Neuroscience 291 (2015) 260–271

EARLY APPLIED ELECTRIC FIELD STIMULATION ATTENUATES SECONDARY APOPTOTIC RESPONSES AND EXERTS NEUROPROTECTIVE EFFECTS IN ACUTE SPINAL CORD INJURY OF RATS C. ZHANG, a G. ZHANG, a W. RONG, b A. WANG, a C. WU a AND X. HUO a*

functional and historical recovery. Furthermore, these neuroprotective effects may be related to the inhibition of secondary apoptotic responses after SCI. These findings support further investigation of the future clinical application of EFS after SCI. Ó 2015 IBRO. Published by Elsevier Ltd. All rights reserved.

a Beijing Key Laboratory of Bioelectromagnetism, Institute of Electrical Engineering, Chinese Academy of Sciences, No. 6 Beiertiao, Zhongguancun, Haidian District, Beijing 100190, China b Department of Orthopedics, Beijing Tsinghua Changgung Hospital Medical Center, Tsinghua University, Li Tang Road No. 168, Dongxiaokou Town, Changping District, Beijing 102218, China

Key words: electric field stimulation, spinal cord injury, secondary injury, apoptosis, injury potential.

Abstract—Injury potential, which refers to a direct current voltage between intact and injured nerve ends, is mainly caused by injury-induced Ca2+ influx. Our previous studies revealed that injury potential increased with the onset and severity of spinal cord injury (SCI), and an application of applied electric field stimulation (EFS) with the cathode distal to the lesion could delay and attenuate injury potential formation. As Ca2+ influx is also considered as a major trigger for secondary injury after SCI, we hypothesize that EFS would protect an injured spinal cord from secondary injury and consequently improve functional and pathological outcomes. In this study, rats were divided into three groups: (1) sham group, laminectomy only; (2) control group, subjected to SCI only; and (3) EFS group, received EFS immediately post-injury with the injury potential modulated to 0 ± 0.5 mV by EFS. Functional recovery of the hind limbs was assessed using the Basso, Beattie, and Bresnahan (BBB) locomotor scale. Results revealed that EFS-treated rats exhibited significantly better locomotor function recovery. Luxol fast blue staining was performed to assess the spared myelin area. Immunofluorescence was used to observe the number of myelinated nerve fibers. Ultrastructural analysis was performed to evaluate the size of myelinated nerve fibers. Findings showed that the EFS group rats exhibited significantly less myelin loss and had larger and more myelinated nerve fibers than the control group rats in dorsal corticospinal tract (dCST) 8 weeks after SCI. Furthermore, we found that EFS inhibited the activation of calpain and caspase-3, as well as the expression of Bax, as detected by Western blot analysis. Moreover, EFS decreased cellular apoptosis, as measured by TUNEL, within 4 weeks post-injury. Results suggest that early EFS could significantly reduce spinal cord degeneration and improve

INTRODUCTION Spinal cord injury (SCI), which involves primary and secondary injury mechanisms, is the most devastating injury to the spinal cord. The initial impact directly results in immediate hemorrhage and rapid tissue injury at the injury site (Oyinbo, 2011). Structural damage, such as plasma membrane integrity impairment and sodium– calcium exchanger dysfunction, causes a marked increase in Ca2+ influx within injured axons at the injury site after acute injury (Mikkelsen et al., 2004). Injury potential, which refers to a direct current potential gradient between intact and injured nerve ends, is mainly caused by injury-induced Ca2+ influx (Geddes and Hoff, 1971; Goodman et al., 1985; Zuberi et al., 2008). Our previous papers accordingly revealed that injury potential increased with the onset and severity of SCI (Pan et al., 2011). Injury potential increased dramatically within 30-min post-injury and then decreased gradually to a normal value of 0 mV after several hours post-injury. The initial amplitude of injury potential positively correlated with the severity of SCI. The injury-induced influx of excessive Ca2+ into cells is also considered, among others, as a major mechanism for secondary injury (Imaizumi et al., 1997; Xiong et al., 2007). Ca2+ influx not only diffuses to adjacent regions, which results in further tissue breakdown (Ray et al., 2003; Beirowski et al., 2005), but also inappropriately stimulates a variety of apoptosis-related proteins to activate the apoptotic pathway, thereby aggravating irreversible tissue loss and dysfunction to worsen the primary lesion (Beattie et al., 2000; Huff et al., 2011). Previous investigations indicated that early Ca2+ blocking is a viable therapeutic strategy that could reduce the degree of secondary injury and prevent

*Corresponding author. Tel: +86-10-82547242; fax: +86-1082547164. E-mail address: [email protected] (X. Huo). Abbreviations: BBB, Basso, Beattie, and Bresnahan; dCST, dorsal corticospinal tract; EFS, electric field stimulation; IF, immunofluorescent; LFB, Luxol fast blue; OFS, oscillating electrical fields; SBDP, spectrin breakdown product; SCI, spinal cord injury; WB, Western blot; TEM, transmission electron microscopy. http://dx.doi.org/10.1016/j.neuroscience.2015.02.020 0306-4522/Ó 2015 IBRO. Published by Elsevier Ltd. All rights reserved. 260

C. Zhang et al. / Neuroscience 291 (2015) 260–271

further damage to the spinal cord axons (Liverman et al., 2005). An approach for reducing Ca2+ influx by applied electric field stimulation (EFS) was reported by Strautman et al. (1990). The study showed that the movement of Ca2+ was greatly reduced by the applied field with the cathode distal to the lesion, and the movement of Ca2+ increased upon application of a field of the opposite polarity. However, no proper indexes have been established to determine the optimal EFS parameters, such as the stimulating voltage and duration. Based on previous approaches, we further investigated the optimal parameters of EFS for therapy in our laboratory (Zhang et al., 2013). In a rat model of SCI, stimulating EFS voltages were established to offset the rostral and caudal injury potential (injury potential compensation). After 30 min of stimulation, results showed that the injury potential was significantly delayed during EFS and was attenuated after EFS. Therefore, this study further determined the efficacy of early EFS in terms of protecting the spinal cord from degeneration and accelerating regeneration after SCI. We also examined the expression and activation of calcium-activated apoptotic proteins to determine the underlying therapy mechanism of EFS following SCI.

261

(n = 53), which underwent SCI only; and EF group rats (n = 53), which received EFS after SCI. The rats were housed in a temperature-controlled room (23 ± 1 °C) with a 12:12 dark/light cycle and were given free access to food and ultrapure water. All attempts were made to minimize the number of animals used and to avoid undue animal suffering. Surgical protocol and applied electric field procedure Surgical protocol. The spinal cords of all three groups of rats were exposed by three small laminectomies located at T8, T10, and T12. For EFS group rats, two electric field stimulators were applied for each rat (Fig. 1). Two anodes were sutured at both sides of the paravertebral muscle of T10, whereas the two cathodes of one stimulator were sutured at the paravertebral muscle of T8. Two cathodes of another stimulator were sutured beside T12. The distances between adjacent electrodes were approximately 1 cm. SCI was induced by dropping a weight of 10 g from a height of 50 mm onto an impounder (with a diameter of 0.2 cm), which was gently placed on the spinal cord. The sham group rats underwent laminectomy only. The control group rats underwent SCI only.

EXPERIMENTAL PROCEDURES Animal uses and groups Adult female Sprague–Dawley (SD) rats weighting 200– 250 g were purchased from Beijing HFK Bio-Technology Co., Ltd. (Beijing, China). Animal experimental procedures were performed according to the National Guidelines for Experimental Animal Welfare (Ministry of Science and Technology of People’s Republic of China, 2006) and were approved by the Animal Welfare Committee of the Beijing Key Laboratory of Bioelectromagnetism. The rats were randomly assigned to three groups: sham group rats, (n = 53) which underwent laminectomy only; control group rats

Applied electric field procedure. The injury potentials were measured by glass electrodes immediately after SCI (Fig. 2). The glass electrodes, which consisted of upper and lower glass tubes, were described in our previous article (Zhang et al., 2013). The upper tube was filled with 3 M KCl solution and contained a calomel electrode that was connected to voltmeters via conducting wires. The lower tube was filled with 0.9% saline and plugged by a small bulk of porous ceramic. The two solutions in the upper and lower tubes were separated by agar. The tip of one glass electrode, which connected the negative input of the voltmeter, was gently placed on

Fig. 1. Experimental setup for applied electric field stimulation in rats SCI. (A) Schematic diagram of the applied electric field stimulation. (B) Image of the electric field stimulation procedure. The electric field stimulator was packaged in a plastic box, and the rostral and caudal stimulating voltages were regulated through the knobs on the box. (C) Enlarged image of electrode suture. The anodes were sutured at the paravertebral muscle of T10, whereas the cathodes were sutured beside T8 and T12. The distances between the adjacent electrodes were approximately 1 cm.

262

C. Zhang et al. / Neuroscience 291 (2015) 260–271

caudal injury potential was also tuned to 0 ± 0.5 mV by adjusting the caudal knob. Then the compensation voltages between rostral electrode and reference electrode as well as between caudal electrode and reference electrode were measured and recorded. The EFS lasted for 30 min, the stimulating voltage was evaluated every 10 min (3 times total per animal) and adjusted if necessary during the EFS. After the stimulation, the stimulating electrodes were removed, and the wounds were sutured in layers.

Electrical stimulator fabrication

Fig. 2. Injury potential detection method. (A) Salt-bridge electrodes were used for injury potential detection. (B) Electrode tips were gently placed on the exposed spinal cord tissue. The potential differences between T8 and T10 represented rostral injury potential, whereas the potential differences between T10 and T12 represented caudal injury potential. The injury potential was determined by a voltmeter. The stimulating voltages were adjusted according to the injury potential.

the lesion at the T10 segment of the spinal cord. The tip of another glass electrode, which connected the positive input of the voltmeter, was gently placed on the T8 or T12 segment of the spinal cord. The voltages between the two electrodes represented the rostral or caudal injury potentials. The electric field was applied by opening the switch on the box (Fig. 1). The rostral injury potential was shown by a voltmeter and was tuned to 0 ± 0.5 mV by adjusting the rostral knob on the box. Using the same method, the

The proposed electrical schematic of the stimulator, which produced the applied electric field, was shown in Fig. 3. The stimulator was powered by two seriesconnected 9-V batteries. The voltage regulation unit, which contained a 100-X potentiometer (R0) and two 1-kX resistors (R1 and R2), was connected to the two terminals of the batteries. The amplification unit included a non-inverting amplifier and a voltage follower. The electric potential of the central terminal of the potentiometer was amplified by the non-inverting amplifier, which contained a 1-kX resistor (R3), a 20-kX variable resistor (R4), and an operational amplifier (LM 358a). The amplification factor of the non-inverting amplifier, which was determined based on the value of 1 + R4/R3, could be adjusted by varying the resistance of R4. The voltage follower, which was formed by LM358b, was used to eliminate the influence of the impedance caused by the tissue between electrodes on the former non-inverting amplifier. The electrode system contained two spiral stimulating electrodes and one spiral reference electrode. To measure and limit the current flow in the tissue, a 100-X resistor R5 was placed between the stimulating electrode and the output of the voltage follower. The reference electrode was connected to the ‘‘ground’’. The spiral electrodes used in this study were platinum/iridium 90%/10% which are the most common electrode materials (Norlin et al., 2002). The selection of this metal type was chosen for its biocompatibility, electrical, mechanical and chemical properties. The radius of the spiral electrode was 1 mm.

Fig. 3. Schematic of the applied electric field stimulator. The electric field stimulator was powered by two 9-V batteries, which provided ±9-V sources. A voltage regulation unit comprising R0, R1, and R2 provided a voltage range from 0.43 V to 0.43 V. The amplification unit included a non-inverting amplifier and a voltage follower. The amplification factor of the non-inverting amplifier, which ranges from 1 to 21, could be adjusted by varying the resistance of R4. The voltage follower was used to eliminate the influence of tissue impedance. Two stimulating electrodes and one reference electrode comprised the electrode system. The reference electrode was connected to the ‘‘ground.’’

C. Zhang et al. / Neuroscience 291 (2015) 260–271

The electrodes were connected to the circuit through insulated copper wires, the length of the connection between the Pt–Ir wire and the copper wire from the tissue was about 11 mm. The connection was covered by silicone in case of the effects produced by the dissimilar metal connection.

Behavioral assessment The Basso, Beattie, and Bresnahan (BBB) locomotor score was used to assess the functional recovery of the rats’ hind limbs (n = 8/group) (Basso et al., 1995). A score of 0 represented no spontaneous movement, whereas a score of 21 represented complete mobility. An increasing score indicated the use of individual joints, coordinated joint movement, coordinated limb movement, weight-bearing, and other functions. Two independent examiners who were blinded to the experimental procedure observed the rats during the tests. Scoring was performed before injury, on days 1 and 3, and subsequently, once weekly until 8 weeks after SCI.

Animal sacrifices and assessments At the indicated time point, the rats were sacrificed with an overdose of pentobarbital (80 mg/kg). The time of sacrifice was determined according to the different parameters measured as follows (Fig. 4): Luxol fast blue (LFB) staining, transmission electron microscopy (TEM), and immunofluorescent (IF) labeling of myelinated axons (8 weeks after SCI); Western blot (WB) analysis for caspase-3 and Bax (1, 2, and 4 weeks post-injury); TUNEL staining (1 and 2 weeks post-injury); IF labeling of calpain (12, 24, and 48 h after SCI); and WB analysis for calpain and a-spectrin (24 h after SCI). LFB staining The animals (n = 5/group) were sacrificed, and the spinal cords encompassing the injury site were then removed. After fixation and dehydration, the spinal cords were cut into 20-lm sections and stained with LFB. The area of spared white matter was expressed as the percentage of LFB-positive area in the total area of the spinal cord and measured using Image-Pro Plus 5.0 software (Cybernetics, Bethesda, MD, USA).

263

Immunofluorescence assessment For tissue staining, 5 lm of the spinal cord frozen sections were incubated in 10% normal goat serum for 1 h at room temperature. For myelinated axon number assessment (n = 5/group), a monoclonal mouse antiNF-H antibody (1:800; Sigma–Aldrich, St. Louis., MO, USA) with a polyclonal rabbit anti-MBP antibody (1:100, Sigma–Aldrich, St. Louis., MO, USA) was applied to the sections at 4 °C overnight. Sections were then incubated with an FITC-conjugated goat anti-rabbit IgG antibody (1:100; ZSGB-Bio, Beijing, China) and a Texas Red-conjugated goat anti-mouse antibody (1:100; ZSGB-Bio, Beijing, China). For calpain expression analysis (n = 5/group), a polyclonal rabbit anti-calpain antibody (1:1000: Beyotime, Shanghai, China) was applied to the sections at 4 °C overnight and incubated with an FITC-conjugated goat anti-rabbit IgG antibody (1:100; ZSGB-Bio, Bejing, China). Photos from the dorsal corticospinal tract (dCST) were taken using a fluorescence microscope (Nikon E600, Japan). The slides were measured by two investigators who were blinded to the treatment of the animals. Electron microscopy and morphometric analysis The spinal cords from the dCST of T10 (n = 3/group) were post-fixed in 2.5% glutaraldehyde in 0.1 M PB for 1 d and in 2% osmic acid for 2 h, after which they were embedded in epoxy resin. Ultrathin sections were mounted on copper grids. Images with a magnification of 50,000 and 100,000 were taken using TEM (Hitachi7650, Tokyo, Japan). Photos were digitized, and at least 100 myelinated fibers were analyzed by using a computer-aided morphometric system (Mac Scope, Mitani, Fukui, Japan). Axonal diameters were measured along the inner border of the myelin sheath, nerve fiber diameters were measured along the outer border, and myelin thickness values were calculated from the obtained data. TUNEL assay Sections (n = 5/group) were prepared as described for immunofluorescence staining and were used for cell death analysis. The directions for operation supplied by Roche Molecular Biochemicals were strictly followed. The apoptotic cells were observed under a Leica

Fig. 4. Timeline of the experimental detection. The time of sacrifice was determined according to the different parameters measured. LFB: Luxol fast blue staining, TEM: transmission electron microscopy, IF: immunofluorescent labeling, WB: Western blot analysis.

264

C. Zhang et al. / Neuroscience 291 (2015) 260–271

photograph microscope (Leica DM 4000B, Germany). The number of apoptotic cells in the dCST of the spinal cord with five unfolded continuous fields in each section was counted by two observers who were blinded to the protocol. The number of apoptotic cells was expressed as the average of TUNEL-positive cells per field in each group. WB analysis For WB analysis, the animals (n = 5/group) were sacrificed, a 1-cm-long spinal cord centered at the injury site was removed, and dorsal white matter strips were separated as described in a previous paper (Ouyang et al., 2010). The protein concentrations were determined by a BCA protein assay (Pierce, Rockford, IL, USA). The tissue homogenates (approximately 100 lg of protein) were separated by 12% SDS–PAGE gel and transferred to PVDF membranes (Millipore). The membranes were blocked in 10% non-fat milk and incubated overnight at 4 °C with anti-pro-caspase-3 (Abcam, Cambridge, MA,

USA, 1:500), anti-active-caspase-3 (Cell Signaling, Danvers, MA, USA, 1:500), anti-Bax (Bioworld Technology,, St. Louis Park, MN, USA, 1:500), anti-calpain (Cell Signaling, Danvers, MA, USA, 1:500), anti-a-145KD spectrin (Santa Cruz, Paso Robles, CA, USA, 1:500), and anti-bactin (Sigma- Aldrich, St. Louis., MO, USA, 1:10000) primary antibody and then with proper secondary IgG antibody (1:5000). Membranes were shortly incubated with enhanced chemiluminescent reagents (ECL Plus, Amersham Biosciences, Waltham, MA, USA) and imaged on an Odyssey-Infrared Imaging System (LICOR Biosciences, Lincoln, NE, USA). Statistical analysis Statistical analysis was performed using SPSS software 13.0. All data were expressed as mean ± SD. These data underwent a normality test (Shapiro–Wilk test) and homogeneity of variance test (Levene’s test) before statistical analysis. For comparison of groups over time (BBB behavioral test), repeated-measures analysis of

Fig. 5. Change pattern of rostral and caudal injury potential and compensation voltages. (A) Rostral injury potential in two groups. The injury potential was slightly larger than 0 mV during stimulation and was then restarted when EFS was completed. The rostral compensation voltages were approximately 2.7 V during the 30-min-EFS. (B) Caudal injury potential in two groups. The changing trends of caudal injury potential were almost the same as rostral injury potential. The Caudal compensation voltages were approximately 2.7 V during the 30-min-EFS. Dotted line a: injury potential immediately after injury; dotted line b: injury potential immediately after EFS; dotted line c: injury potential immediately after EFS, n = 53 per group.

C. Zhang et al. / Neuroscience 291 (2015) 260–271

265

variance (ANOVA) followed by Bonferroni’s post hoc test was used. For comparison of simple effects, t tests were used. P < 0.05 was considered significant for all statistical tests.

RESULTS Injury potential compensation by EFS The injury potential of the control group increased immediately within 30-min post-injury and then decreased gradually to almost 0 mV after several hours post-injury (Fig. 5A). Rats in the the EFS group underwent EFS immediately following SCI. The rostral compensation voltages were approximately 2.7 V during EFS (Fig. 5A). The amount of current was about 400 lA, and the resistance of the spinal cord was about 6.75 kX after calculation. Furthermore, the distances between adjacent electrodes were approximately 1 cm, so the electric fields through the tissue were about 270 mV/mm. The rostral injury potential in the EFS group was slightly larger than 0 mV because of injury potential compensation. When EFS was finished, injury potential restarted (Fig. 5A, b). Although the amplitude of restarted injury potential was slightly larger than that in the control group (Fig. 5A, c), it was significantly smaller than the initial amplitude in the control group (Fig. 5A, a). The injury potential of the EFS group gradually decreased over several hours. The changing trends of caudal injury potential (Fig. 5B) were almost the same as those of rostral injury potential, and the caudal compensation voltages were also approximately 2.7 V during EFS. The rostral and caudal compensation voltages were evaluated every 10 min during the EFS (3 times total) and changed slightly (ranged from 2.4 V to 2.9 V), so we did not adjust at each 10-min point. After the stimulation, the stimulating electrodes were removed, and no significant discoloration and gas evolution were

Fig. 7. Spinal cord myelin sparing was evaluated by LFB staining after SCI. (A) Representative transverse spinal cord sections stained with LFB at 8 weeks post-injury taken within 2 mm rostral (R) and caudal (C) to the injury site in all groups. Scale bar = 500 lm. (B) Quantitative data on the spared white matter. The area of LFBpositive spared white matter was significantly larger in the EFS group than in the control group. ⁄P < 0.05 vs. control group, ⁄⁄P < 0.01 vs. control group, n = 5 per group.

found and no significant changes were seen in muscle and spinal cord tissues around the electrodes. Locomotor function testing after SCI All animals had normal initial BBB scores (Fig. 6). The injured rats exhibited a severe locomotion deficit at days 1 and 3 post-injury. The EFS group rats showed a significant function improvement (Fig. 6). By contrast, the control group rats also regained some motor functions, but the mean scores were significantly lower than those of the EFS group rats (F = 69.24, P < 0.01). The BBB score of the EFS group was 14.8 ± 1.40 at day 56 post-injury, indicating consistent forelimb–hindlimb coordination (Basso et al., 1996). Meanwhile, the BBB score of the control group was 10.2 ± 2.66, which demonstrated only occasional weight-supported plantar stepping (Basso et al., 1996).

Fig. 6. Functional assessment of rats with BBB testing at the indicated times. Repeated-measures ANOVA followed by the Bonferroni’s post hoc test showed that the EFS group rats exhibited significantly better improvements in score over time (F = 69.24, P < 0.01). The blue dotted line indicates a BBB score of 10, representing occasional weight-supported plantar steps but no forelimb–hindlimb coordination. The black dotted line indicates a BBB score 15, representing consistent plantar stepping and consistent forelimb–hindlimb coordination.⁄P < 0.05 vs. control group, ⁄⁄ P < 0.01 vs. control group, n = 8 per group. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Histological evaluation of spared myelin after SCI The myelin content of intact and injured spinal cords was determined by LFB staining 8 weeks after injury. In the sections of sham group rats, well-defined butterflyshaped areas of gray matter could be observed, and white matter containing myelin was stained blue    For interpretation of color in Fig. 7A, the reader is referred to the web version of this article.

266

C. Zhang et al. / Neuroscience 291 (2015) 260–271

Fig. 8. Double labeling and merged photomicrographs of NF-H and MBP at 8 weeks after SCI. The number of myelinated axons was quantified in the dCST (A and B). Double immunostaining for MBP and NF of sham (C), control (D), and EFS (E) group rats showed axons with surrounding myelin. Bar graph (F) showed quantitative data that the number of myelinated axons in the EFS group was significantly greater than that in the control group. ⁄P < 0.05 vs. control group, Scale bar = 50 lm (B, C, D, and E), n = 5 per group.

(Fig. 7A). Axonal degeneration, myelin loss, and vacuoles could be identified within the stained section in the control and EFS group rats (Fig. 7A). As shown in Fig. 7B, the area of myelin in the EFS group was significantly larger than that in the control group at R 2 mm (t = 2.390, P = 0.044), R 1.5 mm (t = 2.653, P = 0.029), R 1.0 mm (t = 5.420, P = 0.001), R 0.5 mm (t = 4.384, P = 0.002), C 0.5 mm (t = 4.821, P = 0.001), C 1.0 mm (t = 4.986, P = 0.001), C 1.5 mm (t = 3.930, P = 0.005), and C 2.0 mm (t = 2.573, P = 0.033) relative to the injury site and at the epicenter (t = 2.590, P = 0.032). Immunofluorescence assessment of myelinated axons in dCST after SCI To assess the protective effect of EFS on myelinated axons, we counted the number of myelinated axons at the injury site (Fig. 8A, B). Double labeling photomicrographs showed axons with surrounding myelin (Fig. 8C–E). The statistical results showed that the rats treated with EFS presented better preservation of myelinated axons than the control group rats (t = 2.590, P = 0.032) (Fig. 8F). Electron microscopic analysis of axons in dCST after SCI To examine further the effect of EFS on ultrastructure of myelinated axons, the spinal cords of dCST at the injury site were observed under an electron microscope. Morphologically, axons in the control and EFS groups of rats were wrapped by thinner myelin sheaths with scattered profiles of abnormally swollen sheaths and axons (Fig. 9A–D), whereas the myelin sheath healthy appearance with intact membranes and good density (Fig. 9E). The diameters of axons and the thickness of myelin were measured. The frequency distribution of axonal diameters revealed a shift in the diameters of

myelinated axons toward larger sizes in the EFS group (Fig. 9F). The histogram demonstrated a significant increase in the mean myelin thickness in EFS rats compared with control rats (t = 3.393, P = 0.009) (Fig. 9G). Changes in calpain expression, content, and activity in the spinal cord Calpain, a Ca2+-dependent neutral protease, triggers apoptotic cascade post-injury. Calpain expression in the lesion tissue of the control group was elevated after the onset of SCI within 48 h, as revealed by fluorescence staining (Fig. 10A) in dCST. Meanwhile, injured tissue of the EFS group did show an increase in calpain staining after injury. The intensity of staining presented a minimal increase as compared with that in the control group. In the sham group, low calpain expression was noted. Calpain content and activity were examined via WB analysis using antibodies against 80-kDa calpain and antibodies against calpain-specific 145-kDa spectrin breakdown product (SBDP) (Figs. 10B, C). A small amount of calpain and 145-kDa SBDP was found in the sham tissue. At 24-h post-SCI, significant increases in 80-kDa calpain and 145-kDa SBDP expressions were observed in the lesion tissue in control rats. EFS significantly reduced 80-kDa calpain (t = 2.613, P = 0.043) and 145 kDa SBDP (t = 2.570, P = 0.038) expression compared with that in the control group in dorsal white matter. Changes in the Bax and active-caspase-3 content of the spinal cord To confirm further the effect of EFS on secondary apoptotic responses in the dorsal white matter, we measured two calpain-related apoptosis factors by WB

267

C. Zhang et al. / Neuroscience 291 (2015) 260–271

Fig. 9. Ultrastructural analysis of myelinated axons of dCST in the spinal cord white matter. Abnormally swollen axons and axons with thin myelin sheaths were observed under an electron microscope in the control (A) and EFS groups (B). Portions of myelin sheaths of the control (C), EFS (D), and sham groups (E) at higher magnification are shown. Electron microscopy was quantified, and a frequency histogram was created (F). We note the marked shift in distribution of axonal diameters in the EFS group rats toward larger fibers as compared with the control group rats. The thicknesses of the myelin sheath of EFS rats increased significantly in comparison with those of rats in the control group (G). ⁄P < 0.05 vs. control group, Scale bar = 1000 nm (A and B), 250 nm (C, D, and E), n = 3 per group.

at different time points (Fig. 11A). First, the results showed that SCI resulted in a significant elevation in the expression of Bax in the control group as compared with the EFS group at week 1 (t = 2.696, P = 0.027) and week 2 (t = 2.569, P = 0.033) (Fig. 11B). However, no difference in Bax expression was observed 4 weeks (t = 1.111, P = 0.299) after SCI between the control and EFS groups (Fig. 11B). Second, the amounts of procaspase-3 and active-caspase-3 at the injury site were increased in the control group within 4 weeks (Fig. 11A). The administration of EFS significantly suppressed the up-regulation of active-caspase-3 at week 1 (t = 4.752, P = 0.001), week 2 (t = 4.417, P = 0.002), and week 4 (t = 2.399, P = 0.043) postinjury (Fig. 11C).

representative microphotographs, apoptotic cells were characterized by reduced cell size, shrunken rounded cell bodies, and brown staining (Fig. 12A). In the sham operation group, almost no TUNEL-positive cells were observed (Fig. 12A). The numbers of TUNEL-positive cells in the EFS and control groups were significantly increased after SCI, indicating increased apoptotic activity. We found that compared with that in the control group, the number of TUNEL-positive cells significantly decreased at 2 mm rostral from the epicenter (week 1: t = 2.426, P = 0.041; week 2: t = 2.449, P = 0.041), at the epicenter (week 1: t = 3.747, P = 0.006; week 2: t = 2.324, P = 0.045), and at caudal from the epicenter (week 1: t = 3.568, P = 0.007; week 2: t = 3.548, P = 0.008) in the EFS group (Fig. 12B).

TUNEL assay for apoptotic cells in dCST after SCI The effect of EFS on cellular apoptosis was analyzed by TUNEL staining in dCST after SCI. The sections at 0–2 mm rostral and caudal from the epicenter in three groups were stained with TUNEL. As revealed by

DISCUSSION In this study, rats were subjected to injury potential compensation by early EFS after SCI, and satisfactory results were obtained.

268

C. Zhang et al. / Neuroscience 291 (2015) 260–271

group rats from SCI was demonstrated by BBB testing in EFS group rats. The early motor recovery at day 7 post-injury may be due to the reduction in inflammation reaction and free-radical response caused by the decreased Ca2+ influx (Borgens, 2012). However, this effect was unsustained until day 35, thereby suggesting the potential long-term benefits of EFS with improvements lasting for several weeks after SCI (Maybhate et al., 2012). Accordingly, the functional and historical results in this study demonstrated that EFS therapy could significantly exert neuroprotective (reduction of myelin loss) and stimulatory effects (regeneration of axon and myelin) after SCI. Effect of EFS on the cellular apoptosis in SCI rats

Fig. 10. Expression and activity of Ca2+-dependent neutral protease calpain. (A) Immunofluorescent staining showed an increase in fluorescence intensity in the control group as compared with the EFS group within 48 h after SCI. (B) Calpain expression in lesion tissues was confirmed by Western blot analysis. Quantification of OD of protein bands showed a reduction in protein levels of calpain in the EFS group rats 24 h after SCI. (C) Calpain activity was measured by 145-kDa spectrin SBDP. Bar graph showed the reduction information on 145-kDa SBDP in EFS group rats 24 h after SCI. b-Actin expression was monitored for equal loading. ⁄P < 0.05 vs. control group, Scale bar = 50 lm, n = 5 per group.

Effect of EFS on behavioral and morphological changes in SCI rats Myelin loss peaked at the epicenter and decreased with increasing distance in the control group as shown in previous works (Rong et al., 2012). With the loss of myelin, axons were directly exposed to damaging effects resulting in further neuronal loss (Tsutsui and Stys, 2013). Thereby, the main strategy was to arrest the myelin degeneration after SCI. The extent of LFB-positive staining of the myelin showed that EFS significantly increased the myelin area within 2 mm cranial and caudal from the epicenter at 8 weeks post-injury. Promoting remyelination is another important strategy for SCI treatment. Compared with the case of the control rats, the number of myelinated axons in dCST in the EFS group rats was significantly higher, the regenerative improvements in the axonal diameters and myelin thickness of myelinated axon were also supported by EM observation in the EFS group. Improved behavioral recovery of EFS

Calpain was a Ca2+-dependent neutral protease and the elevation of intracellular Ca2+ triggered the activation of the calpain (Stys, 2005; Zuberi et al., 2008). Calpain has been reported to serve several functions in activating factors that may be involved in apoptosis and has a variety of possible interactions with pro-apoptotic caspase-3 and Bax (Talbert et al., 2013; Yu et al., 2013). Given that EFS may reduce the Ca2+ influx, we investigated whether EFS affected the activation of calpain after SCI and thereby attenuated the activation of pro-apoptotic factors, eventually limited cell death in SCI. Calpain expression and 145-kDa spectrin increased in the lesion area were confirmed in the control rats. EFS treatment effectively inhibited both the expression and proteolytic activity of calpain within 48 h post-injury, as determined by immunofluorescence staining and WB after SCI. Activated-caspase-3 contributes to the execution of apoptosis, as reported in many previous studies (Samantaray et al., 2011; Wang et al., 2012). Consequently, we found that EFS effectively attenuated the activation of caspase-3 within 4 weeks post-injury. In addition, the expression of Bax, another important protein in the apoptosis pathway (Lan et al., 2014), was suppressed by EFS treatment within 2 weeks after SCI. Decreased apoptosis in EFS treatments was further verified by TUNEL. Thus, attenuating the secondary apoptotic response after SCI provided evidence for reducing of Ca2+ influx and contributed to the mechanism by which EFS exerts early effects. Application of EFS Several studies have shown that applied electric field crossing the injury site could promote nerve regeneration. Borgens reported that oscillating electrical fields (OFS, approximately 500–600 mV/mm) improve the function recovery in neurologically complete paraplegic dogs 18 days after the treatment (Borgens et al., 1999). Borgens implied that the function recovery after SCI may be due to the regeneration of spinal axons around and through the lesion. Moriarty reported that they imposed a voltage gradient (320 lV/mm) on injured spinal cord and fibroglial scar was relatively suppressed by the electric field (Moriarty and Borgens, 2001). Moreover, our team previously found OFS (500 mV/mm) could

C. Zhang et al. / Neuroscience 291 (2015) 260–271

269

Fig. 11. Effects of EFS on Bax and active-caspase-3 protein levels in the dorsal white matter. (A) Representative immunoblots of 32-kDa procaspase-3, 17-kDa, and 12-kDa active-caspase-3 and Bax at 1, 2, and 4 weeks post-injury. The protein level was detected by Western blot, and the intensity of each band was estimated by densitometric analysis. b-Actin was used as a protein-loading control. The statistical analysis showed a significant reduction in protein levels of 17-kDa and 12-kDa active-caspase-3 (B) and Bax (C) in the dorsal white matter of the EFS group rats within 4 weeks after SCI. ⁄P < 0.05 vs. control group, ⁄⁄P < 0.01 vs. control group, n = 5 per group.

promote locomotor recovery and remyelination in SCI rats after 12-weeks treatment, and this effect may be related to the improved differentiation of OPCs in the spinal cord (Zhang et al., 2014).

Moreover, Strautman and colleagues reported a particular applied electric field (10 mV/mm) stimulation on injured lamprey spinal axons in the photomultiplier tube (Strautman et al., 1990). The results showed that

Fig. 12. TUNEL analysis in the spinal cord white matter of dCST. (A) Representative images were taken from the selected sections 2 mm rostral (R) and caudal (C) to the injury epicenter (Epi) 1 and 2 weeks post-injury in the three groups of rats. (B) Bar graph showed quantification of the TUNELpositive cell counts. ⁄P < 0.05, ⁄⁄P < 0.01, Scale bar = 10 lm, n = 5 per group.

270

C. Zhang et al. / Neuroscience 291 (2015) 260–271

applied electrical fields could modulate Ca2+ movement into the axons, and the electric field of appropriate polarity could nearly eliminate the diffusion of Ca2+ into the cut end. Accordingly, in the present study, we have conducted the applied electrical field stimulation with the cathode distal to the lesion and the anode at the lesion site on the spinal cord after SCI. The compensation voltages were adjusted to approximately 2.7 V according to the injury potentials. The distances between adjacent electrodes were approximately 1 cm, so the electric fields through the tissue were about 270 mV/mm. These larger applied electrical fields may help offset the Ca2+ flux at the injury site and contributed to attenuating the secondary apoptotic response after SCI. Previous studies have showed that influx in the injured cord begins within minutes following transection (Ray et al., 2003). Our previous study showed that the sudden Ca2+ influx rapidly decreased at 30 min post-injury. These results suggested that there was a short window of time available for intervention, and this period was preferably less than 30 min after SCI. Limitations and prospect We are at the initial stage of this study. This study has limitations, including the unavailability of the direct evidence of the Ca2+ change pattern in the spinal cord. To solve this problem, related experiments are in progress. We conducted in vitro experiments of the spinal cord by using the double sucrose gap technique (Velumian et al., 2010, 2011), through which we tried to detect the distribution and content of Ca2+ before and after direct electrical stimulation using Ca-imaging methods and Ca2+ ion meter. However, these areas require further work.

CONCLUSION We conclude that injury potential compensation using EFS is an effective strategy that shows neuroprotective effects in acute SCI of rats. This type of electrical stimulation has the potential to be a feasible and efficient therapy for acute SCI.

AUTHOR DISCLOSURE STATEMENT No competing financial interests exist. Acknowledgments—The authors thank Dr. Chen Huang and Li Chen from the third hospital of the Peking University for valuable discussions and experimental help. We are grateful to our technical staff in our lab for assistance. We also wish to acknowledge support from National Natural Science Foundation of China (Nos. 51177162, 51307166, & 31400717).

REFERENCES Basso DM, Beattie MS, Bresnahan JC (1995) A sensitive and reliable locomotor rating scale for open field testing in rats. J Neurotrauma 12:1–21. Basso DM, Beattie MS, Bresnahan JC (1996) Graded histological and locomotor outcomes after spinal cord contusion using the

NYU weight-drop device versus transection. Exp Neurol 139:244–256. Beattie MS, Farooqui AA, Bresnahan JC (2000) Review of current evidence for apoptosis after spinal cord injury. J Neurotrauma 17:915–925. Beirowski B, Adalbert R, Wagner D, Grumme DS, Addicks K, Ribchester RR, Coleman MP (2005) The progressive nature of Wallerian degeneration in wild-type and slow Wallerian degeneration (WldS) nerves. BMC Neurosci 6:6. Borgens RB (2012) Understanding second injury. Q Rev Biol 87:89–127. Borgens RB, Toombs JP, Breur G, Widmer WR, Waters D, Harbath AM, March P, Adams LG (1999) An imposed oscillating electrical field improves the recovery of function in neurologically complete paraplegic dogs. J Neurotrauma 16:639–657. Geddes LA, Hoff HE (1971) The discovery of bioelectricity and current electricity The Galvani Volta controversy. IEEE Spectr 8:38–46. Goodman RM, Wachs K, Keller S, Black P (1985) Spontaneous spinal cord ‘‘injury potential’’ in the rat. Neurosurgery 17:757–759. Huff TB, Shi Y, Sun W, Wu W, Shi R, Cheng JX (2011) Real-time CARS imaging reveals a calpain-dependent pathway for paranodal myelin retraction during high-frequency stimulation. PLoS One 6:e17176. Imaizumi T, Kocsis JD, Waxman SG (1997) Anoxic injury in the rat spinal cord: pharmacological evidence for multiple steps in Ca(2+)-dependent injury of the dorsal columns. J Neurotrauma 14:299–311. Lan WB, Lin JH, Chen XW, Wu CY, Zhong GX, Zhang LQ, Lin WP, Liu WN, Li X, Lin JL (2014) Overexpressing neuroglobin improves functional recovery by inhibiting neuronal apoptosis after spinal cord injury. Brain Res 1562:100–108. Liverman TC, Altevogt MB, Joy EJ, Johnson TR (2005) Spinal cord injury: progress, promise, and priorities. National Academy Press. Maybhate A, Hu C, Bazley FA, Yu Q, Thakor NV, Kerr CL, All AH (2012) Potential long-term benefits of acute hypothermia after spinal cord injury: assessments with somatosensory-evoked potentials. Crit Care Med 40:573–579. Mikkelsen UR, Fredsted A, Gissel H, Clausen T (2004) Excitationinduced Ca2+ influx and muscle damage in the rat: loss of membrane integrity and impaired force recovery. J Physiol 559:271–285. Moriarty LJ, Borgens RB (2001) An oscillating extracellular voltage gradient reduces the density and influences the orientation of astrocytes in injured mammalian spinal cord. J Neurocytol 30:45–57. Norlin A, Pan J, Leygraf C (2002) Investigation of interfacial capacitance of Pt, Ti and TiN coated electrodes by electrochemical impedance spectroscopy. Biomol Eng 19(2):67–71. Ouyang H, Sun W, Fu Y, Li J, Cheng JX, Nauman E, Shi R (2010) Compression induces acute demyelination and potassium channel exposure in spinal cord. J Neurotrauma 27:1109–1120. Oyinbo CA (2011) Secondary injury mechanisms in traumatic spinal cord injury: a nugget of this multiply cascade. Acta Neurobiol Exp (Wars) 71:281–299. Pan SY, Zhang GH, Huo XL, Bai JZ, Song T (2011) Injury potentials associated with severity of acute spinal cord injury in an experimental rat model. Neural Regen Res 6:1780–1785. Ray SK, Matzelle DD, Sribnick EA, Guyton MK, Wingrave JM, Banik NL (2003) Calpain inhibitor prevented apoptosis and maintained transcription of proteolipid protein and myelin basic protein genes in rat spinal cord injury. J Chem Neuroanat 26:119–124. Rong W, Wang J, Liu X, Jiang L, Wei F, Hu X, Han X, Liu Z (2012) Naringin treatment improves functional recovery by increasing BDNF and VEGF expression, inhibiting neuronal apoptosis after spinal cord injury. Neurochem Res 37:1615–1623. Samantaray S, Knaryan VH, Le Gal C, Ray SK, Banik NL (2011) Calpain inhibition protected spinal cord motoneurons against 1methyl-4-phenylpyridinium ion and rotenone. Neuroscience 192:263–274.

C. Zhang et al. / Neuroscience 291 (2015) 260–271 Strautman AF, Cork RJ, Robinson KR (1990) The distribution of free calcium in transected spinal axons and its modulation by applied electrical fields. J Neurosci Methods 10:3564–3575. Stys PK (2005) General mechanisms of axonal damage and its prevention. J Neurol Sci 233:3–13. Talbert EE, Smuder AJ, Min K, Kwon OS, Powers SK (2013) Calpain and caspase-3 play required roles in immobilization-induced limb muscle atrophy. J Appl Physiol 114:1482–1489. Tsutsui S, Stys PK (2013) Metabolic injury to axons and myelin. Exp Neurol 246:26–34. Velumian AA, Wan Y, Samoilova M, Fehlings MG (2010) Modular double sucrose gap apparatus for improved recording of compound action potentials from rat and mouse spinal cord white matter preparations. J Neurosci Methods 187:33–40. Velumian AA, Wan Y, Samoilova M, Fehlings MG (2011) Contribution of fast and slow conducting myelinated axons to single-peak compound action potentials in rat spinal cord white matter preparations. J Neurophysiol 105:929–941. Wang J, Rong W, Hu X, Liu X, Jiang L, Ma Y, Dang G, Liu ZJ, Wei F (2012) Hyaluronan tetrasaccharide in the cerebrospinal fluid is

271

associated with self-repair of rats after chronic spinal cord compression. Neuroscience 210:467–480. Xiong Y, Rabchevsky AG, Hall ED (2007) Role of peroxynitrite in secondary oxidative damage after spinal cord injury. J Neurochem 100:639–649. Yu CG, Li Y, Raza K, Yu XX, Ghoshal S, Geddes JW (2013) Calpain 1 knockdown improves tissue sparing and functional outcomes after spinal cord injury in rats. J Neurotrauma 30:427–433. Zhang GH, Huo XL, Wang AH, Wu CZ, Zhang C, Bai JZ (2013) Electrical stimulation modulates injury potentials in rats after spinal cord injury. Neural Regen Res 8:2531–2539. Zhang C, Zhang GH, Rong W, Wang AH, Wu CZ, Huo XL (2014) Oscillating field stimulation promotes spinal cord remyelination by inducing differentiation of oligodendrocyte precursor cells after spinal cord injury. Biomed Mater Eng 24:3629–3636. Zuberi M, Liu-Snyder P, Ul Haque A, Porterfield DM, Borgens RB (2008) Large naturally-produced electric currents and voltage traverse damaged mammalian spinal cord. J Biol Eng 2:17–27.

(Accepted 10 February 2015) (Available online 18 February 2015)