Animals and surgeries
All animal studies were approved by the University of California, San Diego Institutional Animal Care and Use Committee (Protocol No.: S01193). The study design is outlined in Figure . Twelve-week-old Female SD rats were used. The rationale for choosing female rats was based on our previous experience which demonstrates better tolerability of female rats to spinal trauma-related side effects, such as urinary retention. Animals were anesthetized with isoflurane (5% induction, 1.5% to 2% maintenance; in room air) and placed into a Lab Standard Stereotaxic frame (Stoelting, Cat# 51600, Wood Dale, IL, USA). The animal was elevated 2 cm by placing it on a homeothermic heating blanket (set at 37°C with feedback from a rectal thermometer (Harvard Apparatus, Cat# 507214, Holliston, MA, USA) which sits on a plastic rectangular block. The animal was then placed in Spine Adaptors (Stoelting, Cat# 51695, Wood Dale, IL, USA) and a wide Th13 laminectomy was performed using an air-powered dental drill and binocular microscope (exposing the dorsal surface of spinal segment L3). An acrylic rod (Ø 2.9 mm, length 15 cm; 35 g) was then slowly lowered onto the exposed L3 segment until it slightly touched the spinal cord but without inducing any compression. The laminectomy site was then filled with mineral oil in which the tip of a small thermocouple (Physitemp, Cat# IT-14, Clifton, NJ, USA) was submerged and touched the dura. The light from the two fiber optic light pipes of the surgical light (Fiber-Lite, Cat# MI-150 & BGG1823M, Dolan-Jenner, Boxborough, MA, USA) was focused on the surgical site (and directly illuminating the temperature probe). Next, the light intensity was manually regulated so that the spinal cord/mineral oil was warmed to 37°C and remained at 37 ± 0.3°C. If necessary, a 100 W infrared lamp was used to gradually adjust and maintain the animal’s core temperature at 37°C (rectal). When both temperatures (that is, paraspinal and rectal) were at 37 ± 0.3°C for at least five minutes, the rod was slowly lowered until its weight fully rested, perpendicularly, on the spinal cord. The rod was kept in place for 15 minutes, while both temperatures were maintained at 37 ± 0.3°C. After spinal compression, the rod and mineral oil were removed and the wound sutured in anatomical layers.
Figure 1 Schematic diagram of experimental design. A: To induce spinal cord injury, a 35 g circular rod was placed on the exposed L3 spinal segment and the spinal cord compressed in the dorso-ventral direction for 15 minutes. B: Three days after injury, the animals (more ...)
Buprenorphine (0.05 mg/kg, s.c., Reckitt Benckiser, Richmond, VA, USA), 5 mL of lactated Ringer’s, 10 mg/kg of cefazolin (Novaplus/Sandoz, Holzkirchen, Germany) and standard triple antibiotic ointment to cover the incision site (bacitracin, neomycin, Polymyxin B) was given after every surgery. Bladders were manually emptied twice daily (if full). Sulfamethoxazole and trimethoprim USP oral suspension (200 mg and 40 mg per 250 mL drinking water, Hi-Tech Pharmacal, Amityville, NY, USA) was given for at least 10 to 14 days after spinal cord injury (SCI) or until autonomic bladder voiding occurred and for 1 to 2 days after any other surgery (sham or grafting). Food was provided by placing it at the bottom of cage and water bottles with an elongated drinking tube were used, until regular overhead supplies could be reached by the animal. Animals diagnosed with bacterial infections throughout the study were treated with sulfamethoxazole (as above), 10 mg/kg/day of cefazolin, and lactated Ringer’s 5 mL/0.5 day.
Cell derivation and preparation
The cells, named ‘NSI-566RSC’, were produced by Neuralstem Inc. (Rockville, MD, USA), as described before [33
]. Briefly, human spinal cord neural precursors (HSSC) were prepared from the cervical-upper thoracic region obtained from a single eight week fetus. The fetal tissue was donated by the mother in a manner fully compliant with the guidelines of NIH and FDA and approved by an outside independent review board and by the University of California, San Diego Human Research Protection Program (Project# 101323ZX). Meninges and dorsal root ganglia were removed and dissociated into a single cell suspension by mechanical trituration in serum-free, modified N2 media (human plasma apo-transferrin, recombinant human insulin, glucose, progesterone, putrescine, and sodium selenite in (Dulbecco’s) modified Eagle’s medium ((D)MEM)/F12). For growth of the HSSC, 10 ng/ml basic fibroblast growth factor (bFGF) was added to the modified N2 media and expanded serially as a monolayer culture on poly-D-lysine and fibronectin [34
]. Approximately 6.1 × 106
total cells were obtained upon the initial dissociation of the spinal cord tissue. The growth medium was changed every other day. The first passage was conducted 16 days after plating. At this point, the culture was composed mostly of post-mitotic neurons and mitotic HSSC. Mainly the mitotic cells were harvested through brief treatment with trypsin and subsequent use of soybean trypsin inhibitor. The cells were harvested at approximately 75% confluence, which occurred every five to six days (20 passages). At various passages, the cells were frozen in the growth medium plus 10% dimethyl sulfoxide at 5 to10 × 106
cells/ml. The frozen cells were stored in liquid nitrogen. Upon thawing, the overall viability and recovery was typically 80% to 95%. A cell bank of passage 16 cells was prepared and used for this study.
For the production of eGFP-labeled NSI-566RSC, a Lentiviral vector was constructed containing the human Ubiquitin C promoter driving expression of enhanced GFP. Viral particles produced by infected 293FT cells were collected after overnight incubation, then concentrated by centrifugation and stored frozen. Neural stem cell cultures were infected by overnight incubation in growth medium supplemented with viral supernatant. Infected stem cells were washed with phosphate-buffered saline (PBS) and cultured as described above. After multiple passages, >90% of the cells were GFP positive (assessed after immunohistochemical staining). A cell bank of passage 17 cells was prepared and used for this study.
One day prior to each grafting day, one cryopreserved vial of the previously prepared cells was thawed, washed, concentrated in hibernation buffer, and shipped from the cell preparation site (Neuralstem, Inc., Rockville, MD, USA) to the surgery site (University of California, San Diego, CA, USA) at 2 to 8°C by overnight delivery. Upon receipt the following day, the cells were used directly for implantation without further manipulation. Before and after implantation, the viability of cells was measured with trypan blue (0.4%; Sigma-Aldrich, St. Louis, MO, USA). Typically, a >85% viability rate was recorded.
Inclusion and exclusion criteria, randomization and blinding
Three days following SCI and prior to grafting, animals were randomly divided into three groups: the vehicle-injected group, non-injected group, or the HSSC-injected group. SCI animals with an open-field locomotion score of ≤1 and appearing healthy enough were included. Animals found moribund or automutilating at any point during the study were excluded and euthanized. A total of 42 animals were employed and divided into 6 experimental groups, as follows:
Group A (n = 14): SCI animals-NSI-566RSC-grafted,
Group B (n = 10): SCI animals-vehicle-injected,
Group C (n = 8): SCI animals-non-injected,
Group D (n = 6): sham operated (laminectomy only),
Group E (n = 6): naïve animals (no surgical manipulation)
Group F (n = 2): SCI athymic animals-ubiquitin.eGFP+ NSI-566RSCs-grafted.
One animal was excluded in Group A because of automutilation of the hind paw; two animals were excluded in Group C, one because of automutilation of the hind paw and one because of bacterial infection. Six animals had been replaced before dosing/randomization, five due to inadequate injuries and one because of bacterial infection.
For the intraparenchymal injections, the animals were placed in the stereotactic frame as described above. The L3 spinal cord (that is, the dura mater) was then re-exposed at the previous laminectomy site. Injections were performed using a 33 gauge beveled needle and 100 μL Nanofil syringe (World Precision Instruments, Cat# NF33BV and Nanofil-100, Sarasota, FL, USA) connected to a microinjection unit (Kopf Instruments, Cat# 5000 and 5001, Tujunga, CA, USA). The duration of each injection was ≥45 seconds followed by a ≥30 second pause before slow needle withdrawal. The center of the injection was targeted intermediate of the ventral and dorsal horn and close to the lateral funiculus (distance from the dorsal surface of the spinal cord at the L3 level: 0.80 mm). Twelve injections (20,000 cells/μL) were done; four injections (0.5 μL each, 0.8 to 1.0 mm apart, rostrocaudally) at each lateral boundary of the injury (eight in total), plus two (bilateral) injections (0.5 μL each) 1.5 mm caudal from the previous, most caudal injections, and two injections at the core of the epicenter (1 μL at each side of the dorsal vein, bilaterally; see diagram in Figure ). After the injections, the incision was cleaned with penicillin-streptomycin solution and sutured in two layers.
Two days after injury (that is, one day before grafting), a methylprednisolone acetate (Depo-Medrol, 10 mg/kg, i.m.) was given, which was repeated thereafter three times with 1 mg/kg/week i.m. Starting directly after grafting, all animals received 1.5 mg/kg/BID s.c. of tacrolimus (Prograf/FK506, Astellas, Deerfield, IL, USA) until the end of the study. For post-transplant days 0 to 10, the animals also received 30 mg/kg/day s.c. of mycophenolate mofetil (CellCept, Genentech, CA, USA). Immunosuppression was also given to the non-grafted Sprague–Dawley animals (that is, the naïve, sham operated, and all SCI-control animals).
Open field locomotion testing
Locomotion recovery after spinal cord contusion injury was monitored using a modified BBB open field locomotion rating scale [35
]. The BBB score was modified to reflect the distinct locomotor recovery stages observed after L3 SCI. The modified score entailed eight well-defined degrees of locomotor recovery: 0 to 1: are identical to the BBB-score, 2: is cumulative score of 2 and 3 of the BBB score, 3: is cumulative score of 4, 5 and 6 of the BBB score, 4: is cumulative score of 7 and 8 of the BBB score, 5: reflects weight support with poor paw clearance, 6: is broadened and/or shortened stepping, and 7: is normal walking. In the present study, the locomotor score was obtained before grafting and weekly after injury until the end of the study (that is, 8.5 weeks post-injury). In addition to a modified BBB score, a regular full 21 scale BBB score was periodically assessed.
The CatWalk apparatus (CatWalk 7.1, Noldus Technology, Wageningen, The Netherlands) was used to quantify gait parameters during walkway crossings (for example, paw positioning, base of support, stride length, front limb versus hind limb coordination) by footprint analysis [36
]. Animals had to walk down a horizontal glass walkway (109 × 15 × 0.6 cm, L × W × H), the glass of which is illuminated along the long edge. At the end of the walkway, animals had access to their home cage and were given a treat upon arrival (Certified Supreme Mini-Treats™, Cat# F05472-1, Frenchtown, NJ, USA). The light only enters the (side of the) glass and reflects merely internally (when the glass is bordered by air). As an animal walks on the glass walkway, light reflects off of the animal’s paws, producing a series of bright footprints when viewed through the glass, from below the walkway. The illuminated footprints were then recorded by a video camera with a wide-angle objective that was located underneath the elevated glass walkway. In order to get an optimal contrast between the paws and the surroundings; the test was performed in a room that was totally darkened. The animals were trained for smooth walkway crossing on the five days prior to the video acquisitions. To obtain accurate and meaningful data, the following criteria concerning walkway crossings needed to be met: (1) the animal needed to walk uninterrupted across the walkway, at a constant pace and (2) a minimum of three such crossings per animal were required. Animals without bilateral paw clearance could not be analyzed (n = 4 control-SCI animals, and 3 HSSC-treated animals). Digital data analysis consisted of assigning labels (left-fore, left-hind, right-fore, or right-hind) to the animal’s paw prints in a recorded walkway crossing, using dedicated CatWalk software. Next, the software calculated gait parameters. Data from three proper crossings were averaged for statistical analysis.
Inclined ladder test
The inclined ladder test was performed as described before [37
]. An inclined ladder (55°) with twenty 120 mm wide rungs (diameter: 1/4″), spaced at equal intervals (60 mm) and having 150 mm-high side walls was used. The rats were trained for this test so that smooth runs were recorded. At the end of the ladder, the animals had access to their home cage and received a treat (as above). The rats were placed at the bottom, and in front, of the ladder. The bottom of the ladder was placed on a 20 cm elevated platform. Climbing was video recorded from a position below the ladder, so that the ventral aspect of the animal is recorded. All animals were able to climb up the ladder. The correct placing of a hind paw and sustained position until its next forward move was counted over 18 rungs (placement on first and last rung were not counted).
Single frame hind limb motion analysis
Two parameters were measured in bilateral video captures of animals crossing a runway: the foot-stepping angle (FSA) and the rump-height index (RHI), as described before [37
]. The FSA is the angle at which the hind paw is placed on the ground just after the swing phase. The angle is defined by a line parallel to the dorsal surface of the paw and a horizontal line behind the paw. Four to six measurements were made for each hind limb (a total of 8 to 12 step cycles). The RHI was defined as the highest point of the base of the tail during the (recorded part of the) run. The values for the left and right paw of each animal were averaged. The elevated runway bar was made of a wooden plate/beam (1500 × 150 × 20 mm, L × W × H). The animals were trained to smoothly walk the beam. Once more, at the end of the beam the animals had access to their home cage and received a treat (as above). The videos (that is, the selected frames) were selected and analyzed using the video tool VirtualDub 1.9.11 (Written by Avery Lee, http://www.virtualdub.org
) and the on-screen measurement tool Screen Ruler V1.0.1a (http://www.caveworks.net
Myogenic motor evoked potentials
Animals were anesthetized with ketamine (80 mg/kg i.p., Ketaset, Fort Dodge Animal Health, Overland Park, KS, USA). Myogenic motor evoked potentials (MEPs) were elicited by transcranial electrical stimulation (with a pulse duration of 1 ms at 7 mA using a DS3 constant current isolated stimulator (Digitimer LTD., Welwyn Garden City, UK) of the motor cortex using two percutaneously placed 30G stainless steel stimulation electrodes. Responses were recorded from the gastrocnemius muscle using 30G platinum transcutaneous needle electrodes (distance between recording electrodes approximately 1 cm; Grass Technologies, Astro-Med, Inc., West Warwick, RI, USA). Recording electrodes were connected to an active headstage (3110 W Headstage, Warner Instruments LLS, Hamden, CT, USA) and signal amplified using a DP-311 differential amplifier (Warner Instruments LLS). An amplified signal was acquired by the PowerLab 8/30 data acquisition system (AD Instruments, Inc., Colorado Springs, CO, USA) at a sampling frequency of 20 kHz, digitized and stored in a PC for analysis. MEPs were measured until the three to five highest (stable) recorded potentials were similar. Those traces were averaged per animal and multiplied by one thousand (μV; all values >1). Next, for data normalization, a logarithmical transformation was applied for further analysis (amplitudes of MEP traces tended to vary much more in animals with higher MEPs amplitudes).
Measurement of muscle spasticity
At 1.5 weeks and 2 months post-injury, the presence of muscle spasticity in the lower extremities was measured using a previously described system [39
]. Briefly, fully awake animals were placed in a restrainer and a hindpaw was taped to a rotational metal plate driven by a computer-controlled stepping motor. The metal plate is interconnected loosely to the ‘bridging’ digital force transducer (LCL454G, 0–454 g range; Omega, Stamford, CT, USA). The resistance of the ankle to dorsiflexion was measured during stepping motor-driven ankle dorsiflexion (40°; MDrive 34 with onboard electronics; microstep resolution to 256 microsteps/full step; Intelligent Motion Systems, Marlborough, CT, USA) at three different ankle-rotational velocities (40, 60 or 80°/second). The electromyography (EMG) signal was recorded from the ipsilateral gastrocnemius muscle during the same time frame. To record EMG activity, a pair of tungsten electrodes was inserted percutaneously into the gastrocnemius muscle 1 cm apart. EMG signals were bandpass filtered (100 Hz to 10 kHz) and recorded before, during, and after ankle dorsiflexion. EMG responses were recorded with an alternating current-coupled differential amplifier (model DB4; World Precision Instruments, Sarasota, FL, USA). EMG was recorded concurrently with ankle resistance measurements, both with a sample rate of 1 kHz. Both muscle resistance and EMG data were collected directly to the computer using custom software (Spasticity version 2.01; Ellipse, Kosice, Slovak Republic). Each recorded value was the average of three repetitions. The presence of spasticity response was identified as an increased ankle resistance and concurrent increase in recorded EMG activity during computer-controlled ankle dorsiflexion. To measure the contribution of the ‘mechanical’ component in the measured resistance (that is, caused by ankle ankylosis in chronically paraplegic animals), animals were anesthetized with isoflurane at the end of each recording session and the relative contribution of the neurogenic (that is, isoflurane-sensitive) and the mechanical (that is, isoflurane non-sensitive) component identified. The magnitude of the anti-spasticity effect was then expressed as the maximum possible anti-spasticity effect measured under isoflurane anesthesia minus the value of the mechanical component.
Recovery of sensory function was assessed through quantification of supraspinal ‘above-level’ escape response (AL-ER; that is, an escape or escape-attempt with incorporation of the forelimbs) thresholds to 1) a gradually increasing force to the hind paws (using the Analgesy-Meter, no disc weights added; Cat# 37215, Ugo-Basile, Collegeville, PA, USA), and 2) AL-ER latencies to a constant heat stimulus (intensity 17, cut-off at 30 seconds) to the hind paws (using a constant infrared heat source; Cat# 37360, Ugo-Basile,). The hind paw tested was gently restrained by the investigator to prevent withdrawal. For the heat perception test the apparatus was switched on ≥15 minutes prior to testing, to allow it to warm up.
For the AL-ER tests, both hind paws were tested four times, alternately, for each test, with a testing interval of ≥1 hour. No more than four measurements per day were performed, rendering two testing days per test. Maximum cut-off values for the stimuli or latency were at approximately two times that of the response threshold of uninjured animals, to prevent tissue damage. Prior to (one week) and during the experimental period, the animals are extensively habituated to the experimenter so that the animals can be held upright (loosely) during all sensory assessments. Habituation consists of picking the animal up and holding/handling it twice daily for ≥3 minutes. Subsequently, in the absence of a stimulus, animals only rarely showed escape behavior when held for the time it would take to reach cut-off values. We measured the AL-ER thresholds/latencies before injury (baseline) and every second week after injury. The final measurement was done at eight weeks post-injury. Two or less (out of the total of eight, bilateral) measurements could manually be assigned as outliers and be excluded per time point (done while blinded for time point, animal, and treatment group). In addition, individual scores were log transformed before analysis and we calculated the Maximal Possible Effect, using these log scores, as previously suggested [40
]. Hence, we used the standard formula to calculate the Maximal Possible Effect, and assuming a logarithmic relation between stimulus intensity and perceived intensity:
Here, xy is the average AL-ER threshold of an individual animal at time point y (either for a thermal or mechanical stimulus).
Magnetic resonance imaging
Eight weeks after cell grafting, rats were deeply anesthetized with 2 mg pentobarbital and 0.25 mg phenytoin (0.5 mL of Beuthanasia-D, Intervet/Schering-Plough Animal Health Corp., Union, NJ, USA) and transcardially perfused with 200 ml of heparinized saline followed by 250 ml of 4% paraformaldehyde (PFA) in PBS. A 3 cm piece of the vertebral column (Th8-L1) was placed in a tight small latex container filled with 4% PFA to prevent the formation of air bubble/tissue interface artifacts. Samples were scanned using Magnetic Resonance Imaging (MRI). Images were acquired using a 7 Tesla Bruker (Bruker Biospin Billerica, MA, USA) horizontal bore small animal magnet and a 2.5 cm imaging volume transmit/receive coil. A 3D turboRARE sequence was used with the following imaging parameters: echo time/repetition time 45/1500 ms, flip angle 180 degrees, field of view 16 × 16 × 16 mm, matrix 256 × 256 × 70 with a resulting voxel size of 62 × 62 × 229 microns. The imaging time was 84 minutes per sample.
Volume reconstructions and calculations were done using Amira software (Visage Imaging GmbH, Berlin, Germany).
Axon counting in plastic semi-thin sections
After MRI imaging, spinal cords were dissected from the spine and a transverse (1.5-mm-thick) spinal cord block cut from the injury epicenter and prepared for plastic embedding as previously described [41
]. Briefly, dissected tissue blocks were treated with 0.1% osmium tetroxide in 0.1 M non-saline phosphate buffer (pH 7.4) for 12 hours, followed by adequate rinsing in non-saline phosphate buffer. This was followed by progressive alcohol dehydration according to standard procedures up to 100% ethanol, with the addition of further dehydration in a 1:1 solution of ethanol/propylene oxide, and lastly in 100% propylene oxide. Dehydrated blocks were then prepared for resin infiltration by incubation in a 1:1 solution of resin/propylene oxide on a rotator in a fume hood overnight. The resin solution used consisted of: Eponate 12, Araldite 502, dodecenyl succinic anhydride, and 2,4,6-tri (dimethylamino-methyl) phenol (DMP-30; Ted Pella, Inc., Redding, CA, USA), mixed in ratios of 10:10:25:1, respectively. The blocks were then transferred to 100% resin for subsequent overnight infiltration on a rotator. Finally, the tissue blocks were embedded using fresh resin in multi-chamber silicone rubber molds made from a Silastic® E RVT Silicone Rubber Kit (Dow Corning Corp., Midland Township, MI, USA). The molds with embedded sections were placed in an oven at 60°C for 1 day to facilitate resin polymerization. Semi-thin (1 μm) transverse sections were then cut using a microtome (Leica Supercut RM 2065) with a 8-mm diamond knife (Histo Diamond Knife, Cat# LM 7045, DiATOME, Hatfield, PA, USA). The sections were mounted on slides with distilled water and allowed to dry on a slide warmer. Prior to staining, the slides were incubated at 60°C in an oven for 10 to 15 minutes and then contrast-stained with 4% para-phenylene-diamine (PPD).
Mosaic images were taken of two sections per animal at 20X using a Zeiss Imager. M2 fitted with a Zeiss MRm camera (Carl Zeiss Microscopy, Thornwood, NY, USA), a BioPrecision2 stage (Cat# 96S100, Ludl Electronic Products, Hawthorne, NY, USA), and Stereo Investigator software (MBF Biosciences, Williston, VT, USA). Complete mosaic images were loaded into ImageJ 1.45s. Axonal quantification involved manual definition of pixel threshold (0 to 255, grayscale; using the Triangle method). Next, ImageJ’s Analyze Particles option was used to find particles with a size of 0.20 to 250 μm2 and a circularity of 0.5 to 1.0 (which corresponded to axons). All acquisition and analysis values were held consistent throughout the study. Final measurements acquired were the minimal diameter (Feret’s) of each particle (and particle counts). Particles with a minimum diameter >10 μm were excluded. Employment of this parameter allowed for further axonal analysis, in which axons were divided into empirically-derived caliber sizes of small, medium, and large axons (0.3 to 1.0 μm, 1.0 to 2.5 μm, and 2.5 to 10 μm, respectively). Data were acquired per spinal region (that is, dorsal, ventral, and lateral funiculi).
After removing the 1.5 mm block from the spinal cord at the injury epicenter, the remaining caudal and rostral parts of the spinal cord (±1 cm each) were placed in 30% sucrose for cryoprotection for a minimum of five to seven days. Transverse spinal cord sections were then prepared from the L6 segment. The segment(s) in between the L6 and the injury epicenter and the one rostral to the injury epicenter were sectioned coronally and used for identification of grafted human cells. All sections were cut on a cryostat and stored free-floating in PBS with thimerosal (0.05 wt%). Sections were stained overnight at 4°C with primary human-specific (h) or non-specific antibodies in PBS with 0.2% Triton X-100: mouse anti-nuclear mitotic apparatus (hNUMA, 1:100; Millipore, Billerica, MA, USA), mouse anti-neuron specific enolase (hNSE, 1:500; Vector Labs, Burlingame, CA, USA), mouse anti-synaptophysin (hSYN, 1:2,000; Millipore), rabbit anti-glial fibrillary acidic protein (hGFAP, 1:500; Origene, Rockville, MD, USA), mouse anti-neuronal nuclei (NeuN, 1:1,000; Millipore), chicken anti-GFP (1:1,000; Aves Labs, Tigard, OR, USA), rabbit anti-anti-glutamate decarboxylase 65 and 67 (GAD65 and 67; 1:300; Millipore), mouse anti-GFAP (Cy3-labeled; 1:500; Sigma-Aldrich; St. Louis, MO, USA), rabbit anti-Ki67 antibody (mitotic marker, 1:100; Abcam, Cambridge, MA, USA), goat anti-doublecortin (DCX, 1:1000, Millipore), goat anti-choline acetyltransferase (CHAT, 1:50, Millipore/Chemicon), and rat anti human axonal neurofilament antibody (hHO14; 1:100; gift from Dr. Virginia Lee; University of Pennsylvania, Philadelphia, PA, USA). Mouse anti-growth associated protein 43 (GAP43, 1:16,000; Millipore), rabbit anti-calcitonin gene-related peptide (CGRP, 1:1,000; Biotrend, Destin, FL, USA), and rabbit anti-ionized calcium binding adaptor molecule 1 (Iba1, 1:1,000; Wako, Richmond, VA, USA), were used on the L6 transverse sections. Following washing in PBS for three to five minutes, sections were incubated with fluorescent-conjugated secondary donkey antibodies (Alexa® Fluor 488 & 647; 1:500; Jackson Immuno Research, West Grove, PA, USA; and Alexa® Fluor 555, 1:500; Invitrogen, Carlsbad, CA, USA). Sections were then mounted on slides, dried at room temperature, and covered with Prolong anti-fade kit (Invitrogen).
Confocal images (1024 × 1024 pixels) were captured with a Fluoview FV1000 microscope (Olympus, Center Valley, PA, USA) with a 20X or 40X objective, optical section spacing of 0.5 μm, and pulse speed of 20 μsec/pixel. Other images were taken using the Zeiss Imager. M2 setup as described above, using a 10, 20 or 63X magnification. CGRP, GAP43, and Iba1 stainings on L6 transverse sections were quantified using densitometry measurements of the main dorsal horn region (Laminae I through IV; area as marked in Figure B). ImageJ software was used for quantification by using the Background Subtraction function.
Figure 2 Significant decrease in the dorsal horn CGRP immunoreactivity caudal to the injury epicenter in SCI-HSSC-treated versus SCI-control animals. CGRP- (A), GAP-43- (B), and Iba1- (C) immunoreactivity in the dorsal horns (DH) caudal of the injury epicenter (more ...)
Behavioral data were analyzed using analysis of variance (ANOVA) one-way, or two-way group × time repeated measures, using a fixed-effect model, and a Bonferroni post hoc test for multiple comparisons). A P value of 0.05 was considered significant. Unequal variances were explored prior to using ANOVA analyses using the Bartlett’s test, but were not identified. Post hoc tests were only calculated if overall group differences were found. Results are expressed as means with the standard error of the mean (SEM). To analyze differences between the two groups (for example, vehicle injected versus non-injected SCI animals), we used Student’s t-tests (unequal variances were explored with the F-test, but not found) or repeated measures ANOVA. Naïve and sham operated animals were grouped (and named ‘non-injured’) in all outcomes besides the sensory tests. All statistical analyses were done using GraphPad Prism (La Jolla, CA, USA), SPSS statistics 17 (for K-Means clustering; IBM, Armonk, NY, USA), or STATA 12 (for precise post-hoc test P-value calculations; StataCorp LP, College Station, TX, USA) and performed two-tailed.