Anti-allodynic effect of Buja in a rat model of oxaliplatin-induced peripheral neuropathy via spinal astrocytes and pro-inflammatory cytokines suppression

Young adult male Sprague-Dawley rats (Daehan Biolink, Chungbuk, Korea), weighing approximately 200–220 g, at the beginning of the experimental procedure were used. Animals were housed in cages (3–4 rats per cage), and fed with water and food ad libitum. The room was maintained with a 12 h-light/dark cycle (a light cycle; 08:00–20:00, a dark cycle; 20:00–08:00) and kept at 23 ± 2 °C. All animals were acclimated in their cages for 1 week prior to any experiments. All procedures involving animals were approved by the Institutional Animal Care and Use Committee of Kyung Hee University (KHUASP(SE)-15-088) and were conducted in accordance with the guidelines of the International Association for the Study of Pain [31].

Oxaliplatin (Sigma-Aldrich, St Louis, MO, USA) was delivered at an amount of 6 mg/kg [32, 33], dissolved in 5% glucose (Sigma-Aldrich) solution at a concentration of 2 mg/ml. Oxaliplatin was administered intraperitoneally (i.p.). Control animals received an equivalent volume of 5% glucose solution i.p. as a vehicle.

Buja (Bushi in Japanese; TJ-3023, Tsumura Co. Ltd., Ibaraki, Japan) was obtained by a generous gift from Prof. Schuichi Koizumi (Department of Neuropharmacology, Faculty of medicine, University of Yamanashi). The quality of Buja is strictly controlled by the manufacturer. Buja was suspended and diluted with distilled water (30 mg/ml, 300 mg/kg, approximately 2–2.2 ml/rat) [26]. Our preliminary study using several doses of Buja confirmed that 300 mg/kg was the optimal concentration. Buja was treated orally for 5 consecutive days after an oxaliplatin injection. Equivalent volume of distilled water (DW) was administered to control animals.

Animals were arbitrarily divided into 4 groups: Vehicle + DW, group1; Vehicle + Buja, group2; Oxaliplatin + DW, group3; Oxaliplatin + Buja; group4.

For assessment of oxaliplatin-induced neuropathic pain before and after Buja administraion, cold and mechanical test were performed. Cold allodynia were determined by cold immersion test as previously described [34, 35]. In brief, rats were placed into an acrylic cylinder holder with the tail protruding and were adapted to the testing environment at least 30 min prior to testing. After immersing the tail in 4 °C water, the tail withdrawal latency (TWL) was measured with a 15 s cut-off time. This test was repeated 5 times at 5 min intervals to prevent tissue damage. The average of each latency was used to represent cold allodynia (i.e. the shorter latency was considered the more severe cold allodynia). Mechanical allodynia was evaluated the withdrawal response of tail using a series of von Frey filaments (bending forces to 0.4, 0.6, 1.0, 2.0, 4.0, 6.0, 8.0 and 15.0 g; equivalent in log units: 3.61, 3.84, 4.08, 4.31, 4.56, 4.74, 4.93 and 5.18; Stoelting, IL, USA), as previously described [25]. Using the up-down method, the 50% withdrawal threshold was determined [36]. Rats were immobilized in an acrylic cylinder holder as mentioned above. Test was initiated with a filament having 2.0 g bending force. When withdrawal of the tail was observed during or right after stimulation, this was considered as a positive response. The filament with the next lower bending force was applied after a positive response was observed, whereas the filament with the next higher bending force was applied when there was no response (i.e. negative response). This procedure continued until the sixth von Frey filament stimulation from the first stimulation (2.0 g) or until the third change of response (positive or negative) was observed. The extreme values of series of von Frey filaments were set as a cut-off value. These responses were converted into 50% threshold value using the following formula: 50% threshold = Xf + κδ (Xf is the value of the final von Frey filament [log units], κ is the correction factor from calibration table, and δ is the mean difference of log units between stimuli) [37].

At the end of the experiment (day 5), the L4/L5 segments of the spinal cord were exposed from the lumbar vertebral column via laminectomy and identified by tracing the dorsal roots from their respective dorsal root ganglia (DRG). Animals were perfused using 0.1 M phosphate buffered saline (PBS), followed by 4% paraformaldehyde (BBC Biochemical, WA, USA). Sampled tissues were post-fixed in 4% paraformaldehyde (BBC Biochemical) for 24 h at 4 °C, and then permeated with 30% sucrose (Sigma-Aldrich) in 0.1 M PBS for 48 h at 4 °C. Lumbar spinal cord segments were embedded in optimal cutting temperature (OCT) compound (Sakura Finetek, Tokyo, Japan) on dry ice. Using cryostat (Microm HM 505N; Thermo Scientific, MA, USA), frozen spinal cord segments were cut at a 20 μm thickness. Sections were collected in 0.1 M PBS at 4 °C. These sections were mounted on slide glass (Matsunami, Osaka, Japan) and incubated for 1 h in 0.2% Triton X-100 in 0.5% bovine serum albumin (BSA; BOVOGEN biologics, East Keilor, Australia) solution at room temperature (RT). After rinsing the slide glass with 0.5% BSA solution, double immunostaining using primary antibodies raised in different species was carried out. The sections were incubated overnight at 4 °C with primary antibodies: mouse anti-glial fibrillary acidic protein (GFAP 1:500; Millipore, CA, USA), rabbit anti-Iba-1 (1:500; Wako, Osaka, Japan). After rinsing in 0.5% BSA solution, sections were incubated at RT in dark for 1 h with secondary antibodies: anti-mouse and anti-rabbit-immunoglobulin G (IgG) labeled with Alexa Fluor 488 and Alexa Fluor 546 (1:200; Invitrogen, USA). Confocal laser microscope (LSM 5 Pascal, Zeiss, Oberkochen, Germany) was used to obtain immunofluorescent images. Quantitative analysis of GFAP and Iba-1 positive cells were performed on the spinal dorsal horn images taken through a 20X 0.5NA objective. ImageJ (https://​imagej.​nih.​gov/​ij/​, National Institutes of Health, USA) was used for quantifying the GFAP positive cells and Iba-1 positive cells [22]. To quantify GFAP or Iba-1 positive cells, number of GFAP or Iba-1 positive cells from six lumbar spinal cord section images of each animal were averaged. Six animals were allocated in each group.

To investigate whether Buja administration decreases the quantity of TNF-α or IL-1β in the spinal cord, each cytokine was measured by enzyme linked immunosorbent assay (ELISA). The animals were sacrificed at the end of the experiment. After perfusion with 0.1 M PBS, the lumbar spinal cord segments were obtained as described above. Every collected tissue was stored in 1 ml RIPA buffer (Thermo Scientific) with protease inhibitor cocktail (Roche, Basel, Switzerland). Samples were assayed using a commercial rat TNF (BD OptEIA Set Rat TNF, BD biosciences, CA, USA) and mouse IL-1β (BD OptEIA Set mouse IL-1β, BD biosciences) ELISA kit following the manufacturer’s protocol. In brief, 1:10 dilution of serum was used for the quantification of both cytokines. Microtiter plates were coated overnight at 4 °C with anti-rat TNF or anti-mouse IL-1β monoclonal antibodies (mAbs). Each well was blocked with 10% fetal bovine serum (FBS; Gibco, Thermo Scientific) for 1 h at RT. Samples and standards were loaded after washing out FBS and incubated for 2 h at RT. Biotinylated anti-rat TNF and anti-mouse IL-1β mAbs were added for 1 h at RT. Streptavidin-horseradish peroxidase conjugate was treated and incubated for 30 min. TMB substrate solution (BD Bioscience) was treated for 30 mins, and then Stop solution was added. Washing each well with PBST (PBS with Tween-20; Sigma-Aldrich) was performed between every step. Optical density (O.D.) was measured at 450 nm with λ correction 570 nm. O.D. was measured in a microplate reader (Tecan). Total amount of protein in samples were measured using Bio-Rad protein assay kit (Bio-Rad, CA, USA). All results were normalized to the total amount of protein in each sample.

To investigate whether Buja alleviates oxaliplatin-induced neuropathic pain, cold and mechanical allodynia were assessed using tail immersion test and von Frey hair test, respectively, in randomly divided 4 groups of animals (see Methods). Significant cold allodynia sign (i.e. decreased TWL in response to cold stimuli) was observed since day 3 after an oxaliplatin (6 mg/kg, i.p.) injection (p < 0.05 at D + 3, p < 0.001 at D + 5, group1 vs. group3, Fig. 1a). Daily oral dministration of Buja (300 mg/kg) for 5 consecutive days following an oxaliplatin injection reversed such decrease in TWL to normal level (p < 0.001 at D + 5, group3 vs. group4, Fig. 1a). For mechanical allodynia, an oxaliplatin injection induced a significant decrease in 50% threshold since day 3 (p < 0.001 at D + 3 and D + 5, group1 vs. group3, Fig. 1b). Buja administration also reversed this mechanical allodynia sign to normal level (p < 0.001 at D + 3 and D + 5, group3 vs. group4, Fig. 1b). Buja administration in vehicle-injected rats (group2) showed no effect on behavioral responses to cold and mechanical stimuli (p > 0.05, vs group1, Fig. 1a and b). These results suggest that oral administration of Buja potently inhibits oxaliplatin-induced cold and mechanical allodynia in rats.

To determine whether Buja suppresses glial activation in the dorsal horn of spinal cord after an oxaliplatin injection, activation of spinal glial cells (i.e. astrocytes and microglia) in laminae I-II of the dorsal horn was quantified using immunohistochemical analysis. As shown in Fig. 2, an oxaliplatin injection significantly increased GFAP-positive cells (astrocytes) in the spinal dorsal horn (p < 0.001, group1 vs. group3) and these cells in the group3 exhibited somatic hypertrophy with thick processes (Fig. 2a), which is also represented by enhanced intensity of immunoreactivity (Additional file 1: Figure S2), indicating oxaliplatin-induced activation of spinal astrocytes. Administration of Buja significantly reduced such activation of spinal astrocytes following an oxaliplatin injection (p < 0.001, group3 vs. group4, Fig. 2). Co-immunolabelings of GFAP (astrocytes) and Iba-1 (microglia) positive cells in the same spinal cord samples were performed and these cells were not co-localized (Additional file 2: Figure S3). The number of Iba-I positive cells (microglia) in the spinal dorsal horn is also increased following an oxaliplatin injection (p < 0.001, group1 vs. group3, Additional file 3: Figure S1) and these cells in the group3 showed amoeboid shapes with thick processes (Additional file 3: Figure S1A), which is represented by enhanced intensity of immunoreactivity (Additional file 1: Figure S2). However, Buja administration did not change such microglial activation, i.e. increased Iba-1 positive cells and altered morphology (p > 0.05, group3 vs. group4, Additional file 3: Figure S1). These results suggest that Buja suppresses activation of astrocytes in the spinal dorsal horn following an oxaliplatin injection without affecting microglial activation.

Pro-inflammatory cytokines, IL-1β and TNF-α, were measured with ELISA carried out at the end of the experiment (day 5). As shown in Fig. 3, the levels of IL-1β and TNF-α in the spinal cord were significantly increased after an oxaliplatin injection (p < 0.001 for IL-1β, p < 0.05 for TNF-α, group1 vs. group3). Treatment of Buja reversed this up-regulation of IL-1β and TNF-α in the spinal cord to normal level (p < 0.001 for IL-1β, p < 0.01 for TNF-α, group3 vs. group4). These results suggest that Buja can suppress the oxaliplatin-induced increase in the levels of spinal pro-inflammatory cytokines.