TRPA1 has a key role in the somatic pro-nociceptive actions of hydrogen sulfide

Abstract

Hydrogen sulfide (H(2)S), which is produced endogenously from L-cysteine, is an irritant with pro-nociceptive actions. We have used measurements of intracellular calcium concentration, electrophysiology and behavioral measurements to show that the somatic pronociceptive actions of H(2)S require TRPA1. A H(2)S donor, NaHS, activated TRPA1 expressed in CHO cells and stimulated DRG neurons isolated from Trpa1(+/+) but not Trpa1(-/-) mice. TRPA1 activation by NaHS was pH dependent with increased activity at acidic pH. The midpoint of the relationship between NaHS EC(50) values and external pH was pH 7.21, close to the expected dissociation constant for H(2)S (pK(a) 7.04). NaHS evoked single channel currents in inside-out and cell-attached membrane patches consistent with an intracellular site of action. In behavioral experiments, intraplantar administration of NaHS and L-cysteine evoked mechanical and cold hypersensitivities in Trpa1(+/+) but not in Trpa1(-/-) mice. The sensitizing effects of L-cysteine in wild-type mice were inhibited by a cystathionine β-synthase inhibitor, D,L-propargylglycine (PAG), which inhibits H(2)S formation. Mechanical hypersensitivity evoked by intraplantar injections of LPS was prevented by PAG and the TRPA1 antagonist AP-18 and was absent in Trpa1(-/-) mice, indicating that H(2)S mediated stimulation of TRPA1 is necessary for the local pronociceptive effects of LPS. The pro-nociceptive effects of intraplantar NaHS were retained in Trpv1(-/-) mice ruling out TRPV1 as a molecular target. In behavioral studies, NaHS mediated sensitization was also inhibited by a T-type calcium channel inhibitor, mibefradil. In contrast to the effects of NaHS on somatic sensitivity, intracolonic NaHS administration evoked similar nociceptive effects in Trpa1(+/+) and Trpa1(-/-) mice, suggesting that the visceral pro-nociceptive effects of H(2)S are independent of TRPA1. In electrophysiological studies, the depolarizing actions of H(2)S on isolated DRG neurons were inhibited by AP-18, but not by mibefradil indicating that the primary excitatory effect of H(2)S on DRG neurons is TRPA1 mediated depolarization.

Figure 1. NaHS activation of TRPA1.

A) NaHS evoked a concentration-dependent increase in [Ca²⁺]_i in Fura-2 loaded mTRPA1 expressing CHO cells but not in cells expressing TRPV1, TRPM8 or TRPV4 (mean ± s.e.m. , n = 3 for each point). B) Potentiation of NaHS evoked Ca²⁺ response at lower external pH. C) NaHS concentration-Ca²⁺ response relationships at physiological and acidic external pH showed a reduction in EC₅₀ value at pH 6 (each point is mean ± s.e.m of triplicate samples). D) Relationship between external pH and EC₅₀ values for NaHS activation of TRPA1. Logistic curve fitted to the data has a mid-point at pH 7.21. The slope factor for the relationship between [H+] and EC₅₀ values was 1.86±0.46. Each point is mean ± s.e.m of EC₅₀ values from between 5–16 individual concentration-response curves for each pH value.

Figure 2. Membrane currents evoked by NaHS in TRPA1 expressing cells.

Whole cell currents evoked by 5 mM NaHS in A) TRPA1 CHO cell and B) DRG neuron. TRPA1 expression in DRG neurons was confirmed by subsequent activation by AITC. C) Single channel currents evoked by a relatively low concentration of NaHS (100 µM) applied to the intracellular side of an inside-out membrane patch from a TRPA1 CHO cell. D) NaHS evoked single channel currents in a cell-attached patch evoked by extracellularly applied NaHS (2 mM).

Figure 3. NaHS activation of DRG neurons is mediated by TRPA1.

A) Pseudocoloured images of isolated mouse DRG neurons loaded with Fura-2 sequentially challenged with NaHS and capsaicin and depolarized by a high K⁺ solution NaHS increased [Ca²⁺]_i in a subset of capsaicin-sensitive neurons from wild-type mice (top) but not in neurons from Trpa1 ^−/− mice Bottom). B) Fura-2 responses showing NaHS-evoked increases in [Ca²⁺]_i in DRG neurons from Trpa1 ^−/− and wild-type mice. NaHS evoked clear responses in AITC-sensitive neurons from wild-type mice, but either no response or very slow, small responses in Trpa1 ^−/− DRG neurons. C) Percentage of capsaicin-sensitive DRG neurons from Trpa1 ^−/− and wild-type mice that responded to NaHS.

Figure 4. The pronociceptive effects of NaHS require TRPA1.

Intraplantar administration of NaHS (1 nmole) reduced A) mechanical paw pressure threshold and B) latency for paw withdrawal from a cold (10°C) plate stimulus in wild type C57Bl/6 mice (n = 5 per group). NaHS (1 nmole intraplantar) evoked mechanical hypersensitivity (C) and cold hypersensitivity (D) were inhibited by intraplantar co-administration of TRPA1 antagonist AP-18 (25 nmole, n = 6) in wild-type mice and were absent in Trpa1 ^−/− mice, n = 6 (E, F).

Figure 5. TRPA1 is required for the pro-nociceptive effects of cysteine.

Intraplantar administration of L-cysteine (100 nmole) reduced A) mechanical paw pressure threshold and B) latency for paw withdrawal from a cold plate stimulus in wild type mice (n = 6 per group). Cysteine-evoked mechanical hypersensitivity (C) and cold hypersensitivity (D) were inhibited by systemic administration of a cystathionine β-synthase inhibitor (PAG, 11.25 mg/kg i.p. 60 minute pretreatment.), but PAG was without effect on NaHS-evoked mechanical (E) and cold (F) hypersensitivities (n = 6 per group).

Figure 6. NaHS-induced hypersensitivity is independent of TRPV1.

Intraplantar administration of NaHS (1 nmole) evoked similar mechanical (A) and cold (B) hypersensitivities in Trpv1 ^−/− and wild-type mice (n = 6 per group).

Figure 7. The pronociceptive effect of visceral NaHS is independent of TRPA1.

A) The number of pain-related behaviors produced by intracolonic administration of NaHS (5 nmoles) or vehicle did not differ between Trpa1 ^−/− and Trpa1 ^−/− mice (n = 7–9). B) Compared to vehicle, intracolonic NaHS did not evoke referred hyperalgesia, measured as the number of withdrawal responses produced by abdominal stimulation with von Frey filaments, in Trpa1 ^−/− and Trpa1 ^−/− mice (n = 7–9).

Figure 8. T-type calcium channels play a role in NaHS-evoked hypersensitivity but TRPA1 mediates DRG neuron depolarization.

A) T-type calcium channel blocker, mibefradil (9 mg/kg i.p.) inhibited intraplantar NaHS-evoked mechanical hypersensitivity (n = 7 per group). B) Mibefradil (10 µM) did not antagonize TRPA1 activation evoked by AITC in TRPA1 CHO cells. C) NaHS (5 mM) depolarized a subset of DRG neurons. D) NaHS depolarization of DRG neurons was blocked by the TRPA1 antagonist AP-18 (10 µM) but not by mibefradil (10 µM).

Figure 9. The nociceptive effect of local LPS requires stimulation of TRPA1.

Intraplantar administration of LPS (0.1–10 µg) evoked mechanical hyperalgesia in Trpa1^+/+ (A) but not in Trpa1^−/− mice (B). Intraplantar injection of 10 µg produced a mechanical hypersensitivity in C57Bl/6 mice (C). Pretreatment with PAG (11.25 mg/kg i.p., 1 h before LPS) or AP-18 (3 mg/kg i.p., 30 min before LPS), completely prevented the development of the LPS-induced mechanical hyperalgesia (** p<0.01, *** p<0.001, ANOVA followed by Tukey's HSD test).