Activation of transient receptor potential ankyrin 1 evokes nociception through substance P release from primary sensory neurons

Authors


Address correspondence and reprint requests to Y. Nakata, Department of Pharmacology, Graduate School of Biomedical Sciences, Hiroshima University, Kasumi 1-2-3, Minami-ku, Hiroshima 734-8553, Japan. E-mail: ynakata@hiroshima-u.ac.jp

Abstract

J. Neurochem. (2012) 120, 1036–1047.

Abstract

To examine mechanisms underlying substance P (SP) release from primary sensory neurons in response to activation of the non-selective cation channel transient receptor potential ankyrin 1 (TRPA1), SP release from cultured rat dorsal root ganglion neurons was measured, using radioimmunoassay, by stimulating TRPA1 with allyl isothiocyanate (AITC), a TRPA1 agonist. AITC-evoked SP release occurred in a concentration- and time-dependent manner. Interestingly, p38 mitogen-activated protein kinase (p38) inhibitor SB203580 significantly attenuated AITC-evoked SP release. The in vivo effect of AITC-evoked SP release from primary sensory neurons in mice was evaluated. Hind paw intraplantar injection of AITC induced nociceptive behaviors and inflammation (edema, thermal hyperalgesia). AITC-induced thermal hyperalgesia and edema were inhibited by intraplantar pre-treatment with either SB203580 or neurokinin-1 receptor antagonist CP96345. Moreover, intrathecal pre-treatment with either CP96345 or SB203580 inhibited AITC-induced nociceptive behaviors and thermal hyperalgesia. Immunohistochemical studies demonstrated that intraplantar AITC injection induced the phosphorylation of p38 in mouse dorsal root ganglion neurons containing SP. These findings suggest that activation of TRPA1 evokes SP release from the primary sensory neurons through phosphorylation of p38, subsequent nociceptive behaviors and inflammatory responses. Furthermore, the data also indicate that blocking the effects of TRPA1 activation at the periphery leads to significant antinociception.

Abbreviations used
AITC

allyl isothiocyanate

DRG

dorsal root ganglion

DMEM

Dulbecco’s modified Eagle’s medium

i.pl.

intraplantar

i.t.

intrathecal

JNK

c-Jun N-terminal kinase

MAP

mitogen-activated protein

MEK

MAP kinase kinase

NK

neurokinin

PI3K

phosphoinositide 3-kinase

PKA

protein kinase A

PKC

protein kinase C

PTK

protein tyrosine kinase

SDS

sodium dodecyl sulfate

SP

substance P

TRP

transient receptor potential

TRPA1

TRP ankyrin 1

TRPV1

TRP vanilloid 1

Substance P (SP) is synthesized from the pre-protachykinin-A gene in dorsal root ganglion (DRG) neurons (Inoue et al. 1999) and released from both central and peripheral terminals of primary afferent sensory neurons, transmitting nociceptive information via an extraordinarily integrated and complex process. Substance P released from central terminals of primary sensory neurons activates neurokinin (NK) receptors in lamina I, II of the dorsal horn of spinal cord (Wang and Marvizón 2002). In addition, SP released in peripheral tissues from primary afferents induces vasodilation, lymphocyte proliferation and cytokine secretion from inflammatory cells (Severini et al. 2002; O’Connor et al. 2004).

Transient receptor potential ankyrin 1 (TRPA1) is a member of the transient receptor potential (TRP) family of non-selective cation channels. TRPA1 is expressed on small diameter peptide-containing DRG neurons but is not found in other tissues such as the brain, liver or lung (Story et al. 2003). Recent immunohistochemical studies have shown that TRPA1 co-exists with SP and/or transient receptor potential vanilloid 1 (TRPV1) expressing DRG neurons (Kondo et al. 2009; Kim et al. 2010). Thus, TRPA1 is found exclusively on nociceptive primary afferent neurons.

Treatment with a TRPA1 inhibitor significantly blocked complete Freund’s adjuvant-induced hind paw cold allodynia and mechanical hyperalgesia and TRPA1 knockout mice treated with adjuvant did not demonstrate cutaneous hypersensitivity to hind paw stimulation (Petrus et al. 2007; del Camino et al. 2010), indicating TRPA1 involvement in the induction of chronic inflammatory pain. Moreover, some studies have suggested that TRPA1-evoked SP release is involved in inflammatory bowel disease and acetaminophen-induced chronic obstructive pulmonary disease (Kondo et al. 2009; Engel et al. 2010; Nassini et al. 2010). TRPA1 is activated by noxious cold temperatures (<17°C) and mechanical stimulation. In addition, irritants such as allyl isothiocyanate (AITC), allicin, cinnamaldehyde and formalin activate TRPA1 and also evoke nociceptive behaviors and edema in rodents (Story et al. 2003; Corey et al. 2004; Bautista et al. 2006; McNamara et al. 2007). However, the mechanisms underlying SP release from peripheral and central terminals of primary sensory neurons in response to activation of TRPA1 with TRPA1 agonists are as yet unclear.

The current study investigated the mechanisms that might be associated with TRPA1-evoked SP release from cultured rat DRG neurons using AITC, a TRPA1 agonist, and whether AITC-induced SP release at either peripheral or central primary afferent terminals was crucial for the expression AITC-evoked nociceptive behaviors and inflammation in mice.

Materials and methods

All animal procedures were performed in accordance with the Guidelines for Animal Experimentation, Hiroshima University and reviewed and approved by the Committee of Research Facilities for Laboratory Animal Sciences, Graduate School of Biomedical Sciences, Hiroshima University, Japan.

Isolation and culture of DRG neurons

According to a previously described method, DRGs were removed from male Wistar rats (6–9 weeks) and dissociated into single cells by enzyme treatment with 0.125% collagenase and 0.25% trypsin followed by mechanical trituration (Miyano et al. 2009).

Measurement of SP release by radioimmunoassay

All experiments were performed using Krebs–HEPES buffer [NaCl 110, CaCl2 2, MgSO4 1.2, KH2PO4 1.2, NaHCO3 25, D-glucose 11.7, HEPES 5 (mM)] containing 0.1% bovine serum albumin and peptidase inhibitors (1 mM phosphoramidon, 1 mM captopril and 4 mg/mL bacitracin). Ca2+-free Krebs–HEPES buffer was prepared by removal of CaCl2. After washing, the cultured DRG neurons were pre-treated with inhibitors for 10 min at 37°C. Thereafter, neurons were treated with AITC for the designated periods of time (10, 30 and 60 min) at 37°C. Substance P content in the Krebs–HEPES buffer was measured using a sensitive radioimmunoassay as previously described (Miyano et al. 2009).

Western blot analysis

After treatment of cultured DRG neurons with the various drugs in Krebs–HEPES buffer, the cell samples were processed for western blot analysis as previously described (Morioka et al. 2010). Primary antibodies were raised against phospho-p38 mitogen-activated protein (MAP) kinase (1 : 1000) and total-p38 MAP kinase (1 : 1000), respectively. The horseradish peroxidase-conjugated anti-rabbit secondary antibodies (1 : 500) were used for chemiluminescence detection according to the manufacturer’s instructions.

AITC-induced nociceptive behaviors, paw edema and thermal hyperalgesia in mice

Behavioral testing was performed in adult, male ddY mice (5–6 weeks) after 1 week of acclimation to the housing facility. Following habituation to the testing apparatus, 25 μL AITC (0.025–2.5 nmol/paw) was injected into the plantar (i.pl.) left hind paw and the total duration of nociceptive behaviors (licking and flinching of the injected paw, measured in seconds) was recorded over a period of 5 min. In control mice, the hind paw was injected with an equal volume of saline. Paw edema was defined as the percent increase in hind paw thickness compared with the baseline value (prior to the injection of AITC or saline). Paw thickness was measured 30 min after either AITC or saline injection. Hind paw responses to noxious heat was assessed with the hot-plate test (Ugo Basile, Comerio, VA, Italy) and expressed as the paw withdrawal latency. Mice were placed on the hot-plate apparatus (52°C) 30 min after AITC injection, and the latency of the injected hind paw to respond (licking and lifting) was measured in seconds. A cut-off of 20 s was used to prevent tissue damage. Mice were pre-treated, either i.pl. or intrathecally (i.t.), with inhibitors 10 min (capsazepine, HC-030031) or 30 min (CP96345, SB203580) prior to i.pl. AITC injection.

Immunohistochemistry

After pre-treatment with either SB203580 (1.5 nmol, i.t., 15 μL) or saline (i.t., 15 μL), mice were injected 10 min later with either AITC (0.25 nmol, i.pl., 25 μL) or saline (i.pl., 25 μL) and euthanized 10 min following the second injection. Under sodium pentobarbital anesthesia (50 mg/kg, i.p.; Dainippon Sumitomo Pharma, Osaka, Japan), mice were transcardially perfused with 20 mL of saline followed by 50 ml of freshly prepared 4% (w/v) paraformaldehyde in 0.1 M phosphate buffer (pH = 7.4). The lumbar (L4–L6) segments of spinal cord and DRG neurons were quickly removed and post-fixed in the same fixative solution for three days at 4°C and then cryoprotected overnight in 30% (w/v) sucrose in 0.1 M phosphate buffer at 4°C. Tissues were embedded in Tissue-Tek OCT compound 4583 (Sakura Finetech, Tokyo, Japan) and frozen in liquid nitrogen, then cut serially (20 μm thickness) by cryostat, and collected onto glass slides (MAS-GP type A; Matsunami Glass, Osaka, Japan). After slides were dried at 25°C, tissue sections were processed for double-labeling immunohistochemistry for phospho-p38 MAP kinase and SP, or NK1 receptor and NeuN (a marker for neurons). The tissue sections were rinsed with phosphate-buffered saline, incubated in 10% goat serum with 3% bovine serum albumin, 0.1% Triton X and 0.05% Tween-20 in phosphate-buffered saline for 2 h at 25°C, and then incubated with a mixture of rabbit anti-phospho-p38 MAP kinase antibody (1 : 1000) and mouse anti-SP antibody (1 : 1000) or rabbit anti-NK1 receptor antibody (1 : 1000) and mouse anti-NeuN antibody (1 : 1000) for 72 h at 4°C followed by a corresponding secondary antibody (1 : 200) conjugated with either Alexa 488 or 546 for 2 h at 4°C in a dark chamber. The sections were then extensively washed in phosphate-buffered saline and then coverslipped. Sections were examined with a LSM5 PASCAL laser scanning microscope (Carl Zeiss, Jena, Germany).

Drugs

Unless otherwise indicated, all materials obtained from Sigma Chemical Co. and were of the highest purity (St Louis, MO, USA).

AITC, a TRPA1 agonist (Nakarai Tesque, Kyoto, Japan); CP96345, a NK1 receptor antagonist (Pfizer Central Research, Groton, CT, USA); genistein, a protein tyrosine kinase (PTK) inhibitor (Tocris Cookson, Bristol, UK); Gö6983, a protein kinase C (PKC) inhibitor (Calbiochem, La Jolla, CA, USA); H89, a protein kinase A (PKA) inhibitor (Seikagaku, Tokyo, Japan); LY294002, a phosphoinositide 3-kinase (PI3K) inhibitor, U0126, a MAP kinase (MEK) inhibitor, and wortmannin, a PI3K inhibitor (Promega, Madison, WI, USA); capsazepine, a TRPV1 antagonist, HC-030031, a TRPA1 antagonist, SB203580, a p38 MAP kinase inhibitor, and SP600125, a c-Jun N-terminal kinase (JNK) inhibitor.

Antibodies

Goat anti-rabbit IgG antibody (Alexa Fluor 555) and goat anti-mouse IgG antibody (Alexa Fluor 488; Invitrogen, Burlington, ON, Canada); anti-rabbit SP antibody (Santa Cruz biotechnology, Santa Cruz, CA, USA); anti-mouse NeuN antibody (Millipore Corporation, Bedford, MA, USA); anti-rabbit IgG horseradish peroxidase-linked antibody, anti-rabbit phospho-p38 mitogen-activated protein (MAP) kinase antibody and anti-rabbit total-p38 MAP kinase antibody (Cell Signaling Technology, Beverly, MA, USA); anti-rabbit NK1 receptor antibody.

Tissue culture, buffer and radioimmunoassay materials

Bacitracin, bovine serum albumin, captopril and Dulbecco’s modified Eagle’s medium (Nissui, Tokyo, Japan); glutamine and horse serum (Gibco-BRL, Gaithersburg, MD, USA); 2.5% trypsin (Invitrogen, Burlington, ON, Canada); mouse laminin (BD Biosciences, Bedford, MA, USA); nerve growth factor (2.5 S), penicillin/streptomycin, phosphoramidon and polyethyleneimine (Promega, Madison, WI, USA); [125I–Try8]-SP (81.4 TBq/mmol; New England Nuclear, Boston, MA, USA).

Statistical analysis

The data are presented as mean ± SEM of at least three independent experiments. Statistical analysis of the data was performed by one-way analysis of variance (ANOVA) followed by Student’s t-test and two-tailed t-test with Bonferroni correction following one-way ANOVA. A probability value (p) of less than 0.05 was considered to be statistically significant.

Results

Activation of TRPA1 evokes SP release from cultured rat DRG neurons

To clarify the possible molecular mechanism of SP release evoked by activation of TRPA1, the effects of AITC, a specific agonist of TRPA1 (Jordt et al. 2004), on SP release from cultured rat DRG neurons were examined. As shown in Fig. 1a, significant concentrations of SP were found in the buffer over time following treatment with 10 μM AITC and a significant concentration of SP could be measured 60 min following stimulation. Significant release of SP was observed with AITC treatment for 10 min at concentrations of 10 and 100 μM (Fig. 1b). Thus, treatment with 10 μM AITC for 10 min was used to further investigate the mechanism of SP release from cultured DRG neurons. Furthermore, pre-treatment with HC-030031, a specific TRPA1 antagonist (Eid et al. 2008), and removal of extracellular Ca2+ from the Krebs–HEPES buffer, completely inhibited AITC-evoked SP release (Fig. 2a and b). However, pre-treatment with capsazepine (100 nM), a specific TRPV1 antagonist (Tang and Nakata 2008), did not inhibit AITC-evoked SP release (Fig. 2a).

Figure 1.

 Effect of TRPA1 agonist AITC on SP release from cultured rat DRG neurons. (a) Time course of AITC-evoked SP release. Cells were treated with either 10 μM AITC or control buffer for the indicated periods of time in minutes. (b) Concentration-dependent SP release evoked by AITC. Cells were treated with increasing concentrations of AITC (μM) for 10 min. Data are expressed as mean ± SEM. n = 3–5 independent experiments. **p < 0.01 compared with control-treated DRG neurons.

Figure 2.

 Effects of TRPA1 antagonist and removal of extracellular Ca2+ on AITC-evoked SP release from cultured rat DRG neurons. (a) DRG neurons were pre-treated with 10–50 μM TRPA1 antagonist HC-030031 or 100 nM of TRPV1 antagonist capsazepine for 10 min. After pre-treatment, cells were stimulated with 10 μM AITC for 10 min. When given alone, 50 μM HC-030031 and 100 nM capsazepine did not affect SP release. (b) DRG neurons were treated with 10 μM AITC in the presence or absence of Ca2+ in Krebs-HEPES buffer. Data are expressed as mean ± SEM. n = 4–6 independent experiments. *p < 0.05, **p < 0.01 compared with control-treated DRG neurons. p < 0.05 compared with DRG neurons treated with AITC alone. NS, not significant.

Effects of kinase inhibitors on AITC-evoked SP release

Previous studies have demonstrated the importance of various intracellular signaling factors, including MAP kinases, in the regulation of SP release from neurons (Suzuki et al. 1998; Tang et al. 2006, 2007; Miyano et al. 2009). Therefore, the role of PKA, PKC, PTK, PI3K, MEK, JNK, and p38 MAP kinase in AITC-evoked SP release from cultured rat DRG neurons was investigated. As shown in Fig. 3a, 10 μM of SB203580, a p38 MAP kinase inhibitor (Tang and Nakata 2008), significantly attenuated AITC-evoked SP release. However, H89 (10 μM), a PKA inhibitor (Morioka et al. 2010); Gö6983 (3 μM), a PKC inhibitor (Morioka et al. 2010); genistein (50 μM), a PTK inhibitor (Morioka et al. 2010); wortmannin (10 μM) and LY294002 (10 μM), PI3K inhibitors (Tang and Nakata 2008); U0126 (10 μM), a MEK1/2 inhibitor (Tang and Nakata 2008); and SP600125 (20 μM), a JNK inhibitor (Tang and Nakata 2008), had no effect on AITC-evoked SP release (Fig. 3a–c).

Figure 3.

 Effects of kinase inhibitors on AITC-evoked SP release from cultured rat DRG neurons. (a) DRG neurons were pre-treated with either buffer, 10 μM of MEK1/2 inhibitor U0126, 10 μM of the p38 MAP kinase inhibitor SB203580, or 20 μM of the JNK inhibitor SP600125 for 10 min. (b) DRG neurons were pre-treated with either buffer, 10 μM of the PKA inhibitor H89, 3 μM of the PKC inhibitor Go6983 or 50 μM of the PTK inhibitor genistein for 10 min. (c) DRG neurons were pre-treated with either buffer or PI3K inhibitors, wortmannin (10 μM) or LY294002 (10 μM), for 30 min. After pre-treatment, DRG neurons were treated with 10 μM AITC for 10 min. The kinase inhibitors alone did not affect SP release. Data are expressed as mean ± SEM. n = 6 independent experiments. **p < 0.01 compared with control-treated DRG neurons. p < 0.05 compared with DRG neurons treated with AITC alone.

TRPA1-induced phosphorylation of p38 MAP kinase in cultured rat DRG neurons

The results shown in Fig. 3 suggest that activation of TRPA1 might induce phosphorylation of p38 MAP kinase, which in turn leads to the release of SP from cultured rat DRG neurons. As demonstrated by western blotting (Fig. 4a), a time-dependent phosphorylation of p38 MAP kinase was observed following treatment with 10 μM AITC. This response was dependent on levels of extracellular Ca2+, as removal of Ca2+ from the buffer prevented AITC-evoked phosphorylation of p38 MAP kinase. Furthermore, pre-treatment with either HC-030031 or SB203850 completely inhibited SP release, indicating that the effect of AITC is TRPA1-mediated and p38 MAP kinase-dependent (Fig. 4b).

Figure 4.

 Phosphorylation of p38 MAP kinase induced by AITC in cultured rat DRG neurons. (a) DRG neurons were treated with 10 μM AITC for the indicated periods of time (0, 1, 3, 5 and 10 min). (b) DRG neurons were pre-treated with either buffer, 50 μM of the TRPA1 antagonist HC-030031 (HC) or 10 μM of the p38 MAP kinase inhibitor SB203580 (SB) for 10 min. After pre-treatment, cells were treated with 10 μM AITC for 10 min. In a separate experiment, DRG neurons were treated with 10 μM of AITC in either the presence or absence of Ca2+ in Krebs-HEPES buffer. Treatment with either HC, SB, or Ca2+-free Krebs-HEPES buffer alone did not affect SP release. The upper panel indicates representative western blots. The graph in the lower panel quantifies levels of phospho-p38 (p-p38) MAP kinase against total-p38 (t-p38) MAP kinase as a ratio. Data are expressed as the mean ± SEM. n = 4–5 independent experiments. *p < 0.05 compared with DRG neurons treated with AITC at 0 min (a) or to control-treated (b). ††p < 0.01 compared with DRG neurons treated with AITC alone.

AITC-induced nociceptive and inflammatory responses in mice

To confirm whether AITC actually induces SP release from sensory nerve endings in peripheral tissue in vivo, AITC was injected into the hind paw of mice. Injection with AITC into the hind paw induced significant acute nociceptive behaviors (licking and flinching) and paw edema (Fig. 5a and b). Moreover, i.pl. injection with AITC significantly reduced the hind paw withdrawal latency (thermal hyperalgesia) in the hot-plate test (Fig. 5c). Intraplantar pre-treatment with HC-030031 (0.25–2.5 nmol) significantly inhibited AITC-induced nociceptive behaviors, paw edema and thermal hyperalgesia (Fig. 5d–f). By contrast, i.t. pre-treatment with HC-030031 (1.5 nmol) did not affect these responses (data not shown). In addition, i.pl. pre-treatment with capsazepine (0.25 nmol) did not affect nociceptive behaviors or paw edema (Fig. 5d and e), but capsazepine pre-treatment significantly ameliorated thermal hyperalgesia (Fig. 5f).

Figure 5.

 AITC-induced nociceptive and inflammatory responses in mice. Intraplantar (i.pl.) injection of AITC (25 μL, 0.025–2.5 nmol/paw) induced nociceptive behaviors (a), paw edema (b) and thermal hyperalgesia (c). Mice were i.pl. pre-treated with either saline (25 μL), TRPA1 antagonist HC-030031 (25 μL, 0.25–2.5 nmol/paw) or TRPV1 antagonist capsazepine (25 μL, 0.25 nmol) for 10 min (d–f). Following pre-treatment, AITC (25 μL, 0.25 nmol/paw) was injected i.pl. Pre-treatment with either HC-030031 or capsazepine alone did not induce nociceptive behaviors or an inflammatory response. Data are expressed as mean ± SEM. n = 9/treatment group. *p < 0.05 and **p < 0.01 compared with saline-treated mice. p < 0.05, ††p < 0.01 compared with mice treated with AITC alone.

To confirm that the effect of AITC was mediated through the release of SP from sensory nerve endings in the skin, the NK1 receptor antagonist CP96345 (Tang et al. 2008) was injected i.pl. prior to i.pl. AITC injection. Pre-treatment with CP96345 did not affect subsequent AITC-induced nociceptive behaviors (Fig. 6a), but inhibited paw edema and thermal hyperalgesia (Fig. 6b and c). By contrast, i.t. pre-treatment with CP96345 (0.15 or 1.5 nmol) significantly inhibited AITC-induced nociceptive behaviors and thermal hyperalgesia but not paw edema (Fig. 6d–f).

Figure 6.

 Effect of NK1 receptor antagonism on AITC-induced nociceptive and inflammatory responses in mice. After pre-treatment with either intraplantar (0.25 or 2.5 nmol, 25 μL) (a–c) or intrathecal (0.15 or 1.5 nmol, 15 μL) (d–f) NK1 receptor antagonist CP96345 for 30 min, AITC (25 μL, 0.25 nmol/paw) was injected intraplantar and spontaneous nociceptive behaviors (a, d), paw edema (b, e) and withdrawal latency to noxious heat (c, f) were measured. Neither intraplantar nor intrathecal CP96345, when given alone, affected nociceptive behaviors or inflammatory responses. Data are expressed as mean ± SEM. n = 8–9/treatment group. *p < 0.05, **p < 0.01 compared with saline-treated mice. p < 0.05 and ††p < 0.01 compared with mice treated with AITC alone. NS, not significant

Finally, the role of p38 MAP kinase in AITC-induced nociceptive and inflammatory responses was investigated in vivo. Intraplantar pre-treatment with SB203850 (0.025 or 0.25 nmol) significantly inhibited AITC-induced nociceptive behaviors, paw edema and thermal hyperalgesia (Fig. 7a–c). Moreover, i.t. pre-treatment with SB203850 (0.015 or 0.15 nmol) significantly inhibited AITC-induced nociceptive behaviors and thermal hyperalgesia but not paw edema (Fig. 7d–f).

Figure 7.

 Effects of p38 MAP kinase inhibition on AITC-induced nociceptive or inflammatory responses in mice. After pre-treatment with either intraplantar (0.025 or 0.25 nmol, 25 μL) (a–c) or intrathecal (0.015 or 0.15 nmol, 15 μL) (d–f) p38 MAP kinase inhibitor SB203580 for 30 min, AITC (25 μL, 0.25 nmol/paw) was injected intraplantar and spontaneous nociceptive behaviors (a, d), paw edema (b, e) and withdrawal latency to noxious heat (c, f) were measured. Neither intraplantar nor intrathecal SB203580, when given alone, affected nociceptive behaviors or inflammatory responses. Data are expressed as mean ± SEM. n = 7–9/treatment group. *p < 0.05 and **p < 0.01 compared with saline-treated mice. p < 0.05 and ††p < 0.01 compared with mice treated with AITC alone.

AITC-induced phosphorylation of p38 MAP kinase in SP-containing DRG neurons

Immunohistochemical studies were performed to determine whether activation of TRPA1 induced phosphorylation of p38 MAP kinase in mouse DRG neurons. Hind paw injection with saline had no effect on p38 MAP kinase phosphorylation (Fig. 8a). By contrast, 10 min after hind paw AITC injection (0.25 nmol, i.pl.), phosphorylation of p38 MAP kinase was observed in DRG neurons (Fig 8b). Also, some phospho-p38 MAP kinase positive neurons were double-labeled for SP. Furthermore, pre-treatment with SB203580 (1.5 nmol, i.t.) inhibited AITC-induced phosphorylation of p38 MAP kinase (Fig. 8c).

Figure 8.

 Fluorescent photomicrographs of mouse DRG neurons following AITC injection into the hind paw. Lumbar (L4-L6) DRG sections from either saline (a) or AITC (b, c)-injected mice (25 μL, i.pl.) were incubated with anti-phospho-p38 (p-p38) MAP kinase and anti-substance P (SP) antibodies. AITC induced phosphorylation of p38 MAP kinase in SP containing DRG neurons. In a separate experiment, mice were intrathecally (i.t.) pre-treated (15 μL) with either saline (a, b) or the p38 MAP kinase inhibitor SB203580 (c). Ten minutes following the first injection, AITC (or saline) was injected i.pl. The photomicrographs in the ‘Merge’ column are merged images of the p-p38 MAP kinase and SP panels. Scale bars are 50 μm. Yellow arrow heads indicate expression of SP in the DRG neurons. Yellow arrows indicate co-localization of p-p38 MAP kinase and SP in DRG neurons.

AITC-induced NK1 receptor internalization in the spinal dorsal horn neurons

Previous studies showed NK1 receptor internalization upon receptor stimulation in spinal cord slices and in vivo studies (Chen et al. 2009; Zhang et al. 2010; Huang et al. 2011). Therefore, internalization of NK1 receptors on spinal dorsal horn neurons was investigated following i.pl. injection with AITC. Figure 9a and b show representative images of mouse spinal dorsal horn after injection with either ip.l. saline (25 μL) or AITC (0.25 nmol, 25 μL), respectively. Neurons, labeled with NeuN, in spinal dorsal horn after injection with saline show a distinct ring of NK1 receptor staining on the cell membrane. By contrast, in spinal dorsal horn neurons from mice injected with AITC, NK1 receptor immunostaining was abolished from the cell membrane, and was found instead within the neurons, indicating receptor internalization. Pre-treatment with SB203580 (1.5 nmol, i.t.) inhibited AITC-induced NK1 receptor internalization in spinal dorsal horn neurons (Fig. 9c). Each graph shows the fluorescence intensities of either NK1 receptor (red) or NeuN (green) immunostaining of the white line drawn across the neurons (last column of figures). Note that NK1 receptor fluorescence intensity is diminished from the edges of the neuron following i.pl. AITC treatment compared with that of i.pl. saline-injected mice.

Figure 9.

 The blockade of AITC-induced NK1 receptor internalization by p38 MAP kinase inhibitor in mouse spinal dorsal horn neurons. Lumbar (L4–L6) spinal cord sections of either saline (a) or AITC (b, c)-injected mice (25 μL, i.pl.) were incubated with anti-NeuN and anti-NK1 receptor antibodies. Note the presence of NK1 receptor immunostaining in the cell membrane from saline-treated mice (a).Intraplantar injection of saline did not lead to NK1 receptor internalization. In a separate study, mice were i.t. pre-treated (15 μL) with either saline (a, b) or SB203580 (c). Ten minutes later, mice were i.pl. injected with AITC (or saline). The inserts in the upper left-hand corners show spinal neurons magnified from the white boxes. Fluorescent intensity was measured across the white lines shown in the ‘Merge’ column. The photomicrographs in the ‘Merge’ column are merged images of the ‘NK1 receptor’ and ‘NeuN’ panels. Scale bars are 50 μm for the main panels and 5 μm for the insets. Yellow arrows indicate NK1 receptor-positive neurons without receptor internalization.

Discussion

The current study demonstrated key steps in the intracellular-signaling mechanisms by which activation of TRPA1 on primary sensory neurons leads to the release of SP. In addition, activation of peripherally expressed TRPA1 through i.pl. injection of AITC in mice produced robust inflammatory nociceptive responses involving SP release from the peripheral as well as the central terminals of primary sensory neurons. The current data suggest that SP is released from the terminals of primary sensory neurons following activation of TRPA1, and that release is dependent on an influx of extracellular Ca2+ and phosphorylation of p38 MAP kinase. Moreover, TRPA1 activation triggers NK1 receptor-dependent inflammatory responses in peripheral tissues, including edema and thermal hyperalgesia, and nociceptive responses that are both peripherally and centrally mediated by NK1 receptors. Figure 10 summarizes the findings of the current study.

Figure 10.

 Schematic representation of the mechanism of TRPA1-mediated SP release and subsequent events in the central and peripheral terminals of primary sensory neurons.

In the DRG neuron preparation, it was found that SP release was preceded by phosphorylation of p38 MAP kinase and either pre-treatment with HC-030031 or removal of extracellular Ca2+ inhibited phosphorylation of p38 MAP kinase evoked by AITC. In addition, the present immunohistochemical studies demonstrated that i.pl. injection with AITC induced phosphorylation of p38 MAP kinase in SP containing DRG neurons of mice. Recent in vivo studies also showed that activation of TRPA1 by noxious cold stimulation-induced phosphorylation of p38 MAP kinase in DRG neurons colocalized with TRPA1, which was inhibited by pre-treatment with TRPA1 antisense oligodeoxynucleotide (Obata et al. 2005; Mizushima et al. 2006). Some authors also reported that activation of other ion channels induced phosphorylation of p38 MAP kinase through the elevation of the intracellular concentration of Ca2+ (Zhu et al. 2002; Trang et al. 2009). These results confirm that activation of TRPA1 leads to phosphorylation of p38 MAP kinase via an increase in intracellular Ca2+ in DRG neurons.

Phosphorylation of p38 MAP kinase is an important intracellular process in the regulation of various cell responses, including the release of inflammatory mediators (Severini et al. 2002; Suzuki et al. 2004; Clark et al. 2006; Trang et al. 2009). Stimulation of the ionotropic purinoceptor P2X4s by ATP causes the release of brain-derived neurotrophic factor from brain microglia and this response is also dependent on the influx of extracellular Ca2+ and subsequent phosphorylation of p38 MAP kinase. The present study showed that pre-treatment with SB203580 or removal of extracellular Ca2+ inhibited SP release from cultured rat DRG neurons. Furthermore, although potential involvement of other protein kinases, including PKA, PKC, PTK, PI3K, ERK and JNK, were evaluated with the use of selective inhibitors, these kinases did not contribute to AITC-evoked SP release. Moreover, i.pl. treatment with AITC induced NK1 receptor internalization in the spinal dorsal horn, an indication that the NK1 receptor was activated because of release of SP from central primary afferent terminals, and i.t. pre-treatment with SB203580 inhibited receptor internalization, indicating that terminal release of SP was mediated through a mechanism dependent upon p38 MAP kinase. The current results clearly demonstrate, using pharmacological and immunohistochemical approaches, that activation of p38 MAP kinase plays an important role in TRPA1-evoked SP release from DRG neurons.

Recent studies showed that activation of TRPA1 on central and peripheral nerve terminals contributed to cutaneous hypersensitivity (Merrill et al. 2008; da Costa et al. 2010; Wei et al. 2011). In this study, i.pl. injection of AITC induced nociceptive behaviors, such as licking and flinching of the injected paw, thermal hyperalgesia and paw edema. Intraplantar, but not i.t., pre-treatment with HC-030031significantly reduced theses responses. Intraplantar pre-treatment with CP96345 inhibited AITC-induced paw edema and thermal hyperalgesia but not nociceptive behaviors, and i.t. pre-treatment with CP96345 inhibited nociceptive behaviors and thermal hyperalgesia but not paw edema, as shown in Figs 5 and 6. Moreover, i.t. pre-treatment with SB203580 inhibited both AITC-induced pain behaviors and NK1 receptor internalization in mouse dorsal horn (Figs 7 and 9). These results indicated that activation of TRPA1 on peripheral nerve terminals (but not central nerve terminals) by i.pl. injection with AITC evoked SP release from both the peripheral and central terminals of primary sensory neurons though phosphorylation of p38 MAP kinase, with SP acting as an inflammatory mediator in peripheral tissues to induce edema and thermal hyperalgesia, and as a ‘pain transmitter’, transmitting nociceptive signals from primary afferents to spinal dorsal horn neurons expressing the NK1 receptor. The in vitro data indicated that activation of TRPA1 in DRG neurons can lead to release of SP. However, it is not known if release would be via either peripheral or central primary afferent terminals. The in vivo data suggest that TRPA1-induced release of SP could be bi-directional, from either end of the primary afferent sensory neuron.

This study demonstrated that i.pl. pre-treatment with capsazepine inhibited AITC-induced thermal hyperalgesia but not nociceptive behaviors or paw edema (Fig. 5d–f). Therefore, these results suggest that TRPV1 might play an important role in AITC-induced thermal hyperalgesia and that hind paw edema but nociceptive behaviors are independent of TRPV1 activation. One possible explanation is that SP is released after stimulation of TRPA1 might directly or indirectly sensitize TRPV1 channels located on primary afferent sensory neurons, as shown in Fig. 10, and this response might be involved in the induction of thermal hyperalgesia. A previous study reported that primary sensory neurons that expressed substance P also expressed NK1 receptors (von Banchet and Schaible 1999), and we have previously demonstrated that exposure of cultured rat DRG neurons to a NK1 receptor agonist sensitized TRPV1 channels through the phosphorylation of the NK1 receptor, a direct intracellular mechanism of action of SP on the TRPV1 channel (von Banchet and Schaible 1999; Tang et al. 2008). Furthermore, some studies have reported that SP released from peripheral nerve terminals stimulated the release of several inflammatory mediators, such as histamine and prostaglandins, from epithelial or mast cells (Petersen et al. 1994; Szarek et al. 1998; Chiang et al. 2000), and that these mediators also regulate TRPV1 function, indicating an indirect action of SP (Petersen et al. 1994; Szarek et al. 1998; Chiang et al. 2000; Moriyama et al. 2005; Hudmon et al. 2008; Kajihara et al. 2010). Furthermore, these inflammatory mediators alone sensitize primary afferents, leading to increased sensitivity to cutaneous stimulation (e.g. hyperalgesia). These observations raise the possibility that SP release by activation of TRPA1 in peripheral tissue may induce thermal hyperalgesia through direct as well as indirect actions on the functional regulation of TRPV1. Further investigation would be required to clarify the interactions of the various substances in tissue that are released following injury. AITC could serve as a pharmacological tool to dissect the mechanism of tissue injury-induced hyperalgesia.

In conclusion, the current study found that activation of TRPA1 is Ca2+ dependent and leads to phosphorylation of p38 MAP kinase, which in turn leads to SP release from DRG neurons. Other intracellular kinases do not appear to play crucial roles in the release of SP following TRPA1 activation. Moreover, as shown in vivo, stimulation of TRPA1 at peripheral nerve terminals produced inflammatory nociceptive responses that were dependent on SP. These observations provide new evidence that SP is a key molecule in TRPA1-associated nociceptive and inflammatory responses at the peripheral primary afferent terminal. Thus, modulation of TRPA1 could prove to be a potent therapeutic target for both inflammatory disease states and peripherally mediated pain states.

Acknowledgements

This study was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology and by grants from the Japanese Smoking Research Association. The authors declare no potential conflict of interests. Experiments were carried out using equipment at the Analysis Center of Life Science, Hiroshima University and the Research Center for Molecular Medicine, Faculty of Medicine, Hiroshima University. We also thank Jeffrey L. Hart (RERF, Japan) and Dr Aldric T. Hama for their critical reading of the manuscript.

Ancillary