Protease-activated receptor 2 sensitizes the capsaicin receptor transient receptor potential vanilloid receptor 1 to induce hyperalgesia

Abstract

Inflammatory proteases (mast cell tryptase and trypsins) cleave protease-activated receptor 2 (PAR2) on spinal afferent neurons and cause persistent inflammation and hyperalgesia by unknown mechanisms. We determined whether transient receptor potential vanilloid receptor 1 (TRPV1), a cation channel activated by capsaicin, protons, and noxious heat, mediates PAR2-induced hyperalgesia. PAR2 was coexpressed with TRPV1 in small- to medium-diameter neurons of the dorsal root ganglia (DRG), as determined by immunofluorescence. PAR2 agonists increased intracellular [Ca2+] ([Ca2+]i) in these neurons in culture, and PAR2-responsive neurons also responded to the TRPV1 agonist capsaicin, confirming coexpression of PAR2 and TRPV1. PAR2 agonists potentiated capsaicin-induced increases in [Ca2+]i in TRPV1-transfected human embryonic kidney (HEK) cells and DRG neurons and potentiated capsaicin-induced currents in DRG neurons. Inhibitors of phospholipase C and protein kinase C (PKC) suppressed PAR2-induced sensitization of TRPV1-mediated changes in [Ca2+]i and TRPV1 currents. Activation of PAR2 or PKC induced phosphorylation of TRPV1 in HEK cells, suggesting a direct regulation of the channel. Intraplantar injection of a PAR2 agonist caused persistent thermal hyperalgesia that was prevented by antagonism or deletion of TRPV1. Coinjection of nonhyperalgesic doses of PAR2 agonist and capsaicin induced hyperalgesia that was inhibited by deletion of TRPV1 or antagonism of PKC. PAR2 activation also potentiated capsaicin-induced release of substance P and calcitonin gene-related peptide from superfused segments of the dorsal horn of the spinal cord, where they mediate hyperalgesia. We have identified a novel mechanism by which proteases that activate PAR2 sensitize TRPV1 through PKC. Antagonism of PAR2, TRPV1, or PKC may abrogate protease-induced thermal hyperalgesia.

Figure 1.

PAR₂ activation sensitizes TRPV1 in transfected cells. a, HEK-TRPV1 cells express PAR₁ and PAR₂. mRNA for PAR₁ (lane 2) and PAR₂ (lane 4) was detected in HEK-TRPV1 cells by RT-PCR. Lanes 1 and 3 are controls (no RT). Results are representative of three experiments. b–g, Cells were exposed to PAR₂-AP or -RP (100 μm), trypsin (10 nm), PMA (5 μm), or vehicle for 5 min and were then challenged with capsaicin. PAR₂ agonists (c, e–g) and PMA (g) potentiated responses to capsaicin. Inhibition of PLC (U73122, 5 μm) and PKC (GFX and Gö6976, 1 and 0.1 μm, respectively) suppressed the potentiation (g). Results in g are expressed as a percentage of the positive control (cells treated with PAR₂ agonists or PMA). *p < 0.05 compared with PAR₂-RP or saline (f), PMA, PAR₂-AP, or trypsin (g). Try, Trypsin; AP, PAR₂-AP; n = 120–1040 cells.

Figure 2.

PAR₂ activation sensitizes TRPV1 in DRG neurons. a, Colocalization (arrows) of immunoreactive PAR₂ and TRPV1 in tissue sections of DRG neurons (top panels) and in cultured DRG neurons (bottom panels). Scale bars: top panels, 50 μm; bottom panels, 25 μm. b–h, Neurons were exposed to PAR₂-AP or -RP (100 μm), trypsin (10 nm), PMA (5 μm) or vehicle for 5 min, and were then challenged with capsaicin (100 nm). PAR₂ agonists (c, e–h) and PMA (g) potentiated responses to capsaicin. Inhibition of PLC (U73122, 5 μm) and PKC (GFX and Gö6976, 1 and 0.1 μm, respectively) suppressed the potentiation (g). PAR₂ activation also increased the proportion of neurons that gave detectable responses to capsaicin (h). Results in g are expressed as a percentage of the positive control (magnitude of capsaicin response in cells after PAR₂ activation). *p<0.05 compared with PAR₂-RP or saline (f, h) and PMA or PAR₂-AP (g). AP, PAR₂-AP; n = 17–231 neurons.

Figure 3.

PAR₁ activation does not sensitize TRPV1 in transfected cells (a, b, e) or DRG neurons (c, d, f, g). HEK-TRPV1 cells (a, b, e) or DRG neurons (c, d, f, g) were exposed to PAR₁-AP or -RP (100 μm) for 5 min and were then challenged with capsaicin. In HEK-TRPV1 cells, PAR₁-AP failed to potentiate the response to 3 or 10 nm capsaicin (b, e). In DRG, PAR₁-AP failed to potentiate the magnitude of the response to 100 nm capsaicin (d, f) or affect the proportion of neurons that gave a detectable response to capsaicin (g); n = 160–320 cells, 28–53 neurons.

Figure 4.

PAR₂ activation excites DRG neurons in culture. Neurons were superfused for 3 min with PAR₂-AP or -RP (50 μm) or trypsin or SBTI-treated trypsin (10 nm) or puffed with tryptase (3.8 μm). The rheobase (a, b) and membrane potential (c, d) were measured at the indicated times. Trypsin and tryptase caused a sustained decrease in rheobase that was maintained for at least 50 min. PAR₂-AP caused a transient increase in rheobase at 10 min followed by a sustained decrease (a). This effect was nonspecific because PAR₂-RP produced an identical transient increase (data not shown) without a sustained effect on rheobase (b) or a transient decrease in membrane potential (d). PAR₂ agonists caused a transient decrease in membrane potential (c, d). *p < 0.01 compared with basal (a, b) or PAR₂-RP and trypsin (Try) + SBTI (c, d). Numbers in parentheses are numbers of neurons.

Figure 5.

PAR₂ activation sensitizes TRPV1 currents. Neurons were superfused for 3 min with PAR₂-AP or -RP (50 μm), trypsin (10 nm), or vehicle. They were challenged at the indicated times with capsaicin (500 nm), and currents were measured. a, Representative currents under basal conditions and at 30 min after activation of PAR₂. PAR₂ activation caused a marked and sustained increase in the capsaicin-induced current (a, b), and the increase observed 30 min after activation is shown in c. *p < 0.01 compared with basal (b) or vehicle and PAR₂-RP (c). Numbers in parentheses are numbers of neurons.

Figure 6.

PAR₂ activation sensitizes TRPV1 currents by a PKC-dependent mechanism. Neurons were superfused for 3 min with PAR₂-AP or -RP (50 μm) and then challenged with capsaicin (500 nm). PAR₂ activation caused a marked increase in the capsaicin current, and the increase was abolished by GFX (1 μm). GFX had no effect in neurons treated with PAR₂-RP. Representative traces are shown in a and c, and pooled data are shown in b and d. *p<0.01 compared with vehicle and PAR₂-AP. Numbers in parentheses are numbers of neurons.

Figure 7.

Agonist of PAR₂ induced phosphorylation of TRPV1. HEK-TRPV1 cells were exposed to PAR₂-AP or -RP (100 μm), PMA (10 μm), or DMSO (1:250) for 10 min. TRPV1 was immunoprecipitated (IP), and Western blots (WB) were probed for phosphoserine (a). Blots were then stripped and reprobed to TRPV1 to ensure equal loading (b). Activation of PAR₂ and PKC induced serine phosphorylation of TRPV1. Representative blots are shown.

Figure 8.

Mechanisms of PAR₂-induced thermal hyperalgesia in mice. PAR₂-AP or PAR₂-RP was administered by intraplantar injection, and latency of paw withdrawal was measured in C57BL/6 mice (a, c, d, f) or TRPV1^+/+ and TRPV1^–/– mice (b, e). a, PAR₂-AP decreased the latency of withdrawal, which was inhibited by systemic administration of capsazepine (capsaz; 35 mg/kg, s.c.). b, PAR₂-AP also decreased the latency of withdrawal in TRPV1^+/+ mice but not TRPV1^–/– mice. c, The effect of PAR₂-AP on withdrawal latency was dose-related. d, Coinjection of PAR₂-AP and capsaicin (Cap) at doses that alone did not cause hyperalgesia decreased the latency of paw withdrawal. e, The synergistic effects of coinjected PAR₂-AP and capsaicin were observed in TRPV1^+/+ but not TRPV1^–/– mice. f, Intraplantar injection of GFX (1 μg/paw) inhibited the synergistic effects of PAR₂-AP and capsaicin. veh, Vehicle. *p < 0.05 compared with basal withdrawal latency (b, d–f) or PAR₂-AP-induced hyperalgesia (a, c); n = 6–8 mice per group.

Figure 9.

PAR₂ activation sensitizes TRPV1-mediated release of SP (a) and CGRP (b). Slices of dorsal horn of the spinal cord were pretreated with vehicle (Veh.) or PAR₂-AP or -RP (10 μm) for 20 min and then challenged with capsaicin (0.5 or 1 μm). Pretreatment with PAR₂-AP enhances capsaicin-stimulated release of SP-LI and CGRP-LI. *p < 0.05 compared with vehicle or PAR₂-RP; n = 6–8 rats.