All animal experiments were performed in accordance with the National Institute of Health guidelines on animal care, and were approved by the Institutional Animal Care Committee. All efforts were made to minimize the number of animals used and animal suffering. Adult male Sprague-Dawley rats weighing 290-310 grams (Samtako Bio, Osan, Korea) were randomly and blindly allocated into three groups (n=12 per group). In group 1 (sham), laminectomy was performed. In group 2 (SCI-VPA), the animals received single doses of VPA (Sigma-Aldrich, St. Louis, MO, USA). In group 3 (SCI-saline), animals received 1.0 mL of the saline vehicle solution. Initially, rats were anesthetized intraperitoneally with a mixture of xylazine (10 mg/kg) and ketamine (60 mg/kg). After laminectomy at T9, the extradural plane between the dura and adjacent vertebrae was carefully dissected. A modified aneurysm clip with a closing force of 30 grams (Aesculap, Tuttlingen, Germany) was held in an applicator in the open position. The clip was rapidly released from the applicator and applied vertically onto the exposed spinal cord for a 2-minute compression. For the sham controls the same surgical procedure was followed, but clip compression was not applied. After surgery, the muscle, fascia, and skin were sutured using a 4-0 silk suture. The rectal temperature was maintained at 37.0±0.5℃ by a thermostatically-regulated heating pad during surgery, and during recovery, animals were placed overnight in a temperature- and humidity-controlled chamber. To reduce post-surgery isolation-induced stress, rats were housed in pairs at an ambient temperature of 22-25℃ in an alternating 12-hour light/dark cycle. Bladders were manually emptied twice daily until spontaneous voiding occurred (usually within 7-10 days). A dose of 200 mg/kg of VPA or normal saline as a vehicle control was intraperitoneally injected twice daily at 12 hours intervals for 7 days. The total daily VPA dose of 400 mg/kg/day was similar to doses used in previous studies15,52). To evaluate histological changes, the animals were sacrificed and the spinal cords were collected 2 weeks after SCI.

The rats were tested for functional deficits each week for 2 weeks after the surgery using the open field locomotor rating scale developed by Basso, Beattie, and Bresnahan (the BBB score)3). Two evaluators who were unaware of the group allocations and previous functional scores observed each animal for 1 minute. Functional scores for each hind limb were recorded and averaged.

Six rats in three groups were deeply anesthetized by intraperitoneal injection of ketamine prior to decapitation 14 days after SCI. Following decapitation, a 1.5 cm segment of the spinal cord centered at the injury site was immediately harvested from the vertebral canal and postfixed in 10% formalin overnight. The portion of the spinal cord was divided into seven segments at 2-mm intervals from the lesion epicenter, and seven segments were embedded in paraffin. The segments were (6 mm, 4 mm, and 2 mm rostral to the lesion; lesion epicenter; and 2 mm, 4 mm, and 6 mm caudal to the lesion). Seven spinal cords from each of the two injury groups and the sham group were randomly selected. Representative sections were sliced into 5 µm-thick sections on the horizontal plane and stained with hematoxylin-eosin. For quantitative evaluation of spared tissue and cavity areas, 20 sequential slides of the serial sections were obtained from representative segments. The tissues were examined and photographed using a Zeiss Axioplan microscope (Carl Zeiss Meditec Incorporation, Jena, Germany) with high power differential interference contrast (DIC) optics. The images were viewed on a computer monitor using a Zeiss Plan-Apochromat 5x objective and a Zeiss AxioCam HRc digital camera (Carl Zeiss). The area of cavitation and total spared tissue area of each section were traced and measured using Axio vision 4 software (Carl Zeiss). Due to variable shrinkage of the lesion cavity, we chose to measure the area of remaining white and gray matter, in addition to measuring the area of the lesion itself. Any necrotic tissue within the cavities was counted as part of the lesion. The total cavity volume was calculated by a summation of the measured cavity area of each section multiplied by the intersection distance.

Six rats in three groups were deeply anesthetized by an intraperitoneal injection of ketamin and were perfused intracardially with 4% paraformaldehyde in 0.1 M sodium phosphate buffer (PB, pH=7.4). The thoracic spinal cord was excised, postfixed for 24 hours, and maintained overnight in 30% sucrose in 0.1 M PB. Spinal cord tissues were sectioned at a thickness of 30 µm on a cryostat, and sections were floated on the surface of 0.1 M PB. To detect ED-1 (marker for activated macrophages) and histone acetylation, spinal cord sections were blocked with 4% normal serum in 0.5% Triton X-100 for 1 hour at room temperature and incubated overnight at 4℃ with a 1 : 500 dilution of mouse monoclonal anti-rat ED-1 (Serotec, Oxford, UK) and a 1 : 1000 dilution of polyclonal anti-rat acetyl-histone H3 (K9, Ac-H3/K9; Cell Signaling Technology, Danvers, MA, USA) and a 1 : 1000 dilution of polyclonal anti-rat acetyl-histone H3 (K18, Ac-H3/K18; Cell Signaling Technology), and rinsed for 3×10 min in 0.1 M PB. Sections were then incubated in 0.1 M PB containing 4% normal serum and 0.5% Triton X-100 for 2 hours at 25℃ on a shaker, and then in primary antiserum in 0.1 M PB containing 4% normal serum and 0.5% Triton X-100 for 12 hours at 25℃. After rinsing (3×10 min) in 0.1 M PB, sections were incubated in a 1 : 200 dilution of biotinylated anti-mouse IgG (Sigma-Aldrich) and a 1 : 200 dilution of anti-rabbit IgG (Vector Laboratory, Burlingame, CA, USA) in 0.1 M PB containing 4% normal serum and 0.5% Triton X-100 at 25℃ for 2 hours. The sections were then incubated in a 1 : 50 dilution of avidin-biotinylated horseradish peroxidase (Vector Laboratory) in 0.1 M PB for 2 hours and rinsed (3×10 min) in 0.25 M Tris. Finally, staining was visualized by reaction with 3, 3'-diaminobenzidine tetrahydrochloride (DAB) and hydrogen peroxide in 0.25 M Tris for 3-10 min using a DAB reagent set (Kirkegaard & Perry, Gaithersburg, MD, USA). All the sections were then rinsed in 0.1 M PB and mounted on Superfrost Plus slides (Fisher, Pittsburgh, PA, USA) and dried overnight at 37℃. The mounted sections were then dehydrated with alcohol, cleared with xylene, and coverslipped with Permount mounting medium (Fisher).

The labeled cells were identified and counted with separation of each antibody from two sites at four tissues in six different animals. The labeled tissues were photographed using a Zeiss Axiopan microscope with high power DIC optics (Carl Zeiss). The images were viewed on a computer monitor using a Zeiss Plan-Apochromat 40x objective (Carl Zeiss) and a Zeiss AxioCam HRc digital camera (Carl Zeiss). For comparison, labeled cells were respectively counted in 48 sampled areas in both the gray and white matter (each 250×250 µm field). Enumeration of immune-positive cells used a Labworks, version 4.5, computer-assisted image analyzer (UVP, Upland, CA, USA).

All statistical comparisons were computed using SPSS 17.0 (SPSS, Chicago, IL, USA). Data are expressed as mean±standard error of the mean (SEM). Repeated measure ANOVA was used to compare groups. Significance was accepted for p-values <0.05.

SCI in rats was followed by an injection regimen of VPA or saline (n=12/group). The injured rats were assessed for 2 weeks after surgery according to the open field motor testing using the BBB locomotor rating scale (Fig. 1). While all rats exhibited severe functional impairment the week following SCI, the motor function of the VPA-injected rats was markedly better than their saline-injected counterparts. The average BBB scores (mean±SEM) of the saline-injected rats were 4.33±0.27 on day 1, 5.06±0.21 on day 3, 6.22±0.54 on day 7, 8.44±0.53 on day 10, and 9.83±0.56 on day 4. The corresponding BBB scores of the VPA-injected rats were 4.9±0.19, 5.45±0.19, 8.20±0.47, 10.80±0.42, and 12.10±0.34. A difference in BBB scores at day 7 between the two groups was significant (p<0.05).

Two weeks following SCI, histological examination of the injured spinal cords revealed a central cavity with severe necrosis at the lesion epicenter. The lesions extended to over 4 mm rostrally and caudally, tapering gradually to cavities affecting the central and dorsal areas of the spinal cord gray and white matter (Fig. 2). In VPA-treated rats, the area of the preserved tissue was significantly increased compared to that of the rats that received saline (Fig. 3A). The spared area (µm²) of the spinal cord was 3.64±0.13 at 6 mm rostral section from epicenter, 3.65±0.10 at rostral 4 mm, 3.65±0.07 at rostral 2 mm, 3.57±0.08 at epicenter, 3.71±0.08 at caudal 2 mm, 4.09±0.19 at caudal 4 mm, and 4.43±0.15 at caudal 6 mm in the control rats. In VPA-treated rats, the corresponding values were 3.37±0.05, 3.40±0.70, 3.35±0.08, 2.84±0.16, 3.56±0.19, 3.81±0.11, and 4.19±0.11 µm². In case of the saline-injected groups, the corresponding values were 3.37±0.09, 3.23±0.11, 3.05±0.11, 2.39±0.16, 3.08±0.21, 3.53±0.13, and 4.09±0.13 µm². The cavitation volume was 3.17±0.28 µm³ and 1.83±0.27 µm³ in the saline- and VPA-treated group, respectively. The difference was significant (p<0.05) (Fig. 3B)

SCI causes significant hyperacetylation of histones. However, VPA injection significantly alleviated reduction of Ac-H3/K9, and Ac-H3/K18 at the 6 mm rostral and caudal segments from the lesion sites (Fig. 4, 5). Separate analysis of the gray and white matter revealed a quantitatively similar level of histone acetylation in the same group, but a significantly dissimilarity between the treated and control rats (p<0.05). In addition, decreased immunoreactivity of the ED-1 macrophage marker was evident in the VPA-injected group, while immunoreactivity was pronounced in the control group (Fig. 6, 7). Furthermore, the separate analyses of the gray and white matter revealed no disparity of immunoreactivity, although the number of ED-1 immunoreactive cells was considerably dissimilar in VPA-treated and control rats (p<0.05). Within 4 mm rostral and caudal from the injury site, expression of histone acetylation was too low and ED-1 expression was too great to allow comparison of the two groups. Besides, injured spinal cords had injury-induced cavities <4 mm from each epicenter. Thus, we compared IHC staining at distances further removed from the injury sites (6 mm).