Up-regulation of matrix metalloproteinase-3 in the dorsal root ganglion of rats with paclitaxel-induced neuropathy

Abstract

Paclitaxel-induced painful peripheral neuropathy is a major dose-limiting factor. Recently, it has been reported that macrophages accumulated in the dorsal root ganglion of paclitaxel-treated rats, and their activation is suggested to contribute to generation and development of the neuropathy. However, the mechanism for macrophage activation is still unknown. In this study, to explore candidate genes involved in the mechanism for macrophage activation in the dorsal root ganglion of paclitaxel-treated rats, we developed model rats for paclitaxel-induced neuropathic pain and performed a microarray assay to analyze the changes of gene expressions in the dorsal root ganglion. Among the genes with changed expression levels, we focused on matrix metalloproteinase-3 (MMP-3, stromelysin-1) and CD163, a macrophage marker. By reverse transcription-polymerase chain reaction, the expression levels of MMP-3 and CD163 were markedly up-regulated in paclitaxel-treated dorsal root ganglion. As a result of immunohistochemical study, large ganglion neurons, but neither Schwann cells nor macrophages, predominantly expressed MMP-3. This MMP-3 up-regulation occurred prior to macrophage accumulation in the dorsal root ganglion. In addition, recombinant MMP-3 led to the activation of RAW264 macrophages in vitro. Taken together, the up-regulation of MMP-3 and following macrophage activation caused in the dorsal root ganglion might be a significant event to trigger a series of reactions developing paclitaxel-induced peripheral neuropathic pain.

Figure 1

Paclitaxel‐induced mechanical hyperalgesia in rats and expression of activation transcription factor‐3 (ATF3) in lumbar dorsal root ganglions (DRGs). (a) Behavioral analysis was performed with a dynamic plantar aesthesiometer on day 1 and 10. Data show the mean ± SE (n = 5) *P < 0.05, versus control. (b) On day 10 following the dosing schedule of paclitaxel, isolated lumbar DRGs were sectioned and immunostained with the specific antibody against ATF3. Panel (b) shows the quantitative results of ATF3‐positive nuclei. Each column represents the mean ± SE (n = 6–7) **P < 0.01, versus control. (c, d) Microphotographs show representative immunostaining of ATF3 in the nuclei of DRG of either control (c) or paclitaxel‐treated rats (d). Arrowheads indicate ATF3‐positive nuclei.

Figure 2

Expression of matrix metalloproteinase‐3 (MMP‐3), CD163, and arginase‐1 in the lumbar dorsal root ganglion (DRG) of paclitaxel‐treated rats. On day 10 following the dosing schedule of paclitaxel, total RNA was extracted from lumbar DRGs, and the expressions of MMP‐3, CD163, and arginase‐1 were determined by semiquantitative reverse transcription–polymerase chain reaction (a,b). Panel (b) shows the quantitative results after correction by the corresponding glyceraldehyde 3‐phosphate dehydrogenase (GAPDH) mRNA expression shown in panel (a). Each column represents the mean ± SD (n = 3). *P < 0.05, versus control. In panel (c), latent and active forms of MMP‐3 were detected as bands of 57, 45, and 28 kDa, respectively. Panel (d) shows the quantitative results after correction by the corresponding GAPDH protein expression shown in panel (c). Each column represents the mean ± SD (n = 6). **P < 0.01, versus control.

Figure 3

Matrix metalloproteinase‐3 (MMP‐3) immunoreactivity in lumbar dorsal root ganglions (DRGs). On day 10 following the dosing schedule of paclitaxel, longitudinal sections of lumbar DRGs were fluorescently immunostained with anti‐MMP‐3 antibody (a,b) and double‐immunostained in combination with antibodies against MMP‐3 and S100 (c–e). Arrowheads indicate MMP‐3‐positive cells (b). S100 immunoreactivity was detected in round neurons (asterisks, c and e), in spindly Schwann cells (arrows, d and e), and in satellite cells (arrowheads, d and e). Scale bar is 50 µm.

Figure 4

Accumulation and localization of macrophages in lumber dorsal root ganglions (DRGs). On day 6, 8, and 10 following the dosing schedule of paclitaxel, longitudinal sections of lumbar DRGs were fluorescently double‐immunostained in combination with antibodies against matrix metalloproteinase‐3 (MMP‐3) and Iba1, a macrophage marker. Left and middle panels represent representative immunostaining for MMP‐3 (a1, b1, c1, and d1) and Iba1 (a2, b2, c2 and d2), respectively, and right panels show their merged microphotographs (a3, b3, c3, and d3) for at least three independent experiments. MMP‐3 immunoreactivity was detected in large ganglion neurons (arrowheads). Scale bar is 50 µm.

Figure 5

Effect of recombinant matrix metalloproteinase‐3 (MMP‐3) on reactive oxygen species (ROS) generation and cell viability in macrophage‐like cells. After RAW264 cells were incubated in horseradish peroxidase containing Hank's balanced salt solution (HBSS) for 15 min, they were exposed to the indicated concentrations of recombinant MMP‐3 in the same media for 1 h. To inhibit MMP‐3 enzymatic activity and scavenge generated ROS, 60 µM N‐isobutyl‐N‐(4‐methoxyphenylsulfonyl)glycyl hydroxamic acid (NNGH), 1 mM N‐acethyl‐L‐cysteine (NAC) and 20 µM EUK‐8 were added to media during MMP‐3 treatment. Thereafter, the ROS level was determined using fluorescence intensity. Panels (a) and (b) show the dose‐dependent effect of MMP‐3, and the effect of an MMP‐3 inhibitor and antioxidant on ROS generation, respectively, in RAW264 cells, microphotographs shown in the panels being representative ones of at least three independent experiments. Panel (c) shows the effect of MMP‐3 and NAC treatment on the cell viability determined by a 3[4,5‐dimethyl‐2‐thiazolyl]‐2,5‐diphenyl‐2H‐tetrazolium bromide (MTT) assay. After cells were treated with 12.5 ng/mL recombinant MMP‐3 in HBSS for 1 h, they were incubated in fresh Dulbecco's modified Eagle's medium for 4 h, and then MTT assays were performed. Each point represents the mean ± SE (n = 4). Scale bar is 50 µm.