Adult (3 to 4 months) female GFAP-IκBα-dn (IκBα-dn) transgenic mice were generated and characterized in our laboratory . All animals, IκBα-dn and wild-type (WT) littermates (LM), were kept as a colony in a virus/antigen-free environment at the University of Miami Miller School of Medicine, Miami, FL, USA. IκBα-dn mice were obtained by breeding heterozygous IκBα-dn males with WT females. Mice were housed under diurnal lightning conditions and allowed free access to food and water.
Induction of spinal cord injury
Surgeries were performed at the Animal and Surgical Core Facility of the Miami Project to Cure Paralysis according to protocols approved by the Institutional Animal Care and Use Committee of the University of Miami. Contusion injury was induced with the Infinite Horizon Device (Precision Systems and Instrumentation LLC, Kentucky, USA). Female IκBα-dn (21.5 ± 2.7 g) and WT LM (21.0 ± 2.8 g) mice were anesthetized intraperitoneally (i.p.) using a ketamine (100 mg/kg, VEDCO Inc., Saint Joseph, MO, USA)/xylazine (10 mg/kg, VEDCO) cocktail, and a laminectomy was performed at the vertebral level T9. The contusion device was lowered onto the spinal cord at a predetermined impact force of 50 kdynes (moderate injury) and the mice were injured by a rapid displacement of the impounder resulting in a spinal cord displacement of 400 to 500 μm. Immediately after surgery, mice were sutured and injected subcutaneously (s.c.) with 1 ml lactated Ringer’s Injection USP (B. Braun, L7502, Bethlehem, PA, USA) to prevent dehydration and housed separately in a recovery room, where their post-surgical health status was observed. Thereafter, mice were returned to the conventional animal facility, where they were observed bi-daily for activity level and general physical condition. Manual bladder expression was performed twice a day until bladder function was regained. In addition, mice received s.c. prophylactic injections of antibiotic gentamicin (40 mg/kg, Hospira Inc., Lake Forest, IL, USA) for 7 days following SCI to prevent urinary tract infections. Mice were allowed 3 days, 3, 6 or 7 weeks survival.
Bromodeoxyuridine injections and tissue processing
Mice in the 7 weeks survival group were injected i.p. with bromodeoxyuridine (BrdU; 50 μg/g body weight; Sigma, St. Louis, MO, USA) once a day for 7 days starting at week 5 post-SCI and were allowed to survive for 1 more week. Then the mice, naïve, 3 days, 6 and 7 weeks survival, were deeply anesthetized and perfused through the left ventricle using ice cold 0.01 M PBS followed by ice cold 4% paraformaldehyde (PFA) in PBS. The spinal cords were post-fixed in 4% PFA followed by immersion in 25% sucrose in PBS overnight. Spinal cords were cut into 1-cm segments centered on the injury site and then embedded in optimal cutting temperature (OCT) compound (VWR International, Arlington Heights, IL, USA), frozen and cut into 10 series of 25 μm transverse cryostat sections. Sections were stored at -20°C until further use.
Antibodies used for immunohistochemical staining were rat anti-mouse CD11b (1:600, AbDSerotec, Hercules, CA, USA, MCA711 clone 5C6) and rabbit anti-NG2 (1:500, Chemicon, Billerica, MA, USA, AB5320). Isotype control antibodies were rabbit immunoglobulin (Ig)G (1:20,000, DakoCytomation, Carpinteria, CA, USA, X0903) and rat IgG2b (1:600, Biosite, Plymouth Meeting, PA, USA, IG-851125). Visualization of CD11b+ microglia-macrophages was performed using the three-step biotin-streptavidin-horseradish peroxidase technique described by Lambertsen and colleagues, 2001 . Visualization of NG2+ OPCs was performed using peroxidase-labeled “ready-to-use” EnVision+ polymer (K4300, DakoCytomation) according to the manufacturer’s instructions on spinal cord sections demasked using 0.5% Pepsin (Sigma-Aldrich, P-7012) in HCl and H2O for 10 minutes at 37°C. Sections were counterstained using Hematoxylin Gills or Toluidine blue. Isotype controls were devoid of staining (not shown).
Estimation of the total number of CD11b+ and NG2+ cells
Using an approximated stereological counting technique unaffected by shrinkage/tissue resorption , we estimated the total number of CD11b+ and NG2+ cells in the spinal cord of naïve IκBα-dn and WT mice and the total number of CD11b+ cells in IκBα-dn and WT mice that had survived 3 days and 6 weeks after SCI. Briefly, cells with a clearly identifiable H&E or Toluidine Blue stained nucleus in conjunction with a detectable immunohistochemical signal were counted on approximately 13 sections in naïve cords and at 3 days, and on 17 sections 6 weeks after injury separated by 250 μm from each animal, using a 100× objective and a 2,470 μm2 frame area stepping 150 μm/150 μm in the XY-position using the CAST Grid System from Olympus (Ballerup, Denmark). The total number (N) of cells in each animal was estimated using the formula: Estimate of N = ∑Q × (1/ssf) × (1/asf) × (1/tsf), where 1/tsf is the thickness sampling fraction (1/tsf = 1), 1/ssf the sampling section fraction (1/ssf = 10), and 1/asf the area sampling fraction (22,500/2,470) as previously described . In naïve mice and for the time point of 3 days we, for consistency, analyzed a total of 3.25 mm long piece of mouse spinal cord, 1.625 mm on pre- and post-epicenter. For the time point of 6 weeks we analyzed a 4.25 mm long piece of mouse spinal cord, 2.125 mm on both sides of the epicenter.
Estimation of the lesion and white matter volumes
The lesion volume and the white matter volume were estimated on Luxol Fast Blue serial sections counterstained with H&E using the Neurolucida software (MBF Bioscience, Williston, VT, USA) as previously described .
For BrdU immunofluorescent staining, cryostat sections were thawed at room temperature for 5 minutes, rinsed in 1X PBS, and processed for antigen retrieval using 2N HCl for 30 minutes at 37°C. The sections were then neutralized for 10 minutes in 0.1 M sodium borate (pH 8.5) and rinsed in 1X PBS. After blocking 30 minutes in 5% BSA/5% normal goat serum (NGS)/0.3% Triton X100/PBS, rat anti-BrdU antibody (1:200, Novus Biologicals, Littleton, CO, USA; diluted in 4% BSA/3% NGS/0.1% Triton X100/PBS) was applied to the sections in combination with either mouse anti-adenomatous polyposis coli (APC; clone CC1) antibody (1:500, Calbiochem, Billerica, MA, USA) or rabbit anti-NG2 antibody (1:500, Chemicon), and incubated overnight at 4°C. For triple immunostaining we used rat anti-BrdU (1:200, Novus Biologicals) and rabbit anti-Olig2 (1:500, Millipore, Billerica, MA, USA) with either mouse anti-NG2 (1:200, Millipore) or mouse anti-APC (1:500, Calbiochem). Following extensive rinses in 1X PBS, Alexa-conjugated secondary antibodies (1:500, Molecular Probe, Grand Island, NY, USA) were applied for 30 min at room temperature. Sections were finally rinsed and mounted with Vectashield (Vector Laboratories, Burlingame, CA, USA). To estimate the number of BrdU+/CC1+, BrdU+/NG2+, and total CC1+-cells following SCI, serial sections were counted using Zeiss Axiovert 200M fluorescent microscope (63X objective; Thornwood, NY, USA) and Stereo Investigator software (MicroBrightField, Williston, VT, USA) for unbiased stereological estimation of cell numbers. For each section a 50 × 50 μm counting frame and a 120 × 120 μm grid was used to count the cells at 250 μm intervals. A total number of 11 sections, centered on the lesion site, were counted. For the number of CC1+ cells in the naïve thoracic spinal cord, a total number of 5 sections were counted.
For CXCR4 immunostaining, thawed cryostat sections were fixed and permeabilized in ice-cold acetone for 10 minutes at −20°C, then rinsed in PBS and blocked for 1 hour in 10% NGS/PBS and 30 minutes in 5% BSA/PBS. Sections were then incubated overnight with rabbit anti-CXCR4 antibody (1:500, Abcam, Cambridge, MA, USA) diluted in 5% BSA/1% NGS/PBS in combination with either mouse anti-GFAP (1:500, BD Pharmingen, San Jose, CA, USA) or mouse anti-APC (1:500, Calbiochem) antibodies. Alexa-conjugated secondary antibodies (1:500, Molecular Probes) diluted in 5% BSA/1% NGS/PBS were applied to the rinsed sections for 30 minutes at room temperature. Then sections were rinsed and mounted with Vectashield with 4',6-diamidino-2-phenylindole (DAPI) (Vector Laboratories). For toll-like receptor 4 (TLR4; 1:50, Santa Cruz, Dallas, TX, USA) and TNF receptor 2 (TNFR2; 1:200, Santa Cruz), a similar protocol was used except that the sections were permeabilized and blocked in 5% BSA/5% NGS/0.3% Triton X100/PBS. Nuclei were visualized using a DAPI counterstain. Images were obtained with an Olympus FluoView 1000 confocal microscope.
Total RNA isolation
Total RNA was isolated from spinal cord samples (1.5 cm centered on the lesion site) using TRIzol reagent (Invitrogen, Grand Island, NY, USA) according to the manufacturer’s directions. Precautions were taken to preserve RNA integrity during the isolation, including rapid dissection on ice with RNase-free dissecting tools followed by flash-freezing in liquid nitrogen of the spinal cord segment sample as previously described by Brambilla and colleagues . RNA integrity was determined with the Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA, USA).
Microarray analysis and data processing
Microarray experiments were conducted at the University of Miami DNA and Microarray Core Facility (http://www.mihg.org/weblog/core_resources/2007/11/microarray-and-gene-expression.html) using Agilent Whole Mouse Genome Oligo microarrays (Agilent Technologies). Arrays were scanned at a 5 μm resolution using a GenePix 4000B scanner (Axon Instruments at Molecular Devices) and images analyzed with the software GenePix Pro 6.1 (Axon Instruments at Molecular Devices, LLC, Sunnyvale, CA, USA). Extracted data were transferred to the software Acuity 4.0 (Axon Instruments at Molecular Devices) for quality control. Features for further analysis were selected according to the following quality criteria: at least 90% of the pixels in the spot with intensity higher than background plus two standard deviations; less than 2% saturated pixels in the spot; signal to noise ratio (ratio of the background subtracted mean pixel intensity to standard deviation of background) 3 or above for each channel; spot diameter between 80 and 110 μm; regression coefficient of ratios of pixel intensity 0.6 or above. To identify significantly expressed genes the R software LIMMA (Bioconductor, open source software at http://www.bioconductor.org)  was used. “Within array” normalization was carried out with Lowess normalization and “between arrays” normalization with the “quantile” algorithm in the LIMMA package. Differential expression and false discovery rate (FDR) were assessed using a linear model and empirical Bayes moderated F statistics [18, 19]. Genes with FDR below 1% were considered statistically significant. All primary microarray data were submitted to the public database at the GEO website (http://www.ncbi.nih.gov/geo; record number: GSE46695). Selected genes were classified according to Gene Ontology category “biological process” using Onto-Express . Pathway analysis was performed with WebGestalt . Hierachical clustering was performed using GeneSpring 10.0 (Agilent Technologies). All experiments were performed in three replicates/groups/time points.
Quantitative real-time PCR
An aliquot of 2 μg of spinal cord RNA from each time point was reverse transcribed using the omniscript RT-PCR kit (Qiagen, Valencia, CA, USA) as previously described . qPCR was performed with the Rotor-Gene 3000 Real Time Cycler (Corbett Research, Valencia, CA, USA) on cDNA samples with TAQurate GREEN Real-Time PCR MasterMix (Epicentre Biotechnologies, Madison, WI, USA) as previously described  for the following genes: CXCR4 (forward primer: TGT GAC CGC CTT TAC CCC GAT AGC, reverse primer: TTC TGG TGG CCC TTG GAG TGT GAC), TLR4 (forward primer: TGC CCC GCT TTC ACC TC, reverse primer: ACC AAC GGC TCT GAA TAA AGT GT), Lingo-1 (forward primer: GAC TGC CGG CTG CTG TGG GTG TT, reverse primer: CCG GCG GCA GGT GAA GTA GTT GG), Sox17 (forward primer: CGG CGC AAG CAG GTG AAG, reverse primer: GGC TCC GGG AAA GGC AGA C), CNPase (forward primer: AGA TGG TGT CCG CTG ATGCTT AC, reverse primer: CTC CCG CTC GTG GTT GGT), CD11b (forward primer: GCC CCA AGA AAG TAG CAA GGA GTG, reverse primer: TAC GTG AGC GGC CAG GGT CTA AAG) and ICAM1 (forward primer: TGA GCG AGA TCG GGG AGG ACA G, reverse primer: GTG GCA GCG CAG GGT GAG GT). Relative expression was calculated by comparison with a standard curve after normalization to β-actin .
Spinal cords (1.5 cm centered on the injury site) were homogenized in 300 μl radio immunoprecipitation assay buffer (0.01 M sodium phosphate pH 7.2, 0.15 M NaCl, 1% NP40, 1% sodium deoxycholate, 0.1% SDS, 2 mM EDTA) supplemented with complete protease inhibitor cocktail (Roche, Indianapolis, IN, USA), incubated for 30 minutes at 4°C on an end-over-end rotator, and centrifuged at 4°C for 10 minutes at 14,000 rpm. The supernatant was then transferred to a fresh tube on ice and an aliquot was used for protein quantification using the DC Protein Assay (Biorad, Hercules, CA, USA). Equal amounts of proteins were resolved by SDS-PAGE on 10% or 15% gels, transferred to nitrocellulose membranes, and blocked in 5% nonfat milk in 0.1 M Tris buffered saline-triton (TBS-T) for 1 hour at room temperature. Membranes were probed with an antibody recognizing either proteolipid protein (PLP; mouse monoclonal, Millipore, 1:250), CXCR4 (rabbit polyclonal, Abcam, 1:500), Foxc2 (mouse monoclonal, Santa Cruz, 1:500), TLR4 (mouse monoclonal, Santa Cruz, 1:200), TNFR2 (rabbit polyclonal, Santa Cruz, 1:200), CXCR7 (rabbit polyclonal, GeneTex, Irvine, CA, USA, 1:1000) followed by horseradish peroxidase–conjugated secondary antibody (GE Healthcare, Little Chalfont, Buckinghamshire, UK, 1:2000). Proteins were visualized with a chemiluminescent kit (ECL; GE Healthcare). Blots were also probed for β-actin (mouse monoclonal, Santa Cruz, 1:500) as a loading control. The data were analyzed using Quantity One software (Biorad).
One-way or two-way analysis of variance (ANOVA) followed by the appropriate post hoc test and Student’s t-test (one-tailed and two-tailed). Statistical analyses were performed using Prism 4.0b software for Macintosh, GraphPad Software, San Diego, CA, USA, http://www.graphpad.com. Data are presented as mean ± SEM. Statistical significance was established for P < 0.05.