Adult male Sprague Dawley [Sam:TacN (SD) BR; Samtako, Osan, Korea] rats were subjected to moderate contusion injury (10 g×25 mm) as described previously (Yune et al., 2007). For the sham-operated controls, the animals underwent a T10 laminectomy without weight-drop injury. Surgical interventions and postoperative animal care were performed in accordance with the Guidelines and Policies for Rodent Survival Surgery provided by the Animal Care Committee of the Kyung Hee University.

The dried roots of Bupleurum falcatum were extracted with 70% ethanol as described previously (Kim et al., 2001). The ethanol filtrate was evaporated in vacuo, and powdered BF was stored at -20℃ until use. By HPLC analysis, BF contains saikosaponin A, B1, B2, B3, B4, C, D, G, H and I.

Powdered BF were suspended in sterile deionized water and administrated orally at a dose of (100 mg/kg) beginning at 2 h after SCI and then once a day for 14 d.

RNA isolation using TRIZOL Reagent (Invitrogen, Carlsbad, CA) and cDNA synthesis were performed as described previously (Lee et al., 2010). The primers used for MMP-2, MMP-9, iNOS, COX-2, TNF-α, IL-1β and GAPDH were synthesized by the Genotech Corp (Daejeon, Korea). Details of primers used for PCR in this study are shown in Table 1. GAPDH was used as an internal control. Experiments were repeated three times and the values obtained for the relative intensity were subjected to statistical analysis. The gels shown in figures are representatives of results from three separate experiments.

Total protein extracts from spinal cord samples at 1 d after injury were prepared as described previously (Yune et al., 2007). Spinal cord tissues (1 cm) were homogenized in a lysis buffer and the tissue homogenates were incubated for 20 min at 4℃, and centrifuged at 25,000×g for 30 min at 4℃. The protein level of the supernatant was determined using the BCA assay (Pierce, Rockford, IL). Gelatin zymography was performed by using 10% zymogram gel (Invitrogen) as previously described (Choi et al., 2010). In brief, 50µg proteins were mixed with sample buffer and loaded without boiling. After electrophoresis, the gel were soaked in 2.5% Triton X-100, rinsed, and incubated for 24 h at 37℃ in developing buffer (50 mM Tris-HCl, pH 8.5, 0.2 M NaCl, 5 mM CaCl₂, 0.02% Brii35). After incubation, gels were stained with 0.5% Coomassie blue and then destained. Clear bands on the zymogram were indicative of gelatinase activity.

Examination of functional deficits after injury was conducted by using the 21-point Basso-Beattie-Bresnahan (BBB) locomotion scale, inclined plane test and foot print analysis as previously described (Yune et al., 2007).

After SCI, spinal cord sections were prepared as described previously (Yune et al., 2007). In brief, animals were perfused via cardiac puncture with 4% paraformaldehyde. A 20-mm section of the spinal cord, centered at the lesion site, was dissected out, post-fixed and placed in 30% sucrose. The segment was embedded in OCT for frozen sections, and longitudinal or transverse sections were then cut at 10 or 20µm on a cryostat (CM1850; Leica, Germany).

One day after injury, serial spinal cord sections (20µm thickness) were collected and every 100 µm section was processed for TUNEL as previously described (Yune et al., 2007). Only those cells showing morphological features of nuclear condensation and/or compartmentalization only in the gray matter (GM) were counted as a TUNEL-positive. Quantitation of TUNEL-positive cells was accomplished by counting the number of cells labeled positively as described previously (Yune et al., 2009).

Lesion volume, using rats employed for behavioral analyses, was assessed as described previously (Yune et al., 2008). In brief, serial longitudinal sections (10µm) through the dorsoventral axis of the spinal cord were used to determine lesion volume. Every 50µm, sections were stained with Cresyl violet acetate and studied with light microscopy. Areas at each longitudinal level were determined, and the total lesion volume derived by means of numerical integration of sequential areas.

Data presented as the mean±SD values. Comparisons between vehicle and BF treated groups were made by unpaired Student's t test. Behavioral scores from BBB analysis and inclined plane tests were analyzed by repeated measures ANOVA (time vs treatment). Tukey's multiple comparison was used as Post hoc analysis. Statistical significance was accepted with p<0.05. All statistical analyses were performed by using SPSS 15.0 (SPSS Science, Chicago, IL).

It has been known that saikosaponins modulates the gene expression or activation of MMP-2 in endothelial cells and spleen cells from picryl chloride-induced ear contact sensitivity mice (Shyu et al., 2004; Zhang et al., 2006). To test the hypothesis whether BF may inhibit MMP-2 and MMP-9 activation after SCI, we first examined the effect of BF on the expression of MMP-2 and MMP-9 by RT-PCR at 1 d after injury. A basal level of MMP-2 expression was detected in sham control. A significant upregulation of MMP-2 expression was observed at 1 d after SCI and BF treatment significantly reduced the MMP-2 mRNA expression (Fig. 1A, B). Similarly, MMP-9 mRNA was markedly increased at 1 d after SCI and BF treatment also significantly attenuated the MMP-9 mRNA expression (Fig. 1A, B). However, MMP-9 expression was not observed in the sham control. Next, we examined the enzyme activities of MMP-2 and MMP-9 by gelatin zymography using spinal cord total extract from 1 d after injury. As shown in Figure 1C, active form of MMP-2 and MMP-9 was strongly induced at 1 d after injury. Pro-MMP-2 and pro-MMP-9 were also induced by SCI. Furthermore, BF treatment significantly reduced the degree of activation of MMP-2 and MMP-9 when compared to vehicle-treated control (e.g., active MMP-2; BF 1.66±0.08 vs vehicle 3.22±0.22, p<0.05/active MMP-9; BF 5.3±0.11 vs vehicle 7.5±0.25, p<0.05) (Fig. 1B, D).

It has been known that BBB disruption by MMPs including MMP-2 and MMP-9 induces blood infiltration and thereby contributes to inflammation after SCI. Since BF reduced the expression and activity of MMP-2 and MMP-9 after SCI (Fig. 1), we hypothesized that BF would inhibit the expression of inflammatory factors after SCI. As shown in Fig. 2A, TNF-α, IL-1β, COX-2 and iNOS mRNA expression were markedly increased after SCI as reported (Lee et al., 2004; Yune et al., 2009). In particular, BF treatment significantly alleviated the expression levels of pro-inflammatory cytokines and mediators when compared with those of vehicletreated control at 1 d after injury (Fig. 2).

Trauma to the spinal cord results in extensive apoptotic cell death (Liu et al., 1997; Lee et al., 2003). Based on the anti-inflammatory effects of BF in the injured spinal cord (Fig. 2), we hypothesized that BF would inhibit apoptotic cell death after SCI. Many TUNEL-positive cells were seen mostly within the lesion area in the gray matter (GM) at 1 d after injury. As shown in Fig. 3, BF treatment significantly decreased the number of TUNEL-positive cells when compared to the vehicle-treated control (BF 132±9 vs. Vehicle 205±17 cells, p<0.05) (Fig. 3B). Thus, our results indicate that BF inhibits apoptotic cell death after injury.

Previous report showed that pharmacological inhibition of MMPs at a early stage after SCI improves locomotor recovery (Noble et al., 2002). Since BF treatment inhibited the expression and activity of MMP-2 and MMP-9 and also attenuated apoptotic cell death after injury (Fig. 1 and 3), we expected that BF would improve functional recovery following injury. Injured rats were treated with BF (100 mg/kg) for 14 consecutive days and functional recovery was then evaluated for 35 d using the BBB rating scale, inclined plane test and footprint analysis. The hindlimbs paralyzed immediately after injury were spontaneously recovered within 7 to 14 d after injury (Fig. 4A) as in previous report (Yune et al., 2007). The BBB scores were significantly higher in rats treated with 100 mg/kg of BF by 14 to 35 d after SCI than in vehicle-treated rats (35 d, BF 11.25±0.5 vs Vehicle 9.0±1.0, p<0.05) (Fig. 4A). The angle of incline, determined 1~4 weeks after injury, was also significantly higher in BFtreated rats than in vehicle-treated rats at 4 weeks after SCI (4 weeks, BF 60±2.1 vs Vehicle 53±2.2, p<0.05) (Fig. 4B). Finally, footprint analyses obtained from sham-treated, vehicle-treated and BFtreated rats at 35 d after SCI revealed that BFtreated rats show more consistent forelimb-hindlimb coordination and less toe drag than in vehicle control (Fig. 4C). With the 25 mm insult used in the present study, the hind paws as well as the central pads of vehicle-treated animals were not clearly recorded because of the hindlimb drags (Fig. 4C). Thus, we didn't quantify footprint analysis using toe spread and ipsilateral distances (limb coordination) in this study.

The functional deficit after SCI is known to be correlated with the massive tissue loss (Basso et al., 1996). To examine whether BF preserves spinal cord tissue after injury, we measured the lesion volume using rats tested for behavioral analysis. As shown in Fig. 4D, BF treatment significantly reduced the lesion volume at 38 d after injury than in vehicle-treated rats (BF 4.2±1.2 mm³ vs Vehicle 8.5±0.7 mm³, p<0.05).