Effects of valproic acid, a histone deacetylase inhibitor, on improvement of locomotor function in rat spinal cord injury based on epigenetic science

Abstract

Background: The primary phase of traumatic spinal cord injury (SCI) starts by a complex local inflammatory reaction such as secretion of pro-inflammatory cytokines from microglia and injured cells that substantially contribute to exacerbating pathogenic events in secondary phase. Valproic acid (VPA) is a histone deacetylase inhibitor. Acetylation of histones is critical to cellular inflammatory and repair processes.

Methods: In this study, rats were randomly assigned to five experimental groups (laminectomy, untreated, and three VPA-treated groups). For SCI, severe contusion was used. In treated groups, VPA was administered intraperitoneally at doses of 100, 200 and 400 mg/kg daily three hours after injury for 7 days. To compare locomotor improvement among experimental groups, behavioral assessments were performed by the Basso, Beattie and Bresnahan (BBB) rating scale. The expression of neurotrophins was evaluated by RT-PCR and real-time PCR.

Results: VPA administration increased regional brain-derived neurotrophic factor and glial cell-derived neurotrophic factor mRNA levels. Local inflammation and the expression of the lysosomal marker ED1 by activated macrophages/microglial cells were reduced by VPA and immunoreactivity of acetylated histone and microtubule-associated protein were increased.

Conclusion: The results showed a reduction in the development of secondary damage in rat spinal cord trauma with an improvement in the open field test (BBB scale) with rapid recovery.

Keywords: Inflammation; Epigenetics; Valproic acid.

Fig. 1

Time course of open field Basso-Beattie Bresnahan (BBB) locomotor scores. The BBB scores were obtained starting from day 3 post injury until day 28 (5 time points). (A) Post hoc one-way ANOVA at 14, 21, 28 days post spinal cord injury revealed significant differences of 200 and 400 mg/kg VPA-treated groups as compared to 100 mg/kg VPA-treated and untreated-SCI groups. (B) Bar graph shows numerical difference (delta number) of BBB scores between day 3 and day 28 post injury. There are significant differences between groups treated with 200 and 400 mg/kg VPA as compared to the other groups. Bar graphs indicate the mean ± SEM (n = 6) for each time point. *P<0.05.

Fig. 2

Product analysis of RT-PCR for glial cell-derived neurotrophic factor (GDNF) and brain-derived neurotrophic factor (BDNF) and mRNA of 400 mg/kg Valproic acid (VPA)-treated (V) and -untreated groups (C) using RT-PCR and Image J software. (A) RT-PCR products of GDNF (122 bp). (B) RT-PCR products of BDNF (157 bp). Glyceraldehyde-3-phosphate dehydrogenase (496 bp) was used as a positive control of RNA integrity. The brightness of the product bands were measured using Image J software and results showed increasing thickness of product bands of both genes in 400 mg/kg VPA-treated group compared to the untreated spinal cord injury group. In negative controls (V) (without cDNA), no amplification products were seen. For sizing of fragments, a ladder of 200 bp and multiples was used.

Fig. 3

Fold change ratio of glial cell-derived neurotrophic factor (GDNF) and brain-derived neurotrophic factor (BDNF) mRNA. Real-time PCR results are presented as relative expression normalized to B2M mRNA amplification. (A) Amplification of BDNF and GDNF mRNA derived from 400 mg/kg VPA-treated and -untreated groups 28 days post injury showing increased levels of both mRNA after 400 mg/kg VPA treatment. The results demonstrate significant differences between BDNF and GDNF mRNA levels. (B) Total RNA analyzed on a 1.5% denaturing agarose gel. The 18S and 28S ribosomal RNA bands were clearly visible in the intact RNA sample. (C and D) The amplification plot and melting curves for both amplification products. Bar graphs indicate the mean ± SEM. *P<0.05.

Fig. 4

Quantification of cell density and cavity percentage in gray and white matter 28 days post injury. (A and B) Means of cavity percentage (A) and cell density (B) in the 3000 µm length of injured spinal cord. Significant differences between 400 mg/kg VPA-treated and -untreated SCI groups, but not between the sham control and 400 mg/kg VPA-treated group. (C and D) Representative photomicrographs and line graphs showing the lesion extending from the epicenter rostrally end caudally. A massive destruction of gray matter and tissue loss at the epicenter in the untreated SCI group was seen in comparison to 400 mg/kg VPA-treated group. From epicenter to caudally and rostrally increase of tissue density and decrease of cavity percentage is seen. 4x magnification; Bar graphs indicate mean ± SEM; *P<0.05 (untreated group compared to sham control and 400 mg/kg VPA groups).

Fig. 5

Quantification of cell percentage in gray and white matter 28 days post injury. (A) Graph shows a significant increase of glial cells and decrease of neuronal cell percentages in untreated SCI and 400 mg/kg VPA-treated groups compared to the sham control group. (B, C and D) Representative hematoxylin/eosin stained photomicrographs showing the cell population in ventral horn of the spinal cord in VPA 400 mg/kg treated (B), sham control (C) and untreated (D) groups. Scale bar 50 and 200 µm; Bar graphs indicate the mean ± SEM; *P<0.05 (compared to 400 mg/kg VPA groups); **P<0.05 (compared to sham control group).

Fig. 6

Quantification of ED1 immunoreactivity 28 days post injury. (A and B) This graph shows the mean of positive pixel percentage at the lesion (above the central canal) (A) and in contralateral lesion side (below the central canal) (B). A significant difference between 400 mg/kg VPA-treated group compared to the untreated SCI and sham control groups was observed. (C) A significant difference between the mean sizes of the lesion area was seen for all groups and all regions. (D and E) Representative photomicrographs of ED1 immunoreactivity at the lesion side and contralateral of untreated SCI (D) and 400 mg/kg VPA-treated groups (E). Graphs indicate the mean ± SEM; **P<0.05 (compared to the sham control and untreated SCI groups); *P<0.05 (compared to the sham control and VPA 400 mg/kg groups).

Fig. 7

Quantification of MAP2 immunoreactivity 28 days post injury. Representative photomicrographs showing MAP-2 immunoreactivity in ventral and dorsal horns of the spinal cord of sham control (A), VPA 400 mg/kg treated (B) and untreated SCI (C) groups. The brown pixels are immunopositive and the blue pixels are hematoxylin staining. (D) Percentage of MAP2 positive pixels. A significant difference between 400 mg/kg VPA treated, untreated SCI and sham control groups were observed. Scale bar 50 and 200 µm; Graphs indicate the mean ± SEM; **P<0.05 (compared to the sham control and 400 mg/kg VPA groups); *P<0.05 (compared to sham control and untreated SCI groups).

Fig. 8

Quantification of acetylated histone (H3) immunoreactivity 28 days post injury. Representative photomicrographs showing H3 immunoreactivity in ventral and dorsal horn spinal cord areas of sham control (A), VPA 400 mg/kg treated (B) and untreated SCI groups (C). The brown nuclei are immunopositive and are counterstainesd by hematoxylin (blue). (D) Graph showing the percentage of H3 positive cells. There is no significant difference between VPA treated and sham control groups, but there is significant difference between the 400 mg/kg VPA treated and the untreated SCI group. Scale bar 50 and 200 µm; Graphs show mean ± SEM; *P<0.05 (compared to the sham control and 400 mg/kg VPA treatment groups).