Astrocyte activation in the periaqueductal gray promotes descending facilitation to cancer-induced bone pain through the JNK MAPK signaling pathway

Abstract

Descending nociceptive modulation from the supraspinal structures has an important role in cancer-induced bone pain (CIBP). Midbrain ventrolateral periaqueductal gray (vlPAG) is a critical component of descending nociceptive circuits; nevertheless, its precise cellular and molecular mechanisms involved in descending facilitation remain elusive. Our previous study has shown that the activation of p38 MAPK in vlPAG microglia is essential for the neuropathic pain sensitization. However, the existence of potential connection between astrocytes and c-Jun N-terminal kinase (JNK) pathway in CIBP has not yet been elucidated. The following study examines the involvement of astrocyte activation and upregulation of p-JNK in vlPAG, using a CIBP rat model. Briefly, CIBP was mimicked by an intramedullary injection of Walker 256 mammary gland carcinoma cells into the animal tibia. A significant increase in expression levels of astrocytes in the vlPAG of CIBP rats was observed. Furthermore, stereotaxic microinjection of the astrocytic cytotoxin L-α-aminoadipic acid decreased the mechanical allodynia as well as established and reversed the astrocyte activation in CIBP rats. A significant increase in expression levels of p-JNK in astrocytes in vlPAG of CIBP rats was also observed. Moreover, the intrathecal administration of JNK inhibitors SP600125 reduced the expression of glial fibrillary acidic protein, while microinjection of the SP600125 decreased the mechanical allodynia of CIBP rats. These results suggested that CIBP is associated with astrocyte activation in the vlPAG that probably participates in driving descending pain facilitation through the JNK MAPK signaling pathway. To sum up, these findings reveal a novel site of astrocytes modulation of CIBP.

Keywords: Cancer-induced bone pain; astrocyte; c-Jun N-terminal kinase; descending facilitation; hyperalgesia; periaqueductal gray.

Figure 1.

A representative photomicrograph of a coronal slice (vlPAG) stained with red fluorescent stain. The arrow indicates the tip position of the cannulation sites and the circle shows the site of injection. Scale bars = 100 μm. Aq: aqueduct.

Figure 2.

Radiological, behavioral, and histochemical analysis of tumor development in the right tibia. On day 18 after inoculation, intact bone was observed in both naïve group (a) and sham-operated rats (b); while mild (c) and evident (c and d) bone destruction were observed in the CIBP group on 6th, 12th, and 18th day post-surgery, respectively. (f) The ipsilateral PWT progressively decreased from day 6 to day 18 in CIBP rats. Sham group rats showed no significant change in pain sensitivity. Data were expressed as mean ± SEM. n = 10 rats in each group (***P < 0.001; vs. Baseline, ^###P < 0.001; vs. Sham group). (g) The body weight was gradually increased in both sham rats and BCP rats during an 18-day observation period. (h) Hematoxylin and eosin staining of the right tibia showed that bone marrow spaces were infiltrated with malignant tumor (see the arrow) on day 18 after Walker 256 cell inoculation. CIBP: cancer-induced bone pain.

Figure 3.

Time-course-dependent hyperactivation of astrocytes in the VL-PAG after inoculation. (a) Low-power view of the PAG. The dotted lines show the location of the VL-PAG. The rectangle represents the location of photomicrographs (b to f) showing GFAP immunostaining in the VL-PAG of naive rats, sham rats, and at 6, 12, and 18 days after inoculation. (g) Western blot analysis of changes in GFAP protein levels over time in the VL-PAG (***P < 0.001 vs. naïve group; n = 4). (h and i) RT-PCR and the number of GFAP immunoreactive cells revealed a parallel and significant increase in PAG at 6, 12, and 18 days in CIBP model (***P < 0.001 vs. naïve group, n = 4). Scale bars = 100 μm. Aq: aqueduct; VL-PAG: ventrolateral periaqueductal gray; GFAP: glial fibrillary acidic protein; GAPDH: glyceraldehyde 3-phosphate dehydrogenase.

Figure 4.

Tumor cell inoculation-induced hypersensitivity and GFAP expression are reversed by astrocyte cytotoxin. Midbrain sections obtained from Sham rat injections of Veh (a) or CIBP and receiving microinjections of Veh (b) or 100 nmol of LAA (c) in the PAG were labeled with GFAP for immunofluorescent visualization of astrocytes. (d) LAA or Veh were administered once a day on POD 12 to 18. Compared with CIBP + veh group, microadministration of LAA significantly elevated the PWTs of the CIBP rats in a dose-dependent manner (***P < 0.001 vs. CIBP+Veh group; ^###P < 0.001, ^##P < 0.01 vs. CIBP+LAA 1 nmol group; n = 10). (e) PAG tissues were collected 1 h after the last injection. The WB shows continuous VL-PAG injection of LAA reversed the upregulation of GFAP in CIBP rats. GAPDH was used as a loading control (***P < 0.001 vs. Sham+Veh; ^###P < 0.001 vs. CIBP +Veh; n = 4 for each group). (f) The numbers of cells showing GFAP in each group is quantified (***P < 0.001 vs. Sham+Veh; ^###P < 0.001 vs. CIBP +Veh; n = 4 for each group n = 4). Midbrain sections obtained from Sham rat injected with Veh (g) or CIBP rats microinjected with Veh (h) or 100 nmol LAA (i) in the PAG were labeled with Iba1 for immunofluorescent visualization of microglia. (j and k) Intra-vlPAG microinjection of LAA did not block tumor cell inoculation-induced enhancement of Iba1 expression. Scale bar: 100 μm. Aq: aqueduct; CIBP: cancer-induced bone pain; LAA; L-α-aminoadipic acid.

Figure 5.

Tumor cell inoculation induces time-dependent activation of JNK in PAG. (a and b) Expression levels of pJNK in the PAG in naïve groups and Sham-Saline. (c) Upregulation of pJNK in PAG on postoperative day 12. (d) A high magnification image from the rectangle area of C shows a high density of pJNK-immunoreactive staining in PAG after inoculation. (e) Western blot results show an increase of pJNK/JNK in the PAG after tumor inoculation on days 12–18. pJNK expression gradually increased from day 12 to day 18; (***P < 0.001; vs. naïve group; n = 4) (f) Real-time PCR revealed a persistent pJNK upregulation in PAG from day 12 to day 18; *** indicates a statistically significant difference (P < 0.001) vs. naive group. Four rats in each group. Scale bars: 100 μm in (c), (d), and (e), 20 μm in (d). Aq: aqueduct; CIBP: cancer-induced bone pain; JNK: c-Jun N-terminal kinase; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; mRNA: messenger RNA.

Figure 6.

Cellular localization of pJNK expression in PAG after tumor cell inoculation. High-magnification images showing double immunofluorescent histochemical staining with OX-42 and pJNK (a to c), with NeuN and pJNK (d to f), or with GFAP and pJNK (g to i). Scale bar: 50 μm. Aq: aqueduct; GFAP: glial fibrillary acidic protein; JNK: c-Jun N-terminal kinase.

Figure 7.

Effects of microinjection JNK inhibitor on tumor cell inoculation-induced mechanical allodynia. (a) Tumor cell inoculation induced significant mechanical allodynia as shown by von-Frey tests. Administration of 50 nmol/0.5 μl SP600125 did not change the normal pain threshold of the sham operated group. Injection of (50 nmol/0.5 μl, 10 nmol/0.5 μl SP600125 showed an effective and reliable anti-allodynia effect in a dose-dependent manner on CIBP, whereas, injection of 2 nmol/0.5 μl SP600125 did not influence pain threshold after inoculation at all. Drugs were given once a day from POD 12 to POD 16. ***P < 0.001, compared with that of CIBP+Veh group. ^###P < 0.001, compared with that of CIBP+SP 10 nmol group. There were 10 rats in each group (SP600125:SP). (b) VL-PAG injection of SP600125 reversed upregulation of pJNK in CIBP. ***P < 0.001 vs. Sham+Veh, ^###P < 0.001 vs. CIBP+Veh (n = 4 in each group). CIBP: cancer-induced bone pain; JNK: c-Jun N-terminal kinase; GAPDH: glyceraldehyde 3-phosphate dehydrogenase.

Figure 8.

Suppression of GFAP expression by Intra-PAG JNK inhibitor. (a and b) There was a significant increase in the number of GFAP-IR cells in PAG of CIBP model rats treated with Veh (b) compared with sham rats treated with Veh (a) on day 16. (c) Intra-PAG JNK inhibitor provided a significant decrease of the GFAP positive cells in CIBP rats. (d and e) The number of GFAP immunoreactive cells and Western blot analysis following VL-PAG injection of SP600125 in CIBP rats (n = 4). ***P < 0.001 vs. sham+Veh group; ^###P < 0.001, ^##P < 0.01 vs. CIBP+Veh group. Scale bar = 100 μm. CIBP: cancer-induced bone pain; Aq: aqueduct; GFAP: glial fibrillary acidic protein; GAPDH: glyceraldehyde 3-phosphate dehydrogenase.