Neuroprotective Effects of Direct Intrathecal Administration of Granulocyte Colony-Stimulating Factor in Rats with Spinal Cord Injury

Abstract

Aims: To date, no reliable methods have proven effective for treating spinal cord injury (SCI). Even systemic administration of methylprednisolone (MP) remains controversial. We previously reported that intrathecal (i.t.) administration of granulocyte colony-stimulating factor (G-CSF) improves outcome after experimental spinal cord ischemic insults in rats. The present study aimed to examine the neuroprotective efficacy of i.t. G-CSF or MP in rats with SCI.

Methods: Female rats were subjected to spinal cord contusion injury at T10 using NYU impactor. We i.t. administered G-CSF (10 μg) or MP (one bolus of 100 μg, followed by 18 μg/h infusion for 23 h) immediately after SCI.

Results: Both G-CSF and MP significantly improved the rats' motor function after SCI. Immunofluorescence staining revealed suppressed expression of transforming growth factor-beta 1 (TGF-β1), chondroitin sulfate proteoglycans (neurocan and phosphacan), OX-42 and tumor necrosis factor alpha after i.t. G-CSF, but not MP, in rats with SCI. In addition, G-CSF significantly decreased the expression of astrocytic TGF-β1 and glial fibrillary acidic protein around the injury site. Furthermore, rats with G-CSF treatment showed increased neurofilament expression beyond the glial scars.

Conclusion: Direct i.t. administration of G-CSF provides a promising therapeutic option for SCI or related spinal diseases.

Keywords: G-CSF; Intrathecal; Methylprednisolone; Spinal cord injury; Transforming growth factor.

Figure 1

Administration of i.t. G‐CSF or MP improves locomotor functions after SCI. A significant improvement in hind limb motor function was observed in the G‐CSF‐treated group compared with the SCI control or SCI + MP groups after SCI. The AUC (bar chart) presents the evolution of the BBB score over time for each group. Both treatments showed significantly improved recovery chronically after SCI. *P < 0.05, compared with the SCI control group; ^# P < 0.05, compared with the SCI + MP group.

Figure 2

Immunohistochemistry and IR analyses of the effects of i.t. G‐CSF on spinal TGF‐β1 expression in rats with SCI. (A) Spinal cord sections from the sham, SCI control, SCI + MP, and SCI + G‐CSF groups were harvested at 6 h, 3 days, and 7 days after SCI and incubated with anti‐TGF‐β1 antibody. These results show SCI‐induced upregulation of spinal TGF‐β1 IR at 6 h, 3 days, and 7 days after injury. The i.t. G‐CSF clearly inhibited the SCI‐induced TGF‐β1 upregulation. (B) Quantification of the IR of TGF‐β1 was expressed as the fold change compared to control animals (at each time point), which were considered to be 1. Administration of i.t. G‐CSF significantly attenuated the SCI‐induced upregulation of TGF‐β1 at 6 h, 3 days, and 7 days after injury. (C) Double immunofluorescence of spinal TGF‐β1 (red) plus the astrocyte‐specific marker GFAP (green) in the SCI and G‐CSF groups at 3 days after SCI. Confocal images showed that astrocytes were colocalized with TGF‐β1 in the SCI control group. G‐CSF attenuated the SCI‐induced TGF‐β1 expression in astrocytes. ⁺ P < 0.05, compared with the sham group; *P < 0.05, compared with the SCI control group; ^# P < 0.05, compared with the SCI + MP group. Scale bar: 50 μm.

Figure 3

The effect of i.t. G‐CSF on the expression of spinal CSPGs in SCI rats. The spinal cord tissues were immunofluorescence stained for neurocan (A) and phosphacan (B) at 3 days, 7 days, and 30 days after SCI. The results show SCI‐induced upregulation of spinal neurocan and phosphacan IR from 3 to 30 days after injury. The i.t. G‐CSF inhibited both neurocan and phosphacan upregulation at 7 days and 30 days after SCI. Quantification of the IR of neurocan and phosphacan is expressed as the fold change compared to control animals (at each time point), which was considered to be 1. Administration of i.t. G‐CSF significantly attenuated the SCI‐induced upregulation of neurocan and phosphacan at 6 and 30 days after injury. ▵ indicates the lesion area. ⁺ P < 0.05, compared with the sham group; *P < 0.05, compared with the SCI control group; ^# P < 0.05, compared with the SCI + MP group. Scale bar: 200 μm.

Figure 4

The effect of i.t. G‐CSF on neurofilaments (NFs) in the lesion core at 30 days after SCI. Double immunostaining showing the relationship between the astrocyte‐specific marker GFAP (green) and the axon marker NF (red) expression at the lesion epicenter in the SCI control, SCI + MP, and SCI + G‐CSF groups. Quantification of the same area of IR of NFs is expressed as the fold change compared to SCI animals, which was considered to be 100%. Administration of i.t. G‐CSF, but not MP, significantly attenuated the SCI‐induced downregulation of NF IR. *P < 0.05, compared with the SCI control group; ^# P < 0.05, compared with the SCI + MP group. Scale bar: 500 μm.

Figure 5

Immunohistochemistry and IR analyses of the effect of i.t. G‐CSF on astrocyte and microglial cell activation after SCI. Spinal cord sections from the sham, SCI control, SCI + MP, and SCI + G‐CSF groups were harvested at 3, 7, and 14 days after SCI and incubated with antibodies for the astrocyte‐specific marker GFAP (A) or the microglial cell‐specific maker OX‐42 (B). Immunohistochemistry for proliferating cell nuclear antigen (PCNA) and vimentin was performed to evaluate gliosis at 7 days after SCI, and the images were captured by confocal microscopy. The quantification was performed from nonconfocal immunofluorescence images (C). The results show SCI‐induced upregulation of spinal GFAP and OX‐42 IR from 3 to 14 days after injury. The i.t. G‐CSF clearly enhanced GFAP and inhibited OX‐42 expression after SCI. Quantification of the IR of GFAP and OX‐42 is expressed as the fold change compared to control animals (at each time point), which was considered to be 100%. G‐CSF significantly upregulated GFAP and downregulated OX‐42 expression after SCI. ⁺ P < 0.05, compared with the sham group; *P < 0.05, compared with the SCI control group; ^# P < 0.05, compared with the SCI + MP group. Scale bar: 200 μm; 50 μm in 30 days; 100 μm in C.

Figure 6

The effect of G‐CSF on the upregulation of spinal TNF‐α expression after SCI. The spinal cord tissues were immunostained for TNF‐α at 3, 7, and 14 days after SCI. The expression of TNF‐α IR was increased from 3 to 14 days in the SCI and MP groups, as compared to the TNF‐α IR of the sham or SCI + G‐CSF groups. Quantification of the TNF‐α IR is expressed as the fold change compared to control animals (at each time point), which was considered to be 1. G‐CSF, but not MP, significantly attenuated the SCI‐induced upregulation of TNF‐α from 3 to 14 days after injury. ⁺ P < 0.05, compared with the sham group; *P < 0.05, compared with the SCI control group; ^# P < 0.05, compared with the SCI + MP group. Scale bar: 50 μm.