Endothelial-dependent dilation following chronic hypoxia involves TRPV4-mediated activation of endothelial BK channels

Abstract

Following chronic hypoxia (CH), the systemic vasculature exhibits blunted vasoconstriction due to endothelial-dependent hyperpolarization (EDH). Previous data demonstrate that subsequent to CH, EDH-mediated vasodilation switches from a reliance on SK_ca and IK_ca channels to activation of the endothelial BK_ca channels (eBK). The mechanism by which endothelial cell stimulation activates eBK channels following CH is not known. We hypothesized that following CH, EDH-dependent vasodilation involves a TRPV4-dependent activation of eBK channels. ACh induced concentration-dependent dilation in pressurized gracilis arteries from both normoxic and CH rats. Inhibition of TRPV4 (RN-1734) attenuated the ACh response in arteries from CH rats but had no effect in normoxic animals. In the presence of L-NNA and indomethacin, TRPV4 blockade attenuated ACh-induced vasodilation in arteries from CH rats. ACh elicited endothelial TRPV4-mediated Ca²⁺ events in arteries from both groups. GSK1016790A (GSK101, TRPV4 agonist) elicited vasodilation in arteries from normoxic and CH rats. In arteries from normoxic animals, TRAM-34/apamin abolished the dilation to TRPV4 activation, whereas luminal iberiotoxin had no effect. In CH rats, only administration of all three K_ca channel inhibitors abolished the dilation to TRPV4 activation. Using Duolink®, we observed co-localization between Cav-1, TRPV4, and BK channels in gracilis arteries and in RAECs. Disruption of endothelial caveolae with methyl-β-cyclodextrin significantly decreased ACh-induced vasodilation in arteries from both groups. In gracilis arteries, endothelial membrane cholesterol was significantly decreased following 48 h of CH. In conclusion, CH results in a functional coupling between muscarinic receptors, TRPV4 and K_ca channels in gracilis arteries.

Keywords: BK channel; Chronic hypoxia; Endothelium; TRPV4.

Fig. 1

Endothelial membrane cholesterol was quantified using filipin staining in aortic endothelial cell sheets from normoxic controls and animals exposed to either 48 h or 4 weeks of CH. Sample size is indicated by the number in parentheses. *Different from normoxia (P < 0.05). ^#Different from 48 h CH (P < 0.05)

Fig. 2

Role of caveolae in Ach-induced vasodilation was assessed in gracilis arteries from normoxic and chronic hypoxic (CH) rats. Arteries were pretreated with N-nitro-L-arginine (100 μM, L-NNA) and indomethacin (10 μM, INDO). A subset of arteries from both groups was treated intraluminally with 10 mM methyl-β-cyclodextrin (MbCD) for 30 min. Representative traces of the ACh-induced vasodilation in gracilis arteries from normoxic and CH rats pretreated with either vehicle or luminal MβCD (a). N = 5/group. *Different from Veh within groups (P < 0.05). ^#Different from Veh between groups

Fig. 3

Representative traces of aceylcholine (ACh)-induced vasodilation in the presence of vehicle (black lines), RN-1734 (gray lines), or L-NNA and indomethacin (dark gray lines) in gracilis arteries from normoxic and CH rats. ACh-induced dilation was assessed in arteries from normoxic and CH rats plus and minus the TRPV4 antagonist, RN-1734 (30 μM) under control conditions [a and b] or in the presence of N-nitro-L-arginine (100 μM, L-NNA) and indomethacin (10 μM, INDO) [c and d]. Sample size is indicated by the number in parentheses. ^↑Indicates approximate time point for increasing concentrations of ACh administration. *Different from RN-1734 (P < 0.05). ^†Different from L-NNA/INDO/RN-1734 (P < 0.05). ^#Different from L-NNA/INDO (P < 0.05)

Fig. 4

TRPV4-dependent vasodilation in gracilis arteries. Representative traces of GSK101-induced vasodilation in endothelium intact (gray lines) and denuded (black lines) gracilis arteries from normoxic rats [a]. Summary of the GSK1016790A (GSK101)-induced dilation under endothelium intact and denuded conditions [b]. GSK101 elicits a concentration-dependent vasodilation in arteries from normoxic and CH rats. TRPV4-dependent vasodilation was inhibited by disruption of the endothelium in arteries from both groups. n = 5/group. *Different from endothelium intact within groups (P < 0.05). ^#Different from 10 nM GSK101 within groups (P < 0.05)

Fig. 5

Role of K_ca channels in TRPV4-mediated vasodilation was assessed in gracilis arteries from normoxic and CH rats. Arteries were pretreated with N-nitro-L-arginine (100 μM, L-NNA) and indomethacin (10 μM, INDO). GSK1016790A (GSK101)-induced dilation was examined in artery segments pretreated with either vehicle, the BK channel inhibitor, iberiotoxin administered in the lumen (100 nM, IBTx), the IK and SK channel blockers, TRAM-34 (1 μM) and apamin (100 nM), respectively, or the combination of all three K_ca channel inhibitors. Sample size is indicated by the number in parentheses. *Different than vehicle

Fig. 6

Representative confocal images of the protein expression of Cav-1 [a], TRPV4 [b], and BK channels [c] in gracilis arteries of control rats. Alpha actin (green), protein of interest (red), and internal elastic lamina (blue)

Fig. 7

Co-localization of TRPV4, eBK channels, and Cav-1 in the endothelium of gracilis arteries from normoxic controls and CH rats. a Representative confocal images of the Duolink PLA interactions (black puncta). The number of puncta within the borders of the internal elastic lamina was quantified and normalized to the number of observed endothelial cell nuclei (n = 5 rats/group). For negative controls [b], tissue sections were incubated with each primary alone and both PLA probes. Nuclei are labeled with SYTOX. The internal elastic lamina was stained with 633 hydrazide. Bar = 10 μm

Fig. 8

Co-localization of TRPV4, eBK channels, and Cav-1 in RAEC. a Representative confocal images of the Duolink PLA interactions (black puncta). For negative controls [b], RAEC were incubated with each primary alone and both PLA probes. Nuclei are labeled with SYTOX

Fig. 9

Representative image of a gracilis artery loaded with Ca²⁺ indicators [a]. Total basal and ACh-induced Ca²⁺ events [b]. Arteries were pretreated with N-nitro-L-arginine (100 μM, L-NNA) and indomethacin (10 μM, INDO). Endothelial cells were selectively loaded with the fluorescent calcium indicators fluo-4 and Oregon Green 488 BAPTA-1. 600 images/artery were collected at 50–60 Hz using a spinning disk confocal microscope (× 40 oil). ACh-induced Ca²⁺ events were assessed in arteries pretreated with vehicle or the TRPV4 antagonist GSK2193874 (300 nM). *Different from baseline or baseline GSK219 within group (P < 0.05). ^#Different from ACh within group (P < 0.05). ^&Different from normoxic baseline and normoxic baseline GSK219 (P < 0.05). ^%Different from normoxic ACh (P < 0.05). N = 5–6/group

Fig. 10

In arteries from normoxic animals, EDHF-dependent dilation does not involve activation of TRPV4 channels. Muscarinic receptor activation does not appear to elicit TRPV4-mediated calcium events in the endothelium of arteries from normoxic animals. However, intact caveolae appear to be required for ACh-induced dilation in arteries from normoxic and CH rats. Direct activation of TRPV4 with GSK1016790A elicits an SK/IK-dependent dilation, suggesting that TRPV4 channels are functionally available in the endothelium. In arteries from animals exposed to CH, endothelial membrane cholesterol is reduced and EDHF-mediated dilation is partially dependent on TRPV4 channels that activate SK, IK, and BK channels. Muscarinic receptor activation increases TRPV4-dependent Ca²⁺ events. Transient receptor potential channel V4 (TRPV4), acetylcholine (ACh), large conductance Ca²⁺-activated K⁺ channel (BK), intermediate conductance Ca²⁺-activated K⁺ channel (IK), small conductance Ca²⁺-activated K⁺ channel (SK), muscarinic receptor (M), inositol trisphosphate (IP₃), IP₃ receptor (IP₃R), endoplasmic reticulum (ER), phospholipase C (PLC). TRPV4 agonist (GSK1016790A)