Timed sulfonylurea modulation improves locomotor and sensory dysfunction following spinal cord injury

Abstract

Traumatic spinal cord injury (SCI) results in immediate tissue necrosis and delayed secondary expansion of neurological damage, often resulting in lifelong paralysis, neurosensory dysfunction, and chronic pain. Progressive hemorrhagic necrosis (PHN) and excessive excitation are the main sources of secondary neural injury. Recent approaches to attenuate PHN by glibenclamide can improve locomotor function after SCI. However, use of glibenclamide can exacerbate development of SCI-induced chronic pain by inhibiting K_ATP channels to increase neuronal excitation and glial activation. In this study, we explored a treatment strategy involving administration of glibenclamide, which suppresses PHN, and diazoxide, which protects against neuronal excitation and inflammation, at different time intervals following spinal cord contusion. Our goal was to determine whether this combined approach enhances both sensory and motor function. Contusive SCI was induced at spinal segment T10 in adult rats. We found that K_ATP channels opener, diazoxide, decreased the hyperexcitability of primary sensory neurons after SCI by electrophysiology. Timed application of glibenclamide and diazoxide following contusion significantly improved locomotor function and mitigated development of SCI-induced chronic pain, as shown by behavioral evidence. Finally, we found that timed application of glibenclamide and diazoxide attenuates the inflammatory activity in the spinal cord and increases the survival of spinal matters following SCI. These preclinical studies introduce a promising potential treatment strategy to address SCI-induced dysfunction.

Keywords: ATP-gated potassium channels; chronic pain; diazoxide; glibenclamide; spinal cord injury.

FIGURE 1

Effect of diazoxide on the hyperexcitability of dorsal root ganglion (DRG ) neurons induced by SCI. (A) Representative recordings showing that spontaneous activity of DRG neurons from a contusive animal were largely blocked after cells were exposed to diazoxide for 1 min. Note that cell membrane potential is hyperpolarized from about −46 mV to −53 mV. (B) Summary data indicate the baseline membrane potential before vehicle (Veh) or diazoxide (Diaz) application (left panel), as well as changes in membrane potential upon local application of vehicle or diazoxide (100 μM). Each circle represents one DRG neuron. **p < 0.05, One way ANOVA. (C) Table indicates the effects of diazoxide on firing of DRG neurons from SCI rats.

FIGURE 2

Diazoxide attenuates chronic pain-like behaviors in SCI rats. (A) Effects of diazoxide (Diaz, 5 mg/kg, i. p.) on chronic SCI-induced mechanical hypersensitivity tested 120 min after injections. (B) Effects of diazoxide on chronic thermal hypersensitivity following SCI. Animal numbers are indicated on columns. **p < 0.01, ***p < 0.001, one-way ANOVA followed by Tukey’s post hoc test.

FIGURE 3

Effects of glibenclamide (Gliben; starting 30 min after contusion for 24 h) and diazoxide (Diaz; starting 3 days post-injury for 10 days) on development of pain behaviors after spinal cord injury (SCI) compared to vehicle control (Veh). (A) Mechanical sensitivity of hindpaws was measured by von Frey test 5 weeks after initial traumatic SCI. Animal numbers are indicated on columns. *p < 0.05, **p < 0.01, two-tailed paired t-tests (before and after contusion of same group) or two-tailed unpaired t-tests (different treatment groups). (B) Thermal sensitivity of hindpaws was measured 5 weeks after SCI. Animal numbers are indicated on columns. *p < 0.05, **p < 0.01, two-tailed unpaired t-tests (different treatment groups). (C) Conditioned place preference apparatus was used for CPP tests, which include one black and one white chambers. (D) Spontaneous pain was evaluated with conditioned place preference (CPP) tests 42 days after contusion. Animal numbers are indicated on columns. *p < 0.05, **p < 0.01, ***p < 0.001, one-way ANOVA followed by Tukey’s post hoc test.

FIGURE 4

Effects of glibenclamide (Gliben; starting 30 min after contusion for 24 h) and diazoxide (Diaz; starting 3 days post-injury for 10 days) on development of locomotor behaviors after spinal cord injury (SCI) compared to vehicle control (Veh). (A) Baso, Beattie, Bresnahan (BBB) locomotor test scoring was performed on days –1, 1, 2, 3, 7, 14, 21, 28, and 35 after contusion. *, #p < 0.05, **, ##p < 0.01; ###p < 0.001; *, ** glibenclamide + diazoxide vs. vehicle + vehicle at same timepoints; #, ##, ### glibenclamide + vehicle vs. vehicle + vehicle at same timepoints; two-tailed unpaired t-tests. (B) Horizontal ladder test was performed 21, 28, and 35 days after contusion. ***p < 0.001; ##p < 0.05; * glibenclamide + vehicle vs. vehicle + vehicle at same timepoints; ## glibenclamide + vehicle vs. glibenclamide + diazoxide at P35; two-tailed unpaired t-tests.

FIGURE 5

Effects of glibenclamide (Gliben; starting 30 min after contusion for 6 days) and diazoxide (Diaz; starting 6 days post-injury for 7 days) on development of locomotor and sensory behaviors after spinal cord injury (SCI). (A) Mechanical sensitivity of hindpaws was measured by von Frey test 35 days after initial traumatic SCI. Animal numbers for each group are indicated on columns. **p < 0.01, ***p < 0.001, two-tailed paired t-tests (before and after treatment of the same group) and two-tailed unpaired t-tests (different groups). (B) Thermal sensitivity of hindpaws was measured 35 days after contusion. Animal numbers for each group are indicated on columns. *p < 0.05, ***p < 0.001, two-tailed unpaired t-tests (different groups). (C) Baso, Beattie, Bresnahan (BBB) locomotor testing was performed 42 days after contusion. **p < 0.01, one-way ANOVA followed by Tukey’s post hoc test. (D) Horizontal ladder test was performed 42 days after contusion. Animal numbers are indicated on columns. **p < 0.01, ***p < 0.001, one-way ANOVA followed by Tukey’s post hoc test.

FIGURE 6

Effects of glibenclamide (Gliben; starting 30 min after contusion for 24 h) and diazoxide (Diaz; starting 3 days post-injury for 10 days) on spinal cord injury (SCI)-induced activation of glial cells and hemorrhage in the spinal cord. (A) Quantification of GFAP expression normalized to β-actin in L4/L5 spinal cords. #p < 0.05, ##p < 0.01, ***p < 0.001, one-way ANOVA followed by Tukey’s post hoc test. (B) Quantification of IBA-1 expression normalized to β-actin in L4/L5 spinal cords. (C) Quantification of extravascular blood in the spinal cord lesion site. ***p < 0.001, one-way ANOVA followed by Tukey’s post hoc test. Filled symbols in each column represent individual animals.

FIGURE 7

Effects of glibenclamide (Gliben; starting 30 min after contusion for 24 h) and diazoxide (Diaz; starting 3 days post-injury for 10 days) on spinal cord injury (SCI)-induced tissue loss in the spinal cord compared to vehicle control (Veh). (A) Representative images showing eriochrome–cyanine staining of spinal cord sections of treated SCI rats. (B) Quantification and (C) normalization of spared gray matter and white matter of spinal cords at the epicenter of contusion site from glibenclamide and/or diazoxide-treated SCI rats. Animal numbers are indicated on columns. *p < 0.05, one-way ANOVA followed by Tukey’s post hoc test. Scale is 400 μm.