Combination of epidural electrical stimulation with ex vivo triple gene therapy for spinal cord injury: a proof of principle study

Abstract

Despite emerging contemporary biotechnological methods such as gene- and stem cell-based therapy, there are no clinically established therapeutic strategies for neural regeneration after spinal cord injury. Our previous studies have demonstrated that transplantation of genetically engineered human umbilical cord blood mononuclear cells producing three recombinant therapeutic molecules, including vascular endothelial growth factor (VEGF), glial cell-line derived neurotrophic factor (GDNF), and neural cell adhesion molecule (NCAM) can improve morpho-functional recovery of injured spinal cord in rats and mini-pigs. To investigate the efficacy of human umbilical cord blood mononuclear cells-mediated triple-gene therapy combined with epidural electrical stimulation in the treatment of spinal cord injury, in this study, rats with moderate spinal cord contusion injury were intrathecally infused with human umbilical cord blood mononuclear cells expressing recombinant genes VEGF165, GDNF, NCAM1 at 4 hours after spinal cord injury. Three days after injury, epidural stimulations were given simultaneously above the lesion site at C5 (to stimulate the cervical network related to forelimb functions) and below the lesion site at L2 (to activate the central pattern generators) every other day for 4 weeks. Rats subjected to the combined treatment showed a limited functional improvement of the knee joint, high preservation of muscle fiber area in tibialis anterior muscle and increased H/M ratio in gastrocnemius muscle 30 days after spinal cord injury. However, beneficial cellular outcomes such as reduced apoptosis and increased sparing of the gray and white matters, and enhanced expression of heat shock and synaptic proteins were found in rats with spinal cord injury subjected to the combined epidural electrical stimulation with gene therapy. This study presents the first proof of principle study of combination of the multisite epidural electrical stimulation with ex vivo triple gene therapy (VEGF, GDNF and NCAM) for treatment of spinal cord injury in rat models. The animal protocols were approved by the Kazan State Medical University Animal Care and Use Committee (approval No. 2.20.02.18) on February 20, 2018.

Keywords: adenoviral vector; epidural electrical stimulation; gene therapy; glial cell-line derived neurotrophic factor; human umbilical cord blood mononuclear cell; neural cell adhesion molecule; spinal cord injury; vascular endothelial growth factor.

Figure 1

Experimental design. (A) Preparation of gene engendered human umbilical cord blood mononuclear cells (UCBMCs) expressing recombinant VEGF, GDNF, and NCAM. (B) Spinal cord injury (SCI), intrathecal injection of UCBMCs + Ad5-VEGF/GDNF/NCAM 4 hours after surgery, epidural electrical stimulation at C5 and L2 vertebral levels 3 days after SCI. (C) For histological study, spinal cord fragment (30 mm) was divided in three segments: the injury epicenter (10 mm) in the middle and two segments rostral (10 mm) and caudal (10 mm) to the epicenter. (D) Spinal cord zones used for immunofluorescence staining and morphometric analysis. Seven spinal cord areas were selected: ventral horn (VH), ventral corticospinal tract (CST), dorsal root entry zone (DREZ), area of the central canal (CC), dorsal funiculi (DF), ventral funiculi (VF), outer area of the lateral funiculi at the line passing through the central canal (LF). GDNF: Glial cell-line derived neurotrophic factor; NCAM: neural cell adhesion molecule; VEGF: vascular endothelial growth factor.

Figure 2

Locomotor activity of hindlimbs 30 days after spinal cord injury (SCI). Volume of movement in hip (a), knee (b) and ankle (c) joints was estimated by measuring the angles in degrees. Data are visualized using box plots, and expressed as median (interquartile range), *P < 0.05 (Dunn’s test). I group: Intact rats (n = 6); C group: control SCI rats without treatment (n = 6); ES group: SCI rats subjected to epidural electrical stimulation alone (n = 4); GT + ES group: SCI rats subjected to electrical stimulation combined with gene therapy (n = 5).

Figure 3

Evaluation of muscle fiber area in the hindlimb skeletal muscles 30 days after spinal cord injury (SCI). Frozen cross sections of tibialis anterior muscle (A) and medial head of gastrocnemius muscle (B) were stained with hematoxylin and eosin and the mean area of the skeletal muscle fibers was measured. The mean area of the tibialis anterior muscle (C) and gastrocnemius muscle (D) fibers was compared between groups. Data are visualized using box plots and expressed as median (interquartile range), *P < 0.05 (Dunn’s test). I group: Intact rats (n = 6); C group: control SCI rats without treatment (n = 6); ES group: SCI rats subjected to epidural electrical stimulation alone (n = 4); GT + ES group: SCI rats subjected to electrical stimulation combined with gene therapy (n = 5).

Figure 4

Electrophysiological investigation of the gastrocnemius muscle 30 days after spinal cord injury (SCI). M-response of the muscle in control (C; SCI only) group demonstrates several phases (polyphasic M-response). The ratio of maximal sizes of H- and M-responses (H/M ratio) % = [H_max/M_max] × 100. Data are visualized using box plots and expressed as median (interquartile range), *P < 0.05 (Dunn’s test).I group: Intact rats (n = 6); C group: control SCI rats without treatment (n = 6); ES group: SCI rats subjected to epidural electrical stimulation alone (n = 4); GT + ES group: SCI rats subjected to electrical stimulation combined with gene therapy (n = 5).

Figure 5

Sparing of spinal cord gray matter 30 days after spinal cord injury (SCI). (A) Cross section of a caudal segment of the spinal cord stained with hematoxylin and eosin from the GT + ES group. Volume of the cysts and total area of gray matter were measured and converted into the percent data. (B) Comparative analysis of the gray matter preservation in experimental groups. Data are visualized using box plots, and expressed as median (interquartile range), *P < 0.05 (Dunn’s test). C group: Control SCI rats without treatment (n = 6); ES group: SCI rats subjected to epidural electrical stimulation alone (n = 4); GT + ES group: SCI rats subjected to electrical stimulation combined with gene therapy (n = 5).

Figure 6

Sparing of spinal cord white matter 30 days after spinal cord injury (SCI). The average numbers of the myelinated fibers were counted in the ventral, lateral, and dorsal columns at the site of contusion injury and 5 mm caudal to the lesion epicenter. (A) Semi-thin transverse sections of the spinal cords from epicenter and caudal part stained with methylene blue dye. Presented is a white matter (ventral funiculus) of a rat from the GT + ES group. (B) Comparative analysis of the white matter preservation in the experimental groups. Data are expressed as mean ± SEM. C group: Control SCI rats without treatment (n = 6); ES group: SCI rats subjected to epidural electrical stimulation alone (n = 4); GT + ES group: SCI rats subjected to electrical stimulation combined with gene therapy (n = 5).

Figure 7

Immunoexpression of Hsp27 and Caspase 3 in spinal cord 30 days after spinal cord injury (SCI). (A) Immunofluorescence staining with antibodies against Hsp27 (red). Arrows indicate nuclei. (B) Fluorescence density value of Hsp27 immunoreactivity. Data are visualized using box plots and expressed as median (interquartile range), *P < 0.05 (Dunn’s test). (C) Immunofluorescence staining with antibodies against Caspase 3 (red). Arrows indicate the apoptotic cells. Nuclei (blue) were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). I group: Intact rats (n = 6); C group: control SCI rats without treatment (n = 6); ES group: SCI rats subjected to epidural electrical stimulation alone (n = 4); GT + ES group: SCI rats subjected to electrical stimulation combined with gene therapy (n = 5).

Figure 8

Immunoexpression of synaptic proteins synaptophysin and PSD95 in spinal cord ventral horns (VH) 30 days after spinal cord injury (SCI). Upper panel: Immunofluorescence staining with antibodies against synaptophysin (red) and PSD95 (green). Nuclei (blue) were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Lower panel: Fluorescence density value of synaptophysin and PSD95. Arrows indicate nuclei. Data are visualized using box plots and expressed as median (interquartile range), *P < 0.05 (Dunn’s test). I group: Intact rats (n = 6); C group: control SCI rats without treatment (n = 6); ES group: SCI rats subjected to epidural electrical stimulation alone (n = 4); GT + ES group: SCI rats subjected to electrical stimulation combined with gene therapy (n = 5).

Figure 9

Recombinant GDNF, VEGF, and NCAM produced by gene-modified UCBMCs in vitro. Enzyme-linked immunosorbent assay was used to analyze soluble human VEGF, GDNF, and NCAM in the conditioned culture media. The level of recombinant molecules was estimated in supernatant 72 hours after incubation of naïve UCBMCs, UCBMCs transduced with Ad5 carrying gene encoding reporter green fluorescent protein (UCBMCs + Ad5-GFP) and UCBMCs simultaneously transduced with Ad5-VEGF, Ad5-GDNF, and Ad5-NCAM (UCBMCs + Ad5-VEGF/GDNF/NCAM). Data are presented as the mean ± SEM. GDNF: Glial cell-line derived neurotrophic factor; GFP: green fluorescent protein; NCAM: neural cell adhesion molecule; UCBMCs: umbilical cord blood mononuclear cells; VEGF: vascular endothelial growth factor.

Figure 10

Immunoexpression of recombinant GDNF, VEGF, and NCAM in gene-modified UCBMCs in vivo. Double immunofluorescence staining of UCBMCs in injured spinal cord 30 days after intrathecal infusion. Anti-HNA antibodies were used for identification of UCBMCs. Human recombinant molecules were identified with antibodies against human specific VEGF, GDNF, and NCAM. (A) HNA-positive cells (red) expressing VEGF (green). (B) HNA-positive cells (red) expressing GDNF (green). (C) HNA-positive cells (red) expressing NCAM (green). Nuclei (blue) were counterstained with DAPI. Arrows indicate UCBMCs. DAPI: 4′,6-Diamidino-2-phenylindole; GDNF: glial cell-line derived neurotrophic factor; HNA: human nuclear antigen; NCAM: neural cell adhesion molecule; UCBMCs: umbilical cord blood mononuclear cells; VEGF: vascular endothelial growth factor.