of surface K_v1.5 in myocytes (Fig. 4E). These data indicated that membrane depolarization inhibits K_v1.5 channel internalization and degradation in mesenteric artery myocytes.

Fig. 4. Membrane depolarization maintains K_v1.5 at the cell surface.

(A) Representative Western blots and quantitative data of total K_v1.5 and K_v2.1 in arteries exposed to 6 or 30 mM K⁺ PSS for 3 hours. n = 6 for each. *P < 0.05 versus 0 hour. (B) Western blot of arterial biotinylation samples showing the abundance of surface andintracellular K_v1.5 and K_v2.1 from arteries exposed to 6 or 30 mM K⁺ PSS for 3 hours. (C) Mean data showing the percent of K_v1.5 and K_v2.1 at the cell surface compared to the amount at isolation (0 hour) after 3 hours in 6 or 30 mM K⁺ PSS. n = 6 for each. *P < 0.05 versus 0 hour. (D) Mean data showing K_v1.5 and K_v2.1 at the cell surface as a percent of the total of each channel protein in arteries after 3 hours in 6 or 30 mM K⁺ PSS. n = 5 for each. (E) Immunofluorescence images and quantitative analysis of colocalization of wheat germ agglutinin (WGA) (green) and K_v1.5 (red) in myocytes isolated from arteries at 0 or 3 hours after maintenance in either 6 or 30 mM K⁺ PSS. Yellow is pixel colocalization analyzed using the RG2B plug-in for ImageJ. Scale bars, 5 μm. n = 7 for each. *P < 0.05 versus 0 hour.

Voltage-dependent Ca²⁺ channel activity controls arterial myocyte K_v1.5 channel degradation

Membrane depolarization stimulates K_v channels, which increases K⁺ current mediated by the efflux of K⁺ ions, and Ca_v channels, which permits Ca²⁺ influx, increasing [Ca²⁺]_i (1–9). To investigate contributions of intravascular pressure–induced depolarization and depolarization-induced Ca²⁺ influx to K_v1.5 degradation and surface abundance, we applied nimodipine, a Ca_v channel blocker to arteries maintained at 10 or 80 mmHg or in the presence of 6 or 30 mM K⁺ PSS. Nimodipine blocked both pressure- and depolarization-induced stabilization of K_v1.5 (Fig. 5, A and B). In arterial myocytes, depolarization-induced Ca²⁺ influx stimulates ryanodine-sensitive Ca²⁺-release (RyR) channels, leading to release of Ca²⁺ from the sarcoplasmic reticulum (16). Ryanodine, an RyR channel inhibitor, did not alter depolarization-induced stabilization of K_v1.5 (Fig. 5B). Bay K8644, a Ca_v channel activator, partially prevented isolation-induced K_v1.5 degradation that occurred after 3 hours in depressurized conditions (Fig. 5C). Nimodipine, ryanodine, and Bay K8644 did not change K_v2.1 abundance (Fig. 5, A to C). These data indicated that physiological intravascular pressure and depolarization-induced Ca_v channel activation inhibits K_v1.5 degradation in arterial myocytes.

Fig. 5. Voltage-dependent Ca²⁺ channel activity controls K_v1.5 channel degradation.

(A) Representative Western blots and quantitative data showing the amount of total K_v1.5 and K_v2.1 in arteries exposed to 10 and 80 mmHg intravascular pressure for 3 hours in the presence or absence of nimodipine (Nimo; 1 μM). n = 5 for each. *P < 0.05 versus 0 hour. (B) Western blot and quantitative data showing the amount of total K_v1.5 or K_v2.1 in arteries maintained for 3 hours in 6 or 30 mM K⁺ PSS in the presence or absence of nimodipine (1 μM) or ryanodine (Ryn; 10 μM). n = 6 for each. *P < 0.05 versus 0 hour. (C) Western blot and quantitative data showing the amount of total K_v1.5 and K_v2.1 from arteries exposed to 6 mM K⁺ PSS for the indicated times with or without Bay K8644 (BayK; 100 nM). n = 5 for each. *P < 0.05 versus 0 hour, ^#P < 0.05 versus 3 hour.

We tested the hypothesis that K_v1.5 continuously recycles and that voltage-dependent Ca²⁺ influx inhibits both internalization and degradation, thereby enhancing surface abundance. Concanavalin A, an internalization inhibitor, prevented isolation-induced degradation of K_v1.5 at 3 hours, with-out affecting K_v2.1 (Fig. 6A). Concanavalin A also increased the proportion of total K_v1.5 that was located in the plasma membrane from ~45 to 72% in 6 mM K⁺ PSS and from ~48 to 74% in 30 mM K⁺ PSS (Fig. 6B). These data suggested that K_v1.5 channels continuously recycle to and from the plasma membrane and that membrane potential does not directly control this process. Instead, Ca_v-mediated Ca²⁺ influx inhibits K_v1.5 degradation to enhance K_v1.5 surface abundance.

Fig. 6. Concanavalin A prevents K_v1.5 degradation and increases surface abundance.

(A) Representative Western blot and quantitative data showing the amount of total K_v1.5 and K_v2.1 in arteries at 0 hour, 3 hours after isolation, and 3 hours after isolation with concanavalin A (ConA; 10 μM). n = 4 for each. *P < 0.05 versus 0 hour. (B) Western blot of arterial biotinylation samples showing the abundance of surface and intracellular K_v1.5 and K_v2.1 from arteries 3 hours after isolation in 6 or 30 mM K⁺ PSS in the presence or absence of ConA. Mean data show the percentage of total K_v1.5 or K_v2.1 located at the cell surface. n = 5 for each. *P < 0.05 versus respective samples in the absence of ConA.

K_v1.5 degradation reduces whole-cell K_v1.5 currents in arterial myocytes

We analyzed the effect of K_v1.5 abundance on myocytes isolated from mesenteric arteries by patch-clamp electrophysiology using conditions designed to isolate whole-cell K_v currents and reduce contamination from other current types, including that from large-conductance Ca²⁺-activated K⁺ channels. Mean K_v current density (at +50 mV) in myocytes from arteries isolated for 3 hours was significantly lower than the K_v current density in myocytes from freshly isolated (0 hour) arteries (Fig. 7, A to C). To evaluate the portion of the total K_v current that was mediated by K_v1.5-containing channels, we applied Psora-4, a K_v1.5 channel inhibitor (17), to myocytes and monitored the Psora-4–sensitive current (Fig. 7A). Psora-4 reduced the mean current density in freshly isolated myocytes from ~25.5 to ~8.8 pA/pF (Fig. 7, B and C), suggesting that K_v1.5-containing channels carried ~65% of the current at +50 mV. In contrast, the effect of Psora-4 on the current at +50 mV in myocytes 3 hours after isolation was less, reducing the current density from ~12.6 to ~8.9 pA/pF (~29%) (Fig. 7, A to C). Bafilomycin prevented the isolation-induced reduction in total K_v current and the decrease in Psora-4–sensitive currents in myocytes (Fig. 7, A to C). We obtained similar results with diphenyl phosphine oxide-1 (DPO-1) (Fig. 7, B and C), a putative K_v1.5 channel–specific inhibitor (18, 19). These data indicated that K_v1.5-containing channels are a major contributor to whole-cell K_v currents and that isolation-induced K_v1.5 degradation reduces K_v1.5-mediated current density in arterial myocytes.

Fig. 7. Arterial isolation reduces Psora-4– and DPO-1–sensitive whole-cell K_v currents in arterial smooth muscle cells.

(A) Representative K_v current recordings and the effect of Psora-4 (100 nM) in myocytes either isolated immediately from arteries (0 hour) or 3 hours after isolation and maintenance at 37°C without (3 hour) or with 50 nM bafilomycin (3 hours + Baf). Bottom panel shows voltage step protocol and scale. (B) Mean current-voltage relationships for whole-cell (top, n = 12 for each), Psora-4 (100 nM)–sensitive (n = 5 to 6), and DPO-1 (1 μM)–sensitive (n = 5 to 6) currents at time points indicated with and without bafilomycin. (C) Mean current density values at +50 mV. *P < 0.05 versus 0 hour.

Intravascular pressure stimulates functional K_v1.5 channels in mesenteric arteries

To test the hypothesis that intravascular pressure controls the amount of functional K_v1.5 channels in arterial myocytes, we performed arterial myography on pressurized single third-order, abluminal mesenteric arteries in the presence of Psora-4 or the general K_v inhibitor 4-aminopyridine (4-AP). The arteries were maintained in a temperature-controlled perfusion chamber at pre-experimental intravascular pressures of either 10 or 80 mmHg for 3 hours. Arteries maintained at 10 mmHg pre-experimental pressure were then pressurized to 80 mmHg to stimulate the development of myogenic tone. Arteries in the 80 mmHg pre-experimental pressure group were continued at 80 mmHg, providing an equivalent time course for both experimental groups. Arteries from either the 10 or the 80 mmHg group responded to Psora-4 or 4-AP with a reduction in diameter, indicating vessel constriction (Fig. 8A). However, Psora-4 stimulated the contraction of arteries from the 80 mmHg pre-experimental pressure group by ~29 μm and from the 10 mmHg pre-experimental pressure group by ~10 μm, which was ~65% less (Fig. 8, A and B). As expected, 4-AP caused a greater reduction in diameter, consistent with a stronger constriction response when more classes of K_v channel were inhibited, than that observed for Psora-4 (Fig. 8B). However, both Psora-4 and 4-AP produced significantly greater constriction responses in the 80 mmHg pre-experimental pressuregroup, consistent with pressure maintaining K_v channels and K_v1.5-containing channels at the cell surface to counter vasoconstriction.

Fig. 8. Intravascular pressure maintains functional K_v1.5 channels in arterial myocytes.

(A) Representative diameter traces showing vasoconstriction induced by Psora-4 (100 nM) or 4-AP (1 mM) at 80 mmHg for arteries maintained for 3 hours at pre-experimental pressures indicated. (B) Mean data showing the reduction in diameter in arteries exposed to the indicated drugs and held at the indicated pre-experimental pressures. n = 5 for each. *P < 0.05 versus 10 mmHg pre-experimental pressure group.

DISCUSSION

Myogenic responsiveness requires a balance of ion channels that mediate the vasoconstriction response and those that mediate vasodilation. As critical mediators of membrane hyperpolarization, K_v channels are components that counter vasoconstriction in response to depolarization. Here, we found that membrane potential selectively controls the amount of functional K_v1.5-containing channels at the myocyte surface (N, the number of channels) to modulate K_v currents and, hence, vasoconstrictive tone in resistance-sized arteries. Quantitative PCR experiments revealed that K_v1.5-encoding mRNA accounts for most K_v channel transcripts in mesenteric artery myocytes. Our data indicated that K_v1.5 continuously recycled between the intracellular compartment and the plasma membrane and that membrane potential through Ca²⁺ signaling modulated K_v1.5 surface abundance in arterial myocytes. Loss of intravascular pressure and the associated membrane hyperpolarization stimulated degradation of internalized K_v1.5 through a mechanism dependent on both proteasomal and lysosomal pathways, resulting in a reduction in both total and plasma membrane–localized K_v1.5. We propose that physiological membrane depolarization activates Ca_v channels, which inhibited K_v1.5 degradation, enabling the internalized K_v1.5-containing channels to return to the plasma membrane. Our data indicated that the regulation of K_v1.5 degradation by intravascular pressure and membrane potential determined K_v1.5 current density and functionality. In contrast, K_v2.1, another K_v channel subtype detected in arterial myocytes, did not appear to internalize over the same time course, and total and surface proteins were not modulated by intravascular pressure, membrane potential, or Ca_v channel activity. Mechanisms that underlie the K_v1.5 specificity of this regulatory pathway may involve amino acid sequence dissimilarity of the channel proteins, differences in regulation by auxiliary subunits, or modulation by signal messengers. In summary, our data indicated that intravascular pressure selectively controls functional K_v1.5 surface abundance to regulate myocyte K⁺ currents and arterial contractility.

Membrane depolarization activates Ca_v channels in arterial myocytes, leading to an increase in [Ca²⁺]_i and vasoconstriction (1, 2). Concomitant K_v channel activation limits depolarization and, thus, Ca_v channel activity and the associated vasoconstriction (1, 2). Here, our data showed that physiological intravascular pressure and membrane depolarization inhibit K_v1.5 degradation by activating Ca_v channels. In contrast, ryanodine did not inhibit depolarization-induced K_v1.5 protection. In arterial myocytes, Ca_v channel activation leads to global, nanomolar increases in [Ca²⁺]_i, whereas ryanodine receptors generatelocalized, micromolarintracellular Ca²⁺transients, termed Ca²⁺ sparks, and propagating cytosolic Ca²⁺ waves. Thus, our data indicated that the global nanomolar increase in [Ca²⁺]_i inhibited K_v1.5 degradation, which may involve Ca²⁺-sensing proteins, such as Ca²⁺/calmodulin (20,21), that have affinities for Ca²⁺ in the nanomolar range. Evidence exists linking K_v channels and calmodulin because apocalmodulin associateswith K_v7 channels in neurons to promote trafficking to the axonal surface (22). Cells degrade proteins through lysosomal and proteasomal pathways (23, 24). Here, inhibitors of either lysosomal or proteasomal degradation stabilized K_v1.5, suggesting that lysosomal and proteasomal pathways act in series, rather than in parallel, to degrade K_v1.5 in arterial myocytes. In summary, our data indicated that membrane depolarization stimulates K_v channels both directly through voltage-dependent activation and indirectly by inhibiting K_v1.5 channel degradation.

Although the detailed mechanisms that control K_v channel trafficking in arterial myocytes are unclear, experiments with concanavalin A, an internalization inhibitor, suggested that K_v1.5 constantly recycles to the plasma membrane through a membrane potential–independent mechanism. Concanavalin A prevented the loss of 65% of total K_v1.5, suggesting that this amount of the protein may be available for internalization and either recycling to the plasma membrane or degradation. Degradation inhibitors prevented loss of both surface and total K_v1.5 proteins. Because the cell surface/intracellular ratio of K_v1.5 was similar in pressurized and nonpressurized arteries, our data suggested that membrane potential controls K_v1.5 degradation but does not appear to regulate surface trafficking or internalization in arterial myocytes. In HL-1 mouse atrial myocytes, H9c2 rat cardiac myoblasts, and HEK293 cells stably expressing K_v1.5, constitutively internalized K_v1.5 is recycled to the plasma membrane through a Rab4- and Rab11-dependent pathway, implicating involvement of both recycling and early endosomes (25, 26). In HL-1 cells exposed to oxidative stress, internalized K_v1.5 colocalized with heat shock protein 70, a chaperone that targets misfolded proteins for ubiquitylation and degradation (27, 28). Oxyhemoglobin also reduced the abundance of K_v1.5 at the cell surface through a protein tyrosine kinase–dependent endocytosis mechanism in rabbit cerebral artery myocytes (29). Future experiments should be designed to elucidate trafficking mechanisms by which membrane potential regulates K_v1.5 degradation and surface abundance in arterial myocytes.

Several studies have provided insights into trafficking as a mechanism to regulate ion channel function in arterial myocytes (30–33). Trafficking of Ca_v1.2 to the surface in arterial myocytes requires the α₂δ-1 auxiliary subunit (30). The abundance of both subunits at the cell surface is increased in rats with genetic hypertension, which increases Ca_v1.2 current density and produces a non-inactivating current that stimulates vasoconstriction (30). Pharmacological and molecular targeting of α₂δ-1 reduces Ca_v1.2 surface abundance in arterial myocytes, leading to vasodilation (30, 34). Protein kinase Cδ increases the surface abundance of TRPM4, a melastatin transient receptor potential channel, in cerebral artery myocytes (32). Rapid delivery of large-conductance Ca²⁺-activated K⁺ (BK) channel β1 subunits to the cell surface activates BK channels, leading to vasodilation (33). This pathway is activated by nitric oxide and is a mechanism by which this endothelium-derived factor activates BK channels to dilate cerebral arteries (33). Here, we showed that intravascular pressure, by controlling membrane potential, selectively enhanced the abundance of functional K_v1.5-containing channels at the myocyte surface. Myocyte K_v1 versus K_v2 function may exhibit some vascular bed specificity, suggesting that the regulated trafficking pathway described here may be more prominent in some vessels than others (3, 4, 14). Similarly, K_v channels in arterial myocytes can comprise heteromultimers containing K_v1.5 or other subunits (7, 14). Regulated K_v1.5 trafficking may not only alter the surface expression of Kv1.5 homomultimeric channels but also modulate K_v channel heteromultimer formation. Future studies should be designed to investigate these hypotheses and mechanisms of trafficking of other K_v subunits that are present in arterial myocytes (13). In atrial myocytes, increased mechanical stress activates an integrin signaling pathway, leading to an increase in functional surface K_v1.5 channels (35). Whether there is overlap in mechanisms by which pressure and shear stress increase the abundance of functional K_v1.5 at the surface of arterial and atrial myocytes is unclear. Together, these studies indicate that multiple, unique trafficking pathways, which differ with respect to time course and mechanisms involved, exist to modulate the activity of different ion channels in myocytes, thereby controlling vascular contractility.

Patch-clamp electrophysiology and myography in the presence of K_v1.5 channel blockers supported the biochemical evidence that changes in K_v1.5 surface abundance control myocyte K_v currents and pressurized artery contractility. Psora-4, a Psoralen compound, blocks K_v1.3 and K_v1.5 channels with similar low nanomolar potency but is a less effective blocker of other K_v1 family members (17). There is little to no evidence for expression of the K_v1.3-encoding gene or K_v1.3 function in contractile arterial myocytes, with studies supporting the presence of K_v1.2, K_v1.5, and K_v1.6 members of the K_v1 family in this cell type (3, 7, 17, 36, 37). Psoralens are also weak blockers of K_v2, Kir2.1, Ca_v1.2, and BK channels, which are present in arterial myocytes but at concentrations 25 to 150 times higher than those used here (8, 17, 33, 34, 38). DPO-1 binds to the open state of K_v1.5 channels, and high selectivity for K_v1.5 over I_to, IK₁, I_Kr, I_Ks, and K_v3.1 has been reported in cardiac myocytes (18, 19). Thus, the data with Psora-4 and DPO-1 indicated that changes in intravascular pressure affect the abundance of functional K_v1.5 channels at the cell surface of arterial myocytes to control K_v currents and vascular tone.

MATERIALS AND METHODS

Tissue preparation and cell isolation

Animal protocols were reviewed and approved by the Animal Care and Use Committee of the University of Tennessee Health Science Center (UTHSC). Male Sprague-Dawley rats (200 to 250 g body mass) were euthanized with sodium pentobarbital (150 mg/kg). The mesenteric artery bed was removed and placed into ice-cold PSS that contained 112 mM NaCl, 6 mM KCl, 24 mM NaHCO₃, 1.8 mM CaCl₂, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄, and 10 mM glucose, gassed with 21% O₂, 5% CO₂, and 74% N₂ (pH 7.4).

Individual third- and fourth-order mesenteric arteries (~100 to 200 μm in diameter) were cleaned of adventitial tissue in ice-cold PSS and were dissected for experiments. Arteries were maintained at 37°C for 3 hours in 6 or 30 mM K⁺ PSS in the absence or presence of bafilomycin, MG132, nimodipine, ryanodine, Bay K8644, or concanavalin A as required.

Myocytes were dissociated from mesenteric arteries in isolation solution containing 55 mM NaCl, 80 mM sodium glutamate, 5.6 mM KCl, 2 mM MgCl₂, 10 mM Hepes, 10 mM glucose, with pH adjusted to 7.4 with NaOH, with papain (0.7 mg/ml), dithioerythreitol (1 mg/ml), and bovine serum albumin (BSA) (1 mg/ml) for 15 to 20 min at 37°C. Arteries were then immediately transferred to isolation solution containing collagenase F (0.66 mg/ml), collagenase H (0.33 mg/ml), BSA (1 mg/ml), and 100 μM CaCl₂ at 37°C for 5 to 10 min. Arteries were washed in ice-cold isolation solution and triturated with a fire-polished glass Pasteur pipette to yield single myocytes.

Cell culture and transfection

HEK293 cells were maintained in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin under standard tissue culture conditions (37°C, 21% O₂/5% CO₂). HEK293 cells were transiently transfected using Effectene (Qiagen) with pRBG4 encoding full-length rat K_v1.5 [a gift from J. Trimmer, University of California (UC) Davis], pEGFP-N1 encoding full-length mouse K_v1.5 tagged with green fluorescent protein (a gift from J. Nerbonne, Washington University, St. Louis), or pCMV6 encoding full-length mouse K_v2.1 (a gift from K. O’Connell, UTHSC, Memphis). Cells were used 72 hours after transfection.

Quantitative real-time PCR

Total RNAwas prepared from ~500 isolated myocytes, individually selected using an enlarged patch pipette and a microscope as we have described previously (39), or from whole mesenteric arteries. First-strand cDNA was synthesized through reverse transcription PCR with AffinityScript Multiple Temperature Reverse Transcriptase (Stratagene). Quantitative PCR was then performed with TaqMan hydrolysis probes (Roche) on a Roche Light-Cycler 480 thermal cycler using the following protocol: 95°C for 5 min, then 40 cycles of 95°C for 10 s, 60°C for 30 s, and 72°C for 10 s. Each PCR contained 5 μl of TaqMan LightCycler 480 Master Mix, 0.1 μl of forward primer, 0.1 μl of reverse primer, 0.1 μl of the hydrolysis probe, 2 μl of cDNA template, and 2.7 μl of ribonuclease-free water. A negative control lacking cDNA template was performed in each experiment. Relative mRNA abundance was calculated as the difference in threshold (C_t) values (ΔC_t) of each K_v isoform using the ΔC_t method. Fold changes in K_v1.5 and K_v2.1 mRNA were calculated as 100 × (2^{− ΔΔC_t}), where ΔΔC_t is the difference between the ΔC_t values. Myh11 was used as the reference gene. Primer sequences and probes were as follows: K_v1.2: GGAAGAGAACCTCAGCTCCTG (forward), AGACCCAGAGCCTTCTGTGA (reverse), probe #75; K_v1.5: AGGCTCCTCAGGATGCAG (forward), GCAACCCGGAGATGTTTATG (reverse), probe #9; K_v2.1: TGCTGTGCAGAGAAGAGGAA (forward), GGACGATGAACATGATCGAG (reverse), probe #120; K_v2.2: TGCCCAGAGAAAAGGAAGAA (forward), TAGACACGATGGCCAGGAT (reverse), probe #67; K_v7.1: GCGAGATGTCATCGAGCAG (forward), TTCCCGATGGACTGATCC (reverse), probe #81; K _v 7.4: TGTTGGGATCGGTGGTCTA (forward), GAAGATGAGCACCAG-GAACC (reverse), probe #17; and Myh11: CCTGCTAGTCCACCCCAGTA (forward), ACTGAGCTGCCCTTTCTGTG (reverse), probe #119. PCR products were separated on 1.5% agarose gels.

Surface biotinylation

Mesenteric arteries were incubated for 1 hour at 4°C with EZ-Link Sulfo-NHS-LC-LC-Biotin and EZ-Link Maleimide-PEG2-Biotin (1 mg/ml each) (Thermo Scientific) in phosphate-buffered saline (PBS) (Gibco). Unbound biotin was quenched with glycine, and arteries were washed with PBS. Arteries were then homogenized in radioimmunoprecipitation assay (RIPA) buffer with 1% Triton X, and biotinylated surface proteins were separated using avidin beads in spin columns. Biotinylated proteins were eluted from the avidin beads by boiling in SDS buffer with 5% 2-mercaptoethanol. The entire biotinylated and nonbiotinylated samples from each preparation were loaded onto separate lanes in the same gel and probed using Western blotting.

Western blotting

Protein lysates were prepared in RIPA buffer with 1× SDS and 5% 2-mercaptoethanol, separated on 7.5% SDS–polyacrylamide gels, and transferred onto nitrocellulose membranes. Membranes were blocked with 5% milk, then incubated with primary antibodies for K_v1.5, K_v2.2 (1:1000; NeuroMab, UC Davis), or actin (1:10000; Millipore) overnight at 4°C in tris-buffered solution with 0.1% Tween (TBS-T) and 5% nonfat dry milk. After washing with TBS-T, membranes were incubated with horseradish peroxidase–conjugated secondary antibodies for 1 hour and washed again. Membranes were then developed using enhanced chemiluminescence (Thermo Scientific) with a Kodak In Vivo F Pro Imaging System. Band intensity was analyzed using ImageJ software.

Mesenteric artery bed pressurization

To investigate the regulation of K_v channel total and surface protein by intravascular pressure, a section of intact mesenteric vasculature was used. A freshly isolated, single superior mesenteric artery was cannulated in a temperature-controlled perfusion chamber (Living Systems Instrumentation), and downstream fourth-order arteries were occluded using cotton ties to create a closed, no-flow vascular network. Intra-vascular pressure was controlled using an attached reservoir and monitored using a pressure transducer. Arterial pressure was maintained at either 10 or 80 mmHg for 3 hours, after which, third- and fourth-order arteries were dissected and harvested from the preparation to evaluate total and surface protein expression.

Immunofluorescence

Isolated myocytes were plated on poly-L-lysine–coated coverslips and fixed immediately (0 hour) or after 3 hours in either 6 or 30 mM K⁺ PSS at 37°C. Myocytes were labeled with Alexa 488–conjugated wheat germ agglutinin (Invitrogen), a plasma membrane stain, before permeabilization with 0.1% Triton X-100. Myocytes were blocked with 5% BSA in PBS and incubated with the same K_v1.5-specific antibody used for Western blotting experiments at 1:100 in 0.5% BSA in PBS overnight at 4°C. Myocytes were washed with PBS and incubated with Alexa 546–conjugated secondary antibody. After washing, coverslips were secured to slides using mounting media (1:1 PBS/glycerol), and images were acquired using a laser scanning confocal microscope (LSM Pascal, Carl Zeiss). Alexa 488 and 546 were excited at 488 and 543 nm with emission detected at 505 to 530 and ≥560 nm, respectively. Pixel colocalization was measured using weighted colocalization coefficient values with Zeiss Pascal system embedded software. The RG2B Colocalization plugin for ImageJ was used to isolate colocalized pixel data with automatic selection threshold values and express the data as the average of the corresponding red and green channels.

Patch-clamp electrophysiology

Patch-clamp electrophysiology was performed on isolated myocytes that were allowed to adhere to a glass coverslip for 15 min before experimentation. Whole-cell currents were measured using an Axopatch 200B amplifier and Clampex 9.2 (Axon Instruments). K_v currents were activated from a holding potential of −70 mV by applying stepwise depolarizations to between −60 and +50 mV. Bath solution contained 120 mM NaCl, 3 mM NaHCO₃, 4.2 mM KCl, 1.2 mM KH₂PO₄, 0.5 mM MgCl₂, 1.8 mM CaCl₂, 10 mM glucose, 1 mM tetraethylammonium, and 10 mM Hepes, with pH adjusted to 7.4 with NaOH. Pipette solution contained 110 mM potassium gluconate, 30 mM KCl, 0.5 mM MgCl₂, 5 mM Na₂ATP, 1 mM guanosine triphosphate, 10 mM EGTA, and 5 mM Hepes, with pH adjusted to 7.2 with KOH. K_v currents were digitized at 5 kHz and filtered at 1 kHz. Offline analysis was performed using Clampfit 9.2 (MDS Analytical Technologies).

Pressurized artery segment myography

Experiments were performed using individual endothelium-denuded, third-order mesenteric artery segments (0.5 to1 mm in length and 150 to 200 μm in diameter) cannulated at each end in a temperature-controlled perfusion chamber (Living Systems Instrumentation). Intravascular pressure was modulated using a reservoir and monitored by a pressure transducer. Pressure was maintained at either 10 or 80 mmHg for 3 hours. Arteries maintained at 10 mmHg pre-experimental pressure were then pressurized to 80 mmHg to stimulate the development of myogenic tone. Arteries in the 80 mmHg pre-experimental pressure group were continued at 80 mmHg, providing an equivalent time course for both experimental groups. Arterial diameter was measured using calibrated IonWizard edge detection software (IonOptix).

Statistical analysis

All datawere analyzed using paired or unpaired Student’s t test or analysis of variance (ANOVA) with Newman-Keuls post hoc test using GraphPad Prism 5. All data are expressed as means ± SEM. P < 0.05 was considered significant.