Amitriptyline Suppresses Neuroinflammation-dependent Interleukin-10-p38 Mitogen-activated Protein Kinase-Heme Oxygenase-1 Signaling Pathway in Chronic Morphine-infused Rats

All experiments conformed to the Guiding Principles in the Care and Use of Animals of the American Physiology Society and were approved by the National Defense Medical Center Animal Care and Use Committee (National Defense Medical Center, Taipei, Taiwan). Animal preparation and intrathecal catheter implantation were as described in our previous study.10Briefly, male Wistar rats (300–350 g) were anesthetized with phenobarbital (50 mg/kg, intra-peritoneal; Sigma, Saint Louis, MO), and two intrathecal catheters were implanted via  the atlantooccipital membrane into the intrathecal space at the level of the lumbar enlargement of the spinal cord (L1–L2). Intrathecal catheters were constructed using an 8-cm polyethylene tube (0.008 inch ID, 0.014 inch OD) and a 3.5-cm silastic tube. The silastic tube was inserted into the polyethylene tube and sealed with epoxy resin and silicon rubber. One was connected to a miniosmotic pump for continuous infusion of saline (1 μl/h), amitriptyline (15 μg/h; Sigma), p38 MAPK inhibitor SB203580 (0.5 μg/h; Sigma), morphine (15 μg/h; Sigma), amitriptyline/morphine, or amitriptyline/morphine/SB203580 for 5 days. The other catheter was used for a single daily injection of the anti–IL-10 antibody (10 μg; 1–10 μg as indicated; Pierce-Endogen, Thermo Fisher Scientific Inc, Waltham, MA) or HO-1 antagonist zinc protoporphyrin (24 μg; 6–24 μg as indicated; Tocris Cookson Ltd., Bristol, United Kingdom) for 5 days. Rats received a mouse isotype-matched irrelevant mAb (Pierce-Endogen) as negative control. On day 5, after a discontinuation of drug infusion for 3 h, when tail-flick latency had returned to the baseline level (around 2 s), morphine’s dose-response curves were constructed. All rats were housed individually and maintained on a 12-h light/dark cycle with food and water freely available. Rats with any neurologic deficits were excluded from the study (< 10%).

Tail-flick latency was measured using the hot water immersion test (52 ± 0.5°C) before drug infusion and daily after the start of infusion for 5 days. All rats were placed in plastic restrainers for the antinociceptive test. The average baseline tail-flick test latency was 2 ± 0.5 s in naïve rats, and the cutoff time was 10 s. The percentage of the maximal possible antinociceptive effect was calculated as (maximum latency − baseline latency)/(cutoff latency − baseline latency) × 100. Antinociceptive dose-response curves were constructed for each study group.

After drug infusion, all rats were sacrificed, and the dorsal portion of the lumbar spinal cord enlargement was immediately removed and homogenized in ice-cold lysis buffer, consisting of 50 mm Tris, pH 7.5, 150 mm NaCl, 2% Triton X-100, 100 μg/ml phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin, and phosphatase inhibitors. The lysate was centrifuged at 12,000g  for 30 min at 4°C, and the supernatant was then used for Western blotting. The protein concentration of the samples was determined by the BCA method (Pierce, Thermo Fisher Scientific Inc, Waltham, MA) using bovine serum albumin as the standard. Equal amounts of total protein (20 μg) were adjusted to a similar volume with loading buffer (10% SDS, 20% glycerin, 125 mm Tris, 1 mm EDTA, 0.002% bromophenol blue, 10% β-mercaptoethanol), denatured by heating at 95°C for 5 min, separated on 10% sodium dodecyl sulfate–polyacrylamide gels, and transferred onto polyvinylidene difluoride membranes (Millipore, Billerica, MA). The membranes were blocked with 5% nonfat milk and incubated overnight at 4°C with polyclonal rabbit anti-p44/p42 MAPK (1:1000), anti-phospho -p44/p42 MAPK (1:500), anti-SAPK/JNK (1:1000), anti-phospho -SAPK/JNK (1:500), anti-p38 MAPK (1:1000), or anti-phosphor-p38 MAPK (1:500) antibodies (all from Cell Signaling, Danvers, MA) or polyclonal rabbit anti-HO-2 (1:2000) or monoclonal mouse anti-HO-1 (1:2000) antibodies (both from Stressgen, Victoria, British Columbia, Canada), and then incubated for 1 h at room temperature with appropriate horse radish peroxidase-conjugated secondary antibodies (1:2500; donkey anti-rabbit or anti-mouse immunoglobulin; Chemicon, Temecula, CA). Membrane-bound secondary antibodies were detected using the Chemiluminescence^plusreagent (PerkinElmer LAS, Boston, MA) and visualized using a chemiluminescence imaging system (Syngene, Cambridge, United Kingdom). Finally, the blots were incubated for 18 min at 56°C in stripping buffer (62.6 mm Tris-HCl, pH 6.7, 2% SDS, 100 mm mercaptoethanol) and reprobed with monoclonal mouse anti-β-actin antibody (1:2500; Sigma) as a loading control. The Western blot analysis was repeated three times. The density of each specific band was measured using a computer-assisted imaging analysis system (Gene Tools Match software, Syngene).

Spinal cord lysates were analyzed for four cytokines in a simultaneous multiplexed format using a microbead-based and flow-based protein detection system (Bio-Plex Suspension Array System, Bio-Rad Laboratories Inc., Hercules, CA) based on the Luminex xMAP technology (Luminex Molecular Diagnostics Inc., Toronto, Ontario, Canada). In this quantitative assay system, the surfaces of fluorescence-coded microbeads are conjugated to specific antibodies directed against four cytokines target. Each sample was first incubated with a mixture of all microbead types for 120 min at room temperature. After wash, samples were incubated with a mixture of secondary biotinylated detection antibodies also directed against each target for 30 min at room temperature, then washed again and incubated with a streptavidin-coupled phycoerythrin reporter system for 10 min at room temperature. After a final wash, the samples were resuspended in assay buffer and subjected to flow cytometric analysis. The Bio-Plex system instrumentation incorporates fluidics, laser excitation, fluorescence detection, and digital signal processing in a manner that allows for individual scanning and identification of microbeads. Each bead taken from the samples was identified on the basis of its internal fluorescence signature, and the phycoerythrin reporter signal associated with that bead was quantitated. A minimum 100 microbeads per each of the four targets were analyzed in each spinal cord sample. All spinal cord lysates were assayed in triplicate. Observed concentrations of each target cytokine were determined on the basis of an appropriate set of recombinant rat cytokine internal standard curves.

After drug infusion, the rats were sacrificed by exsanguination under isoflurane anesthesia, and the lumbar spinal cord was immediately removed, embedded in optimal cutting temperature compound (Sakura Finetec Inc, Torrance, CA), and frozen at −20°C. Sections (5 μm) were prepared, air-dried on microscope slides for 30 min at room temperature, and fixed by immersion for 10 min in ice-cold acetone/methanol (1:1). After washing three times in ice-cold phosphate-buffered saline, the sections were incubated with blocking solution (0.01% Triton, 4% fetal bovine serums; 1 h). They were then double-labeled by incubation overnight at 4°C with (1) fluorescin isothiocyanate-labeled (green fluorescence) mouse monoclonal anti-rat CD11b/c antibodies (OX42 1:10, for macroglia; Chemicon, Temecula, CA) and unlabeled polyclonal goat anti-IL-10 antibodies (1:50; R&D System, Minneapolis, MN) and then for 1 h at room temperature with rhodamine-labeled guinea pig anti-goat immunoglobulin antibodies (1:200; R&D System), (2) FITC-labeled neuron-specific nuclear protein antibodies (NeuN 1:100; Chemicon) and unlabeled monoclonal mouse anti-HO-1 antibodies (1:500; Stressgen, Ann Arbor, MI) and then for 1 h at room temperature with rhodamine-labeled goat anti-mouse immunoglobulin antibodies (1:200; R&D System), or (3) unlabeled monoclonal mouse anti-HO-1 antibodies and unlabeled polyclonal rabbit anti-phopho-p38 MAPK antibodies (source) and then for 1 h at room temperature with FITC-labeled goat anti-mouse immunoglobulin antibodies and rhodamine-labeled anti-rabbit antibodies. After three phosphate-buffered saline rinses, the sections were incubated in 1% methanol in phosphate-buffered saline containing 20 μg/ml of 4′,6-diamidino-2-phenylindole (Sigma) (blue fluorescence). Images were captured using an Olympus BX 50 fluorescence microscope (Olympus Optical, Tokyo, Japan) and a Delta Vision disconsolation microscopic system operated by SPOT software (Diagnostic Instruments Inc. Sterling Heights, MI).

All values are presented as the mean ± SEM and analyzed using SigmaStat 3.0 (SYSTAT Software Inc., San Jose, CA). Tail-flick latency data were analyzed by using two-way analysis of variance (ANOVA) with post hoc  Bonferroni test. Values for the analgesic dose of 50% of the maximal possible antinociceptive effect (AD⁵⁰) were analyzed using a computer-assisted linear regression program SigmaPlot 10.0 (SYSTAT Software Inc.). The 95% confidence interval (CI) was calculated using pharmacologic calculations system PHARM/PCS  version 4.2 (PharmSoft.Net, Wynnewood, PA). For immunoreactivity data, the intensity of each test band was expressed as the relative optical density calculated with respect to the average optical density for the corresponding control band. For statistical analysis, immunoreactivity and cytokine data were analyzed by one-way analysis of variance (ANOVA), followed by multiple comparisons with the Student-Newman-Keuls post hoc  test. A significant difference was defined as a P  value less than 0.05.

As shown in figure 1A, analysis of lumbar spinal cord homogenates demonstrated that intrathecal daily injection of anti-IL-10 antibody (10 μg/5 μl) alone did not affect IL-10 expression compared to saline-infused rats. However, coinfusion amitriptyline/morphine significantly increases the antiinflammatory cytokine IL-10 protein levels compared with saline-, amitriptyline-, and morphine-infused rats. Daily injection of anti-IL-10 antibody significantly neutralized the effect of amitriptyline in morphine-infused rats (F(5,18) = 24.35, P < 0.001). Additional experiments in which the rats received an isotype-matched irrelevant Ig show that the IL-10 protein level (956.7 ± 68.35 ng/ml) compared with amitriptyline/morphine–coinfused rats (906.67 ± 52.14 ng/ml) (P = 0.582, NS). As shown in figure 1B, weak immunocytochemistry staining for microglia (OX42-positive cells; green) was distributed throughout the rat spinal cord, and the stained cells had the classic ramified shape of nonactivated microglia in saline-infused, amitriptyline-infused, and anti-IL-10 antibody injection-only rats. In contrast, strong OX42 staining with the amoeboid shape of activated microglia was observed throughout the section, but no IL-10 protein expression was apparent in chronic morphine-infused rats. Overlay of the sections stained for IL-10 and microglia showed colocalization of IL-10 with OX42 immunoreactivity (white arrows) in the spinal cord of amitriptyline plus morphine-infused rats, and additional treatment with anti-IL-10 antibody resulted in a pattern similar to that of morphine-infused rats.

As shown in figure 2A, measured by the hot water immersion test, daily intrathecal injection of anti-IL-10 antibody, or continuous infusion of amitriptyline (15 μg/h) or SB203580 (0.5 μg/h) alone had no antinociceptive effect throughout the 5 days. As in our previous study,6the maximal antinociceptive effect of morphine was seen on day 1, and maximal tolerance was observed from day 3 in morphine-infused rats (F(5,235) = 23.447, P < 0.001). Amitriptyline (15 μg/h) prevented the reduction of the antinociceptive effect of morphine seen in chronic morphine-infused rats. Daily injection of anti-IL-10 antibody or coinfusion of SB203580 with amitriptyline plus morphine blocked the maintenance of morphine’s antinociceptive effect by amitriptyline (F(47,235) = 6.231, P < 0.001). As shown in figure 2Band table 1, amitriptyline, anti-IL-10 antibody, and SB203580 alone did not affect the dose-response curve of morphine, with an AD⁵⁰of 1.13 μg in saline-infused rats, 1.3 μg in amitriptyline-infused rats, 1.24 μg in anti-IL-10 antibody injection rats, and 1.23 μg in SB203580-infused rats. In morphine-infused rats, the dose–response curve was shifted to the right by 77.8-fold compared to saline-infused rats (AD⁵⁰, 87.96 μg), amitriptyline/morphine coinfusion shifted the dose-response curve to the left compared to morphine infusion alone (AD⁵⁰, 5.47 μg), wheras injection of anti-IL-10 antibody or coinfusion of SB203580 produced a rightward shift compared to amitriptyline/morphine-infused rats (AD⁵⁰, 20.23 μg and 17.03 μg, respectively).

Figure 3shows that expression of external signal-regulated kinase (ERK), phospho -ERK (p -ERK), c-Jun N -terminal kinase (JNK), and phospho -JNK (p -JNK) at day 5 in the spinal cord dorsal horn was unaffected by any of the treatments (P = 0.721, 0.632, 0.541, and 0.729, respectively). These results suggest that neither ERK nor JNK mediated amitriptyline-induced IL-10 expression. Figure 4shows that chronic intrathecal infusion of morphine caused a slight, but significant increase in the level of phosphorylated p38 MAPK (p -p38 MAP) (fig. 4, A and C), but it had no effect on total p38 MAPK, HO-1, or HO-2 levels (fig. 4, B and D). In contrast, amitriptyline/morphine coinfusion significantly increased not only p -p38 MAPK expression (F(4,35) = 820, P < 0.001), but also total p38 MAPK expression (F(4,35) = 25.759, P < 0.001) (fig. 4, A and C) and HO-1 expression (F(4,35) = 84.821, P < 0.001) (fig. 4, B and D), and this injection of anti-IL-10 antibody or infusion of amitriptyline or SB203580 alone had no effect on protein expression (data no shown). As expected, HO-2 expression was not affected by any of the treatments (P = 0.824, NS) (fig. 4, B and D). The effects of amitriptyline/morphine coinfusion on p38 MAPK phosphorylation and HO-1 expression were completely blocked by both daily injection of anti-IL-10 antibody and coinfusion of the p38 MAPK inhibitor SB203580 (fig. 4, A–D). In the current study, we also found that daily intrathecal injection of various doses of anti-IL-10 antibody for 5 days dose-dependently reversed the amitriptyline-induced upregulation of p38 MAPK phosphorylation (F(4,35) = 723.242, P < 0.001) and HO-1 expression (F(4,35) = 39.368, P < 0.001) in morphine-infused rats (fig. 5, A and B). These results suggest that IL-10–induced p38 MAPK phosphorylation is involved in the increase in HO-1 expression in amitriptyline/morphine-infused rats.

Daily intrathecal injection of various doses of zinc protoporphyrin for 5 days dose-dependently reversed the upregulation of HO-1 expression (F(4,35) = 41.056, P < 0.001; fig. 6A), and it blocked the maintenance of morphine’s antinociceptive effect by amitriptyline (F(29,145) = 6.682, P < 0.001) in morphine-infused rats (fig. 6B). The HO-2 expression (P = 0.388, NS) was not affected by zinc protoporphyrin injection. In agreement with the above results, the maximal antinociceptive effect of morphine was seen on day 1, and maximal tolerance was observed from day 3 in morphine-infused rats (F(5,145) = 19.902, P < 0.001). Coadministration of HO-1 antagonist zinc protoporphyrin (24 μg/5 μl) with amitriptyline plus morphine blocked the maintenance of morphine’s antinociceptive effect by amitriptyline and produced a rightward shift in dose-response curves (AD⁵⁰, 22.78 μg; table 2).

As shown in figure 7, protein levels for the proinflammatory cytokines, TNF-α (F(5,18) = 58.817, P < 0.001), IL-1β (F(5,18) = 20.317, P < 0.001), and IL-6 (F(5,18) = 102.068, P < 0.001), were significantly increased in the morphine-infused rats, and this effect was significantly attenuated by amitriptyline coinfusion. Furthermore, anti-IL-10 antibody injection, SB203580 coinfusion, and HO-1 antagonist zinc protoporphyrin injection significantly abolished the suppressive effect of amitriptyline on proinflammatory cytokine production in morphine-infused rats.

Fluorescence images of spinal cords from amitriptyline/morphine-coinfused rats, single-staining for neurons (green), HO-1 (red), or nuclei (4′,6-diamidino-2-phenylindole, blue) are shown in the upper row in figure 8. As seen in the merged images, HO-1 expression colocalized with neurons (yellow on the merged image in panel D) and was around the nuclear membrane, with some translocated into the nucleus (purple in the merged image). The lower row in figure 8shows single-staining for HO-1 (green), p -p38 MAPK (red), or 4′,6-diamidino-2-phenylindole (blue); a similar distribution pattern of HO-1 and p -p38 MAPK was seen (yellow in the merged image in panel H). These results indicate a close association among neurons, HO-1, and p -p38 MAPK in the spinal cord dorsal horn.

In the current study, long-term morphine infusion significantly increased protein levels of proinflammatory cytokines, TNF-α, IL-1β, and IL-6, and caused slight but significant phosphorylation of p38 MAPK. Amitriptyline/morphine coinfusion increased the expression of IL-10 in nonactivated microglia, and increasing of p38 MAPK phosphorylation subsequently induced HO-1 expression in neurons, thus further inhibiting the expression of proinflammatory cytokines; however, the protein expression on ERK, p -ERK, JNK, and p -JNK proteins in the rat spinal cord were not affected. Neutralization of IL-10 expression and blockade of p38 MAPK pathway reverses the amitriptyline-induced increase in HO-1 expression and inhibition of proinflammatory cytokines expression, suggesting that the p38 MAPK pathway, but not the ERK or JNK pathways, is involved in the amitriptyline-mediated induction of HO-1 expression. Moreover, inhibition of IL-10 expression, p38 MAPK pathway, or HO-1 expression partially blocked the inhibitory effect of amitriptyline on the morphine’s antinociceptive tolerance, and it shifted the dose-response curve of morphine to the right from an AD⁵⁰of 5.47 μg (morphine plus amitriptyline) to 20.23 μg (morphine plus amitriptyline plus anti-IL-10Ab injection), 17.03 μg (morphine plus amitriptyline plus SB203580), and 22.8 μg (morphine plus amitriptyline plus zinc protoporphyrin injection). These results suggest that the suppressive effect of amitriptyline coinfusion on proinflammatory cytokine production in morphine-infused rats may act through increasing IL-10 protein expression in microglia, which activates the neuronal p38 MAPK pathway and increases HO-1 expression.

Studies have shown that TCAs inhibit the release of inflammatory mediators. Imipramine, clomipramine, and citalopram at low micromolar concentrations partially blocked IL-6, IL-1β, and TNF-α release from human blood monocytes and IL-2 and interferon-γ from T cells.20Similarly, fluoxetine and amitriptyline were shown to inhibit release of nitric oxide and prostaglandin E2 from human synovial cells.21A previously performed study also showed that production of the antiinflammatory cytokine IL-10 was shown to be stimulated by antidepressants.16These changes resulted in a decrease of hypersensitization of the nociceptive neurons in the spinal cord, and amitriptyline, when applied peripherally, also prevented c-fos expression in spinal cord in the formalin pain model.22In the current study, we also demonstrated that coinfusion of amitriptyline increased the expression of antiinflammatory cytokines IL-10 in morphine-infused rats.

A previous study demonstrated that IL-10 inhibited proinflammatory cytokine production.23Local injection of IL-10 abolishes the mechanical hyperalgesia induced by carrageenan administration in the hind paw of rats.24IL-10–encoding adenoviral vector not only attenuates the development of tolerance, hyperalgesia, and allodynia; it also potentiates the analgesic effect of morphine.7Treatment with morphine has been shown to not alter IL-10 messenger RNA level and even reduces lippolysaccharide-induced IL-10 secretion.25,26In the current study, an increase in IL-10 protein expression was observed in some microglia in the spinal cord of amitriptyline/morphine coinfused rats, these being nonactivated microglia, as opposed to the activated forms seen in morphine-infused rats. Anti-IL-10 antibody neutralized the amitriptyline induced IL-10 in activated microglia. It implies that the suppressive effect of amitriptyline in morphine-infused rats acts by increasing IL-10 protein expression.

A previous study showed that IL-10 exerts its antiinflammatory effect by upregulating HO-1 expression.19Antiinflammatory effect of HO-1 has been demonstrated in several inflammatory models.27,28HO-1 overexpression, via  carbon monoxide-cyclic guanosine monophosphate (CO-cGMP) pathway, produces a significant antinociceptive effect against formalin pain in mice.29–31These results show that HO-1 plays a crucial role in the host defense mechanism against inflammation; HO-1 catabolizes heme and produces antioxidants biliverdin and bilirubin, which also play a role in protection against tissue injury-induced inflammation. Carbon monoxide is a key molecule for the protective effect of HO-1.32,33Scavenging of carbon monoxide by hemoglobin significantly reduces the effect of IL-10 on inhibiting the lippolysaccharide-induced TNF-α and nitric oxide production and increasing MMP-9 expression in macrophages.19In addition, a recent study indicated that nuclear localization of HO-1 protein causes upregulation of AP-1 gene and promotes cytoprotection against oxidative stress.34In our current study, we found that amitriptyline/morphine coinfusion significantly increased IL-10 and HO-1 expression, and it decreased the protein level of proinflammatory cytokines TNF-α, IL-1β, and IL-6. In contrast, the inhibition of IL-10 or HO-1 expression significantly downregulated HO-1 expression and increased TNF-α, IL-1β, and IL-6 protein level. These results suggest a role of HO-1 in the antiinflammatory effect of IL-10 in amitriptyline/morphine-coinfused rats. Taken together, carbon monoxide and HO-1 appear to be responsible for the protective effect of IL-10, and the detailed mechanisms need to be determined.

An earlier study revealed that the JAK1/STAT3-dependent pathway is not sufficient for the antiinflammatory effect of IL-10 receptor signaling,35and the IL-10-activated phosphatidylinositol-3-kinase and p70 S6 kinase pathways are not essential for the antiinflammatory effect of IL-10.36Moreover, studies have suggested that IL-10 and HO-1 produce a positive feedback circuit to amplify the antiinflammatory effect by upregulation of carbon monoxide-dependent and p38 MAPK-dependent mechanisms.19,37The transcriptional activation of HO-1 is mediated by the ERK, JNK, and p38 MAPK pathways,38,39and p38 MAPK is a key mediator of cellular stresses, such as inflammation and apoptosis. Both in vitro  and in vivo  studies have shown that p38 MAPK regulates the production of proinflammatory cytokines, nitric oxide, and prostaglandin E2 by increasing cytokine release or messenger RNA transcription.40–42As known, increasing of proinflammatory cytokine expression in the spinal cord is associated with neuropathic pain and morphine tolerance.7,43p38 MAPK is activated in neuropathic pain by peripheral inflammation and nerve injury.44–46Morphine tolerance-associated microglia activation is also activated via  p38 MAPK signaling pathway.47Inhibition of p38 MAPK was considered to be a potential therapeutic target for antiinflammatory therapy.48However, contradictory results were reported. Treatment with a p38 MAPK inhibitor resulted in an enhancement of TNF-α production in mast cells.49Moreover, inhibition of p38 MAPK resulted in a reduction of cytokine production by mast cells, but increase cytokine release from lippolysaccharide-activated macrophages.50These results suggest that p38 MAPK possesses both proinflammatory and antiinflammatory characteristics, and plays different roles in different inflammatory conditions, depending on the cell type and applied stimulus.

Communication between neurons and glia involves ion flux, neurotransmitters, cell adhesion molecules, and specialized signaling molecules released from synaptic and nonsynaptic regions of neurons.51Morphine-induced glia activation and proinflammatory cytokine release are also caused by the induction of neuronal fractalkine release.7Fractalkine, expressed in spinal neurons, is a diffusible signal that activates nearby microglia and then releases proinflammatory cytokines.52Theses results suggest that morphine indirectly affects glia via  chemokines conducting neuron-glia communication. Activated neuronal protein kinase Cγ (PKCγ) acts as an important factor to modulate the neuron-glia communication to increase astrocyte reactivity after repeated morphine treatment, and this neuron-glia communication may be responsible for the development of morphine tolerance.53In the current study, we suggest that amitriptyline modulates neuroimmune responses via  a two-way communication between glia and neurons. Amitriptyline stimulates microglia release of antiinflammatory cytokine IL-10 via  a paracrine-manner and increases neuron phopho -p38 MAPK expression, which induces upregulation of HO-1 via  carbon monoxide, exerts negative feedback control of microglia activity, and inhibits proinflammatory cytokine expression.

Activation of the neuroimmune system has been demonstrated to modulate morphine tolerance and consequently develops mechanical allodynia and thermal hyperalgesia.7,54Administration of glial metabolic inhibitors, cytokine antagonist IL-10, or IL-1 receptor antagonist IL-1ra attenuates opioid tolerance and associated hyperalgesia and allodynia.4,8Similarly, our previous study also demonstrated that amitriptyline inhibits microglia activation and proinflammatory cytokines expression, thus attenuating morphine tolerance.6In our current study, we found that injection of anti-IL-10 antibody, or p38MAPK inhibitor SB203580, or HO-1 inhibitor zinc protoporphyrin increased proinflammatory cytokine expression in amitriptyline/morphine-coinfused rats and partially reversed the effect of amitriptyline on the antinociceptive effect of morphine seen in chronic morphine-infused rats.

Currently available medications for the management of chronic pain, particularly those for neuropathic pain, are either inadequate or cause unbearable side effects, and TCAs are known to be effective adjuvant for the treatment of chronic neuropathic pain in clinical practice.55,56The current study shows that amitriptyline inhibits pro-inflammatory cytokine expression by increasing IL-10 expression; it leads to phosphorylation of p38 MAPK that increases HO-1 expression and thus inhibits the expression of proinflammatory cytokines TNF-α, IL-1β, and IL-6 in morphine-tolerant rats. From our recent studies and the results of current study,5,6we suggest that amitripyline may be used as an adjuvant in combination with opioids for the treatment of chronic neuropathic pain conditions and patients who need long-term opioid administration for pain management in clinical practice. It explains why TCAs play an important role in the treatment of chronic neuropathic pain via  systemic administration. The systemic effect and mechanisms of action of amitryptline in morphine tolerance development and neuropathic pain should require further investigation.