TRPA1 is essential for the vascular response to environmental cold exposure

Abstract

The cold-induced vascular response, consisting of vasoconstriction followed by vasodilatation, is critical for protecting the cutaneous tissues against cold injury. Whilst this physiological reflex response is historic knowledge, the mechanisms involved are unclear. Here by using a murine model of local environmental cold exposure, we show that TRPA1 acts as a primary vascular cold sensor, as determined through TRPA1 pharmacological antagonism or gene deletion. The initial cold-induced vasoconstriction is mediated via TRPA1-dependent superoxide production that stimulates α2C-adrenoceptors and Rho-kinase-mediated MLC phosphorylation, downstream of TRPA1 activation. The subsequent restorative blood flow component is also dependent on TRPA1 activation being mediated by sensory nerve-derived dilator neuropeptides CGRP and substance P, and also nNOS-derived NO. The results allow a new understanding of the importance of TRPA1 in cold exposure and provide impetus for further research into developing therapeutic agents aimed at the local protection of the skin in disease and adverse climates.

Figure 1. Cold-induced vascular response is dependent on TRPA1.

Blood flow was measured using FLPI in anaesthetized mice following immersion of the ipsilateral hindpaw in cold (10 °C) water and contralateral paw remained untreated. (a) Representative blood flow trace of a cold-induced response in WT and TRPA1 KO mice. (b) Representative FLPI pictures alongside grey/black ‘photo’ showing blood flow at baseline, 2 and 30 min in cold-treated hindpaw. (c) % Change in hindpaw blood flow from baseline to 0–2 min following cold treatment (maximum vasoconstriction) and (d) Restoration of cutaneous blood flow, as assessed by AUC for 30 min following cold treatment in WT (n=12) and TRPA1 KO (n=9) or WT mice pretreated with TRPA1 antagonist HC030031 (100 mg kg⁻¹, i.p., 30 min, n=10) or control (10% DMSO in saline, i.p., 30 min, n=8) before cold treatment. (e) % Maximum change in hindpaw blood flow from baseline to 0–2 min following cold treatment (maximum vasoconstriction) in WT (n=10) and TRPM8 KO (n=14) mice or WT mice pretreated with TRPM8 antagonist AMTB (10 mg kg⁻¹, i.p., 30 min, n=8) or control (10% DMSO in saline, i.p., 30 min, n=10) before cold exposure. (f) Representative trace of a cold-induced response in WT and TRPM8 KO mice. (g) Restoration of cutaneous blood flow, assessed by AUC for 30 min following cold treatment in WT (n=10) and TRPM8 KO (n=14) mice or WT mice pretreated with TRPM8 antagonist AMTB (10 mg kg⁻¹, i.p., 30 min, n=8) or control (10% DMSO in saline, i.p., 30 min, n=10) before cold treatment. (h) % Change in hindpaw blood flow from baseline to 0–2 min following cold treatment (maximum vasoconstriction) and (i) Restoration of cutaneous blood flow, assessed by AUC for 30 min following cold treatment in TRPA1 WT (n=5) and KO (n=5) pretreated with control (10% DMSO in saline, i.p., 30 min) or TRPA1 WT (n=4) and KO (n=4) pretreated with AMTB (10 mg kg⁻¹, i.p., 30 min). All error bars indicate s.e.m. *P<0.05, **P<0.01, ***P<0.001 versus respective untreated, ^#P<0.05, ^##P<0.01, ^###P<0.001 versus cold-treated, $P<0.05 versus cold-treated TRPA1 WT (analysis of variance, Bonferroni post hoc test).

Figure 2. Sympathetic nerves, α_2C-adrenoceptors and superoxide in cold-induced vasoconstriction.

Blood flow was measured using FLPI in anaesthetized mice following immersion of the ipsilateral hindpaw in cold (10 °C) water and the contralateral hindpaw remained untreated. (a) % Maximum change in hindpaw blood flow from baseline to 0–2 min following cold treatment (maximum vasoconstriction) in WT mice pretreated with guanethidine (30 mg kg⁻¹, s.c., 4 days, n=7) or control (saline, s.c., 4 days, n=7) and (b) WT mice pretreated with the α-adrenoceptor antagonist phentolamine (10 mg kg⁻¹, i.v., n=6) or control (saline, i.v., n=6). (c) Representative trace of a cold-induced response and (d) % Change in hindpaw blood flow from baseline to 0–2 min following cold treatment (maximum vasoconstriction) in WT mice treated with the α₂-adrenoceptor antagonist yohimbine (10 mg kg⁻¹, s.c., 30 min, n=5) or control (saline, s.c., 30 min, n=5). (e) Representative trace of a cold-induced response and (f) % Change in hindpaw blood flow from baseline to 0–2 min following cold treatment (maximum vasoconstriction) in WT mice treated with the α_2C-adrenoceptor antagonist JP1302 (3 μg kg⁻¹, s.c., 60 min, n=6) or control (saline, s.c., 60 min, n=6). (g) Representative blood flow trace of a cold-induced response and (h) % Change in blood flow from baseline to 0–2 min following cold treatment (maximum vasoconstriction) in WT mice treated with the SOD mimetic TEMPOL (30 mg kg⁻¹, i.v., 5 min, n=4) or control (saline, i.v., 5 min, n=4). Superoxide levels were measured at 2 min (maximum vasoconstriction) following cold treatment in the hindpaw tissue samples by lucigenin chemiluminescence in (i) naïve, WT mice pretreated with TRPA1 antagonist HC030031 (100 mg kg⁻¹, i.p., 30 min) or control (10% DMSO in saline, i.p., 30 min, n=4–5) and (j) WT mice pretreated with JP1302 (3 μg kg⁻¹, s.c., 60 min) or control (saline, s.c., 30 min, n=6–8). (k) % Change in blood flow from baseline to 0–2 min following cold treatment (maximum vasoconstriction) in WT mice treated with the mitochondria-targeted superoxide scavenger mito-TEMPO (10 mg kg⁻¹, i.p., 60 min, n=9) or control (saline, i.p., 60 min, n=8). All error bars indicate s.e.m. *P<0.05, **P<0.01,***P<0.001 versus respective untreated, ^#P<0.05, ^##P<0.01 versus cold-treated hindpaw (analysis of variance, Bonferroni post hoc test).

Figure 3. Rho-kinase-mediated MLC signalling in the cold-induced vasoconstriction.

Blood flow was measured using FLPI in anaesthetized mice following immersion of the ipsilateral hindpaw in cold (10 °C) water and the contralateral hindpaw remained untreated. (a) Representative blood flow trace of a cold-induced vascular response in WT mice pretreated with the selective Rho-kinase associated protein kinase inhibitor Y27632 or control (saline). (b) % Maximum change in hindpaw blood flow from baseline to 0–2 min following local cold treatment (maximum vasoconstriction) in WT mice pretreated with Y27632 (5 mg kg⁻¹, i.p., 30 min, n=6) and control (saline, i.p., 30 min, n=5). (c) Representative western blot of p-MLC at Ser of MLC and total MLC protein expression (uncropped blots are presented in Supplementary Figure 12). (d) Densitometric analysis of p-MLC and total MLC, normalized to β-actin (n=6). All error bars indicate s.e.m. **P<0.01, ***P<0.001 versus respective untreated, ^##P<0.01, ^###P<0.001 versus cold-treated hindpaw (analysis of variance, Bonferroni post hoc test).

Figure 4. TRPA1 mediates the vasodilator component of cold-induced responses.

Blood flow was measured using FLPI in anaesthetized mice following immersion of the ipsilateral hindpaw in cold (10 °C) water and the contralateral hindpaw remained untreated. (a) Representative blood flow trace of a cold-induced vascular response and (b) restoration of cutaneous blood flow, as assessed by AUC for 30 min following cold water immersion in WT mice pretreated with the TRPA1 antagonist HC030031 (6 mg kg⁻¹, i.v., n=6) or control (8% DMSO in 2% Tween-80 in saline, n=6) at the maximum vasoconstriction phase following local cold water immersion. (c) % Maximum change in hindpaw blood flow from baseline to 0–2 min following local cold treatment (maximum vasoconstriction), (d) Representative blood flow trace of a cold-induced vascular response and (e) Restoration of cutaneous blood flow, as assessed by AUC for 30 min following cold water immersion in WT mice pretreated with the TRPV1 agonist resiniferatoxin (0.3 mg kg⁻¹, s.c. daily., 4 days, n=8) or control (10% ethanol, 10% Tween-80 in saline, s.c. daily, 4 days, n=9). (f) Representative blood flow trace of a cold-induced vascular response, (g) % Maximum change in hindpaw blood flow from baseline to 0–2 min following local cold treatment (maximum vasoconstriction) and (h) Restoration of cutaneous blood flow, as assessed by AUC for 30 min following cold water immersion in WT mice pretreated with the CGRP receptor antagonist BIBN4096 (0.3 mg kg⁻¹, i.v., 5 min, n=9) or control (saline, i.v., 5 min, n=10). All error bars indicate s.e.m. *P<0.05, **P<0.01, ***P<0.001 versus respective untreated, ^##P<0.01, ^###P<0.001 versus cold-treated hindpaw (analysis of variance, Bonferroni post hoc test).

Figure 5. The cold-induced vascular response is dependent on neuropeptides.

Blood flow was measured using FLPI in anaesthetized mice following immersion of the ipsilateral hindpaw in cold (10 °C) water and the contralateral hindpaw remained untreated. Mice were pretreated with pharmacological inhibitors or the respective vehicle (i.v., 5 min) before local cold exposure. (a) Effects of CGRP receptor antagonist CGRP_8–37 (400 nmol kg⁻¹, n=7) on the cold-induced vascular response in WT mice. (b) Cold-induced vascular response in WT mice pretreated with a combination of CGRP_8–37 and NK₁ receptor antagonist SR140333 (480 nmol kg⁻¹, n=8) with or without the non-selective NOS inhibitor L-NAME (15 mg kg⁻¹, n=7) and control (0.01% BSA in saline, n=10). (c) Effects of CGRP_8–37, SR140333 and the selective nNOS inhibitor SMTC (10 mg kg⁻¹, n=6) or control (0.01% BSA in saline, n=7) on the cold-induced vascular response in WT mice. All error bars indicate s.e.m. *P<0.05, ***P<0.001, ****P<0.0001 versus respective untreated, ^###P<0.001, ^####P<0.0001 versus cold-treated hindpaw (analysis of variance, Bonferroni post hoc test). (d) Proposed cold-induced TRPA1 pathway in the regulation of cutaneous vasculature. Local cold exposure (10 °C) causes a transient and rapid decrease in blood flow from baseline (1). This initial phase of cold-induced vasoconstriction consists of activation of TRPA1 which further mediates to (a) the release of noradrenaline (NA) and (b) increased reactive oxygen species release (for example, superoxide (O₂⁻) generation) that can also activate the ROCK-mediated pathways and increase constriction via phosphorylated MLC-induced increase in [Ca²⁺]_i. The O₂⁻ induces the translocation of α_2C-adrenoceptors from the Golgi to the surface membrane and increases adrenergic activity in the VSMC in a ROCK-dependent fashion. (2) The restoration of blood flow following local cold treatment is essential in restoring blood flow to baseline to protect against local cold-induced injuries. This phase is mediated by the release of CGRP, substance P and nitric oxide following TRPA1 activation of sensory neurons. MLC-P, myosin light chain phosphatase; NA, noradrenaline; p-MLC; phosphorylated myosin light chain; VSMC, vascular smooth muscle cell.