Acquired Exchange Protein Directly Activated by Cyclic Adenosine Monophosphate Activity Induced by p38 Mitogen-activated Protein Kinase in Primary Afferent Neurons Contributes to Sustaining Postincisional Nociception

LATE recovery from acute postsurgical pain or the presence of pain during the subacute period is associated with poor psychologic condition, lower functional recovery, and prolonged opioid therapy.^1–3  There is a significant interindividual difference in the recovery from acute postsurgical pain.⁴  Reducing the intensity of pain during the subacute period has been found to be challenging.^5,6  The subacute period might be critical for the transition from acute to sustained postsurgical pain since patients with persistent postsurgical pain reportedly experience continuous or relapsing pain during this period.^7–9  However, the pathophysiology of the sensory nervous system during the subacute postsurgical period is largely unknown.

After a plantar incision or carrageenan injection, experimental animals have been shown to develop a condition known as “nociceptor priming” after the resolution of initial acute nociceptive hypersensitivity.¹⁰  Nociceptor priming is a pathophysiologic condition of the peripheral sensory nervous system characterized by normal baseline behavior but long-lasting nociceptive hypersensitivity after a single treatment with inflammatory mediators.¹¹  Several types of chronic pain, including chronic postsurgical pain, can develop by the mechanism of nociceptor priming.¹²  Altered intracellular signaling downstream to cyclic adenosine monophosphate (cAMP), from protein kinase A to protein kinase C (PKC)-ε, is one of the main characteristics of nociceptor priming.¹³  Transactivation of PKCε during inflammatory,^14,15  neuropathic pain,¹⁶  and nociceptor priming^17,18  is mediated by the cAMP sensor exchange protein directly activated by cAMP (EPAC). However, the functional significance of EPAC in dorsal root ganglion (DRG) neurons during nociceptor priming is not fully understood.

The current study was conducted to characterize the molecular mechanisms of nociceptor priming after plantar incision as a model of the pathophysiology in the subacute postsurgical period. We hypothesized that induction of EPAC in the DRG neurons mediated by p38 mitogen-activated protein kinase (p38MAPK) signaling contributes to the development of nociceptor priming during the subacute period after tissue injury. We investigated the distribution of two isoforms of EPAC, EPAC1 and EPAC2, in the rat DRG. Expression levels of EPAC1 and EPAC2 were analyzed in nociceptor-primed animals 14 days after the plantar incision. We also assessed whether direct inhibition of EPAC activity was sufficient to inhibit prostaglandin E2 (PGE2)-evoked persistent nociceptive hypersensitivity during nociceptor priming using the newly developed selective EPAC inhibitor ESI-09.¹⁹  ESI-09 was also used to evaluate the involvement of EPAC in acute nociceptive hypersensitivity after the incision. In addition, we characterized the contribution of p38MAPK activated by the incision on the increase in EPACs and the development of nociceptor priming after the incision.

The Kyoto Prefectural University of Medicine’s Animal Care Committee (Kyoto, Japan) approved all experimental procedures in this study. All experiments were performed in accordance with the guidelines of the National Institute of Health and International Association for the Study of Pain (Washington, D.C.).²⁰  Male Sprague–Dawley rats (200 to 250 g; Shimizu Laboratory Supplies Co., Ltd., Japan; n = 144) were housed in groups of three per cage with a 12-h light/dark cycle. All surgical procedures and drug injections were performed during isoflurane (2% volume [of solute] per volume [of solvent]) anesthesia. The sample sizes of all experimental protocols were determined based on the results obtained in our previous investigations.^21–23  We considered an approximately two-fold difference in the behavioral study and a 40% difference in immunohistochemistry to be meaningful.

To confirm that plantar incision induces nociceptor priming, animals were randomly assigned to the naive plus PGE2 group (n = 6) or incision 14 plus PGE2 group (n = 6). The hind paw of animals in the incision 14 plus PGE2 group was incised as described in the rat plantar incision model,²⁴  wherein a 1-cm longitudinal incision of the left plantar aspect of the hind paw, beginning 0.5 cm from the end of the heel, was made with a number 11 surgical blade through the skin, fascia, and plantaris muscle, and the skin was closed with a 5-0 nylon suture. Rats in the naive plus PGE2 group received no intervention at this time. Fourteen days later, animals in both groups received intraplantar injection of PGE2 (1 µg/50 µl dissolved in 0.4% ethanol in saline; Sigma-Aldrich, USA) and were tested with behavioral experiments for up to 7 days. Separately, to investigate the existence of nociceptor priming at a later time period after the tissue injury, animals were randomly assigned to the naive plus PGE2 group (n = 6) or incision 21 plus PGE2 group (n = 6). Twenty-one days later, animals in both groups received intraplantar injection of PGE2 (1 µg/50 µl dissolved in 0.4% ethanol in saline) and were tested with behavioral experiments for up to 7 days. The primary outcome of this experiment was the difference of threshold against von Frey stimulation.

Immunohistochemistry and in situ hybridization were performed to evaluate the expression of EPAC1 and EPAC2 in the DRG. Animals were randomly assigned to the naive group (n = 6), incision 3-day (3d) group (n = 6), or incision 14-day (14d) group (n = 6). Animals in the incision 3d and incision 14d groups received a plantar incision as with the experiment 1. Naive animals received no intervention. Three days after incision in the 3d group and 14 days after incision in the 14d group, respectively, animals were perfused with 0.9% NaCl followed by 10% neutralized formalin (Wako Pure Chemical Industries, Ltd., Japan) during terminal anesthesia with isoflurane. L5 DRGs were harvested from the animals to process for immunohistochemistry and in situ hybridization.

Separately, Western blotting was performed to quantify the amount of EPAC1 and EPAC2 expression in the DRG. For this, animals were randomly assigned to naive (n = 5) or incision 14d groups (n = 5). Incision 14d group animals received a plantar incision, and naive animals received no intervention. Fourteen days after the incision, L5 DRGs were removed from all the animals during terminal anesthesia with isoflurane. The primary outcome of this experiment was the percentage of EPAC1- or EPAC2-positive neurons per total profile.

To investigate the effect of an EPAC inhibitor on PGE2-induced persistent hypernociception during the incision-induced nociceptor priming, animals were randomly assigned to incision plus PGE2 plus ESI-09 intraplantar (i.pl.; n = 6), incision plus PGE2 plus vehicle (n = 6), or naive plus PGE2 plus vehicle groups (n = 6). Animals in the incision plus PGE2 plus ESI-09 i.pl. and incision plus PGE2 plus vehicle groups received a plantar incision as with the experiment 1. Animals in the naive plus PGE2 plus vehicle group received no intervention at this time. Fourteen days later, the selective EPAC inhibitor, ESI-09 (25 µg/100 µl dissolved in 10% dimethyl sulfoxide [DMSO] in 0.1 M phosphate-buffered saline [PBS]; Sigma-Aldrich), was administered by intraplantar injection 2 h before PGE2 injection in the incision plus PGE2 plus ESI-09 i.pl. animals. Animals in the incision plus PGE2 plus vehicle and naive plus PGE2 plus vehicle groups received 100 µl DMSO, 10%, in 0.1 M PBS 2 h before PGE2 injection. Behavioral experiments were performed for up to 7 days after the PGE2 injection.

Separately, to investigate the effects of the EPAC inhibitor on the development of acute nociceptive hypersensitivity, animals were randomly assigned to ESI-09 i.pl. (n = 6) or vehicle groups (n = 6). Animals in both groups received plantar incisions as with the experiment 1. ESI-09 i.pl. group rats received intraplantar injection of ESI-09 (25 µg/100 µl dissolved in 10% DMSO in 0.1 M PBS) 2 h before the surgery, while those in the vehicle group received intraplantar injection of 100 µl DMSO, 10%, in 0.1 M PBS. Behavioral experiments were performed for up to 7 days after ESI-09 treatment.

To investigate the effects of the EPAC inhibitor on established acute nociceptive hypersensitivity, other sets of animals were randomly assigned to ESI-09 i.pl. (n = 6) or vehicle groups (n = 6). Animals in both groups received plantar incisions as with the experiment 1. ESI-09 i.pl. group rats received intraplantar injection of ESI-09 (25 µg/100 µl dissolved in 10% DMSO in 0.1 M PBS) 1 day after the surgery, while those in the vehicle group received intraplantar injection of 100 µl DMSO, 10%, in 0.1 M PBS. Behavioral experiments were performed for up to 2 days after ESI-09 treatment.

In addition, to investigate the effects of the EPAC inhibitor on the cell bodies of the DRG neurons of animals with established acute nociceptive hypersensitivity, other sets of animals were randomly assigned to ESI-09 intrathecal (i.t.; n = 6) or vehicle groups (n = 6). In both groups, intrathecal catheters were inserted 10 mm from the L6/S1 intervertebral space. In a preliminary study, using Evans blue injection, we confirmed that catheter insertion by this method successfully delivers the drug to L4/L5 DRGs. After confirming the absence of limb paralysis, the animals in both groups received plantar incisions, as with the experiment 1, 3 days after the catheter insertion. ESI-09 i.t. group rats received injection of ESI-09 via the intrathecal catheter (20 µg/20 µl dissolved in 10% DMSO in 0.1 M PBS and artificial cerebrospinal fluid) 1 day after the surgery, while those in the vehicle group received intrathecal injection of the same dose of 10% DMSO in 0.1 M PBS and artificial cerebrospinal fluid. Behavioral experiments were performed for up to 2 days after ESI-09 treatment. The primary outcome of this experiment was the difference of the threshold against von Frey stimulation.

Using a p38MAPK inhibitor, FR167653, we investigated the impact of p38MAPK on the development of nociceptor priming after the plantar incision and on incision-induced EPAC expression. For behavioral testing, the animals were randomly assigned to incision plus PGE2 plus FR167653 intraperitoneal (i.p.; n = 5) or incision plus PGE2 plus vehicle groups (n = 5). Animals in the incision plus PGE2 plus FR167653 i.p. group received intraperitoneal injection of the p38MAPK inhibitor, FR167653 (10 mg/2 ml dissolved in saline; Astellas Pharma Inc., Japan), just before the plantar incision. The timing and dose of FR167653 used in the current study were shown in our previous study to inhibit the activation of p38MAPK in the DRG after plantar incision.²³  Animals in the incision plus PGE2 plus vehicle group received intraperitoneal injection of 2 ml saline just before the plantar incision. Fourteen days after the treatment, animals were injected with PGE2, and behavioral assessment was performed for up to 7 days thereafter.

Separately, to investigate the effect of p38MAPK on EPAC1 and EPAC2 expression after the plantar incision, the animals were randomly assigned to incision plus vehicle (n = 6), incision plus FR167653 i.p. (n = 6), or naive groups (n = 6). Animals in the incision plus FR167653 i.p. group received intraperitoneal injection of FR167653 (10 mg/2 ml, in saline), and animals in the incision plus vehicle group received intraperitoneal injection of 2 ml saline just before the plantar incision. Naive animals received no intervention. Animals were perfused with 0.9% NaCl followed by 10% neutralized formalin during terminal anesthesia with isoflurane 14 days after the plantar incision. L5 DRGs were obtained from the animals and processed for immunohistochemistry against EPAC1 and EPAC2.

Further, phosphorylation of p38MAPK in the dorsal horn of the spinal cord has been reported to increase after tissue injury.²⁵  We, therefore, investigated the involvement of spinal p38MAPK in the development of nociceptor priming after tissue injury. The animals were randomly assigned to incision plus PGE2 plus FR167653 i.t. (n = 5) or incision plus PGE2 plus vehicle groups (n = 5). In both groups, intrathecal catheters were inserted 30 mm from the L6/S1 intervertebral space. In a preliminary study, using Evans blue injection, we confirmed that catheter insertion by this method successfully delivers the drug to the spinal segment corresponding to L4/L5. After confirming the absence of limb paralysis, animals in both groups received plantar incisions, as with the experiment 1, 3 days after the catheter insertion. Incision plus PGE2 plus FR167653 i.t. group rats received injection of FR167653 via the intrathecal catheter (50 µg/20 µl in artificial cerebrospinal fluid) 30 min before the surgery, while those in the vehicle group received intrathecal injection of 20 µl artificial cerebrospinal fluid. Fourteen days after the treatment, animals were injected with PGE2, and behavioral assessment was performed for up to 7 days thereafter. The primary outcomes of this experiment were the threshold against von Frey stimulation for behavioral experiments and the percentage of EPAC1- or EPAC2-positive neurons per total profile for immunohistochemistry.

All behavioral experiments were performed in a blinded manner. Mechanical sensitivity was assessed as withdrawal responses to von Frey stimulation. Unrestricted animals were placed in a clear plastic chamber on an elevated wire grid. After the animals became acclimatized, the withdrawal response to mechanical stimulation was determined using a calibrated von Frey monofilament set (Muromachi Kikai, Japan). The lowest force that evoked a clear withdrawal response at least twice in 10 applications was accepted as the threshold.

Neurons that were positive for EPAC1 and EPAC2 were visualized using immunohistochemistry. The L5 DRG tissues were cryoprotected in 20% sucrose in 0.1 M PBS (pH 7.4) at 4°C overnight and then frozen and stored at −80°C. The DRG sections (10 µm) were cut using a cryostat (Leica Biosystems, Germany) and mounted on silane-coated glass slides.

Sections were washed with 0.1 M PBS and then incubated with Blocking One (Nacalai Tesque Inc., Japan) at room temperature for 30 min, followed by incubation at 4°C for 3 days with rabbit anti-EPAC1 antibody (1:100; Abcam, USA) or rabbit anti-EPAC2 antibody (1:100; Abcam) in 0.1% Tween 20 in 0.1 M Tris-buffered saline (pH 7.4) containing 1% blocking reagent. After washing with PBS, the sections were incubated overnight at 4°C with rhodamine-conjugated anti-rabbit secondary antibody (1:100; Merck Millipore, USA) in 0.1 M Tris-buffered saline.

To demonstrate the distribution of EPAC1 and EPAC2 expression, double-staining fluorescence immunohistochemistry was performed. Sections were incubated with anti-EPAC1 (1:100) or anti-EPAC2 (1:100) antibody at 4°C for 3 days, followed by incubation with fluorescein isothiocyanate-conjugated isolectin B4 (IB4) from griffonia simplicifolia (1:100; Sigma-Aldrich) and rhodamine-conjugated anti-rabbit secondary antibodies (1:100; Merck Millipore) at 4°C overnight.

Separately, sections were incubated with anti-EPAC1 (1:100) or anti-EPAC2 (1:100) antibody and mouse anti-calcitonin gene-related peptide (CGRP, 1:400; Abcam) at 4°C for 3 days, followed by incubation with fluorescein isothiocyanate-conjugated anti-mouse secondary antibodies (1:1,000; Merck Millipore) and rhodamine-conjugated anti-rabbit secondary antibodies (1:100; Merck Millipore) at 4°C overnight. Sections were washed and visualized using a fluorescence microscope with a digital camera system (Nikon, Japan).

Digital immunohistochemistry images were analyzed using Image-J²⁶  (National Institute of Mental Health, USA) on a Macintosh computer system (Apple, Inc., USA). The mean signal intensity and cross-sectional area of the neurons with visible nuclei were calculated in the DRG of naive, incision 3d, and incision 14d groups (n = 6 each). The signal intensity of each neuron was divided by the background signal intensity to give the signal/noise ratio.²⁷  Based on the scattergram showing the relationship between the distribution of the signal/noise ratio and cross-sectional area of EPAC1 or EPAC2 immunohistochemistry, we defined signal/noise of 1.5 as the borderline to determine positive immunoreactivity in each neuron. The digital profile count obtained by this procedure correlated well with the results of manual counting performed by the investigator who was blinded to the study procedure.

Cell counting for double-labeling immunohistochemistry against EPAC1 or EPAC2 and IB4 or CGRP was performed using the DRGs of the incision 14d group 14 days after the plantar incision (n = 6) by individuals who were unaware of the experimental protocol in a double-blinded manner. The proportions of DRG neurons expressing EPAC1/EPAC2 immunoreactivity were determined by counting the neuronal profiles in the DRG sections that were positive for both EPAC1/EPAC2 and CGRP/IB4. Only neurons with visible nuclei were processed for calculation. Four sections at least 100 µm apart in the DRG were selected from each rat. The overall average of the percentages of EPAC1-/EPAC2-positive profiles was calculated.

Western blotting analysis of the amount of EPAC1 and EPAC2 was performed on the DRG of naive rats (n = 5) and those 14 days after the plantar incision (n = 5). During terminal anesthesia, the DRG was removed and homogenized with brief sonication in a homogenate buffer (20 mM Tris–HCl [pH 8.0] containing 137 mM NaCl, 2 mM EDTA, 10% glycerol, 1% Triton X-100, 1 µM phenylmethane sulfonyl fluoride, 10 µg/ml leupeptin, and 10 µg/ml pepstatin A). After brief centrifugation to remove insoluble tissues, the protein concentration of each homogenate was determined by the Bradford reagent (BioRad, USA). Each cell lysate was then dissolved with an equal amount of 2% sodium dodecyl sulfate (SDS)–Laemmli sample buffer (62.5 mM Tris–HCl [pH 6.8], 2% SDS, 6% 2-mercaptoethanol, and 10% glycerol). Equal amounts of cell lysate were separated on 8.0% SDS–polyacrylamide gels and immunoblotted onto polyvinylidene difluoride membranes (GE Healthcare, United Kingdom).

The membranes were incubated overnight at 4°C with antibodies against EPAC1 (1:1,000; Cell Signaling Technology, Inc., USA), EPAC2 (1:500; Abcam), or actin (1:5,000; Merck Millipore), followed by incubation with horseradish peroxidase-conjugated secondary antibody and visualized using Western Blotting Substrate Plus (Thermo Fisher Scientific, USA) or ECL Detection System (GE Healthcare) and Hyperfilm (GE Healthcare). The intensity of the selected bands was captured and analyzed using Image-J software.

EPAC1 and EPAC2 messenger RNAs (mRNAs) were detected using digoxigenin-labeled in situ hybridization histochemistry. RNA was prepared after DRG homogenization in Trizol (Invitrogen, USA), and reverse transcription was carried out with ReverTra Ace qPCR RT Master Mix (TOYOBO, Japan) using 1 µg total RNA. EPAC1 (651 bp) and EPAC2 (937 bp) complementary DNA fragments were isolated with primers for EPAC1 (5′-TGGTGCTGAAGAGAATGCAC-3′ and 5′-GAAACGGGAACTAGCTGCTG-3′) and EPAC2 (5′-ACTCAAGCTGTTGGCATGTG-3′ and 5′-CCATTCTCTCCTGCTCGGTA-3′) using a Thermal Cycler (GMI Inc., USA) and cloned into the pGEM-T easy vector (Promega Corp., USA). Digoxigenin-labeled RNA probes were transcribed from a linearized vector using a 10 × digoxigenin-labeling mixture and SP6 RNA polymerase (Roche Applied Science, Germany). To confirm the probe specificity of signals, hybridization was performed using sense probes, resulting in no signals on the section. Sections were washed with 0.1 M PBS and acetylated with 0.25% acetic anhydride in 0.1 M triethanolamine at room temperature for 10 min. After washing with 0.1 M PBS once again, the sections were hybridized with digoxigenin-labeled RNA probes (1:100) in a hybridization buffer (Sigma-Aldrich) at 62°C for 12 h in a humidified chamber. Next, the sections were washed with 5 × standard saline citrate (SSC) for 10 min, 0.2 × SSC for 15 min, and 0.1 × SSC for 30 min at 62°C and incubated with buffer 1 (100 mM Tris–HCl [pH 7.5] and 150 mM NaCl) for 5 min, buffer 2 (1% blocking reagent and 0.3% Triton X-100 in buffer 1) for 60 min at room temperature, and anti- digoxigenin-alkaline phosphatase antibody (1:500; Roche Applied Science) in buffer 2 for 12 h at 4°C. Signals were visualized by incubating the sections with 3.5 µl/ml 5-bromo-4-chloro-3-indolyl-phosphate and 4.5 µl/ml 4-nitroblue tetrazolium chloride in 1 ml digoxigenin buffer 3 (100 mM Tris–HCl [pH 9.5], 100 mM NaCl, and 50 mM MgCl₂).

Cell counting for EPAC1/EPAC2 mRNA-positive neurons was performed in the DRG of naive (n = 5) and rats 14 days after the plantar incision (n = 5). Digital images were obtained using a light microscope with a digital camera system (Nikon). Individuals who were unaware of the experimental protocol counted the cells in a double-blinded manner. The proportions of DRG neurons expressing EPAC1/EPAC2 mRNA were determined by counting the neuronal profiles with distinct mRNA labeling compared with the background in DRG sections. Only neurons with visible nuclei were processed for calculation. Four sections at least 100 µm apart in the DRG were selected from each rat.

Statistical analyses were performed using a two-tailed unpaired Student’s t test, one-way ANOVA with Tukey post hoc test, or two-way ANOVA with Bonferroni post hoc test using GraphPad Prism6 (GraphPad Software, USA). Values of P < 0.05 were considered statistically significant. All data are presented as mean ± SD. We did not exclude any data in this study.

Behavior testing revealed that the mechanical threshold against von Frey stimulation after the plantar incision decreased immediately but returned to normal levels 14 days after the treatment. Injection of PGE2 into the plantar tissues produced a transient decline (less than 8 h) in the mechanical threshold against von Frey stimulation in the naive plus PGE2 group. However, the same dose of PGE2 injected into the plantar tissue produced long-lasting nociceptive hypersensitivity in the incision 14 plus PGE2 group. This mechanical hypersensitivity continued for at least 7 days, demonstrating the development of nociceptor priming (fig. 1A). The presence of nociceptor priming was also tested in animals 21 days after the plantar incision. In the incision 21 plus PGE2 group, intraplantar injection of PGE2 resulted in prolonged mechanical hypersensitivity compared to the naive plus PGE2 group, similar to the results of behavioral testing 14 days after the plantar incision (fig. 1B).

To clarify the expression of EPAC in the peripheral nervous system, we performed fluorescent immunohistochemistry against EPAC1 and EPAC2 in the DRG. Low levels of EPAC1 and EPAC2 immunoreactivities were detected in the cell bodies of the DRG neurons in the naive group, but a significant increase in immunoreactivity was observed in the incision 14d group (fig. 2A). The measurement of signal intensity in individual DRG neurons demonstrated that the number of neurons with strong signal intensity for EPAC1 and EPAC2 significantly increased in the incision 14d group (fig. 2B). The percentage of EPAC1- and EPAC2-positive profiles in relation to the total DRG neuronal profiles showed a significant increase in the incision 14d group as compared to the naive group although there was no such increase in the incision 3d group (fig. 2C). Western blotting experiments showed that the amounts of EPAC1 and EPAC2 were significantly increased in the incision 14d group compared to the naive group (fig. 2D).

Expression of EPAC1 and EPAC2 mRNAs in the DRG was also detected by in situ hybridization. Expressions of EPAC1 and EPAC2 mRNAs were low in the naive group but showed a significant increase in the incision 14d group (fig. 3A). The number of neurons expressing EPAC1 and EPAC2 mRNAs significantly increased in the incision 14d group compared to the naive group (fig. 3B).

Double-stained immunohistochemistry for EPAC1 and IB4 or EPAC2 and IB4 in the incision 14d group (fig. 4A) demonstrated that 72% of IB4-expressing neurons were positive for EPAC1 and 68% of IB4-expressing neurons were positive for EPAC2. Double-stained immunohistochemistry for EPAC1 and CGRP or EPAC2 and CGRP in the incision 14d group (fig. 4B) demonstrated that 42% of CGRP-expressing neurons were positive for EPAC1 and 45% of CGRP-expressing neurons were positive for EPAC2.

Size frequency distribution analysis demonstrated that EPAC1 and EPAC2 signals in the incision 14d group were preferentially expressed in the DRG neurons with small-sized (−800 µm²) cell bodies (fig. 4C). Only few EPAC1 and EPAC2 signals were detected in the cell body or nerve terminal in the spinal dorsal horn of both the naive animals and animals that received plantar incisions (data not shown).

To assess the effect of EPAC signaling on PGE2-evoked persistent nociceptive hypersensitivity during nociceptor priming, animals were treated with the EPAC inhibitor ESI-09 before PGE2 injection. Mechanical nociceptive hypersensitivity after the PGE2 injection recovered to normal levels 8 h after the treatment in the incision plus PGE2 plus ESI-09 i.pl. group, while it continued for 7 days in animals in the incision plus PGE2 plus vehicle group. The time course of mechanical nociceptive hypersensitivity in the incision plus PGE2 plus ESI-09 i.pl. group was similar to that in the naive plus PGE2 plus vehicle group (fig. 5A).

To test the effect of ESI-09 on the development of acute nociceptive hypersensitivity after the plantar incision, ESI-09 was injected 2 h before the plantar incision. Behavior testing demonstrated that the duration and intensity of mechanical nociceptive hypersensitivity immediately after the incision were similar between ESI-09 i.pl. and vehicle groups (fig. 5B).

To test the effect of ESI-09 on established acute nociceptive hypersensitivity after the plantar incision, ESI-09 was injected 1 day after the plantar incision. Behavior testing demonstrated that ESI-09 did not have any effect on mechanical nociceptive hypersensitivity 1 day after the incision (fig. 5C). To test the effect of selective delivery of ESI-09 to the DRG neurons on established acute nociceptive hypersensitivity after the plantar incision, ESI-09 was injected intrathecally 1 day after the plantar incision. Behavior testing demonstrated that ESI-09 had no effect on mechanical nociceptive hypersensitivity 1 day after the incision (fig. 5D).

To test the involvement of p38MAPK signaling in the development of nociceptor sensitization, the p38MAPK inhibitor, FR167653, was systemically injected before the plantar incision. The duration of PGE2-induced mechanical nociceptive hypersensitivity 14 days after the incision was significantly reduced in the incision plus PGE2 plus FR167653 i.p. group compared to the incision plus PGE2 plus vehicle group (fig. 6A). Expression of EPAC1 and EPAC2 after the plantar incision increased in the incision plus vehicle group but not in the incision plus FR167653 i.p. group (fig. 6B). The percentage of EPAC1- and EPAC2-positive neurons in the DRG significantly decreased in the incision plus FR167653 i.p. group compared to the incision plus vehicle group (fig. 6C). In order to test the involvement of activation of p38MAPK in the dorsal horn of the spinal cord, FR167653 was intrathecally injected before the incision. The duration of mechanical nociceptive hypersensitivity in the incision plus PGE2 plus FR167653 i.t. group was similar to that in the incision plus PGE2 plus vehicle group (fig. 6D).

The observations in the current study demonstrated the presence of nociceptor priming during the 14- to 21-day period after tissue injury in a rat plantar incision model. Nociceptor priming is characterized by latent nociceptive hypersensitivity, whereby additional inflammation causes long-lasting nociceptive hypersensitivity. Increased expression of EPAC1 and EPAC2 driven by the p38MAPK pathway regulates the development of nociceptor priming.

The etiology of severe pain experienced during the subacute period and/or failure of resolution of initial acute pain after surgery is not fully understood. Based on the findings in peripheral nerve injury models in animals, nerve injury might be one of the possible mechanisms for pain persistence.²⁸  In our study, animals that received a plantar incision, but not peripheral nerve injury, developed long-lasting nociceptive hypersensitivity in response to an inflammatory stimulus 14 to 21 days after the initial injury. Our observations, therefore, might provide a new mechanism for the late recovery from acute pain after surgery.

The subacute period after surgery might be a critical period for the development of persistent postsurgical pain.²⁹  Previous clinical observations have demonstrated that some patients with postmastectomy pain or postthoracotomy pain showed no apparent symptoms associated with nerve injury.^30–32  The transition from acute to sustained pain hypersensitivity without nerve injury observed in our study might explain, at least in part, the pathophysiology of this phenomenon.

EPAC is a novel class of cAMP target proteins.³³  Two isoforms of this protein, EPAC1 and EPAC2, have been identified in humans and rodents.³⁴  Dysregulation of EPAC is associated with the development of various kinds of diseases, including cardiac hypertrophy, atherosclerosis, diabetes, asthma, and Alzheimer disease.³⁵  EPAC mediates signaling from cAMP to PKCε in the DRG neurons¹⁴  and might contribute to the development of nociceptor priming.^17,18  Our current observations indicate that the expression of EPAC1 and EPAC2 markedly increases 14 days after plantar incision, at the time when nociceptor priming is detected. Moreover, this study is the first to report the use of ESI-09, a newly developed inhibitor that has selective inhibitory activity against both EPAC1 and EPAC2,^19,36  in the investigation of the pain pathway. We demonstrated that ESI-09 successfully prevents the persistent hyperalgesia induced by PGE2 during nociceptor priming. The same amount of ESI-09 showed no obvious effect on acute nociceptive hypersensitivity, suggesting a specific role of EPAC for nociceptor priming but not for acute nociceptive hypersensitivity. To the best of our knowledge, this is the first evidence showing that pharmacologic inhibition of EPAC inhibits nociceptor priming.

In the current study, the majority of EPAC1 and EPAC2 signals induced by the plantar incision were detected in small-sized neurons. Double-labeling immunohistochemistry demonstrated that EPAC1 and EPAC2 were preferentially distributed within IB4-positive neurons. This observation is in line with the results of previous studies demonstrating that EPAC-mediated translocation of PKCε¹⁴  or nociceptor priming³⁷  occurs within the IB4-positive neuronal population.

In our study, EPAC1 and EPAC2 expression showed similar distribution in the DRG, and both of them increased after the incision. While reducing EPAC1 expression by antisense oligodeoxynucleotides prevented carrageenan-induced nociceptor priming,¹⁸  the contribution of EPAC2 toward the modulation of nociceptive perception cannot be ignored since EPAC2 mediates prolonged electrical activity after PGE2 treatment in rat DRG cultures.³⁸  EPAC1 and EPAC2 double-knockout mice displayed more pronounced learning impairment compared to EPAC1 or EPAC2 single-knockout mice, suggesting their redundant functions in the nervous system.³⁹  Together, EPAC1 and EPAC2 are likely to receive similar regulation and to work cooperatively in the DRG neurons to develop nociceptor priming after a plantar incision.

Transcriptional or posttranscriptional activation of DRG neurons exposed to inflammatory mediators results in development of injury-induced nociceptor plasticity, leading to acute nociceptive hypersensitivity.⁴⁰  One of the major intracellular signaling pathways responsible for nociceptor plasticity is the phosphorylation of p38MAPK.^41–46  We recently showed that induction of phosphorylated p38MAPK in DRG neurons contributes to the development of acute nociceptive hypersensitivity after plantar incisions.²³  In addition, our present data demonstrated that activation of p38MAPK after the plantar incision increased EPAC1 and EPAC2 expression and induced nociceptor priming. Our observation demonstrates that inhibition of p38MAPK during surgery can successfully prevent the development of pain sustained during the subacute postsurgical period.

While intrathecal injection of FR167653 did not have any effect on the development of nociceptor priming, we could not completely rule out the possible involvement of spinal p38MAPK. A previous study demonstrated that p38MAPK was phosphorylated after the plantar incision and contributed to the development of acute nociceptive hypersensitivity.²⁵ 

In conclusion, we demonstrated that acquired EPAC activity in the DRG neurons via p38MAPK is the key molecular event during the development of nociceptor priming as a model for sustained pain during the subacute period after surgery. The p38MAPK/EPAC pathway is a promising candidate target in acute postoperative pain management for the prevention of pain sustained beyond the initial injury.¹ 

Supported in part by a grant-in-aid from the Japan Society for the Promotion of Science (Tokyo, Japan; KAKENHI, nos. 15H04969 and 26670692).

Dr. Amaya was supported by a grant-in-aid from the Japan Society for the Promotion of Science (Tokyo, Japan; KAKENHI, nos. 15H04969 and 26670692). The other authors declare no competing interests.