Disruption of vascular Ca2+-activated chloride currents lowers blood pressure

Abstract

High blood pressure is the leading risk factor for death worldwide. One of the hallmarks is a rise of peripheral vascular resistance, which largely depends on arteriole tone. Ca2+-activated chloride currents (CaCCs) in vascular smooth muscle cells (VSMCs) are candidates for increasing vascular contractility. We analyzed the vascular tree and identified substantial CaCCs in VSMCs of the aorta and carotid arteries. CaCCs were small or absent in VSMCs of medium-sized vessels such as mesenteric arteries and larger retinal arterioles. In small vessels of the retina, brain, and skeletal muscle, where contractile intermediate cells or pericytes gradually replace VSMCs, CaCCs were particularly large. Targeted disruption of the calcium-activated chloride channel TMEM16A, also known as ANO1, in VSMCs, intermediate cells, and pericytes eliminated CaCCs in all vessels studied. Mice lacking vascular TMEM16A had lower systemic blood pressure and a decreased hypertensive response following vasoconstrictor treatment. There was no difference in contractility of medium-sized mesenteric arteries; however, responsiveness of the aorta and small retinal arterioles to the vasoconstriction-inducing drug U46619 was reduced. TMEM16A also was required for peripheral blood vessel contractility, as the response to U46619 was attenuated in isolated perfused hind limbs from mutant mice. Out data suggest that TMEM16A plays a general role in arteriolar and capillary blood flow and is a promising target for the treatment of hypertension.

Figure 1. Disruption of TMEM16A under the control of the smooth muscle myosin heavy chain promoter abolishes CaCCs in VSMCs of the aorta.

(A) In the targeted Tmem16a locus, exon 21 is flanked by loxP sites (upper panel). After Cre-mediated excision, the floxed fragment is removed (lower). Numbered black squares represent exons and triangles, loxP sites; EcoRI restriction sites used for cloning/screening are indicated. (B) Typical currents recorded from VSMCs isolated from thoracic aortae of non-induced control mice (Cre⁺ no tamoxifen) and induced conditional Tmem16a knockout mice (Cre⁺ tamoxifen). Inset: Voltage protocol. (C) CaCCs were absent in VSMCs isolated from the aorta of induced knockout mice. Mean I-V relationship was recorded after a 1 second test pulse from VSMCs isolated from thoracic aortae of WT, non-induced control (Cre⁺ no tamoxifen), and induced conditional Tmem16a knockout mice (Cre⁺ tamoxifen). CaCCs measured from controls were absent under Ca²⁺-free conditions (Cre⁺ no tamoxifen Ca²⁺ free). CaCCs of VSMCs were blocked by 300 εM niflumic acid (Cre⁺ no tamoxifen + 300 εM niflumic acid). n = 6–15 cells each. (D) Amplitude of various VSMC tail current densities measured after a voltage step to –80 mV from a 1 second test pulse at +80 mV (n = 6–15 cells each). Whereas no difference was detected between non-induced control and wild-type mice, non-induced control mice and mice under other conditions differed significantly. 2-way ANOVA; **P < 0.01.

Figure 2. Disruption of TMEM16A in blood vessels decreases mean arterial blood pressure.

(A) Blood pressure in three cohorts of floxed Tmem16a mice. Cre recombinase expressed under control of the SMMHC promoter was induced by feeding tamoxifen to the experimental cohort (Cre⁺ tamoxifen, n = 14). Littermates of the same genotype that did not receive tamoxifen served as a control (Cre⁺ no tamoxifen, n = 15). A Cre-negative control cohort received tamoxifen to exclude a gene-independent effect of tamoxifen on mean arterial blood pressure (Cre^– tamoxifen, n = 6). Induction with tamoxifen caused a significant decrease in systemic blood pressure in the Cre⁺ cohort (2-way ANOVA; **P < 0.01. (B) At the same time, the pulse pressure decreased (2-way ANOVA; **P < 0.01). (C) Fractional shortening, cardiac output, ejection fraction, and heart rate did not differ between cohorts before and 2 weeks after termination of the tamoxifen induction. Student’s t test. (D) The hypotensive effect of TMEM16A disruption on mean arterial blood pressure was enhanced in the presence of the vasopressor angiotensin II, which was continuously infused (1.5 ng*g^–1*min^–1) with osmotic mini pumps starting 2 weeks after tamoxifen induction (n = 6–7 for each cohort). 2-way ANOVA; **P < 0.01. (E) Mean arterial blood pressure did not further decrease on a low-salt diet. Switching to a high-salt diet eliminated the difference in blood pressure between induced and non-induced mice (n = 4 for each group). 2-way ANOVA; *P < 0.05.

Figure 3. TMEM16A is expressed in the aorta and modulates its contractility but absent from medium-sized arteries.

(A and B) Western blot analysis of protein lysates of isolated arterial blood vessels of non-induced control mice revealed that TMEM16A expression is particularly strong in the aorta as compared with carotid and mesenteric arteries. In tamoxifen-induced conditional Tmem16a knockout mice, the bands corresponding to TMEM16A were shifted to a smaller size. (C) Tail currents as a measure for the size of CaCCs in VSMCs isolated from different blood vessels differed significantly in size. In the induced conditional knockout, currents were abolished (n = 10–15 in each group). mes., mesenteric. (D–F) Whereas vascular smooth muscle cells in cryosections of the aorta of control mice were robustly stained (D), staining was less intense in large (E) and almost absent in small mesenteric arteries (F). Endothelial cells were stained with an antibody recognizing CD31. Scale bars: 30 εm. (G, H, J, and K) Dose-response curves for the contractile effects of angiotensin II (G and H) and U46619 (J and K) for aortic (G and J; n = 6 in each group) and first/second-order mesenteric artery rings (H and K; n = 16–24 in each group), as determined by wire myography. (I and L) Changes in diameter of pressurized third/fourth-order mesenteric arteries in response to different concentrations of angiotensin II (I) and U46619 (L), as determined by videomicroscopy (n = 4–6 in each group). Arteries of the two genotypes showed comparable dilator responses of roughly 25% to Ca²⁺-free solutions (data not shown). 2-way ANOVA; *P < 0.05; **P < 0.01.

Figure 4. TMEM16A is expressed in small arterioles of brain and retina and enhances contractility in the retina.

(A) Immunostaining of brain slices for the endothelial cell marker CD31 (upper) and TMEM16A (middle); the bottom panel shows an overlay. Small but not large arterioles show strong TMEM16A staining. Scale bar: 50 εm. (B) Epifluorescence images of retinal whole mount preparation stained for smooth muscle actin (SMA, upper) and TMEM16A (lower). Primary arterioles (a) and primary veins (v) originate from the optical disc in an alternate manner. Whereas TMEM16A staining in primary arterioles was absent or week, secondary and higher-order arterioles were intensely labeled. Scale bar: 300 εm. (C) Retina whole mount immunostainings for CD31 (upper) and TMEM16A (middle) showing a TMEM16A-negative primary arteriole (arrowhead) and second- and higher order arteriolar branches and overlay (lower). Arrows indicate strongly labeled cells at branching points of second-order arterioles. Scale bar: 30 εm. (D) Typical wild-type currents recorded from retina tissue prints of VSMC-like cells of first-order arterioles and (E) dome-shaped cells primarily found on second/third-order arterioles. The voltage protocol for D and E is shown in D. (F) CaCCs were absent in contractile cells of conditional knockout mice. (G) Mean tail current densities of contractile cells of the retina and brain arterioles for different genotypes and comparison with currents from VSMCs of the aorta (n = 6–17 each). (H) Constriction of second/third-order arterioles was diminished in the absence of CaCCs. Dose-response curves for U46619 in non-induced control and conditional Tmem16a knockout mice (n = 60 each). Student’s t test; **P < 0.01.

Figure 5. TMEM16A modulates peripheral resistance.

(A) Double immunostaining for CD31 (left) and TMEM16A (center) of slices from skeletal muscle. The overlay is shown on the right. Scale bar: 50 εm. (B and C) I-V relationship (B) and tail current density (C) from isolated pericytes from skeletal muscle showed a drastic reduction of CaCCs upon disruption of TMEM16A (n = 13–16 cells each). **P < 0.01. (D) Resting membrane potentials of isolated skeletal muscle pericytes did not differ between genotypes. n = 6; Student’s t test; P > 0.05. (E) U46619-induced depolarization of WT pericytes was almost absent in the mutant. TMEM16A-deficient pericytes hyperpolarized after agonist application, which suggests the presence of Ca²⁺-activated K⁺ channels. Replacing Cl^– with methanesulfonate further increased the depolarization in WT, but not in mutant pericytes, indicating that Cl^– ion flow is important for U46619-induced pericyte depolarization. n = 6; Student’s t test; **P < 0.01. (F) Original tracing of the perfusion pressure of an isolated hind limb perfusion of a control mouse. The bolus injection of either 20 pmol or 60 pmol U46619 (arrows) caused a short pressure peak because of the injected volume and subsequently a protracted pressure increase. The peak responses are indicated by arrowheads. (G) The difference between peak and steady-state pressure at the two doses of U46619 analyzed. perf., perfusion. n = 20. Student’s t test; *P < 0.05.