J. Neurochem. (2010) 115, 363–372.
Prostaglandin E2 (PGE2) is a well known pain and pro-inflammatory mediator abundantly produced in inflamed tissue. It causes pain by directly exciting nociceptive primary sensory neurons (nociceptors) and indirectly stimulating the release of pain-related peptide substance P (SP) and calcitonin gene-related peptide (CGRP). In an ex vivo culture of sensory ganglion explants, we tested the hypothesis that PGE2 could induce the synthesis of SP and CGRP in nociceptors. A stabilized PGE2 analog, 16,16-dimethyl PGE2, in a concentration- and time-dependent manner, significantly increased mRNA and peptide levels of SP and CGRP. The agonists of EP1 and EP4 receptors also significantly increased SP and CGRP levels. Moreover, 16,16-dimethyl PGE2-induced SP and CGRP were blocked by EP1 and EP4 antagonists as well as the inhibitors of both protein kinase A and protein kinase C. Nerve growth factor was partially involved in PGE2-induced SP and CGRP synthesis. Taken together, these results indicate that PGE2 contributes to the synthesis of SP and CGRP in nociceptors, an event mediated by EP1 and EP4 receptors, nerve growth factor and protein kinase A and protein kinase C signalling pathways. We thus conclude that facilitating the synthesis of pain-related peptides in nociceptors is a novel mechanism underlying the role of PGE2 in nociception and chronic pain states.
1-hydroxy prostaglandin E1
calcitonin gene-related peptide
dorsal root ganglion
fetal bovine serum
nerve growth factor
protein kinase A
protein kinase C
As a well known inflammatory and pain mediator, prostaglandin E2 (PGE2) is abundantly produced by inflammatory cells in inflamed tissues. PGE2 induced nociception is mediated through its four EP receptors expressed in nociceptive dorsal root ganglion (DRG) neurons (nociceptors) and in the spinal dorsal horn neurons (Vanegas and Schaible 2001). It causes pain by directly exciting nociceptors (Vanegas and Schaible 2001) and indirectly stimulating the release of pain-related neuropeptides including substance P (SP) and calcitonin gene-related peptide (CGRP) from nociceptors (Vasko et al. 1994; Vasko 1995, White 1996). However, it is entirely unknown whether PGE2 is able to induce the synthesis of SP and CGRP in primary sensory neurons.
Substance P and CGRP are enriched in small size DRG neurons and play important roles in neurogenic inflammation and nociception (Brain 1997; Fernandes et al. 2009). Both peptides are well known pain neurotransmitters in the dorsal horn and the activation of their receptors contribute to the central sensitization and maintenance of hyperalgesia (Seybold 2009). During inflammation, the two peptides in primary sensory neurons are up-regulated (Brain 1997). Complete axotomy of peripheral nerves dramatically reduced both peptides in DRG neurons (Hokfelt et al. 1994) while partial nerve injury increases the two peptides in spared DRG neurons (Ma et al. 1997, 1999). Their receptor antagonists are able to attenuate both inflammatory and neuropathic pain, thus the two peptides and their receptors are potential therapeutic targets to treat the two chronic pain states (Fernandes et al. 2009). Therefore, research to uncover the factors which are able to facilitate the synthesis of the two peptides is of great clinical significance.
We have previously shown that one important enzyme in PGE2 synthesis, cyclooxygenase 2 (COX-2), is dramatically up-regulated in invading macrophages in injured rat nerves (Ma and Eisenach 2002). One of its important end product, PGE2, is thus consequently increased in injured nerves after peripheral nerve injury (Muja and Devries 2004; Ma and Quirion 2005). The up-regulation of COX-2 and over-production of PGE2 last for one and half years in injured nerves of rats (Ma et al. 2008). Consistently, up-regulation of COX-2 was also found in injured nerves of human patients suffering from neuropathic pain up to 2 years (Durrenberger et al. 2004). All these data suggest that the over-production of PGE2 in injured nerves is a chronic and universal phenomenon occurring following nerve injury. Increased PGE2 in injured nerves likely exerts not only acute, but also chronic effects on DRG neurons by stimulating EP receptors present in en passant injured and spared axons, thus contributing to the long term plasticity in DRG which underlies the maintenance of neuropathic pain (Scholz and Woolf 2007).
Our prior studies showed that peripheral injection of a non-selective COX-2 inhibitor (Ma and Eisenach 2003) or a selective COX-2 inhibitor (Ma et al. 2008) dramatically suppressed the expression of SP and CGRP in DRG and in the superficial dorsal horn following nerve injury or inflammation. Others also reported that a selective COX-2 inhibitor, rofecoxib, also suppressed joint inflammation increased CGRP levels in DRG neurons (Staton et al. 2007). These observations suggest that peripherally-derived PGE2 is likely involved in the synthesis of the two pain-related peptides in nociceptors. In the present study, this hypothesis was tested in an ex vivo model of sensory ganglion explant culture. The first purpose of the present study is to determine if chronic PGE2 treatment is able to induce SP and CGRP in primary sensory neurons at the transcriptional and translational levels. As PGE2 has a rather short life-span (2.5 h), a stabilized PGE2 analog, 16,16-dimethyl PGE2 (dmPGE2), was used, which has been widely used as a long-acting substitute of PGE2 in chronic treatment in literature. We next explored the EP receptor subtype(s) and the intracellular signal transduction pathways which are involved in PGE2-induced synthesis of SP and CGRP. As nerve growth factor (NGF) is a well known inducer of SP and CGRP in adult DRG neurons (Lindsay and Harmar 1989; Donnerer et al. 1992), the third objective is to investigate the role of NGF in PGE2-induced synthesis of SP and CGRP.
Materials and methods
Hank’s balanced buffered saline, Dulbecco’s Modified Eagle’s Medium, HEPES buffer, penicillin/streptomycin, fetal bovine serum (FBS), Trizol, Molony murine leukemia virus reverse transcriptase and 4–20% polyacrylamide gel were purchased from Invitrogen, Burlington, ON, Canada. dmPGE2, 17-phenyl trinor prostaglandin E2 (17-PGE2), butaprost, sulprost, 1-hydroxy prostaglandin E1 (1-OH-PGE1), NS-398 and rabbit polyclonal anti-EP1–4 antisera were purchased from Cayman Chemical Inc., Ann Arbour, MI, USA. AH23848, mouse monoclonal anti-β-actin antiserum, dimethylsulfoxide and dNTP mix were purchased from Sigma-Aldrich Canada Ltd., Oakville, ON, Canada. NGF receptor chimera, NGF ELISA kit, ristocetin-induced platelet agglutination lysis buffer and protease inhibitor cocktail were purchased from R&D Systems, Burlington, ON, Canada. SP and CGRP extraction-free ELISA kits were purchased from Bachem, Torrance, CA, USA. GF109203X and H89 were purchased from Biomol, Plymouth Meeting, PA, USA. TaqMan Universal PCR Master Mix, random hexamers, 4352340E, Rn01500392-m1 and Rn01511353-g1 were purchased from Applied Biosystems, Foster City, CA, USA.
Sensory ganglion explant cultures
Adult male Sprague–Dawley rats (300 g in body weight) were decapitated and all DRG and trigeminal ganglia were removed. After rinsing with Hank’s balanced buffered saline twice, all ganglia were incubated in a 35 mm Petri dish containing Dulbecco’s Modified Eagle’s Medium and 1% HEPES buffer solution, penicillin/streptomycin (1 : 200) and 10% heat inactivated FBS in the presence of the following treatments: vehicle (5% dimethylsulfoxide or 0.01% ethanol), dmPGE2 (1–100 μM), 17-PGE2 (EP1 agonist, 1, 25, 50 μM), butaprost (EP2 agonist, 1, 25, 50 μM), sulprostone (EP3 agonist, 1, 25, 50 μM), 1-OH-PGE1 (EP4 agonist, 1, 25, 50 μM), SC19220 (EP1 antagonist, 1, 25, 50 μM), AH23848 (EP4 antagonist, 1, 25, 50 μM), NS-398 (selective COX-2 inhibitor, 1, 10, 50 μM), H89 [protein kinase A (PKA) inhibitor, 1, 25, 50 μM], GF109203X [pan-protein kinase C (PKC) inhibitor, 1, 10, 50 μM] and NGF receptor chimera (1.5 and 5 μg/mL). Only the concentrations that had significant effects on peptide levels or effects on dmPGE2-induced peptides were presented in Results. The treatment lasted for 3, 6, 12, 24, 48 and 72 h at 37°C with 5% CO2. All treated sensory ganglion explants were collected, weighed, homogenized and centrifuged. Supernatants were used for analysis using quantitative real time PCR (qRT-PCR), ELISA or western blotting analysis as mentioned below. Each treatment has been repeated for at least three times.
RNA extraction and quantitative real-time PCR
Total RNA was extracted from treated sensory ganglion explants using Trizol. The first-strand cDNA was synthesized from 1 μg of total RNA using 200 units of Molony murine leukemia virus reverse transcriptase, 5 μM of random hexamers, and 1 mM of dNTP mix in a total reaction volume of 20 μL. The cDNA product was then diluted 10 times with dH2O and stored at −20°C. PCR reaction was run in total 20 μL reaction mix containing TaqMan Universal PCR Master Mix and gene expression assay probes, using a real time thermal cycler (PCR 7500 system, Applied Biosystems). The gene expression assay probes used in this study were 4352340E, Rn01500392-m1 and Rn01511353-g1 for measuring the mRNA of β-actin, substance P precursor pre-protachykinin (PPT) and CGRP, respectively. Samples were run in duplicate. To normalize the expression level of genes between different samples, the levels were estimated as the ratio of gene X/β-actin. Data were expressed relative change from one of untreated sample. The means from all groups were compared statistically using paired Student’s t-test. The significance level was set at p < 0.05.
ELISA of SP, CGRP and NGF
Following treatments, sensory ganglion explants were weighed and homogenized in cell culture medium (RPMI1640 + 10% heat inactivated FBS, 9 mL/g of tissue) on ice. Following centrifugation, supernatants were collected and frozen at −80°C until ELISA. Total protein was assayed for all samples. ELISA kits for SP, CGRP and NGF were used. All procedures were done according to the manufacturer’s instructions and the microplate was read using a microplate reader (Molecular Devices, Sunnyvale, CA, USA). The mean value of SP, CGRP and NGF (ng/g wet weight protein) was compared statistically using one-way anova with Student-Newman-Keuls`s or Dunnet multiple comparison methods. The significance level was set at p < 0.05.
Sensory ganglion explants were homogenized in ristocetin-induced platelet agglutination lysis buffer containing protease inhibitor cocktail. Obtained supernatants or supernatants extracted from treated DRG explants for ELISA were used for western blotting analysis. The concentration of protein in each sample was determined using bicinchoninic acid method with bovine serum albumin as the standard. Samples with equal amounts of protein (20 μg) were loaded and then separated by 4–20% polyacrylamide gel electrophoresis, and the resolved proteins were electrotransferred to Hybond-C nitrocellulose membrane. Membranes were incubated with 5% non-fat milk in Tris-buffered saline containing Tween-20 and then in rabbit polyclonal anti-EP1–4 antisera (1 : 1000), rabbit anti-phosphorylated pan-PKC or phosphorylated PKCα antiserum (Cell Signaling, Pickering, ON, Canada) and horseradish peroxidase (HRP)-conjugated mouse monoclonal antibody against β-actin (1 : 2000, Santa Cruz Biotechnology, Santa Cruz, CA, USA). Membranes were then probed with a goat anti-rabbit IgG conjugated with HRP (1 : 3000, Santa Cruz Biotechnology). Membranes were finally developed using an enhanced chemiluminescent ECL detection kit (Pierce, Rockford, IL, USA) and exposed to X-ray film. Immunoblots were scanned and the average density of each band was quantitated using SigmaScan densitometry system (SPSS, Chicago, IL, USA) and normalized with β-actin or total kinase protein bands. For membranes incubated with EP2–4 antisera, HRP conjugated mouse antiserum against β-actin was used at the same time, as the molecular size of β-actin is far from these of EP2–3. For membranes immunoblotted with EP1 antiserum, EP1 antiserum incubation were performed first, then the membrane were stripped and immunoblotted with HRP-Actin antiserum, as the molecular sizes of β-actin and EP1 are very close (Actin: 43 kDa; EP1: 42 kDa). Folds of EP1–4, phospho-pan-PKC and phospho-PKCα versus β-actin, total pan-PKC and total PKCα were compared statistically among groups using one-way anova with post hoc Student-Newman-Keul`s or Dunnet multiple comparison method. The significance level was set at p < 0.05.
Stabilized PGE2 increased mRNA and peptide levels of SP and CGRP in cultured sensory ganglion explants
To determine if chronic dmPGE2 treatment is able to induce SP and CGRP at the transcriptional level, qRT-PCR was used. Following treatments of sensory ganglion explants for 3–72 h, 100 μM dmPGE2 significantly increased pre-protachykinin mRNA levels between 6 and 48 h of treatment (Fig. 1a, p < 0.05) while CGRP mRNA levels were increased from 6 h to 72 h of treatment (Fig. 1b, p < 0.05). Following treatment for 24 h, 10–100 μM dmPGE2 significantly increased SP peptide levels in sensory ganglion explants compared to vehicle treatment (Fig. 2a, p < 0.05). Following treatment for 72 h, 1–10 μM dmPGE2 significantly increased CGRP peptide levels in sensory ganglion explants (Fig. 2a, p < 0.05–0.001). When ganglion explants were treated for 3–72 h, 100 μM dmPGE2 significantly increased CGRP levels between 6 and 72 h (Fig. 2b, p < 0.05–0.001) while SP levels were increased between 3 and 48 h (Fig. 2b, p < 0.05).
EP1 and EP4 receptors are involved in PGE2-induced synthesis of SP and CGRP in sensory ganglion explants
The PGE2 EP receptor subtype(s) responsible for PGE2-induced SP and CGRP synthesis were explored next. We previously reported that SP and CGRP were predominantly co-localized with EP1 and EP4 receptors, but rarely with EP2 and EP3, in DRG neurons of adult rat (Ma et al. 2008), suggesting that the effects of PGE2 are likely mediated through activation of EP1 and EP4. After a 24 h treatment, dmPGE2 (1, 10 and 100 μM) significantly increased EP1 and EP4 expression compared to vehicle (Fig. 3a,b and e, p < 0.05). dmPGE2 (1 μM) significantly increased EP2 expression (Fig. 3a and c, p < 0.05) while 10 μM had no effects and 100 μM significantly reduced the expression of EP2 receptors (Fig. 3a and c, p < 0.05). Compared to vehicle treatment, dmPGE2 (1, 10, 100 μM) significantly increased the 52 kDa EP3 receptors (Fig. 3a and d, p < 0.05). However, only 1 μM dmPGE2 increased the 65 kDa isoform of EP3 (Fig. 3d, p < 0.05). dmPGE2 (1, 10 and 100 μM) significantly increased the expression of EP4 in cultured sensory ganglion explants at 24 h post-treatment (Fig. 3a and e, p < 0.05).
Selective EP1 and EP4 agonists were used to determine if they can also increase SP and CGRP peptide levels while selective EP1 and EP4 antagonists were used to see if their effects can be reversed. A selective EP1 agonist, 17-PGE2 (50 μM), or a selective EP4 agonist, 1-OH-PGE1 (50 μM), significantly increased the levels of SP (Fig. 4a, p < 0.05) and CGRP (Fig. 4b, p < 0.05–0.001) while a selective EP2 agonist, butaprost (50 μM), or a selective EP3 agonist, sulprostone (50 μM), had no effects (Fig. 4a and b). 1-OH-PGE1 appeared to induce a more prominent increase in CGRP levels than 17-PGE2 (Fig. 4b). Interestingly, co-treatment of 17-PGE2 and 1-OH-PGE1 induced significantly higher levels of SP than 1-OH-PGE1 alone, but not than 17-PGE2 alone (Fig. 4a, p < 0.05) while the co-treatment also induced significantly higher levels of CGRP than 17-PGE2 alone, but not than 1-PGE1 alone (Fig. 4b, p < 0.05). The EP1 antagonist, SC19220 (50 μM), and the EP4 antagonist, AH23848 (50 μM), significantly suppressed 17-PGE2 and 1-OH-PGE1 induced SP and CGRP, respectively (Fig. 4a and b). dmPGE2 increased SP and CGRP levels in ganglion explants were also suppressed by its co-treatment with SC19220 (50 μM) or AH23848 (50 μM) (Fig. 4c and d, p < 0.05). A selective COX-2 inhibitor, NS-398 (50 μM), had no effects on the contents of both SP and CGRP in sensory ganglion explants (Fig. 4c and d), suggesting that endogenous PGE2 is likely not involved in the basal synthesis of SP and CGRP in sensory neurons.
PKA and PKC signalling pathways are involved in PGE2 induced synthesis of SP and CGRP in sensory ganglion explants
The possible signalling transduction events involved in dmPGE2-induced SP and CGRP synthesis in cultured sensory ganglion explants were examined next. Western blotting analysis revealed that dmPGE2, in a concentration-dependent manner, significantly increased the phosphorylation of pan-PKC in sensory ganglion explants only following a 24-h treatment (Fig. 5a and b, p < 0.05). In contrast, dmPGE2 did not significantly alter the phosphorylation of PKCα/βII isoform in cultured sensory ganglion explants (Fig. 5a and b).
After a 24-h exposure, 100 μM dmPGE2 or 50 μM 17-PGE2 significantly increased SP levels in DRG explants (Fig. 5c, p < 0.05–0.01), effects blocked by the pan-PKA inhibitor H89 (50 μM) or the pan-PKC inhibitor GF109203X (10 μM). Following a 72-h exposure to dmPGE2 (100 μM) or 1-OH-PGE1 (50 μM), CGRP levels in ganglion explants were significantly increased (Fig. 5d, p < 0.05–0.001). These increases were dramatically suppressed by H89 (50 μM) or GF109203X (10 μM) (Fig. 5d). H89 or GF109203X alone had no effects on the basal levels of SP or CGRP in sensory ganglion explants (Fig. 5c and d).
NGF is partially involved in dmPGE2 induced synthesis of SP and CGRP
In order to determine if NGF is involved in dmPGE2-induced synthesis of SP and CGRP, a NGF sequester (NGF receptor Fc-chimera) at the concentrations of 1.5 and 5 μg/mL was used. The higher concentration of NGF receptor Fc-chimera significantly suppressed dmPGE2-induced SP, CGRP and NGF in ganglion explants (Fig. 6a and b, p < 0.05–0.01) while the lower concentration had no effect. Although dmPGE2-induced SP and CGRP were suppressed by the NGF sequester, the levels of SP and CGRP still remained significantly higher than vehicle (Fig. 6a, p < 0.05). The higher concentration of NGF receptor Fc-chimera by itself also significantly suppressed the basal levels of SP, CGRP and NGF in sensory ganglion explants (Fig. 6a and b, p < 0.05) while the lower concentration had no effects on its own.
Exogenous dmPGE2 is able to induce the synthesis of pain-related peptides at both transcriptional and translational levels
It has been well recognized that acute PGE2 treatment (10 min) induces the release of SP and CGRP from DRG neurons (Hingtgen et al. 1995; White 1996) or enhances bradykinin and capsaicin induced neuropeptide release (Vasko et al. 1994; Hingtgen et al. 1995). It is entirely unknown whether PGE2 is involved in the synthesis of SP and CGRP in sensory neurons. We and others have previously shown that the peripheral injection of non-selective COX inhibitors or selective COX-2 inhibitors significantly suppressed the expression of SP and CGRP in DRG neurons of rats suffering from neuropathic or inflammatory pain (Ma and Eisenach 2003; Staton et al. 2007; Ma et al. 2008). These observations suggest that peripherally-derived COX-2/prostanoids are likely involved in the synthesis of SP and CGRP in primary sensory neurons. We have also shown before that exogenous PGE2 concentration-dependently increased the levels of CGRP in Raw 264.7 murine macrophage cell line (Ma et al. 2010), supporting the hypothesis that peripherally-derived COX-2/PGE2 is able to stimulate the synthesis of SP and CGRP in both neurons and non-neuron cells. In the present study, we demonstrated that a stabilized PGE2 analogue, dmPGE2, in a concentration- and time-dependent manner, increases mRNA and peptide levels of SP and CGRP in sensory ganglion explants. These data convincingly indicate that PGE2 contributes to the synthesis of SP and CGRP at both transcriptional and translational levels. The concentrations that induced significant increase in peptide levels fell into the μM range, a concentration range causing evident biological effects in vitro (Nicol et al. 1992; North et al. 2007) and in vivo (Malmberg et al. 1997; Tessner et al. 2004) in literature.
It has been well known that NGF can stimulate the production of SP and CGRP in DRG neurons (Lindsay and Harmar 1989; Donnerer et al. 1992). Interestingly, PGE2 was also shown to induce NGF in cultured mouse astrocytes (Toyomoto et al. 2004). Consistently, we found here that dmPGE2 also increased NGF levels in sensory ganglion explants, although it remains to be clarified if NGF is produced by sensory neurons or glia. Therefore, it is necessary to determine if PGE2, directly or via NGF indirectly, induces the synthesis of SP and CGRP in sensory neurons. Here, we found that a NGF sequester, in a concentration-dependent manner, significantly reduced NGF levels and suppressed dmPGE2-increased SP and CGRP levels in ganglion explants. However, the levels of SP and CGRP in ganglion explants co-treated with dmPGE2 and the NGF sequester were still significantly higher than vehicle, suggesting that NGF is only partially involved in dmPGE2 induced SP and CGRP. Interestingly, the NGF receptor Fc-chimera by itself, in a concentration-dependent manner, also significantly reduced the basal levels of SP, CGRP and NGF in sensory ganglion explants, suggesting that endogenous NGF is involved in the maintenance of the basal levels of two peptides in primary sensory neurons.
We also observed here that a selective COX-2 inhibitor, NS-398, did not alter the basal levels of SP and CGRP in sensory ganglion explants, indicating that endogenous PGE2 is not involved in the maintenance of the basal expression of both peptides. Interestingly, perineural injection of NS-398 in naive rats had no effect on both basal pain behaviours and the levels of SP and CGRP in DRG neurons (Ma et al. 2008), suggesting that the effects of the COX-2 inhibitor are unique to nerve injury or inflammation in which PGE2 is abundantly produced in injured nerve or inflamed tissue. Take together, these data suggest that PGE2 not only activates nociceptors, but also plays a role in amplifying nociceptive responses by stimulating the synthesis and release of pain-related neuropeptides from nociceptors.
EP1 and EP4 receptor subtypes as well as PKA and PKC signaling pathways are involved in dmPGE2-induced synthesis of SP and CGRP
Prostaglandin E2 mediates its biological effects by acting through its four EP receptors, EP1–4, all of which belong to the family of G protein coupled receptors and exist in DRG neurons (Vanegas and Schaible 2001; Lin et al. 2006; Ma et al. 2008). In an inflammatory pain model, EP4 receptors, but not EP1–3, were significantly increased in DRG neurons (Lin et al. 2006). We and others have shown that partial sciatic nerve injury increased EP1 and EP4, not EP2 and EP3, receptors in DRG neurons of rats (Durrenberger et al. 2006; Ma et al. 2008). Perineural injection of NS-398 suppressed not only partial sciatic nerve ligation-induced increases in EP1 and EP4 expression, but also the increase in SP and CGRP levels (Ma et al. 2008), suggesting that over-produced PGE2 in injured nerves likely contributes to the synthesis of SP and CGRP via the induction of EP1 and EP4 receptors. Moreover, the extensive co-localization of SP and CGRP with EP1 and EP4 (Ma et al. 2008) supports a role of the two receptor subtypes in PGE2-induced synthesis of neuropeptides. In the present study, we found that dmPGE2-increased SP and CGRP levels were significantly suppressed by EP1 (SC19220) or EP4 (AH23848) receptor antagonists. Moreover, the EP1 agonist (17-PGE2) and/or the EP4 agonist (1-OH-PGE1) were able to increase the levels of SP and CGRP, with EP1 and EP4 activation inducing more SP and CGRP, respectively. The selective EP2 agonist (butaprost) or the selective EP3 agonist (sulprostone) at the μM, a concentration range which is widely used in literature, failed to increase the levels of SP and CGRP. Collectively, these data suggest that EP1 and EP4 are the receptor subtypes responsible for PGE2-induced SP and CGRP synthesis in our model. In an earlier report, it was shown that the EP3 and EP4 are the receptor subtypes mediating PGE2-increased cAMP accumulation and peptide release in cultured DRG neurons (Southall and Vasko 2001). Here dmPGE2, in a concentration-dependent manner, significantly increased not only the expression of EP1 and EP4 but also the expression of EP2 and EP3, suggesting that some of the biological effects induced by PGE2 are likely mediated through the activation of EP2 and EP3 in this ex vivo model.
Although activating cAMP/PKA and PKC signalling pathways are known to mediate PGE2-induced release of SP and CGRP in sensory neurons (Barber and Vasko 1996; Southall and Vasko 2001), it was not clear if PKA and PKC could mediate PGE2-induced synthesis of SP and CGRP in primary sensory neurons. It has been shown that the PKA activator, dibutyryl cAMP, and the PKC activator, phorbol ester, were able to stimulate CGRP synthesis in cultured DRG neurons (Supowit et al. 1995). We observed here that dmPGE2 concentration-dependently increased the phosphorylation of PKC, suggesting a role for this signalling pathway in mediating PGE2-induced biological effects in sensory ganglion explants. However, the phosphorylation of PKCα was not significantly modulated by dmPGE2, suggesting that other isoforms of PKC are involved. Inhibitors of PKA and PKC suppressed dmPGE2- or 17-PGE2-induced SP production as well as dmPGE2- or 1-OH-PGE1-increased CGRP levels, strongly supporting the hypothesis that PKC and PKA play a role in PGE2-induced SP and CGRP synthesis and release in sensory neurons. The receptor subtypes and signalling pathways underlying dmPGE2 induced SP and CGRP are summarized in Fig. 7.
Significance of the findings
Substance P and CGRP are pain-related peptides co-expressed in nociceptors (Wiesenfeld-Hallin et al. 1984) and serve as pain transmitters in the dorsal horn, thus contributing to the augmented synaptic strength between nociceptors and nociceptive dorsal horn neurons and to the generation of mechanical hypersensitivity and central sensitization (Luo and Wiesenfeld-Hallin 1995; Sun et al. 2004; Seybold 2009). Following inflammation, both peptides are dramatically up-regulated in primary sensory neurons (Donnerer et al. 1992, 1993). The antagonists of SP/neurokinin 1 and CGRP receptors are effective to relieve inflammatory pain in various animal models (Moussaoui et al. 1993; Brain and Grant 2004). We have previously shown that SP (Ma and Bisby 1998) and CGRP (Ma et al. 1999) are up-regulated in medium and large size spared DRG neurons following partial sciatic nerve injury. The activation of SP/neurokinin 1 and CGRP receptors-dependent intracellular signalling events in the dorsal horn have been implicated in the genesis of neuropathic pain (Coudore-Civiale et al. 1998; Lee and Kim 2007). Collectively, these data suggest that SP and CGRP are up-regulated in primary sensory neurons following peripheral inflammation or partial nerve injury and that these two peptides and their receptor signalling contribute to the generation of inflammatory and neuropathic pain (Fernandes et al. 2009). We also showed here that in addition to NGF, PGE2 indeed contributes to the synthesis of SP and CGRP at both gene and peptide levels. Facilitating the production of pain-related neuropeptides in primary sensory neurons is a possible novel mechanism underlying the role of PGE2 in nociception as well as in pathological pain conditions such as inflammatory and neuropathic pain. Blocking PGE2/EP receptor signalling could open a novel therapeutic avenue in treating both inflammatory and neuropathic pain.
This study was supported by grants from Canadian Institutes of Health Research to Weiya Ma (grant no. 87372 and 89892). The author sincerely appreciates the technical help from the laboratories of Dr. Remi Quirion and Dr. Luheshi, particularly the help from Dr. Watara Inoue with quantitative real-time PCR.