Peripheral administration of morphine attenuates postincisional pain by regulating macrophage polarization through COX-2-dependent pathway

Abstract

Background: Macrophage infiltration to inflammatory sites promotes wound repair and may be involved in pain hypersensitivity after surgical incision. We recently reported that the development of hyperalgesia during chronic inflammation is regulated by macrophage polarity, often referred to as proinflammatory (M1) or anti-inflammatory (M2) macrophages. Although opioids such as morphine are known to alter the inflammatory milieu of incisional wounds through interactions with immunocytes, the macrophage-mediated effects of morphine on the development of postincisional pain have not been well investigated. In this study, we examined how morphine alters pain hypersensitivity through phenotypic shifts in local macrophages during the course of incision-induced inflammation.

Results: Local administration of morphine in the early phase, but not in the late phase alleviated mechanical hyperalgesia, and this effect was reversed by clodronate-induced peripheral depletion of local macrophages. At the morphine-injected incisional sites, the number of pro-inflammatory F4/80+iNOS+M1 macrophages was decreased during the course of pain development whereas increased infiltration of wound healing F4/80+CD206+M2 macrophages was observed during the early phase. Morphine increased the gene expression of endogenous opioid, proenkephalin, and decreased the pronociceptive cytokine, interleukin-1β. Heme oxygenase (HO)-1 promotes the differentiation of macrophages to the M2 phenotype. An inhibitor of HO-1, tin protoporphyrin reversed morphine-induced analgesic effects and the changes in macrophage phenotype. However, local expression levels of HO-1 were not altered by morphine. Conversely, cyclooxygenase (COX)-2, primarily produced from peripheral macrophages in acute inflammation states, was up-regulated in the early phase at morphine-injected sites. In addition, the analgesic effects and a phenotype switching of infiltrated macrophages by morphine was reversed by local administration of a COX inhibitor, indomethacin.

Conclusions: Local administration of morphine alleviated the development of postincisional pain, possibly by altering macrophage polarity at the incisional sites. A morphine-induced shift in macrophage phenotype may be mediated by a COX-2-dependent mechanism. Therefore, μ-opioid receptor signaling in macrophages may be a potential therapeutic target during the early phase of postincisional pain development.

Figure 1

Local administration of morphine in the early, but not the late phase, alleviates hypersensitivity to mechanical stimuli. Morphine (3 μg or 10 μg) and/or naloxone (5 μg) were injected into the incisional sites during the early phase (1 h after incision and on PODs 1 and 2) or late phase (PODs 5, 6, and 7). The withdrawal threshold to mechanical stimuli (A), withdrawal latency to thermal stimuli (B), and paw edema (C) were examined in groups, receiving morphine in the early phase. The withdrawal threshold to mechanical stimuli was not altered by morphine administered in the late phase (D). *P < 0.001, Mor 10 μg compared with Veh, **P < 0.001, Mor 10 μg compared with Mor 3 μg, ***P < 0.0001, Mor 10 μg compared with (Mor 10 μg + Nal 5 μg), ^#P < 0.001, Mor 10 μg compared with Nal 5 μg (two-way ANOVA followed by Bonferroni post-hoc testing). Data are mean ± SEM (Figure 1A-C, n = 8 for each group, Figure 1D, n = 5 for each group). Veh; vehicle, Mor; morphine, Nal; naloxone.

Figure 2

Depletion of local macrophages resulted in decreased analgesic effects of morphine. (A) Macrophage depletion after intraplantar injection of 10 μL clodronate liposomes was evaluated by immunostaining for local F4/80⁺ macrophages. Green: F4/80; blue: DAPI. Scale bar: 50 μm. (B) The number of F4/80⁺ macrophages was reduced by clodronate liposomes. *P < 0.01 (two-way ANOVA followed by Bonferroni post-hoc testing). Data are mean ± SEM (n = 6 for each group). (C) The elevated withdrawal threshold to mechanical stimuli induced by morphine was reversed by clodronate liposomes. *P < 0.05, Empty-Mor compared with Empty-Veh, **P < 0.05, Empty-Mor compared with Clodronate-Veh, ***P < 0.05, Empty-Mor compared with Clodronate-Mor (two-way ANOVA followed by Bonferroni post-hoc testing). Data are mean ± SEM (n = 8 for each group). Empty-Veh; empty liposomes 10 μL + vehicle (PBS), Clodronate-Veh; clodronate liposomes10μL + vehicle (PBS), Empty-Mor; empty liposomes 10 μL + morphine 10 μg, Clodronate-Mor; clodronate liposomes 10 μL + morphine 10 μg.

Figure 3

Morphine induces phenotype shifts of local macrophages. The infiltration of F4/80⁺iNOS⁺ M1 macrophages (A) and F4/80⁺CD206⁺ M2 macrophages (B) was evaluated on PODs 2 and 7. (C) The numbers of total F4/80, F4/80⁺iNOS⁺, and F4/80⁺CD206⁺ macrophages were counted. Scale bar, 25 μm. *P < 0.05 (two-way ANOVA followed by Bonferroni post-hoc testing). Each column represents mean ± SEM (n = 6 for each group). Veh; vehicle (PBS), Mor; morphine 10 μg, Nal; naloxone 5 μg.

Figure 4

Morphine decreases the expression of pro-nociceptive cytokine, IL-1β and increased the expression of endogenous opioid, enkephalin. The expression levels of Penk were increased at morphine-treated sites. The expression levels of Il1b were reduced at morphine-treated sites, whereas morphine did not alter the expression levels of Tnf. *P < 0.05, **P < 0.01 (two-way ANOVA followed by Bonferroni post-hoc testing). Data are mean ± SEM (n = 8 for each group). Veh; vehicle (PBS), Mor; morphine 10 μg, Nal; naloxone 5 μg, Penk; preproenkephalin, Tnf; tumor necrosis factor, Il1b; interleukin-1β.

Figure 5

HO-1 inhibitor SnPP reverses the attenuation of mechanical hyperalgesia by morphine. (A) The withdrawal threshold to mechanical stimuli was examined. *P < 0.001, (DMSO + Mor) compared with (DMSO + PBS), **P < 0.001, (DMSO + Mor) compared with (SnPP + PBS), ***P < 0.001, (DMSO + Mor) compared with (DMSO + Mor) (two-way ANOVA followed by Bonferroni post hoc testing). Data are mean ± SEM (n = 7 for each group). (B) Morphine did not alter expression levels of Hmox-1. *P < 0.05. Each bar represents mean ± SEM (n = 6 for each) Mor; morphine 10 μg, SnPP; tin protoporphyrin 400 nmol.

Figure 6

HO-1 inhibitor SnPP blocks morphine-induced phenotype shift of local macrophages. The infiltration of F4/80⁺iNOS⁺ M1 macrophages (A) and F4/80⁺CD206⁺ M2 macrophages (B) were evaluated on PODs 2 and 7. (C) The numbers of total F4/80, F4/80⁺iNOS⁺, and F4/80⁺CD206⁺ macrophages were counted. Scale bar, 25 μm. *P < 0.001 (two-way ANOVA followed by Bonferroni post-hoc testing). Each column represents mean ± SEM (n = 6 for each group). Veh; vehicle (PBS + DMSO), Mor; morphine 10 μg, SnPP; tin protoporphyrin 400 nmol.

Figure 7

COX inhibitor, indomethacin reverses morphine-induced elevation of mechanical threshold. (A) The withdrawal threshold to mechanical stimuli was examined. *P < 0.05, (Mor + TB) compared with (PBS + TB), **P < 0.05, (Mor + TB) compared with (PBS + Indo), ***P < 0.01, (Mor + TB) compared with (Mor + Indo) (two-way ANOVA followed by Bonferroni post-hoc testing). Data are mean ± SEM (n = 8 for each group). (B) Morphine increased the expression of Ptgs2 at the incised sites. *P < 0.05 (two-way ANOVA followed by Bonferroni post-hoc testing). Each bar represents mean ± SEM (n = 6–9 for each group). TB; tris buffer, Mor; morphine 10 μg, Indo; indomethacin 50 μg, Veh; vehicle (PBS), Nal; naloxone 5 μg.

Figure 8

Indomethacin reverses morphine-induced phenotype switch of macrophages. Infiltration of F4/80⁺iNOS⁺ M1 macrophages (A) and F4/80⁺CD206⁺ M2 macrophages (B) were evaluated on PODs 2 and 7. (C) The numbers of total F4/80, F4/80⁺iNOS⁺, and F4/80⁺CD206⁺ macrophages were counted. Scale bar, 25 μm. *P < 0.05, **P < 0.01 (two-way ANOVA followed by Bonferroni post-hoc testing). Each column represents mean ± SEM (n = 6 for each group). Veh; vehicle (PBS + TB), Mor; morphine 10 μg, Indo; indomethacin 50 μg.

Figure 9

Schematic diagram of morphine-induced analgesia representing the molecular pathways in macrophages. Peripherally administered morphine alters macrophage polarity through a COX2/PGE2-dependent pathway. Morphine-induced changes in macrophage polarity decrease IL-1β levels, and increase PGE2 and enkephalin in the late phase. HO-1 might be upstream of μ opioid receptor signaling.