Sérgio Henrique Ferreira, Faculdade de Medicina de Ribeirão Preto – USP, Av. Bandeirantes, 3900, Ribeirão Preto, SP, Brazil 14049-900. E-mail: firstname.lastname@example.org
Background and purpose: Protein kinase (PK) A and the ε isoform of PKC (PKCε) are involved in the development of hypernociception (increased sensitivity to noxious or innocuous stimuli) in several animal models of acute and persistent inflammatory pain. The present study evaluated the contribution of PKA and PKCε to the development of prostaglandin E2 (PGE2)-induced mechanical hypernociception.
Experimental approach: Prostaglandin E2-induced mechanical hypernociception was assessed by constant pressure rat paw test. The activation of PKA or PKCε was evaluated by radioactive enzymic assay in the dorsal root ganglia (DRG) of sensory neurons from the hind paws.
Key results: Hypernociception induced by PGE2 (100 ng) by intraplantar (i.pl.) injection, was reduced by i.pl. treatment with inhibitors of PKA [A-kinase-anchoring protein St-Ht31 inhibitor peptide (AKAPI)], PKCε (PKCεI) or adenylyl cyclase. PKA activity was essential in the early phase of the induction of hypernociception, whereas PKC activity was involved in the maintenance of the later phase of hypernociception. In the DRG (L4-L5), activity of PKA increased at 30 min after injection of PGE2 but PKC activity increased only after 180 min. Moreover, i.pl. injection of the catalytic subunit of PKA induced hypernociception which was markedly reduced by pretreatment with an inhibitor of PKCε, while the hypernociception induced by paw injection of PKCε agonist was not affected by an inhibitor of PKA (AKAPI).
Conclusions and implications: Taken together, these findings are consistent with the suggestion that PKA activates PKCε, which is a novel mechanism of interaction between these kinases during the development of PGE2-induced mechanical hypernociception.
A-kinase-anchoring protein St-Ht31 inhibitor peptide
protein kinase A
catalytic subunit of PKA
protein kinase C
ε isoform of protein kinase C
pseudo receptor octapeptide for activated PKCε, a specific agonist of PKCε
PKCεV1–2 peptide, a selective PKCε inhibitor
Tissue injury and inflammation are associated with increased prostanoid synthesis which sensitizes primary sensory neurons (Martin et al., 1987; Schaible and Schmidt, 1988; Rueff and Dray, 1993). This sensitization of the nociceptive neurons in humans results in hyperalgesia (Ferreira, 1972) (an increased response to a stimulus which is normally painful) or allodynia (pain from stimuli that are not normally painful). However, in animal behaviour models of mechanical nociception, hyperalgesia and allodynia can be distinguished by the use of apparently different mechanical stimuli, as in the Randall-Selitto and the electronic von Frey tests. Moreover, the terms hyperalgesia and allodynia have been developed for use in clinical practice rather than for experimental work, physiology or anatomical purposes (see IASP Pain Terminology). Therefore, we have used the term hypernociception to describe the decrease of behavioural nociceptive threshold in experimental animals.
Although the second messenger cascade associated with PGE2-induced hypernociception involves the participation of PKA downstream from cAMP, this evidence does not exclude the participation of PKC in this pathway. Thus, the aim of the present study was to investigate whether, in mechanical hypernociception, PKC participates in the PGE2/cAMP/PKA intracellular signalling pathway in the primary nociceptive neurons.
Animal care and handling procedures were in accordance with International Association for the Study of Pain guidelines for the use of animals in pain research and with the approval of the Ethics Committee of the School of Medicine of Ribeirão Preto (University of São Paulo). All efforts were made to minimize discomfort for the animals. Experiments were performed with 180 to 200 g male Wistar rats kept in a 12 h light–dark cycle, with controlled humidity (60–80%) and temperature (22–25°C). Food and water were available ad libitum. The animals were taken to the testing area at least 1 h before testing. Each experiment used five rats per group, previously found to be the minimum number of animals necessary to detect significant variations of the paw nociceptive sensitivity. All behavioural testing was performed between 9:00 am and 4:00 pm.
Nociceptive test: constant pressure rat paw test
Nociception was measured in the present study by applying a constant pressure via a ring-shaped syringe piston (15 mm2) to the dorsal surface of the hind paw. The piston pressure (20 mm Hg) was measured by a sphygmomanometer connected to the source of compressed air (Ferreira et al., 1978). The end point of the mechanical nociception is a typical behavioural reaction (freezing) characterized by a combination of signs: a brief apnoea, retraction of the head and forepaws and reduction in the escape movements from the position imposed by the experimental situation. Usually, the apnoea is associated with successive waves of muscular tremor. The latency of the freezing reaction is measured before (control reaction time at zero time) and at different times after intraplantar (i.pl.) administration of the tested substance in the same paws. The reduction of the latency obtained by subtracting the value of latency at a given time point from the value at zero time (31.5 ± 0.1 s, n = 50) quantified the intensity of mechanical hypernociception (Ferreira et al., 1978). In this test, local or systemic pretreatment with steroidal and non-steroidal anti-inflammatory drugs reduces the intensity of hypernociception by inflammatory stimuli (carrageenan, endotoxin, cytokines/chemokines) but does not affect hypernociception induced by directly sensitizing mediators like PGE2 or dopamine (Nakamura and Ferreira, 1987; Cunha et al., 1991; 1992; Lorenzetti et al., 2002). Thus this test, using a constant pressure as mechanical nociceptive stimulus, allows the measurement of variations of nociceptive behaviour.
Determination of PKA and PKC activity in DRG
Rats were given a lethal overdose of sodium pentobarbital and the L4 and L5 DRG removed. These samples were frozen on dry ice and homogenized (Homogeneizador Kont). The PKA and PKC activity was measured using a commercial kit (protein kinase A assay kit, protein kinase C assay kit, Calbiochem) according to the manufacturer's instructions. Assays were only carried out using the DRG (L4-L5) ipsilateral to the hind paws receiving the i.pl. injection of PGE2 (100 ng per paw) or N6,2′-O-dibutyryladenosine 3′:5′-cyclic monophosphate (db-cAMP) (100 µg per paw), removed at 30 or 180 min after injection. The assay is based on the reaction of the enzyme sample with a biotinylated peptide substrate and γ32P-ATP, and is expressed as pmoles of phosphate incorporated per minute.
Effect of drug treatments
To investigate the involvement of cAMP, PKA or PKCε in hypernociception induced by PGE2 (100 ng; Ferreira and Nakamura, 1979) or db-cAMP (100 µg; Ferreira and Nakamura, 1979), the animals were pretreated with subcutaneous i.pl. injections into the same (ipsilateral) paw. Note all i.pl. treatments are shown as the dose per paw. Pretreatment with inhibitors of AC (SQ22536; 3-27 µg), of PKA [A-kinase-anchoring protein St-Ht31 inhibitor peptide (AKAPI); 0.03-0.3 µg] or of PKCε (PKCεI, 1-9 µg) was given 5 min before the i.pl. injection of PGE2 or db-cAMP. The participation of PKA or PKCε in later phases of PGE2 or db-cAMP-induced hypernociception was assessed by post-treatment (30 or 90 min) with AKAPI (0.3 µg) or PKCεI (9 µg). To investigate if PKA or PKCε had intrinsic hypernociceptive effects, the animals were injected with the catalytic subunit of PKA (PKAcs 0.3–9 U) or pseudo receptor octapeptide for activated PKCε, a specific agonist of PKCε (ΨεRACK 0.3-9 µg; Dorn et al., 1999). To determine if PKA was able to activate PKCε, the animals were pretreated with inhibitor AKAPI (0.3 µg) or PKCε (PKCεI, 9 µg) 30 min before the PKAcs (1 U) or the specific agonist of PKCε (ΨεRACK 1 µg). In the present study, all hypernociceptive stimuli (PGE2, db-cAMP, ΨεRACK and PKAcs) or antagonists used were given by subcutaneous injection into the hind paw.
Results are presented as means ± SEM for groups of five animals. Analysis of variance (one-way anova) was used, followed by the Bonferroni test. The level of significance was set at P < 0.05. The dose–response relationships for SQ22536, AKAPI and PKCεI were analysed by non-linear regression.
Drugs and reagents
The pseudo receptor for activated PKCε octapeptide (ΨεRACK; Dorn et al., 1999) and PKCεV1–2 peptide (a selective PKCε inhibitor; Gray et al., 1997) were obtained from SynPep Corp (Dublin, CA, USA). SQ22536 was obtained from Biomol (Aley and Levine, 1999). The PKAcs was obtained from Calbiochem (La Jolla, CA, USA). PGE2 and db-cAMP (Taiwo et al., 1989; Taiwo and Levine, 1991; Ouseph et al., 1995) were obtained from Sigma (St Louis, MO, USA). InCELLlect® AKAPI (Moita et al., 2002; Parada et al., 2005) was obtained from Promega Corp (Madison, WI, USA). The stock solution of PGE2 (1 µg·µL−1) was prepared in 10% ethanol, and additional dilutions were made in physiological saline (0.9% NaCl), yielding a final concentration of ethanol of less than 1%. All other drugs were dissolved directly in saline. Protein kinase A assay kit and protein kinase C assay kit were from Calbiochem.
PKA and PKCε participate in the PGE2-activated, cAMP second messenger cascade in primary afferent neurons
The results presented in Figure 1 show that subcutaneous i.pl. pretreatment with an inhibitor of AC (SQ22536; 3, 9 and 27 µg; ID50: 4.01 µg; Figure 1A) or of PKA (AKAPI; 0.03, 0.1 and 0.3 µg; ID50: 0.22 µg; Figure 1B) prevented PGE2-induced hypernociception in a dose-dependent manner. The PKCε inhibitor (PKCεI; 1, 3 and 9 µg; ID50: 2.6 µg; Figure 1C) prevented in a dose-dependent manner the hypernociception induced by PGE2. Pretreatment with PKCεI (9 µg), or AKAPI (0.3 µg), but not with the AC inhibitor SQ22536 (27 µg), inhibited the hypernociception induced by db-cAMP (100 µg, Figure 1D). It should be stressed that SQ22536 (27 µg, Figure 1A), AKAPI (0.3 µg, Figure 1B) or PKCεI (9 µg, Figure 1C) at the highest doses used here had no nociceptive effect when given alone into normal paws.
Differential time effects of PKA and PKC inhibitors on the hypernociception induced by PGE2 or db-cAMP
Treatment with inhibitors of PKA (AKAPI, 0.3 µg) or PKCε (PKCεI, 9 µg), either before or 30 min after the i.pl. injection of PGE2 (100 ng) or db-cAMP (100 µg), reduced the mechanical hypernociception. However, later post-treatment (90 min after PGE2 or db-cAMP i.pl. injection) with AKAPI was ineffective but similar treatment with PKCεI clearly reduced mechanical hypernociception. Together, these results suggest that PKA activity is essential only in the early phase of the establishment of hypernociception, whereas PKC activity is involved in the maintenance of later phases of hypernociception. In support of this suggestion, the radioactive assay for PKA activity in ipsilateral DRG (L4-L5) was increased at 30 min, but not 180 min after paw injection of PGE2 (Figure 2B) or db-cAMP (Figure 3B). However, the PKC activity was enhanced much later, mainly at 180 min after i.pl. injection of PGE2 (Figure 2D) or db-cAMP (Figure 3D).
PKA activates PKCε during PGE2-induced hypernociception
We also investigated whether the hypernociception induced by a PKAcs depends on PKCε activation, and whether the hypernociceptive activity of ψεRACK depends on PKA activation. Both PKAcs (0.3, 1, 3 and 9 U; ED50: 1.31 U; Figure 4A) and ψεRACK (0.3, 1, 3 and 9 µg; ED50: 1.67 µg; Figure 4B) given alone induced dose-dependent mechanical hypernociception. Pretreatment (30 min before) with the inhibitor of PKCε (9 µg) significantly inhibited the hypernociception induced by PKAcs (1 U; Figure 4C). In contrast, the PKA inhibitor AKAPI (0.3 µg) given 30 min before the administration of ψεRACK (1 µg) did not change the hypernociceptive effect (Figure 4D). These results are compatible with the proposition that PKA activates PKCε during PGE2-induced hypernociception.
Enhanced sensitivity to nociceptive stimuli is one of the characteristics of the inflammatory response and results from an increase in the excitability of primary nociceptive neurons. There is substantial experimental evidence demonstrating that PGs, among other mediators (sympathetic amines, endothelins, 5-hydroxytryptamine (5-HT), adenosine and cytokines) sensitize the primary nociceptive neurons to innocuous mechanical (Handwerker and Neher, 1976; Pateromichelakis and Rood, 1982; Schaible and Schmidt, 1988; Mizumura et al., 1993; Moriyama et al., 2005) and chemical stimuli (Nicol and Cui, 1994; Smith et al., 2000). This sensitization is referred in the present study as hypernociception, thus avoiding the use of terms like hyperalgesia or allodynia, because they describe an aggregate of characteristic human pathological nociceptive symptoms (see IASP Pain Terminology).
In the present study, we confirmed that the treatment with cAMP or PKA inhibitors (Figure 1) prevents PGE2-induced hypernociception, as already described (Taiwo and Levine, 1990; 1992; Aley and Levine, 1999; Parada et al., 2005; Kassuya et al., 2007). In addition, we found that inhibition of the PKCε prevented PGE2 or db-cAMP-induced hypernociception. These results suggest that an increase in cAMP levels is necessary for the activation of PKCε (directly or indirectly) for the induction of PGE2 hypernociception. Further, the activation of PKA downstream from cAMP was involved in the onset of hypernociception, mainly in the initial 30 min. After that, a continuous PKCε activity was required for the development of hypernociception induced by PGE2. As we have shown, in contrast with PKA, PKCε inhibitors significantly reduced the hypernociception induced by PGE2 (Figure 2A,C) or db-cAMP (Figure 3A,C), by either early (30 min after) or later (90 min after) post-treatment. Together, these results suggest that PKA activity is essential in the early phase of the establishment of hypernociception, whereas PKC activity is involved in the maintenance of later phases of hypernociception. To support this suggestion, we investigated the activities of PKA and PKCε in the cell bodies of the DRG neurons after PGE2 and db-cAMP injection in the rat paws, assuming that they reflect biochemical events occurring at peripheral terminations of sensory neurons. In fact, PKA activity increased in ipsilateral DRG (L4-L5) mainly in the early phase (30 min after the PGE or db-cAMP injection) and returned to basal levels 180 min after (Figures 2B and 3B). Although the kinetics of PKA activity induced by PGE2 and db-cAMP were similar, the enzyme activity at 30 min after PGE2 injection was higher that that induced by db-cAMP. Experiments using different doses of these mediators might explain this difference. The increase in PKC activity, however, was low when PKA levels were maximal and maximal when the PKA levels returned to the base line (Figures 2D and 3D). These results are in line with the assumption that PKC activation is downstream to the activation of PKA. Zhou et al. (2003) demonstrated that in paw inflammation induced by carrageenan, in which there is a significant activation of PKCε in the DRG, these kinases are synthesized and transcribed in the soma and delivered to the nerve terminals by axonal transport. Finally, the PKCε inhibitor significantly reduced hypernociception induced by the PKAcs (Figure 4C). Nevertheless, the PKA inhibitor AKAPI did not change the hypernociceptive effect of ψεRACK (Figure 4D), thus supporting the suggestion that hypernociceptive effect induced by a PKAcs depends on PKCε activity. Hypernociception induced by PGE2 did not cause variations in PKA and PKC activities in the paw tissues, probably because the nociceptive neurons constitute a relatively small portion of the total tissue of the paw (data not shown).
Besides, PKA and PKC and other kinases or factors such as (c-Jun N-terminalkinase (JNK), extracellular-regulated kinase, mitogen-activated protein kinases or cAMP-activated guanine exchange factor (Epac), all participate in the inflammatory response. Nevertheless, in acute mechanical hyperalgesia, PKCε and PKA (PGE2)-mediated hyperalgesia was independent of extracellular signal-regulated kinase (MEK) activity (Aley et al., 2001). Further, JNK inhibition did not inhibit PGE2-induced mechanical allodynia (Kassuya et al., 2007). Additionally, p38 and JNK, are activated by the peripheral injection of PGE2 and cause paw oedema in mice but no nociceptive behaviours (Claudino et al., 2006). For this reason, we also did not evaluate the participation of those intracellular pathways in the inflammatory sensitization. There is an alternative mechanism by which cAMP might lead to activation of PKCε. The Epac can induce PKCε translocation via phospholipase C or D in DRG neurons (Hucho et al., 2005). However, the Epac activator (CPTOMe) did not induce mechanical hypernociception in our behavioural test (data not shown).
We did not address the mechanism underlying PKA/PKC crosstalk in the present study. However, there is evidence pointing to a role for PKA in the activation and translocation of PKC. In fact, activation of dopamine D1 receptors, known to couple to Gαs, increases PKC activity and translocation in LTK cells (Yu et al., 1996). In addiction, Huang et al. (2001) demonstrated that PKCζ is a key downstream component of a PKA-dependent, anti-apoptotic signalling pathway activated by a G protein-coupled receptor. Recently, Yao et al. (2008) demonstrated that activation of PKA by Sp-cAMPS (PKA activator) was sufficient to induce activity and translocation of PKCε.
In conclusion, the present study shows that both PKA and PKCε participate in acute mechanical hypernociception downstream from PGE2 receptor activation and suggest that PKA may activate PKCε. Thus, our results describe a novel signalling pathway, in addition to the usual cAMP/PKA pathway, in mechanical hypernociception induced by PGE2.
This study was supported by grants from the CNPq and FAPESP, Brazil. We thank IRS Schivo, SR Rosa and FL Mestriner for excellent technical assistance.