Ethyl pyruvate promotes spinal cord repair by ameliorating the glial microenvironment

Abstract

Background and purpose: Spinal cord injury (SCI) triggers a series of endogenous processes, including neuroinflammation and reactive astrogliosis, which may contribute to the failure of neural regeneration and functional recovery. In the present study, the effect of ethyl pyruvate on spinal cord repair was explored.

Experimental approach: Functional assessment and histological analyses of astrogliosis, neuroinflammation, neuronal survival and axonal regeneration were performed to investigate the effects of ethyl pyruvate (0.086, 0.215, 0.431 or 0.646 mmol·kg(-1) ·day(-1) ) on spinal cord repair in a rat model of SCI. The effect of ethyl pyruvate (5, 10 or 15 mM) on astrocytic activation was also evaluated in an in vitro'scratch-wound' model.

Key results: Functional assessment showed evident improvement of behavioural functions in the ethyl pyruvate-treated rats. Reactive astrogliosis was significantly inhibited in vivo, after injection of ethyl pyruvate (0.431 mmol·kg(-1) day(-1) ), and in vitro'scratch-wound' model in the presence of 10 or 15 mM ethyl pyruvate. The difference between effective concentration in vitro and in vivo suggests that the inhibitory effect of ethyl pyruvate on astrogliosis in damaged spinal cord is indirect. In addition, ethyl pyruvate (0.431 mmol·kg(-1) day(-1) ) attenuated SCI-induced neuroinflammation; it decreased the Iba-1-, ED-1- and CD11b-positive cells at the lesion site. Importantly, histological analyses showed a significantly greater number of surviving neurons and regenerative axons in the ethyl pyruvate-treated rats.

Conclusions and implications: Ethyl pyruvate was shown to inhibit astrogliosis and neuroinflammation, promote neuron survival and neural regeneration, and improve the functional recovery of spinal cord, indicating a potential neuroprotective effect of ethyl pyruvate against SCI.

Figure 1

Functional analyses of hindlimb movements. (A) EP (0.086, 0.215, 0.431 or 0.646 mmol·kg⁻¹day⁻¹) was administered i.p. immediately after SCI, then the locomotor BBB score of rats was assessed (n = 5 for sham-operated group, n = 8 for control group and ethyl pyruvate group, respectively). (B) Analyses of motor function of rats that received ethyl pyruvate (0.431 mmol·kg⁻¹day⁻¹) at 0, 12 or 24 h after SCI. (C) By grid-walk analyses, fewer errors in hindlimb placements were observed in animals treated with ethyl pyruvate at 4 weeks after injury compared with the control groups (n = 5 for sham-operated group, n = 6 for control group and n = 7 for ethyl pyruvate group). (D) Representative footprints of sham, control, and ethyl pyruvate rats (n = 5 for sham-operated group, n = 6 for control group and n = 7 for ethyl pyruvate group) showed improvement in stride length, AR and ILC in the ethyl pyruvate group. (E–G) Quantitative analyses of foot-print revealed that ethyl pyruvate significantly reduced the hindlimb angle of rotation at 3 weeks after SCI and improved stride length and interlimb coordination at 3 and 2 weeks after SCI respectively. *P < 0.05, **P < 0.01 versus control.

Figure 2

Quantitative assessment of reactive astrogliosis in the damaged spinal cord. (A, B) Effects of ethyl pryuvate on astrocytic gliosis were histologically analysed by immunostaining with GFAP 10 days after treatment of SCI rats with ethyl pyruvate (0.431 mmol·kg⁻¹ day⁻¹) (A) or normal saline (B) (n = 6 for control group and ethyl pyruvate group respectively). Insets indicate the reactive astrocyte that was revealed to be hypertrophic under higher magnification. Scale bars: 300 µm in left panel; 100 µm in right panel. (C–E) Quantitative analyses of reactive astrogliosis in the injured spinal cord, including the number, the grey-level value, and the size of GFAP-positive astrocytes close to the lesion site or 1 mm proximal to the lesion site. *P < 0.01 versus control, **P < 0.01 versus control. (F) Western blot analyses of GFAP expression in spinal cord. **P < 0.01 versus injured spinal cord.

Figure 3

Inhibition of reactive astrogliosis by ethyl pyruvate in vitro. (A) The morphological change of astrocytes was observed within the scratch area in presence of ethyl pyruvate (0, 5, 10, 15 mM). Immediately after scratch, ethyl pyruvate at indicated concentration was added to the lesioned cultures of astrocyte. Twenty-four hours later, ethyl pyruvate was shown to inhibit the extensive hypertrophy of cell bodies and cytoplasmic processes of astrocytes proximal to injury; an effect which was dose-dependent. (B, C) Immunocytochemical analysis of GFAP and vimentin expression in scratch injury-induced reactive astrocytes in the absence or presence of ethyl pyruvate (10 or 15 mM). A reactive index of astrocytes, described as a ratio of the average fluorescent intensity of GFAP or vimentin in a narrow zone (150 µm wide) next to the scratch injury to that in another narrow zone (150 µm wide) adjacent to it, was used to quantitatively assess scratch-induced reactive astrogliosis. A reactive index of >1 represents the activation of astrocytes next to scratch injury. The histogram illustrates that ethyl pyruvate significantly inhibited astrogliosis. (D) Effect of ethyl pyruvate on GFAP expression in the scratch injury-reactive astrocytes was analysed by Western blot. (E, F) Ethyl pyruvate inhibited the proliferation of reactive astrocytes in the scratch injury model. Using BrdU labelling, proliferation of astrocytes was examined in the scratch injury model in the absence or presence of 10 mM ethyl pyruvate (E). Insets indicate the BrdU-positive astrocytes under higher magnification. Quantitative assessment of BrdU-positive astrocytes proximate to the lesion induced by scratch (F). **P < 0.01 versus control (0 mM). Scale bar = 50 µm.

Figure 4

Quantitative analyses of the glial scar formation after SCI. (A) SCI animals were treated with ethyl pyruvate (0.431 mmol·kg⁻¹ day⁻¹). Four weeks after SCI, glial scar formation was evaluated by immunostaining for CSPG in horizontal sections through the spinal cord with the lesion epicentre in the middle (n = 4 for sham-operated group, n = 5 for control group and n = 4 for ethyl pyruvate group). Insets indicate the image under higher magnification. (B) Quantitative assessment of the size and immunoreactivity of the CSPG-positive area. **P < 0.01 versus control. Scale bar = 400 µm.

Figure 5

Alleviation of SCI-induced neuroinflammation by ethyl pyruvate at 7 days after injury. (A-C) CD11b, Iba-1 or ED-1 was double stained with GFAP or hoechst at the lesion site of spinal cord after treatment with normal saline (control) or ethyl pyruvate (0.431 mmol·kg⁻¹ day⁻¹). (D-F) SCI-induced neuroinflammation was quantitatively assessed by determining the mean immunoreactive density of CD11b at the lesion site and the number of Iba-1 or ED-1 immunoreactive cells in peri-lesion areas (n = 6 for control group and n = 7 for ethyl pyruvate group). The data of ethyl pyruvate group were normalized to that of control group. Scale bar: 100 µm in (A) and (C); 50 µm in (B). **P < 0.01 versus control.

Figure 6

Effect of ethyl pyruvate on neuron survival in the damaged spinal cord. (A) TUNEL labelling of apoptotic nuclei adjacent to the spinal cord hemisection site 24 h after treatment of animals with normal saline (control) or ethyl pyruvate (0.431 mmol·kg⁻¹day⁻¹). Neurons were identified by staining for NeuN. Insets indicate the apoptotic neurons (NeuN⁺/TUNEL⁺) under higher magnification. (B) The number of NeuN⁺/TUNEL⁺ cells was normalized to the total number of NeuN⁺ cells (apoptosis index). The bar graph indicates that ethyl pyruvate significantly inhibited the apoptosis of neurons in the damaged spinal cord (n = 4 for control group and ethyl pyruvate group respectively). Scale bar = 75 µm; **P < 0.01 versus control.

Figure 7

BDA tracing for corticospinal fibres in the animals 8 weeks after SCI. (A) Schematic diagram of BDA anterograde tracing. Hemisection was made at T8 on the right half of the spinal cord, and BDA was injected into the sensorimotor cortex to label endogenous CST. (B–D) Representative photomicrograph of BDA-labelled anterograde corticospinal fibres in horizontal spinal cord section 1 mm caudal to the lesion site. Arrowheads indicate BDA-labelled axons. In the ethyl pyruvate group, SCI rats were injected with 0.431 mmol·kg⁻¹day⁻¹. (E) Quantification of numbers of regenerating BDA⁺ axons in right (ipsilateral to the lesion site) versus left (contralateral to lesion site) half spinal cord among animal groups (n = 4 for sham-operated group, n = 6 for control group and n = 5 for ethyl pyruvate group). Scale bars = 150 µm. **P < 0.01 versus control.