Spinal cord injury (SCI) is a permanently debilitating injury that lowers the quality of life, not only for the patient, but also their family members. Worldwide, an estimated 2.5 million people live with SCI, with more than 130,000 new injuries reported annually [1
]. There is a global effort to decrease the number of disabilities caused by SCI and improve patient recovery by employing strategies such as medication, cell transplantation, and rehabilitation. However, an effective treatment remains elusive.
The pathological process of SCI can be divided into; the primary injury caused by external forces causing physical tissue disruption; and secondary injury, in which biochemical and vascular factors cause delayed damage to tissue that survived the initial injury. During secondary injury, there is widespread apoptosis of neurons, astrocytes, oligodendrocytes, and microglia [2
]. The apoptosis of oligodendrocytes, which begins hours after injury and continues for weeks, is especially detrimental, and contributes to the demyelination of neurons that survived the primary injury [4
]. Although oligodendrocyte precursor cells (OPCs) proliferate at the periphery of the lesion site and remyelinate spinal axons, this process is limited because a large percentage of OPCs succumb to apoptosis before differentiating into mature oligodendrocytes [7
]. Therefore, we have been searching for a treatment that inhibits the apoptosis of OPCs in the injured spinal cord, which in turn would reduce the expansion of secondary injury, and improve function through increased remyelination.
One of the factors implicated in OPC apoptosis is endoplasmic reticulum (ER) stress, which occurs due to the accumulation of misfolded proteins in the ER [9
]. Unfolded or misfolded proteins are a common occurrence in cellular maintenance, and are usually handled by an intricately choreographed process, called the ‘ER stress response’ or ‘unfolded protein response (UPR)’. The ER stress response aims to reduce ER stress and restore equilibrium by slowing protein translation, degrading misfolded proteins, and increasing molecular chaperones that fold proteins, such as glucose-regulated protein 78 (GRP78). However, when ER stress overwhelms the capacity of the UPR to restore equilibrium, apoptosis is induced through multiple pathways, including CCAAT/enhancer-binding protein (C/EBP) homologous transcription factor protein (CHOP), caspase 12, and Jun kinase.
Previous studies have demonstrated that SCI induces an increase of ER stress, with different ER stress responses being observed according to neural cell type [10
]. Furthermore, it was shown that lowering the ER stress response of OPCs leads to increased apoptosis during the secondary injury phase of SCI. Studies have also demonstrated that the administration of amiloride, which enhances the ER stress response, decreased OPC apoptosis [11
]. In our previous study by Kuroiwa, et al. [9
], we demonstrated that amiloride administration increased GRP78 expression and decreased CHOP expression within the spinal cord, which decreased neural cell apoptosis and led to an increase of OPCs in the injured spinal cord. The group treated with amiloride had improved motor function compared to the control group up to 28 days after injury. In this follow-up study, we use the same amiloride treatment and study the tested animals up to 56 days after injury to examine the following points: (1) does the increase in OPCs lead to increased oligodendrocytes; (2) does the increase in oligodendrocytes lead to improved remyelination; and (3) is there further improvement in motor function after 28 days, and what effect does amiloride treatment have on allodynia after SCI?
2. Experimental Section
2.1. SCI Model
Female Sprague-Dawley (SD) rats (8–10 weeks old and weighing 280–320 g) were obtained from CLEA Japan, Inc. (Kanagawa, Japan).
Surgical procedures were performed under aseptic conditions, and the rats were anesthetized by inhaling 4% isoflurane. Laminectomy of the 10th thoracic vertebra was performed to expose the dura mater. A severe spinal cord contusion injury was created using an Infinite Horizon spinal cord impactor device (IH impactor; Precision Systems & Instrumentation, Lexington, KY, USA) with a force of 200 kdyne (2 mN) and a dwell time of zero seconds. To treat the bladder dysfunction that commonly occurs after SCI, abdominal massages were performed twice a day until urinary function was regained.
All experimental procedures were performed in accordance with the National Institute of Health Guidelines for the Care and Use of Laboratory Animals, and were approved by the Animal Experimentation Committee of Tokai University School of Medicine (21591907).
2.2. Administration of Amiloride
The injured rats were randomly divided into two groups: an amiloride group, and a phosphate buffered saline (PBS) control group. At 24 h after injury, and every 24 h thereafter for 14 days, the amiloride group received via intraperitoneal administration 10 mg/kg of amiloride (amiloride hydrochloride hydrate, A7410; Sigma-Aldrich, St. Louis, MO, USA) in accordance with a protocol used for a mouse multiple sclerosis model [13
]. The drug dosage was calculated from the body weight of each rat, which was measured daily. The PBS group received intraperitoneal administrations of PBS using the same protocol.
At 24 h after injury, and every 12 h thereafter for three days, both groups received intraperitoneal administrations of 50 mg/kg of 5-bromo-2-deoxyuridine (BrdU, Sigma-Aldrich, B5002, St. Louis, MO, USA) in order to trace the cells proliferating after SCI.
2.3. Evaluation of Hindlimb Function
2.3.1. Basso, Beattie, Breanahan Locomotor Rating Scale (BBB) Scores
Hindlimb motor function after SCI was assessed weekly using the BBB scale [14
], which is an open-field locomotor test for rats. Locomotor behavior was evaluated immediately before injury and at approximately the same time each week for the following eight weeks after injury (n
= 5 per group).
2.3.2. Dynamic Plantar Esthesiometer
To assess changes in sensation or the development of mechanical allodynia, sensitivity to tactile stimulation was assessed using the dynamic plantar esthesiometer (DPA, UGO BASILE, catalog#37450, Monvalle VA, Italy), which is an automated version of the von Frey hair assessment. Rats were individually placed in a small, enclosed testing arena with a wire mesh floor for 5 min. The DPA device was positioned beneath the rats so that the filament always applied pressure to the center of the sole of the foot, which is defined as the midpoint between both the medial–lateral and the anterior–posterior axes of the foot. If an animal’s movement caused a different part of the foot to be stimulated, the data was disregarded and another attempt was made. When initiated, the device raised the filament to touch the plantar surface of the foot and progressively increased the force until the animal withdrew its foot, or until it reached a maximum force of 50 g. The DPA automatically recorded the force at which the foot was withdrawn as well as the withdrawal latency. Each paw was tested five times per session. This testing was performed prior to surgery, and once a week for eight weeks after SCI (n = 5 per group).
At 7, 14, 28, and 56 days after injury, intracardial perfusion fixation was performed using 0.1 M PBS and 2% paraformaldehyde (PFA) whilst the test subjects were under a 4% isoflurane anesthesia. After fixation, the spinal cord was excised and post-fixed in 2% PFA in 0.1 M PBS for two days at a temperature of 4 °C, and was cryoprotected by soaking it in 15% sucrose for five days. The epicenter was defined as the 2 mm-width of the tip of the IH impactor, and the spinal cord was transected into 3 mm segments: a center segment containing the lesion epicenter, and segments that were both rostral and caudal to the epicenter. Each segment was embedded in an optimal cutting temperature (OCT) compound (Sakura Finetek, Tokyo, Japan), frozen in liquid nitrogen, and sectioned at a thickness of 10 µm using a cryostat microtome (CM3050S, Leica Biosystems, Cincinnati, OH, USA). From this epicenter, sections of the spinal cord were selected from an area approximately 5 mm caudal to the epicenter and underwent immunohistochemical analyses.
The sections were washed three times (5 min each) with PBS, incubated for 10 min in 2N HCl for antigen activation, and rinsed in a boric acid buffer for neutralization. The sections were then washed four times with PBS, blocked for 60 min in PBS with a 5% normal goat serum at a temperature of 24 °C, and incubated overnight at 4 °C with anti-BrdU antibodies (GE Healthcare, anti-mouse; RPN202). After washing with PBS, the sections were incubated in the dark for 60 min at 24 °C, with the respective fluorescent secondary antibodies (Alexa Fluor594, anti-mouse; abcam, ab150116, 1:1000). In the dark, the sections were washed four times and incubated overnight at 4 °C with anti-NG2 (rabbit; a marker for OPCs; Millipore, Bedford, MA, USA, A1B5320, 1:500), or anti-adenomatous polyposis coli (APC) antibodies (rabbit; a marker for mature oligodendrocyte; abcam, ab15270, 1:500). The sections were washed again with PBS and incubated for 60 min at 24 °C with the respective fluorescent secondary antibodies (Alexa Fluor488, anti-Rabbit, abcam, ab150077, 1:1000). Subsequently, nuclei were stained using Vectashield with 4′,6-diamidino-2-phenylindole (DAPI) H-1500 (Vector Laboratories) and then mounted. Images were taken using a Nikon DS-Ri1 camera with mono 10-bit lenses (40 × objective Numerical Aperture (N.A.) = 0.6) on an Olympus IX70 fluorescence microscope and analyzed with Nikon NIS-elements Version 3.1 software (Nikon, Tokyo, Japan). The number of proliferating OPCs (NG2-positive cells) or mature oligodendrocytes (APC-positive cells) were quantified by counting the number of cells positive for both BrdU, and each cell marker within the entire cross-section of the dorsal funiculus; only cells with clearly recognizable nuclei and cytoplasm that had cytoplasmic staining that corresponded to each secondary antibody were scored as positive. A sample size of n = 5 rats per group and per time-point was used, and three sections were observed for each rat. Since a few immunohistochemical labels are not specific to a single cell type, we also relied on morphology to determine the cell type. For example, NG2-positive OPCs were distinguished from NG2-positive macrophages by morphology: OPCs are small cells with multiple processes, whereas macrophages are large, amoeboid cells without processes. Macrophages were not included in the NG2-positive cell counts.
2.5. Western Blot
At 7, 14, 28 and 56 days after injury, rats were anesthetized via the inhalation of 4% isoflurane, and subsequently 5 mm of the injured spinal cord (located 2.5 mm rostral and caudal to the epicenter) was microscopically dissected. The excised spinal cord tissue was immediately washed in ice-cold PBS, after which extracts were prepared using freshly formulated cell lysis buffer (50 mM Tris/HCl pH 7.4, 1 mM CaCl2, 0.5 mM phenylmethylsulfonyl fluoride (PMSF), 1% Nonidet-P (NP)-40) and sonicated using an Ultras Homogenizer VP-5x (Taitec, Saitama, Japan). All protein extraction procedures were performed on ice. These extracts were subjected to electrophoresis on a 15% sodium dodecyl sulfate polyacrylamide gel; 2.5 µg of protein solution was applied to each gel lane. After electrophoresis, the proteins were electrotransferred onto nitrocellulose membranes. The membranes were blocked with 3% bovine serum albumin in Tris-buffered saline with Tween-20 (TBST, 50 mM Tris, pH 7.6, 150 mM NaCl, 0.1% Tween-20), followed by incubation with anti-myelin basic protein (MBP, rabbit, abcam, ab40390, 1:8000) antibodies overnight at 4 °C. The membranes were washed for three hours in PBS containing 0.05% Tween-20 and incubated for 60 min at 24 °C with horseradish peroxidase (HRP)-linked anti-rabbit IgG (catalog #NA9340; GE Healthcare, Amersham, UK, 1: 12,000). Beta-actin was used as an internal control and was labeled with a mouse monoclonal anti-beta-actin antibody (Sigma-Aldrich, A5441, 1:1000). The proteins were labeled with Immobilon Western Chemiluminescent horseradish peroxidase (HRP) (Millipore, Bedford, MA, USA), and protein concentrations were quantified by densitometric scanning assays using the image analysis software CS Analyzer (Atto, Tokyo, Japan). The analysis was performed separately for each lane with a consistently sized designated area with background subtraction and normalization to actin. For each time-point, a sample size of n = 5 was used and the average was calculated for the group.
2.6. Toluidine Blue Staining and Electron Microscopy
At 28 days after injury, rats were anesthetized via the inhalation of 4% isoflurane, and intracardial perfusion fixation was performed with 2% PFA and 2% glutaraldehyde in 0.1 M PBS. After fixation, the spinal cord was excised 5 mm rostral to the epicenter and post-fixed in 2% osmium tetroxide in 0.1 M PBS. The samples were dehydrated in ethanol and embedded. Transverse ultrathin sections were cut with an ultramicrotome (LKB-2088) and examined by electron microscopy (JEM-1400, JEOL, Tokyo, Japan).
To quantify remyelination, a blinded technician was instructed to take 10 images from randomly selected areas on the periphery of the glial scar within the dorsal funiculus surrounding the epicenter of the lesion. The image files were randomized, and the primary investigator performed the quantification of myelination, which is express as G ratio = Myelin sheath thickness/axon diameter. Two ultrathin sections were used from each rat, and a sample size of n = 5 was used per group.
2.7. Statistical Analysis
The BBB scale scores were analyzed by two-way repeated-measures analysis of variance (ANOVA) with Tukey’s post hoc multiple comparison tests. Similarly, the results of the von Frey test were also analyzed by ANOVA with Tukey’s post-hoc multiple comparison tests. Data obtained through immunohistochemistry, MBP expression levels, as well as the ratio of myelin sheath thickness to the axon diameter were analyzed using Mann–Whitney U tests. Statistical significance was determined using IBM SPSS Statistics for Windows, Version 23.0 (IBM Corp., Armonk, NY, USA). Asterisks in figures indicate statistical significance (* p < 0.05, ** p < 0.01).