Protease-activated receptor 2 sensitizes the transient receptor potential vanilloid 4 ion channel to cause mechanical hyperalgesia in mice

Abstract

Exacerbated sensitivity to mechanical stimuli that are normally innocuous or mildly painful (mechanical allodynia and hyperalgesia) occurs during inflammation and underlies painful diseases. Proteases that are generated during inflammation and disease cleave protease-activated receptor 2 (PAR2) on afferent nerves to cause mechanical hyperalgesia in the skin and intestine by unknown mechanisms. We hypothesized that PAR2-mediated mechanical hyperalgesia requires sensitization of the ion channel transient receptor potential vanilloid 4 (TRPV4). Immunoreactive TRPV4 was coexpressed by rat dorsal root ganglia (DRG) neurons with PAR2, substance P (SP) and calcitonin gene-related peptide (CGRP), mediators of pain transmission. In PAR2-expressing cell lines that either naturally expressed TRPV4 (bronchial epithelial cells) or that were transfected to express TRPV4 (HEK cells), pretreatment with a PAR2 agonist enhanced Ca2+ and current responses to the TRPV4 agonists phorbol ester 4alpha-phorbol 12,13-didecanoate (4alphaPDD) and hypotonic solutions. PAR2-agonist similarly sensitized TRPV4 Ca2+ signals and currents in DRG neurons. Antagonists of phospholipase Cbeta and protein kinases A, C and D inhibited PAR2-induced sensitization of TRPV4 Ca2+ signals and currents. 4alphaPDD and hypotonic solutions stimulated SP and CGRP release from dorsal horn of rat spinal cord, and pretreatment with PAR2 agonist sensitized TRPV4-dependent peptide release. Intraplantar injection of PAR2 agonist caused mechanical hyperalgesia in mice and sensitized pain responses to the TRPV4 agonists 4alphaPDD and hypotonic solutions. Deletion of TRPV4 prevented PAR2 agonist-induced mechanical hyperalgesia and sensitization. This novel mechanism, by which PAR2 activates a second messenger to sensitize TRPV4-dependent release of nociceptive peptides and induce mechanical hyperalgesia, may underlie inflammatory hyperalgesia in diseases where proteases are activated and released.

Figure 1

Expression of TRPV4 in HBE cells and HEK cells A, RT-PCR of HBE cells showing amplification of transcripts for PAR₂ and TRPV4 isoforms a, b and c There were no signals when reverse transcriptase (RT) was omitted (control). B, structure of three TRPV4 isoforms cloned from HBE cells. C, transient expression of TRPV4 isoforms with HA.11 tag in HEK cells. Immunoreactive TRPV4 was detected by immunofluorescence (left) and Western blotting (right) using HA.11 antibody. vc, vector control without TRPV4 insert. Scale bar = 10 μm.

Figure 2

Activation of TRPV4 in HBE cells and HEK cells Effects of 4αPDD (0.1–10 μm, upper traces) and hypotonic stimuli (260–310 mosmol l⁻¹, lower traces) on [Ca²⁺]_i in HBE cells (A) and HEK cells transiently expressing TRPV4 isoforms a, b or c or vector control (B). Left panels show records of [Ca²⁺]_i expressed as 340/380 nm emission ratio as a percentage of response measured at 50 s (prestimulus, 100%). Right panels show [Ca²⁺]_i responses as difference in 340/380 nm emission ratio between 50 s (prestimulus) and 300 s (maximal response). 4αPDD and hypotonic stimuli caused graded increases in [Ca²⁺]_i in HBE cells and HEK-TRPV4a cells, but had no effect in HEK cells expressing b or c TRPV4 isoforms or vector control. *P < 0.05, ANOVA and Dunnett's test, n = 9 experiments.

Figure 4

Mechanisms of PAR₂-induced sensitization of TRPV4 Ca²⁺ signals Effects of antagonists of signalling pathways on PAR₂-induced sensitization of Ca²⁺ signals to 4αPDD (0.1 or 1 μm) (A) and hypotonic stimulus (260 mosmol l⁻¹) (B) in HBE cells and to 4αPDD in HEK-TRPV4a cells (C). Cells were pretreated with PAR₂-AP or PAR₂-RP (both 100 μm) for 10 min before challenge with TRPV4 agonists. U73122 (10 μm), H-89 (10 μm), GF109203X (GFX) 10 μm and Gö6976 (0.1 μm), but not Gö6963 (0.1 μm), inhibited PAR₂-induced sensitization of Ca²⁺ responses to 4αPDD in HBE cells and HEK-TRPV4a cells. U73122, H-89 and Gö6976, but not GF109203X or Gö6963, inhibited PAR₂-induced sensitization of Ca²⁺ responses to hypotonic stimulus in HBE cells. Veh, vehicle; *P < 0.05; ANOVA followed by Dunnett's test; n = 8–12 experiments.

Figure 3

PAR-induced sensitization of TRPV4 Ca²⁺ signals A, B, C, left panels show records of [Ca²⁺]_i after challenge with PAR₂-AP, PAR₂-RP (both 100 μm) or vehicle (veh) (A and B) or PAR₁-AP, PAR₁-RP (both 100 μm) or vehicle (C) at 50 s followed by 4αPDD (0.1 or 1 μm) or hypotonic stimulus (260 mosmol l⁻¹) at 350 s. Traces are from HBE cells (A) and HEK-TRPV4a cells (B and C). [Ca²⁺]_i is expressed as 340/380 nm emission ratio as a percentage of response measured at 50 s (prestimulus, 100%). Right panels show [Ca²⁺]_i responses as difference in 340/380 nm emission ratio between 350 s (prestimulus) and 550 s (maximal response). Pretreatment with PAR₂-AP but not PAR₁-AP increased Ca²⁺ responses to 4αPDD and hypotonic stimulus, indicative of TRPV4 sensitization. *P < 0.05, ANOVA and Bonferroni's test, n = 9 experiments.

Figure 5

Localization and expression of TRPV4 in HEK cells and DRG neurons A, localization of TRPV4a transiently expressed in HEK cells by immunofluorescence using antibodies to HA.11 epitope or TRPV4. Both antibodies detected immunoreactive TRPV4. Control shows preabsorption of TRPV4 antibody with antigen used for immunization. B, localization of immunoreactive TRPV4, PAR₂, CGRP or SP in sections of rat DRG. TRPV4 was detected in the soma at the plasma membrane (white arrowheads) and in intracellular locations (white arrows), and also in fibres (white arrows). White arrows show that some neurons coexpressed TRPV4 with PAR₂, CGRP or SP. Yellow arrows show that some neurons expressing CGRP or SP did not express TRPV4. Yellow asterisks show that some neurons expressing TRPV4 did not express PAR2, CGRP or SP. Control shows preabsorption of TRPV4 antibody. Scale bar = 10 μm c, Localization of immunoreactive TRPV4, PAR₂, CGRP or SP in rat DRG after 2 days in culture. White arrows denote colocalization of TRPV4 with PAR₂, CGRP or SP. Yellow arrows show that some neurons expressing CGRP did not express TRPV4. Yellow asterisks show that neurons expressing TRPV4 did not express CGRP or SP. Control shows preabsorption of TRPV4 antibody. Scale bar = 10 μm. D, RT-PCR of mouse and rat DRG showing amplification of transcripts for TRPV4. There were no signals when reverse transcriptase (RT) was omitted (control).

Figure 6

PAR₂-induced sensitization of TRPV4 Ca²⁺ signals and currents in DRG neurons (A–F) and HEK-FLPTREX-TRPV4a cells (G–I) Neurons were pretreated for 20 min and HEK-FLPTREX-TRPV4a cells for 2 min with PAR₂-AP, PAR₂-RP or vehicle (Veh) before challenge with 4αPDD. A–C, [Ca²⁺]_i records in rat DRG neurons. A, records of [Ca²⁺]_i in DRG neurons challenged with 4αPDD (10 μm). B, maximal [Ca²⁺]_i responses to 4αPDD (10 μm) expressed as percentage response in neurons pretreated with PAR₂-RP (100%). Pretreatment with PAR₂-AP (10 μm) increased 4αPDD-stimulated Ca²⁺ response, indicative of TRPV4 sensitization (*P < 0.05, Student's t test). C, effects of inhibitors of second messenger kinases on PAR₂-induced sensitization of Ca²⁺ signals to 4αPDD. Results are expressed as percentage of response to 4αPDD in neurons pretreated with PAR₂-AP (100%). H-89 (10 μm) and GF109203X (GFX) 1 μm inhibited PAR₂-induced sensitization of Ca²⁺ responses to 4αPDD (*P < 0.05, ANOVA and Dunnett's test). D–F, whole-cell currents of mouse DRG neurons. D, whole-cell currents recorded during voltage ramp (−100 mV to +100 mV every 15 s) before and during application of 4αPDD (10 μm). E, current densities at −80 mV and +80 mV. PAR₂-AP (10 μm) increased 4αPDD-mediated current density (**P < 0.01; *P < 0.02, Student's t test). F, proportion of neurons that responded to 4αPDD (10 μm) by activation of an outwardly rectifying current. G–H, whole-cell currents of HEK-FLPTREX-TRPV4a. G, whole-cell currents recorded during voltage ramp (−100 mV to +100 mV every 15 s) before and during application of 4αPDD (0.5 μm). H, current densities at −80 mV and +80 mV. PAR₂-AP (100 μm) increased 4αPDD-mediated current density (**P < 0.01; *P < 0.05, Student's t test). I, effects of antagonists of signalling pathways on PAR₂-induced sensitization of currents to 4αPDD. H-89 (3 μm) and calphostin C (CaC) 2 μm inhibited sensitization (*P < 0.05; Student's t test). In A–I, numbers in parentheses refer to the number of cells.

Figure 7

Localization of immunoreactive TRPV4, CGRP or SP in dorsal horn of rat spinal cord Control shows preabsorption of TRPV4 antibody with antigen used for immunization. White arrows show colocalization of TRPV4 with CGRP and SP in some fibres of superficial laminae I and II. Scale bar = 10 μm.

Figure 8

TRPV4-mediated peptide release from rat dorsal horn Effects of 4αPDD (10–100 μm) (A) and hypotonic stimulus (228 mosmol l⁻¹) (B) on release of CGRP (left panels) and SP (right panels) from superfused slices of dorsal spinal cord. −[Ca²⁺]_o denotes Ca²⁺-free solution and Caps denotes pretreatment with capsaicin. 4αPDD and hypotonic stimulus induced release of CGRP and SP, and release required extracellular Ca²⁺ ions and was prevented by capsaicin pretreatment. *P < 0.05 compared to vehicle; ANOVA and Bonferroni's test. C and D, PAR₂-induced sensitization of neuropeptide release from dorsal spinal cord. Effects of preincubation with PAR₂-AP, PAR₂-RP (both 10 μm) or vehicle for 20 min on release of immunoreactive CGRP (left panels) and SP (right panels) in response to 4αPDD (C) or hypotonic stimulus (D). Pretreatment with PAR₂-AP increased release of CGRP and SP in response to 4αPDD and hypotonic stimulus, indicative of TRPV4 sensitization. *P < 0.05 compared to vehicle; ANOVA followed by Bonferroni's test. n = 6–9 experiments.

Figure 9

Role of TRPV4 in PAR₂-induced mechanical hyperalgesia Frequency of paw withdrawal to stimulation with a von Frey hair under basal conditions, after intraplantar injection of 4αPDD or PAR₂-AP alone, or after injection of PAR₂-AP followed 5 min later by 4αPDD (A), and distilled water or PAR₂-AP alone, or after injection of PAR₂-AP followed 5 min later by distilled water (B). When given alone, 4αPDD (10 μl of 50 μm solution) and PAR₂-AP (1 μg in 10 μl saline), but not hypotonic stimulus (10 μl of 17 mosmol l⁻¹), increased the frequency of paw withdrawal in TRPV4^+/+ but not TRPV4^−/− mice, indicating that TRPV4 mediates 4αPDD- and PAR₂-AP-induced mechanical hyperalgesia. Preinjection of PAR₂-AP enhanced responses to 4αPDD and hypotonic stimulus in TRPV4^+/+ but not TRPV4^−/− mice, indicating that PAR₂ sensitizes TRPV4 to cause mechanical hyperalgesia. ***P < 0.001; *P < 0.05 compared to vehicle; †††P < 0.001 compared to TRPV4^+/+. ANOVA and Bonferroni's test. N = 6–9 animals.

Figure 10

Proposed model of protease-induced mechanical hyperalgesia Proteases that are generated during inflammation (1) activate PAR₂ on afferent nerve endings in peripheral tissues (2). PAR₂ couples to activation of second messenger kinases (3) that may phosphorylate and sensitize TRPV4 (4), resulting in enhanced influx of Na⁺ and Ca²⁺ ions (5) and elevated release of CGRP and SP in the dorsal horn in response to mechanical stimuli (6). CGRP and SP activate their receptors on spinal neurons (CGRP, calcitonin-like receptor (CLR) and receptor activity-modifying protein 1 (RAMP1); SP, neurokinin 1 receptor (NK₁R)), resulting in enhanced transmission of nociceptive signals and mechanical hyperalgesia.