Adult female Sprague Dawley (SD) rats (10-12 weeks) were used under the guidelines of the National Health and Medical Research Council of Australia and approved by the Animal Welfare Committee of Flinders University of South Australia.
Animals were divided into 2 experimental groups, both of which received a SNI (day 0) followed by a dorsal column cut (DCC, day 7). The timing of the CNS injury relates to the optimal conditioning of DRG neurons, known to occur seven days after SNI, resulting in the maximum amount of CNS regeneration possible . Control animals (n = 10) received intravenous (iv) tail vein injections of sterile saline alone (2 ml/injection). Test animals (n = 10) received iv tail vein injections of liposome-encapsulated clodronate (2 ml/injection). In both of these two groups sterile saline was administered as the vehicle during iv delivery and the only difference between them was the use of liposomes in the treatment group. Note that control animals did not receive liposome-encapsulating saline given the possibility that liposomes alone may "block macrophage phagocytosis by saturation and may suppress or activate other macrophage functions" . Consequently, this would have provided inconsistent comparisons between liposome-treated and control group animals.
Sciatic Nerve Injury (SNI)
For all surgical procedures, the toepinch-reflex test was used to determine effectiveness of the anaesthetic prior to surgery using a mixture of ketamine (100 mg/kg) and xylazine (100 mg/kg) delivered intraperitoneally. Briefly, a primary longitudinal cut was made on the skin overlaying the femur of the left hind limb. The incision was extended proximally and distally exposing the thigh muscle. Sharp surgical scissors were inserted and opened into the muscle through the first layer to the level at which the sciatic nerve runs. After locating the branching of the sciatic nerve, the nerve was ligated proximal to its trifurcation and cut below the ligation site with fine surgical scissors. The wound was sutured closed using a 6/0 surgical silk suture and animals were placed into individual cages.
Dorsal Column Cut (DCC)
After laminectomy at T9-T10 and a small incision in the dura mater, the dorsal columns of the spinal cord were crushed with iris scissors inserted at a depth of approximately 1.5-2 mm (marked on the scissors' tip) . A sharp scalpel blade was passed through the wound twice to confirm bilateral DCC . A small piece of gelfoam was temporarily placed over the lesion site to encourage blood clotting and the overlaying muscles were sewn together with a 6/0 surgical suture. The skin was stapled closed and the animals were returned to their cages. After surgery, animals were housed separately and received subcutaneous injections of the analgesic drug buprenorphine (0.03 mg/kg) for a period of up to 5 days to alleviate postoperative pain. For all spinal cord injured animals manual bladder expression was performed two to three times per day and antibiotics were administered if required .
Retrograde Tracer Injection
To investigate the presence of ascending regenerated fibres in the injured spinal cord (i.e. across the injury epicentre), a somatic retrograde tracer Fast Blue (FB, 5% in saline, Sigma) was injected into the dorsal column of the proximal stump, 3-4 mm rostral from the site of spinal cord injury (SCI; injection depth 1-1.1 mm, total volume delivered 0.1 μl). FB was administered 2 weeks after CNS lesion, using a stereotaxic frame and a pulled glass micropipette needle. Note that to alleviate any pain caused by tracer injections into the spinal cord, all animals received injections of buprenorphine (0.03 mg/kg) for up to 3 days after injection.
Anterograde Tracer Injection
To further examine the presence of ascending CNS fibres regenerating across the SCI epicentre, the anterograde tracer Biotinylated Dextran Amine (BDA, 10% in saline, 10000 mw, Molecular Probes) was injected into the dorsal column in the lumbar region of the spinal cord [26–28]. BDA was delivered 2 weeks after CNS lesion as described for retrograde tracing (injection depth 1-1.1 mm, total volume delivered 0.1 μl). All care was taken to keep tracer injections within the dorsal column of the spinal cord. Some injected cords were randomly selected to confirm tracer deposition, however, not all injections sites were examined and further damage to the cord at these sites was not investigated in this study.
Liposome Preparation and Administration
Liposomes were prepared according to previous studies [29–32]. Briefly, 86 mg of egg phosphatidylcholine and 8 mg of cholesterol were dissolved in 5 ml of chloroform in a round-bottom flask. Chloroform was removed by using a low-vacuum rotary evaporator at 37°C to form a thin lipid film around the flask. The lipid was then dispersed with 10 ml sterile phosphate buffered saline (PBS, 0.1 M, pH 7.4) containing 2.5 gm of clodronate (dichloromethylene-diphosphonate-DMDP, Sigma) and incubated on a gentle stirrer at room temperature (RT, 2 hrs). After incubation, the suspension was sonicated (50 Hz) at RT (3 min) and incubated again at RT with no stirring to allow for liposome formation (2 hrs). Liposomes were centrifuged at 10,000 g (15 min) at RT to remove any free clodronate. The remaining pellet was washed twice in sterile PBS at 20,000 g at RT (30 min) and resuspended in 4 ml sterile PBS to be used immediately [29, 30, 33].
For all test animals, liposomal delivery was administered on three separate occasions, once: i) three days prior to; ii) immediately after; and iii) four days after SNI. The specific timing of liposome administration ensured macrophage depletion commencing 3 days prior to SNI and ending 3 days after DCC. It is known that macrophage cells are depleted within 24 hrs after liposomal treatment and begin to slowly repopulate approximately 5-7 days after last administration of clodronate (personal communication with liposome pioneer Dr Nico van Rooijen, Netherlands). No adverse side effects were observed in any of the liposome-treated animals.
Perfusion and Cryosectioning
All animals were injected intraperitoneally with 5% chloral hydrate in distilled water and perfused transcardially with 1% NaNO2/phosphate buffer (PB, 0.1 M, pH 7.4) followed by a 4% paraformaldehyde (PFA)/PB. Perfusions were performed 4 weeks after the CNS lesion with all dissected tissues post-fixed in 4% PFA and cryoprotected in 30% sucrose/PB at 4°C (48 hrs). Spleens were cryosectioned at 20 μm (cross-sections), DRG at 20 μm (coronal sections) and spinal cords at 40 μm (longitudinal sections). All specimens were mounted on 2% gelatine-coated glass slides.
For IHC, DRG and spinal cord sections were washed in 0.5% H2O2/50% ethanol at RT (30 min) to quench endogenous peroxide activity, rinsed in PBS, followed by three washes in PBS containing 1% Tween-20 detergent (PBST). These sections were blocked in 20% normal horse serum (NHS, 2 hrs) before incubation with primary antibodies at 4°C (48 hrs). The primary antibodies used were: mouse-anti-rat cluster differentiation 68 (CD68, macrophage, 1:400, Serotec), rabbit-anti-glial fibrillary acidic protein (GFAP, astrocyte/satellite cells, 1:500, Dako). A combination of the following secondary antibodies for single and/or double labelling included: sheep-anti-mouse-cy3-IgG (1:500, Jackson), donkey-anti-mouse-cy2-IgG (1:500, Jackson), donkey-anti-rabbit-488-IgG (1:500, Jackson), sheep-anti-rabbit-cy3-IgG (1:500, Jackson). The specificity of the observed IHC procedure was validated by omitting the primary antibody and/or by using a non-immune serum instead of the primary antibody .
BDA injected tissue was treated in 3% H2O2/100% methanol at RT (10 min), rehydrated in PBS and thoroughly washed in PBST. This was followed by incubation with streptavidin-HRP conjugated antibody (1:2000, Vector Laboratories) in PBST at RT (60 min). After extensive washing with PBST, sections were developed in a solution containing 0.05% 3'3-diaminobenzidine tetrahydrochloride (DAB, Sigma), 0.06% NiSO4 and 0.005% glucose oxidase [26, 34].
Relative DRG Somatic Count
The relative number of retrograde labelled FB+ DRG neurons was obtained by serially counting every fourth section from the ipsilateral and contralateral L4 DRG (total of 5 sections/animal, n = 10). This method of counting allowed for the relative estimation of FB labelled cell bodies and to avoid the possibility of double counting, only those neurons with visible nuclei were counted [35, 36]. Note that in all instances, FB+ neurons were considered regenerated neurons .
Quantification of Ascending Fibre Regeneration
The number of BDA-labelled fibres was counted at both stumps of the spinal cord including the lesion epicentre (0 mm) and 1 mm rostral and caudal. As we have previously described , the axon index was calculated as a percentage of every fourth section (total of 10 sections/animal, n = 6). Note that no BDA-labelled fibres were found rostrally and/or at the epicentre in liposome-treated animals.
Cellular Quantification Method
Due to the complexity in identifying individual populations of macrophage cells (CD68+) and astrocytes (GFAP+) present in the spinal cord after injury, we determined the percentage area fraction of the section occupied by these stained structures [31, 38, 39]. Briefly, using ImageJ (image processing program, NIH version 1.37) 20× (697.68 × 522.72 μm) magnification images immunostained against the aforementioned antibodies were converted to binary contrast images (black and white). This provided a threshold by subtracting background levels from the immunoreactive stained areas and allowed the determination of the percentage area fraction per image to be collected, tabulated and statistically analysed [22, 38, 40, 41].
Sandwich BDNF ELISA Test
According to manufacturer's instructions (Millipore.com), serum samples collected at the end of the experimental period were analysed for BDNF concentration level using an ELISA kit. Briefly, samples were incubated overnight at 2-4°C in BDNF ELISA plates pre-coated with rabbit anti-human BDNF polyclonal antibody. After incubation, plates were thoroughly washed and incubated at RT (2-3 hrs) with a biotinylated mouse anti-human BDNF monoclonal antibody (1:1000). Plates were then washed and incubated at RT (60 min) with a streptavidin-HRP conjugate (1:1000), washed again, developed with a TMB/E substrate at RT (15 min). The optical density (OD) was measured at 450 nm and plotted on a standard curve.
In all graphs, columns represent an averaged mean (n = 10 and/or as specified per figure) and error bars indicate standard error of mean (+/- S.E.). Comparisons between groups were made using an independent samples t-test. Results were considered significant if P < 0.05.