TRPV4-DEFICIENT MICE EXHIBIT IMPAIRED ENDOTHELIUM-DEPENDENT RELAXATION INDUCED BY ACETYLCHOLINE IN VITRO AND IN VIVO

Abstract

Agonist-induced Ca²⁺ entry is important for the synthesis and release of vasoactive factors in endothelial cells. The transient receptor potential vanilloid type 4 (TRPV4) channel, a Ca²⁺-permeant cation channel, is expressed in endothelial cells and involved in the regulation of vascular tone. Here we investigated the role of TRPV4 channels in acetylcholine-induced vasodilation in vitro and in vivo using the TRPV4 knockout mouse model. The expression of TRPV4 mRNA and protein was detected in both conduit and resistance arteries from wild-type mice. In small mesenteric arteries from wild-type mice, the TRPV4 activator 4α-phorbol-12,13-didecanoate increased endothelial [Ca²⁺]_i in situ, which was reversed by the TRPV4 blocker ruthenium red. In wild-type animals, acetylcholine dilated small mesenteric arteries that involved both nitric oxide (NO) and endothelium-derived hyperpolarizing factor(s) (EDHF). In TRPV4-deficient mice, the NO component of the relaxation was attenuated and the EDHF component was largely eliminated. Compared to their wild-type littermates, TRPV4-deficient mice demonstrated a blunted endothelial Ca²⁺ response to acetylcholine in mesenteric arteries, and reduced NO release in carotid arteries. Acetylcholine (5 mg/kg, iv) decreased blood pressure by 37.0±6.2 mmHg in wild-type animals but only 16.6±2.7 mmHg in knockout mice. We conclude that acetylcholine-induced endothelium-dependent vasodilation is reduced both in vitro and in vivo in TRPV4 knockout mice. These findings may provide novel insight into mechanisms of Ca²⁺ entry evoked by chemical agonists in endothelial cells.

Introduction

A variety of agonists such as acetylcholine, bradykinin and even mechanical stimuli induce a rapid increase in endothelial Ca²⁺, leading to the synthesis and release of relaxing factors, including nitric oxide (NO), prostacyclin, and endothelium-derived hyperpolarizing factor(s) (EDHF) (1). In endothelial and other mammalian cells, the Ca²⁺ increase is usually a consequence of Ca²⁺ release from intracellular stores of the endoplasmic reticulum, and Ca²⁺ influx through Ca²⁺ permeable cation channels in the plasma membrane via store-operated or receptor-operated mechanisms (2). The influx of Ca²⁺ from the extracellular space contributes to the sustained increase of the cytosolic Ca²⁺ concentration. Despite the importance of calcium entry in the synthesis of endothelial relaxing factors, the proximate cause of this critical signaling event remains elusive.

The discovery of transient receptor potential (TRP) channels provides new insights into potential mechanisms of Ca²⁺ entry in endothelial cells. TRP channel-mediated Ca²⁺ entry has been implicated in diverse responses including changes in vascular permeability, angiogenesis, vascular remodeling and vasorelaxation (3,4). Of many subtypes of TRP channels expressed in endothelial cells, TRP vanilloid type 4 (TRPV4) channels have received increasing attention. These channels are widely expressed in vascular endothelial cells of several species, and activated by both chemical and physical stimuli, including hypotonic cell swelling (5,6), moderate heating (>27°C) (7,8), shear stress (9), the synthetic phorbol-derivative 4α-phorbol-12,13-didecanoate (4αPDD) (10), as well as arachidonic acid and its metabolites (11, 12). The TRPV4 channel has also been implicated in the release of endothelial derived relaxing factors and regulation of vascular tone (13,14,15,16).

Study of TRP channels in endothelial cells has been challenging due to the lack of specific channel blockers and co-expression of multiple TRP channels in the endothelium. Recently two lines of TRPV4 deficient mice have been generated and found to exhibit phenotypic changes in several body systems such as altered regulation of systemic tonicity, defects in the alveolar barrier, deficits in renal tubular K⁺ secretion, and blunted arterial shear response (15,16,17,18,19,20). Using this TRPV4^−/− mouse model, the present study examined the role of TRPV4 channels in agonist-induced endothelial Ca²⁺ signaling and endothelium-dependent vasodilation. Both in vitro and in vivo vascular responses were examined.

Methods

An expended Methods section is available in the online data supplement at http://hyper.ahajournals.org.

Animals

Fifty-two male TRPV4 knockout (TRPV4^−/−) (18) and sixty male wild-type (WT) C57BL/6J mice at 2–4 months of age were used in this study. All experiments were conducted in accordance with the Institutional Animals Care and Use Committee guidelines.

RNA extraction and RT-PCR

Total RNA from vascular tissues was extracted with TRIzol, and cDNA was synthesized, followed by PCR amplification of TRPV4 and PECAM-1 fragments using gene-specific primers.

Western blot analysis

Protein samples (20 µg) were subjected to 10% SDS-PAGE, and membranes were blotted with a polyclonal antibody against TRPV4 (1:1000 dilution, MBL International), followed by peroxidase-conjugated secondary antibodies. To ensure equal protein loading, the blots were reprobed with a polyclonal anti-endothelial NO synthase (eNOS) antibody (1:1000 dilution, BD Transduction Laboratories).

Immunohistochemistry

Frozen tissue sections were incubated with a polyclonal antibody against TRPV4 (1:100 dilution, Alomone Labs), followed by a goat anti-rabbit IgG conjugated with Alexafluor 568. Images were captured using a regular fluorescence microscope.

Measurement of intracellular Ca²⁺ ([Ca²⁺]_i)

Endothelial [Ca²⁺]_i was measured in situ in freshly isolated mesenteric arteries using Fura-2 as we described previously (21).

Measurement of endothelial NO

The fluorescent NO indicator 4-amino-5-methylamino-2',7'-difluorofluorescein diacetate (DAF-FM DA) was used to measure endothelial NO in situ in freshly isolated carotid arteries (21).

Isometric tension recording

Small mesenteric arteries (1^st-order branch from superior mesenteric artery, ~200 µm) were dissected, and mounted in a wire myograph as previously described (22).

Measurement of vascular responses in vivo

TRPV4^−/− and WT mice were anesthetized with 12% urethane (1.2 g/kg body weight, ip) or ketamine/xylazine (50 mg/kg/10 mg/kg, ip). The right common carotid artery was cannulated for measurement of arterial blood pressure, and the tail vein for drug administration. Heart rate was monitored by ECG at V6 position. All drugs were given as a single iv bolus, including acetylcholine (15 µg/kg), 4α-PDD (1 µg/kg), phenylephrine (1 mg/kg), sodium nitroprusside (5 mg/kg).

Data analysis

Data are presented as mean ± SEM. Significant differences between mean values were evaluated by Student t test or ANOVA followed by the Student-Newman-Keuls multiple comparison test. A value of p<0.05 was considered statistically significant.

Results

TRPV4 expression in conduit and resistance arteries

The loss of TRPV4 gene in TRPV4^−/− mice was confirmed by genotyping with PCR amplification of genomic DNA (Figure 1A). TRPV4 transcripts and proteins were detected in aorta, carotid and mesenteric arteries of WT but not TRPV4^−/− mice (Figure 1B and 1C). The TRPV4 antibody detected two bands of ~95 and ~110 kDa in WT mice. The 95kDa band is in good agreement with the calculated molecular weight of unprocessed TRPV4 protein (98 kDa). The 110kDa presumably represents the glycosylated form of TRPV4 protein (23). Immunohistochemical analysis revealed a strong staining for TRPV4 in the endothelium of WT carotid sections (Figure 1D).There was much less immunofluoresence in underlying smooth muscles. Hematoxylin and eosin (HE) staining confirmed an intact vascular structure of tissue sections from WT and TRPV4^−/− mice.

Figure 1.

TRPV4 channel expression in conduit and resistance arteries of wild-type (WT) and TRPV4^−/− (KO) mice. A, Targeted disruption of TRPV4 gene was confirmed by genotyping. PCR amplification of genomic DNA was performed using specific primers for TRPV4 and neomycin selection cassette. B, RT-PCR analysis of TRPV4 mRNA in aorta, carotid and mesenteric arteries of WT and TRPV4^−/− mice. Amplification of PECAM-1, an endothelial marker, was performed in parallel as a control. C, Western blot analysis of TRPV4 protein expression in aorta, carotid and mesenteric arteries of WT and TRPV4^−/− mice. Equal loading of proteins was assessed by reprobing with an antibody against the endothelial marker eNOS. N.S.: non-specific band. D, Immunohistochemical analysis of TRPV4 expression in aorta of WT and TRPV4^−/− mice (lower panels). Upper panels: hematoxylin and eosin (HE) staining. Scale bar = 5 µm.

We examined TRPV4-mediated Ca²⁺ response in the endothelium in situ of isolated mesenteric arteries. As shown in figure 2, infusion of 4α-PDD (1 µmol/L), a specific TRPV4 channel opener, elicited a rapid increase in [Ca²⁺]_i in endothelial cells of WT mice (Δ[Ca²⁺]_i, 105.8±13.2 nmol/L). This response was rapidly reversed by the addition of ruthenium red (10 µmol/L), a TRPV4 channel blocker (Δ[Ca²⁺]_i, 30.4±3.3 nmol/L). Preincubation of arteries with ruthenium red also prevented endothelial Ca²⁺ response to 4αPDD, whereas ruthenium red itself had no significant effect on basal [Ca²⁺]_i in WT or TRPV4 ^−/− mice (data not shown). 4α-PDD did not induce significant Ca²⁺ influx in mesenteric arteries of TRPV4^−/− mice (Δ[Ca²⁺]_i, 11.2±1.0 nmol/L). Removal of the endothelium at the end of experiments abolished the fluorescence, confirming that the measured fluorescence is specific to endothelial cells.

Figure 2.

TRPV4-mediated Ca²⁺ responses in endothelial cells in situ of mouse mesenteric arteries. A, Representative images of endothelial Ca²⁺ measured with the fluorescence Ca²⁺ indicator Fura-2. Mesenteric arteries of WT and TRPV4^−/− (KO) were isolated and cannulated, and 4α-PDD (1 µmol/L), a specific TRPV4 agonist, was infused into the lumen of arteries, followed by addition of the TRPV4 blocker ruthenium red (RuR; 10 µmol/L) into the bath. The endothelium was removed [EC(−)] at the end of experiments to confirm that the fluorescence is specific to endothelial cells. B, Representative traces of Ca²⁺ response to 4α-PDD and/or ruthenium red in WT and TRPV4^−/− mice. C, Summarized data of endothelial [Ca²⁺]_i increase. * P<0.05 vs. 4α-PDD in WT (n=5–7 mice, 20–30 endothelial cells/each vessel).

TRPV4 in agonist-induced vasodilation in vitro

In mouse mesenteric arteries, acetylcholine elicited concentration-dependent relaxations (maximal dilation, 93.3±2.2%; −logEC50, 7.7±0.1; Figure 3A). Pretreatment of arteries with a NOS inhibitor, N^G-nitro-L-arginine methyl ester (L-NAME) markedly inhibited acetylcholine-induced relaxations (maximal dilation, 34.8±4.7%; −logEC50, 6.7±0.3). The addition of the cyclooxygenase inhibitor indomethacin had no further effect (maximal dilation, 40.8±4.7%; −logEC50, 6.3±0.5). The residual dilation in the presence of L-NAME and indomethacin was abolished by high K⁺ (maximal dilation of 5.3±9.3%). These results confirm the involvement of both NO and K⁺ channels (or EDHF) in acetylcholine-induced relaxations.

Figure 3.

Role of TRPV4 channels in agonist-induced vasodilation in vitro. A, Acetylcholine-induced vasodilation in mesenteric arteries of WT mice. Arteries were preincubated with in the absence and presence of L-NAME (100 µmol/L) or L-NAME plus indomethacin (10 µmol/L), and precontracted with U46619 or high K⁺ (60 mmol/L). * P<0.05 vs. control (n=5–14 rings from 3–5 mice). B, Acetylcholine-induced vasodilation in mesenteric arteries of WT and TRPV4^−/− (KO) mice in the absence and presence of L-NAME. * P<0.05 vs. WT control; # P<0.05 vs. KO control; n=10–14 rings from 6 mice. C, Endothelium-independent vasodilation to papaverine in mesenteric arteries of WT and TRPV4^−/− mice (n=4 rings from 4 mice).

Compared to WT mice, acetylcholine-induced vasodilation was significantly reduced in TRPV4^−/− mice (maximal dilation, 49.9±8.3% vs. 88.1±3.5% for WT; −logEC50, 6.6±0.3 vs. 7.2±0.1 for WT; Figure 3B). L-NAME largely eliminated acetylcholine-induced vasodilation in TRPV4^−/− animals (maximal dilation of 9.0±2.3%). Endothelium-independent dilation to papaverine was similar in TRPV4^−/− and WT mice (maximal dilation of 94.8±1.1% and 98.8±0.6%, respectively) (Figure 3C). Furthermore, there was no difference in contractile responses to U46619 or high K⁺ between those animals (data not shown). 4α-PDD (1 µmol/L) also induced marked relaxations of intact but not denuded mesenteric arteries, with maximal relaxations of 87.9±5.0% and 1.0±2.5%, respectively (n=4 vessels from 4 mice).

Blood pressure response to agonists

Resting arterial pressures and heart rates were similar in TRPV4^−/− and WT mice, with mean values of 82.5±3.5 versus 92.3±8.3 mmHg (P=0.2; n=14 and 10, respectively), and 497±28 versus 499±31 beats/min (p=NS), respectively. The systolic and diastolic blood pressures in TRPV4^−/− and WT mice were 69.3±4.8 and 99.5±2.9 mmHg, and 75.1±5.6 and 112.6±11.9 mmHg, (p=NS for both) respectively. Intravenous acetylcholine acutely reduced blood pressure in WT mice (mean change of 37.0±6.2 mmHg 3–5 min after acetylcholine injection; Figure 4A). This response was significantly blunted in TRPV4^−/− mice (mean reduction of 16.6±2.7 mmHg). Acetylcholine produced a similar drop in heart rate in TRPV4^−/− and WT mice (mean changes of 216±48 and 224±61 beats/min, respectively) (Figure 4B). When animals are matched for baseline blood pressure (>80 mmHg), the acetylcholine-induced reduction in blood pressure was also significantly lower in TRPV4−/− versus WT animals, with mean decreases of 19.1±4.0 and 38.0±7.4, respectively. Baseline MAPs were found to be 91.0±3.3 and 102±5.3 mmHg for matched TRPV4^−/− and WT mice, respectively (P=0.1; n=8 for both).

Figure 4.

Role of TRPV4 channels in acetylcholine (Ach)-induced vasodilation in vivo. A, Mean arterial blood pressure (MAP) in WT and TRPV4^−/− (KO) mice in response to acetylcholine injection. Acetylcholine was administered as a single iv bolus (15 µg/kg). * P<0.05 vs. MAP value before acetylcholine injection; # P<0.05 vs. MAP change in WT mice; n=10 and 14 for WT and TRPV4^−/−, respectively. B, Heart rate (HR) in WT and TRPV4^−/− mice in response to acetylcholine injection. * P<0.05 vs. HR value before acetylcholine injection. C, MAP responses to 4α-PDD (1 µg/kg, iv) in WT and TRPV4^−/− mice. * P<0.05 vs. WT; n=4–5 mice.

4α-PDD transiently lowered blood pressure in WT but not in TRPV4^−/− mice, with MAP changes of 18.8±7.1 and −3.8±1.7 mmHg, respectively (Figure 4C). Phenylephrine caused similar blood pressure increases in TRPV4^−/− and WT mice (mean changes of 31.5±3.7 and 38.5±3.8 mmHg, respectively). Nitroprusside similarly reduced blood pressure in TRPV4^−/− and WT animals (mean changes of 55.5±7.2 and 61.7±5.5 mmHg, respectively) (n=6–8).

TRPV4 in acetylcholine-induced Ca²⁺ and NO increase

Acetylcholine induced a rapid increase in endothelial [Ca²⁺]_i of mesenteric arteries from WT mice, with [Ca²⁺]_i changes of 47.8±4.4 and 31.2 ±4.4 nmol/L at peak and 1 min after peak, respectively (Figure 5). Compared to WT controls, the Ca²⁺ response was more transient and of less magnitude in TRPV4^−/−, with [Ca²⁺]_i changes of 17.9±1.3 and 5.9 ±0.3 nmol/L at peak and 1 min after peak, respectively. This is consistent with a role for TRPV4 in endothelial Ca²⁺ entry during plateau phase of Ca²⁺ response.

Figure 5.

Acetylcholine-induced Ca²⁺ responses in endothelial cells in situ of mesenteric arteries from WT and TRPV4^−/− (KO) mice. A, Representative traces of endothelial Ca²⁺ measured with the fluorescence Ca²⁺ indicator Fura-2. Mesenteric arteries of WT and TRPV4^−/− were isolated and cannulated, and acetylcholine (1 µmol/L) was added into the bath. B, Summary of endothelial [Ca²⁺]_i increase at 0 and 1 min after peak response to acetylcholine. * P<0.05 vs. WT (n=5–7 mice, 20–30 endothelial cells/each vessel).

We also measured NO production in vascular tissues of TRPV4^−/− and WT mice using DAF fluorescence assay. The carotid arteries were used since acetylcholine-induced relaxations are mainly mediated by NO in this vascular bed (data not shown). As shown in figure 6, acetylcholine induced a rapid increase in DAF fluorescence in the endothelial cell layer of WT carotid arteries. This increase was significantly reduced in the presence of L-NAME, confirming that increased DAF fluorescence is due to NO release. TRPV4^−/− mice exhibited a markedly blunted response to acetylcholine. The basal level of DAF fluorescence was also lower in TRPV4^−/− mice versus WT control.

Figure 6.

Acetylcholine (Ach)-induced NO production in carotid arterial endothelial cells in situ from WT and TRPV4^−/− (KO) mice. A, Representative images of endothelial NO fluorescence before (0 min) and 15 min after treatment with acetylcholine (1 µmol/L) in the presence or absence of L-NAME (100 µmol/L). Endothelial NO was measured with the fluorescence indicator DAF-FM. To help locate the endothelial layer, endothelial Ca²⁺ was simultaneously measured using Fura-2. B, Representative traces of endothelial NO increase in response to acetylcholine. A.U.: arbitrary unit. C, Summarized data of DAF fluorescence intensity before (0 min) and 15 min after acetylcholine addition. * P<0.05 vs. Ach 0 min in WT control; # P<0.05 vs. fluorescence change in WT control; n=3–6 mice.

Discussion

Using a TRPV^−/− mouse model, we provide several lines of evidence supporting TRPV4 channels as a novel mediator of agonist-induced endothelium-dependent vasodilation. For the first time we found that TRPV4 channels are expressed in both resistance and conduit arteries of mice, and activation of these channels increases endothelial Ca²⁺ leading to vasodilation in resistance vascular beds. Acetylcholine-induced vasodilator responses in vitro and in vivo are markedly reduced in TRPV4^−/− mice, which is accompanied by blunted Ca²⁺ and NO responses in endothelial cells. These new findings further extend the functional roles of endothelial TRPV4 channels in regulation of vascular tone and endothelial Ca²⁺ signaling.

Consistent with the results of previous studies (24,25,26), we found that both NO and EDHF contribute to endothelium-dependent relaxation induced by acetylcholine in mouse mesenteric arteries. Compared to WT mice, the L-NAME-sensitive component of acetylcholine-induced relaxation was reduced in TRPV4^−/− mice, whereas the K⁺-sensitive relaxation was virtually abolished, indicting that the TRPV4 channel is involved in both NO- and EDHF-dependent vasodilation. The involvement of TRPV4 channels in NO-mediated dilation was also supported by the observation that acetylcholine-induced NO production was significantly reduced in vascular endothelial cells of TRPV4^−/− mice. These results are generally in agreement with previous reports that activation of TRPV4 channels induces NO- and EDHF-dependent vasodilation in rat carotid and gracilis arteries, and rat cerebral arteries (13,14).

Agonist-induced increase in [Ca²⁺]_i is critical in the synthesis of relaxing factors such as NO and EDHF in endothelial cells (1). However, the [Ca²⁺]_i threshold is higher for EDHF-dependent dilation than for NO-dependent responses (27). Therefore, reduction in endothelial Ca²⁺ would have greater effect on EDHF-mediated than NO-mediated relaxation. This may partially explain our findings that TRPV4 knockout affected the K⁺-sensitive relaxation more than the NO-mediated relaxation in small mesenteric arteries, a resistance vascular bed where EDHF-mediated dilation is more prominent. TRPV4 activation and resulting Ca²⁺ influx may also selectively elicit the generation of EDHF and/or NO through specific signaling systems located in subcellular domains. TRPV4 channels have been shown to form a Ca²⁺ signaling complex with ryanodine receptors and large-conductance Ca²⁺-activated K⁺ channels (BK_Ca) in vascular smooth muscle cells (28). A recent study has also reported a close association of Ca²⁺ influx and EDHF-mediated relaxation in the caveolar microdomain of endothelial cells (26).

In contrast to blood pressure changes, acetylcholine induced similar drops in the heart rate in TRPV4^−/− animals compared to WT control, indicating that TRPV4 plays minimal role in the control of heart rate in these animals. A recent study has also reported that TRPV4 agonists have no significant effect on rate or contractility in the isolated, buffer-perfused rat heart (29).

Acetylcholine-induced Ca²⁺ increase (plateau phase) was reduced but not eliminated in TRPV4^−/− mice, indicating that other Ca²⁺ entry pathways may coexist in vascular endothelial cells. Other TRP channels including TRPC (canonical) and TRPM (Melastatin) subfamilies have been found in endothelial cells (4). Several TRPC channels have been proposed as store-operated Ca²⁺ channels in response to agonist stimulation (4). A previous study indicates that agonist-induced endothelial Ca²⁺ current and vasodilation is reduced in aorta from TRPC4^−/− mice (30). In another recent study, Fleming et al reported that bradykinin induces translocation of TRPC6 to the cell membrane and TRP channel-mediated Ca²⁺ influx in human endothelial cells (31). Future studies are required to determine whether these TRP channels contribute to the remaining Ca²⁺ entry in endothelial cells of TRPV4^−/− animals.

Immunohistochemical analysis of mouse aorta revealed that TRPV4 channels are mainly expressed in the endothelium. However, TRPV4 channel has also been found in vascular smooth muscle cells of rat aortic, cerebral and pulmonary arteries (14,28,32,33). We also found evidence for TRPV4 protein in human and bovine coronary vascular smooth muscle, but in much smaller amounts than in endothelial cells (unpublished observations). Therefore, expression of TRPV4 channels in vascular smooth muscle cells may depend on species and vascular beds. However, denuded mouse mesenteric arteries do not dilate to 4α-PDD, thus we conclude that any TRPV4 channels in vascular smooth muscle do not contribute to the observations made in this study.

Baseline blood pressure in TRPV4^−/− mice was not statistically higher than in their wild-type controls as might be expected from reduced release of endothelial relaxing factors. In contrast, a trend toward lower blood pressure was observed in TRPV4^−/− animals. These results are consistent with those of a previous study in unanesthetized animals (34). Although not further explored in the current study, the absence of baseline blood pressure change in TRPV4^−/− mice could reflect compensatory pressure homeostatic mechanisms that minimize blood pressure changes observed in TRPV4 KO mice. A conditional TRPV4 knockout specific to endothelial cells would help to address this possibility in future studies. Alternatively, compared to the mesenteric circulation examined in this study, TRPV4 expression and function might be different in other vascular beds.

Perspectives

TRPV4 channels are expressed in endothelial cells of various species and vascular beds. Given the complex expression pattern of TRP channels and lack of specific channel blockers, TRPV4^−/− mice provide a good model to study molecular and functional properties of endothelial TRPV4 channels in its native cellular environment. Our data suggest that TRPV4 channels, known to be involved in vascular mechanotransduction, are also involved in chemical agonist-induced increases in endothelial Ca²⁺ and endothelium-dependent vasodilation. However, the cellular mechanisms responsible for TRPV4 activation, i.e. via receptor or store-operated mechanism, remain to be determined. Since activation of TRPV4 by channel agonists reduces blood pressure, endothelial TRPV4 channel might serve as a novel pharmacological target for the treatment of hypertension (13). It will also be of interest to determine whether TRPV4-mediated endothelial responses are altered in other cardiovascular diseases such as atherosclerosis, where pharmacological manipulation of channel function might have beneficial therapeutic effects.