Dynamic regulation of β1 subunit trafficking controls vascular contractility

β1 Subunits Are Primarily Intracellular and Can Be Stimulated to Rapidly Traffic to the Plasma Membrane in Arterial Myocytes.

Arterial surface biotinylation was used to measure the cellular distribution of BKα and β1 subunits in myocytes of intact resistance-size (∼200-μm-diameter) arteries that control regional organ blood flow and systemic blood pressure. In rat cerebral arteries, ∼95.2% of BKα subunits were plasma membrane-localized (Fig. 1A and Fig. S1C), whereas only ∼8.1% of β1 subunits were present at the cell surface (Fig. 1 A and B). A similar distribution of BKα and β1 subunits was observed in rat small mesenteric arteries (Fig. S1B) and in cerebral arteries from an 18-y-old human male collected under an institutional review board-approved protocol (Fig. S2). These data indicate that most BKα subunits are located in the arterial myocyte plasma membrane. In contrast, and contrary to expectations, only a small fraction of total β1 subunits are located at the myocyte surface.

We tested the hypothesis that β1 subunits are mobile proteins and that physiological stimuli can control β1 trafficking to regulate BK channel activity. BK channels are activated by cAMP- and cGMP-dependent protein kinases (PKA and PKG), leading to vasodilation, but involvement of β-subunit trafficking in this response is unclear (17). Surface biotinylation revealed that sodium nitroprusside (SNP) or SIN-1, NO donors, increased mean arterial surface β1 protein ∼2.7- and ∼2.5-fold, respectively, in cerebral arteries (Fig. 1 A and B). Immunofluorescence, Förster resonance energy transfer (FRET) imaging, and coimmunoprecipitation (co-IP) were used to examine BKα and β1 cellular distribution, colocalization, and spatial proximity. The mean Normalized-Förster resonance energy transfer (N-FRET) between BKα- and β1-bound secondary antibodies was ∼7.3% in control myocytes (Fig. 1 C and D). SNP (10 min) increased N-FRET to ∼18.4%, or ∼2.5-fold (Fig. 1 C and D). The Förster coefficient for the Alexa488-Alexa546 FRET pair used is ∼6.3 nm (19), suggesting that surface-trafficked β1 subunits associate with BKα. Co-IP was performed to test the hypothesis that SNP can elevate the macromolecular subunit ratio of β1 to BKα in arterial myocytes. Because of the small size of the resistance-size vessels used in this study, arteries collected from ∼six rats were required for each co-IP experiment. The BKα antibody coimmunoprecipitated β1 protein from arterial lysate (Fig. 1E). SNP increased β1 protein that coimmunoprecipitated with BKα 1.97 ± 0.08-fold (n = 3, 18 rats total; Fig. 1E). These data indicate that surface-trafficked β1 subunits associate with plasma-resident BKα channels in arterial myocytes.

NO-induced plasma membrane trafficking of β1 was rapid, being maximal within 1 min, the shortest time point investigated (Fig. 1 A and B). Carbachol, a muscarinic receptor agonist, which stimulates endothelial cell NO generation, and iloprost, a PGI₂ analog that activates myocyte PKA, both increased surface β1 protein in endothelium-intact arteries (Fig. 1B). Similar data were obtained when experiments were performed at either room temperature or 37 °C (Fig. S1A). BKα and β1 were similarly distributed and SNP stimulated β1 subunit surface expression in myocytes of rat mesenteric and human cerebral arteries (Figs. S1B and S2). ODQ, a soluble guanylyl cyclase (sGC) inhibitor, DT-2, a protein kinase G (PKG) inhibitor, or l-NNA, a NOS inhibitor, alone did not alter β1 cellular distribution, but ODQ (1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one), L-NNA (N_ω-Nitro-L-arginine) or DT-2 blocked the SNP-induced elevation in membrane β1 (Fig. 1 A and B and Fig. S1D). Forskolin, an adenylyl cyclase and thus PKA activator, increased β1 surface expression similarly to NO donors (Fig. 1 A and B). These data indicate that both NO-induced, sGC-mediated PKG activation, and adenylyl cyclase-mediated PKA activation stimulate rapid β1 surface expression in arterial myocytes.

β1 Subunits Are Located in Mobile Rab11A-Positive Recycling Endosomes.

Trafficking pathways that control β1 surface expression were investigated in arterial myocytes. Brefeldin A, an endoplasmic reticulum (ER)-to-Golgi trafficking inhibitor, and nocodazole, a Golgi disruptor, did not alter β1 distribution when applied alone (Fig. S1D). In contrast, concanavalin A (Con A), an endocytosis inhibitor, increased membrane β1 protein ∼3.6-fold (Fig. 1 A and B). Brefeldin A and nocodazole each abolished the SNP-induced increase in surface β1 (Fig. 1 A and B). Brefeldin A also inhibited the SNP-induced elevation in BKα to β1 N-FRET (Fig. 1 C and D). Immunofluorescence indicated that only ∼35% of β1 colocalized with ER-Tracker Green, an ER label, suggesting that β1 subunits reside in an intracellular compartment other than the ER or Golgi (Fig. 2A). Given that β1 surface expression was rapid (≤1 min), we tested the hypothesis that recycling endosomes control β1 trafficking. Rab GTPases modulate endosomal trafficking with either Rab11A or Rab11B expressed on recycling endosomes (20–26). Physiological functions of Rab11 in native cell types, including arterial myocytes, are unclear. Immunofluorescence experiments demonstrated that β1- and Rab11A-bound antibodies generated a mean N-FRET of ∼32.5% in myocytes, indicating close spatial proximity (Fig. 2 B and C). Vectors encoding Rab11A shRNA (Rab11A shV) reduced arterial Rab11A protein by ∼50% but did not alter BKα or β1 expression (Fig. S3 B and C). Rab11A knockdown reduced the mean N-FRET signal between rab11A and β1 from ∼32% in scrambled controls to ∼19.7% (Fig. 2 B and C). Rab11A knockdown also blocked both the SNP-induced elevation in surface β1 and the N-FRET signal between BKα and β1-bound fluorescent antibodies in myocytes (Figs. 2 D and E and 3 A and B). Overexpression of a constitutively active Rab11A mutant (Rab11AQ70L) increased surface β1 ∼1.6 fold, whereas a dominant negative Rab11A mutant (Rab11AS25N) inhibited SNP-induced β1 surface trafficking in arteries (Fig. 3 C and D). In contrast, Rab11B knockdown (by ∼70%; Fig. S4 A and B) did not alter total or surface-localized BKα or β1 protein in control or SNP-treated arteries (Fig. S4 C and D). Immunofluorescence resonance energy transfer (immuno-FRET) experiments indicated that BKα and Rab11A were not localized (Fig. S3A), in stark contrast to the close spatial proximity of β1 and Rab11A. Furthermore, NO, forskolin, brefeldin A or Rab11A knockdown did not alter BKα total protein or cellular distribution (Figs. S1 C and D and S3 B and C). These data indicate that NO stimulates rapid Rab11A-mediated anterograde trafficking of β1 subunits in arterial myocytes. In contrast, BKα is trafficked by a Rab11A- and Rab11B-independent mechanism. These data raised the possibility that regulated anterograde trafficking of β1 subunits is a functional mechanism to control surface BK channel activity in arterial myocytes.

Stimulated Surface Trafficking of β1 Subunits Elevates BK Channel Ca²⁺ Sensitivity.

β1 subunits increase BK channel Ca²⁺ sensitivity (14, 27). Patch-clamp electrophysiology was used to examine the regulation of BK channels by an elevation in surface β1 subunits in arterial myocytes. Inside-out patches were pulled from myocytes that had been exposed to a bath solution either without (control) or with SNP (10 μM, 10 min). BK channel activity was then measured in membrane patches in the absence of SNP at −40 mV, a physiological voltage. The mean apparent dissociation constant (K_d) for Ca²⁺ of BK channels from control myocytes was ∼25 µM, with a maximum P_o of ∼0.80 (Fig. 4B). In patches pulled from SNP-treated myocytes, BK channel mean K_d for Ca²⁺ was ∼11 µM or 2.3-fold higher, without an alteration in maximal P_o. Brefeldin A treatment of myocytes blocked the SNP-induced elevation in BK channel Ca²⁺ sensitivity (Fig. 4B). These data indicate that SNP stimulates surface trafficking of β1 subunits that associate with BKα subunits and increase Ca²⁺ sensitivity.

BK channel regulation by lithocholate, a selective activator of β1 subunit-containing BK channels, was also studied (28). We tested the hypothesis that lithocholate would be a more effective activator of BK channels that contain additional SNP-trafficked β1 subunits. Lithocholate increased BK channel mean P_o in inside-out patches pulled from both control cells and SNP-treated myocytes (Fig. 4C). Importantly, lithocholate elevated mean P_o 0.07 ± 0.01 in control myocytes and 0.12 ± 0.02 in SNP-treated myocytes, or 1.71-fold more, consistent with the hypothesis being tested (Fig. 4C). These data indicate that NO-stimulated surface trafficking of β1 subunits elevates BK channel Ca²⁺ sensitivity and BK channel activation by a β1 ligand. When combined with other data in this study, these data indicate that NO increases the amount of functional β1 subunits associated with BK channels.

β1 Surface Trafficking in Arterial Myocytes Dilates Cerebral Arteries.

The physiological function of regulated β1 subunit trafficking was studied using pressurized (60 mmHg) endothelium-denuded cerebral arteries. SNP (10 µM) dilated arteries on average by ∼23 µm (Fig. 5 A and B). Iberiotoxin alone constricted arteries by ∼21 µm and reduced SNP-induced vasodilation by ∼65%, indicating that BK channel activation is the primary mechanism underlying NO-induced vasodilation (Fig. 5B). Brefeldin A alone did not modify myogenic tone or iberiotoxin- or 60 mM K⁺-induced vasoconstriction but reduced SNP-induced vasodilation by ∼74% (Fig. S5C and Fig. 5 A and B). Consistent with these findings, brefeldin A reduced carbachol-induced vasodilation in endothelium-intact arteries by ∼50% (Fig. S5B). Rab11A knockdown did not alter myogenic tone but reduced SNP-induced vasodilation by ∼60%, compared with scrambled controls (Fig. 5 C and D). Lithocholate dilated control arteries on average by ∼6 μm (Fig. 5 E and F). SNP increased mean lithocholate-induced vasodilation to ∼24 μm, or fourfold (Fig. 5 E and F). These data demonstrate that NO stimulates rapid Rab11A-mediated β1 subunit surface trafficking, leading to BK channel activation and vasodilation.

Tissue Preparation.

Animal protocols were reviewed and approved by the Animal Care and Use Committee at the University of Tennessee Health Science Center. Adult male Sprague–Dawley rats (7–9 wk of age) were used for all experiments on rodents. Small (<200-μm-diameter) cerebral and mesenteric arteries were dissected and cleaned of connective tissue. Myocytes were isolated as previously described (36). A human brain sample, from which small (<200-μm-diameter) cerebral arteries were obtained, was obtained after institutional review board approval, written informed consent, and in accordance with the guidelines of the Declaration of Helsinki.

Surface Biotinylation of Intact Arteries.

Biotinylation of intact arteries was performed as previously described (2). Arteries were incubated in a mixture each of EZ-Link Sulfo-NHS-LC-LC-Biotin and Maleimide-PEG2-Biotin reagents (Thermo Fisher Scientific) for 1 h. Arteries were washed with 100 mM glycine in PBS for 15 min to remove any unbound biotin. Arteries were then washed in PBS, placed in ice-cold lysis buffer, and homogenized in lysis buffer. Cellular debris was removed by centrifugation. Total protein was then determined to allow normalization for avidin pull down of biotinylated surface proteins. Following pull-down, the supernatant comprised of the nonbiotinylated (intracellular) protein fraction. Biotinylated surface proteins were eluted from the avidin beads.

Western Blotting.

Surface and intracellular proteins were analyzed using Western blotting [7.5% (wt/vol) SDS polyacrylamide gels] and probed with either mouse monoclonal anti-BKα (1:500; NeuroMab, University of California), rabbit polyclonal anti-BKβ1 (1:500; Abcam), or mouse monoclonal anti-rab11A (1:500; Abcam) antibodies. Blots were cut at ∼50 kDa to allow simultaneous probing for both BKα and β1 subunits. Proteins were visualized using SuperSignal West Pico Chemiluminescent Substrate (Thermo Fisher Scientific). Band intensities were analyzed using Quantity One software (Bio-Rad).

Confocal Imaging and Immunofluorescence Resonance Energy Transfer Microscopy.

Fixed, permeabilized myocytes were incubated with the same anti-BKα, anti-β1, or anti-rab11A antibodies used for Western blotting experiments or ER-Tracker Green (Life Technologies). Alexa 488- or 546-conjugated secondary antibodies were used. Fluorescence images were acquired using a laser-scanning confocal microscope (LSM5 Pascal; Carl Zeiss). Percentage of weighted colocalization was calculated using Pascal system-embedded software program. For N-FRET analysis, images were background-subtracted and N-FRET was calculated on a pixel-by-pixel basis for the entire image and in regions of interest using the Xia method (37) and Zeiss LSM FRET Macro tool Version 2.5, as previously described (38).

Co-IP.

For each experiment, lysate was harvested from cerebral arteries pooled from six rats using ice-cold radioimmunoprecipitation buffer. Co-IP was performed using the Catch and Release V2.0 Coimmunoprecipitation kit (Millipore) according to the manufacturer’s instructions. Briefly, arterial lysate was incubated with control mouse IgG or BKα mouse monoclonal antibody, antibody-affinity ligand, and capture resin in the column provided. The column was centrifuged and washed, and bound proteins were released using denaturing buffer. Boiled eluate was run on a SDS/PAGE gel, and protein samples were analyzed by Western blotting using mouse monoclonal anti-BKα (NeuroMab) or rabbit polyclonal anti-BKβ1 (Abcam) and horseradish peroxidase-conjugated secondary antibodies.

Rab11A shRNA and Mutant Constructs.

A commercial Rab11A shRNA kit was obtained from OriGene Technologies. Rab11A dominant-negative (Rab11AQ70L) and constitutively active (Rab11AS25N) mutants were generated from rat rab11A cDNA obtained from a clone library (GenScript).

Transfection of Intact Cerebral Arteries.

shRNA, mutant construct, or empty vectors were transfected into arteries using either reverse permeabilization or electroporation (CUY21Vivo-SQ electroporator; Bex), as previously described (39).

Patch-Clamp Electrophysiology.

Single BK channel currents were recorded in inside-out patches from isolated cerebral artery myocytes. The pipette and bath solution both contained: 130 mM KCl, 10 mM Hepes (4-(2-Hydroxyethyl)piperazine-1-ethanesulfonic acid), 1 mM MgCl₂, 5 mM EGTA, 1.6 mM HEDTA (N-(2-Hydroxyethyl)ethylenediamine-N,N′,N′-triacetic acid, pH 7.2 with KOH). Free Ca²⁺ concentration was adjusted to between 1 and 300 µM by the addition of CaCl₂, with free Mg²⁺ maintained at 1 mM by altering MgCl₂. Free Ca²⁺ concentration in solutions was measured using Ca²⁺-sensitive (no. 476041; Corning) and reference (no. 476370; Corning) electrodes. Inside-out patch experiments were performed at a membrane voltage of −40 mV. BK currents were filtered at 1 kHz and digitized at 5 kHz. Analysis was performed offline using Clampfit 9.2 (MDS Analytical Technologies). Ca²⁺-sensitivity data were fit with an unconstrained single Boltzmann function.

Pressurized Artery Diameter Measurements.

Experiments were performed using endothelium-denuded rat cerebral arteries cannulated at each end in a perfusion chamber (Living Systems Instrumentation). Arterial diameter was measured at 1 Hz using a CCD camera. Myogenic tone (percentage) was calculated as follows: 100 × (1 – D_active/D_passive), where D_active is active arterial diameter, and D_passive is the diameter determined in the presence of Ca²⁺-free physiological saline solution supplemented with 5 mmol/L EGTA.

Statistical Analysis.

Values are expressed as means ± SEM. Student t test was used for comparing paired and unpaired data from two populations, and ANOVA with Student–Newman–Keuls post hoc test used for multiple group comparisons. P < 0.05 was considered significant.