Transgene-mediated GDNF expression enhances synaptic connectivity and GABA transmission to improve functional outcome after spinal cord contusion

Abstract

Glial cell line-derived trophic factor (GDNF) is a peptide with pleiotropic survival and growth-promoting effects on neurons. We found that intraspinal injection of a non-replicating herpes simplex virus-based vector coding for GDNF 2 h after blunt trauma to the thoraco-lumbar spinal cord produced sustained improvement in motor behavioral outcomes up to 5 weeks following injury. The improvement in behavior correlated with an increase in synaptophysin and glutamic acid decarboxylase (GAD) in the spinal cord at the level of injury. Addition of recombinant GDNF protein to primary spinal cord neurons in-vitro resulted in enhanced neurite growth and a marked increase in protein levels of GAD65 and GAD67, synapsin I and synaptophysin. GDNF-mediated increases in GAD and the synaptic markers were blocked by the MEK inhibitor UO126, but not by the phosphoinositide 3-kinase inhibitor LY294002. These results suggest that GDNF, acting through the MEK-ERK pathway enhances axonal sprouting, synaptic connectivity, and GABAergic neurotransmission in the spinal cord, that result in improved behavioral outcomes after spinal cord contusion injury.

Figure 1

vG improves locomotor functioning and tissue sparing after T12-T13 SCI compared to control vector vC. (A) BBB score measured each week for 4 weeks and (B) gridwalk measured at 4 weeks following injury. (C) Spinal cord cavity area was significantly reduced at the site of injury from 0.99 ± 0.22 mm² in vC treated animals to 0.54 ± 0.085 mm² in vG treated animals. Data is presented as mean ± SEM *P < 0.05, **P < 0.01, ***P < 0.005; N = 6 animals per group; scale bar = 500 μm.

Figure 2

Delivery of vG 2 hrs after injury resulted in a substantial increase in synaptophysin at 1 week compared to animals injected with the control vector (vC; P < 0.005). (A) Amount of synaptophysin protein expressed as normalized value using GAPDH as a loading control. (B) Representative Western blot. Data is presented as mean ± SEM. N = 3 animals per group. (C) Photomicrograph from the anterior horn of the spinal cord below the center of the lesion showing the distribution of synaptophysin immunoreactivity as boutons 1 week after vG treatment compared to vC-treated animals; scale bar = 70 μm.

Figure 3

Injection of vG 2 hrs after injury showed a substantial increase in GAD65 and GAD67 protein levels at 1 week compared to animals injected with the control vector (vC; P < 0.005). (A) Amount of GAD65 protein expressed as normalized value using β-actin as a loading control. (B) Representative Western blot. (C) Amount of GAD67 determined by Western blot expressed as normalized value using β-actin as a loading control. (D) Representative Western blot. Data is presented as mean ± SEM. N = 3 animals per group.

Figure 4

GDNF increases neurite length in spinal cord neurons in vitro. Treatment of spinal cord neurons at 3 DIV with GDNF resulted in an increase in the total neurite length (A,C) and in the length of the longest neurites (B,D). These neurites were stained with SMI31 antibody against phosphorylated neurofilament and tau epitopes, indicating axonal processes, and with the anti GAD65 antibody, confirming the GABAergic phenotype. There was no change in the length of neurites measured by MAP2 immunostaining. C = control untreated and G = GDNF (100 ng/ml) for 24 h. Data is expressed as normalized value and presented as mean ± SEM; *P < 0.05, **P < 0.01, ***P < 0.005.

Figure 5

GDNF increases GAD65 and GAD67 in spinal cord neurons. (A) The amount of GAD67 protein expressed as normalized value using β-actin as a loading control (P < 0.005). (B) Representative Western blot. (C) Amount of GAD65 protein expressed as normalized value using β-actin as a loading control (P < 0.005). (D) Representative Western blot. C = untreated control and G= GDNF (100 ng/ml) for 48 h. Treatment with the MEK inhibitor U0126 (U; 10 μM) prevented the increased expression of GAD67 (P < 0.001) and GAD65 (P < 0.01). Data is presented as mean ± SEM.

Figure 6

GDNF increases synaptophysin and synapsin I in primary spinal cord neurons. (A) Amount of synaptophysin protein expressed as normalized value using β-actin as a loading control (P < 0.05). (B) Representative Western blot. (C) Amount of synapsin I protein expressed as normalized value using β-actin as a loading control (P < 0.05). (D) Representative Western blot. C= untreated control and G= GDNF (100 ng/ml) for 48 h. Treatment with the MEK inhibitor U0126 (U; 10 μM) prevented the increased expression of synaptophysin (P < 0.05) and synapsin I (P < 0.05). Data is presented as mean ± SEM. (E) Photomicrograph of primary spinal cord neurons treated with GDNF showing colocalization of synapsin 1 (red) and GAD65 (green) immunoreactivity; scale bar = 70 μm.

Figure 7

vG in vivo and recombinant GDNF in vitro activate the ERK pathway. Animals injected with vG 2 hrs after injury had increased pERK in the spinal cord compared to control vector (vC) at 1 week after vector delivery. Western blot normalized to internal control. (A,B; P < 0.05). Spinal cord neurons at 10 DIV treated with GDNF show increases in pERK (C,D; P < 0.01), pCREB (E,F; P < 0.01), and zif 268 (G,H; P < 0.05). All values are expressed as mean ± SEM and normalized to internal control. C= control untreated and G = GDNF (100 ng/ml) for 48 h. Treatment with the MEK inhibitor U0126 (U; 10 μM) prevented the increase in pERK (C,D; P < 0.01), pCREB (E,F; P < 0.05) and zif 268 (G,H; P < 0.05).

Figure 8

Schematic representation of the zif 268 pathway in spinal cord neurons. Activation of receptor tyrosine kinase by GDNF activates the MEK/ERK pathway leading to phosphorylation of the transcription factor CREB, which translocates to the nucleus and promotes the expression of zif 268. Acting as a transcription factor, zif 268 regulates the expression of late response genes, such as glutamic acid decarboxylase (GAD), synapsins and neurofilament 68.