The following mice were used and maintained in Glasgow University Veterinary Research Facility: homozygous shiverer mice (shiv/shiv) on the C3H/101 genetic background and mice expressing enhanced GFP under the β-actin promoter  on the C57BL/6 genetic background (C57BL/6-Tg(ACTB-EGFP)1Osb/J). All experimental mice were bred at the facility. Mice had access to food and water, ad libitum. Furthermore, all procedures were carried out in accordance with the guidelines, set forth by the Animals Scientific Procedures Act, under a project license (No. 6003895) granted by the UK Home Office and with the approval of the University of Glasgow Ethical Review Process Applications Panel.
Isolation and culture of GFP-labelled and wild type neurospheres
Neurospheres were generated from striata of β-actin GFP-transgenic mice expressing cytoplasmic GFP (cGFP) or wild type mice as previously described , based on methods from Reynolds and Weiss [43, 44]. Briefly, the striata were dissected from postnatal day 1 embryos and mechanically dissociated and triturated through a glass Pasteur pipette to produce a cell suspension. The dissociated cells were spun at 800 rpm and the cell pellet reconstituted in 20 ml of neurosphere medium, consisting of DMEM/F12 (1:1, DMEM containing 4,500 mg/L glucose), supplemented with 0.105% NaHCO3, 2 mM glutamine, 5,000 IU/ml penicillin, 5 mg/ml streptomycin, 5.0 mM HEPES, 0.0001% bovine serum albumin, (all Invitrogen), 25 μg/ml insulin, 100 μg/ml apotransferrin, 60 μM putrescine, 20 nM progesterone, and 30 nM sodium selenite (all Sigma). The suspension was then placed into a 75 cm3 tissue culture flask (Greiner, UK) and supplemented with 20 ng/ml mouse submaxillary gland epidermal growth factor (EGF, Peprotech, UK). Every 2-3 days, 5 ml neurosphere medium and 4 μl EGF (20 ng/ml) were added into the flask  and the cultures maintained at 37°C, in a humidified atmosphere of 7% CO2/93% air. GFP neurospheres were used in both the in vitro and in vivo experiments. In some experiments the neurospheres were labelled with farnesylated GFP (fGFP) by lentivral transduction (lentivirus gift from Prof J Verhaagen, Netherlands Institute for Neuroscience, NIN) which remained bound to the plasma membrane as previously described . Briefly triturated neurospheres were incubated with 10 μl of 2.2 × 109 transducing units (TU)/ml of the viral supernatant overnight and maintained in neurosphere medium.
Myelinating cultures were generated from E13.5 (embryonic day) pups as described previously [24, 45, 46]. Briefly the spinal cord was dissected, the meninges removed, and minced using a scalpel blade prior to enzymatic dissociation (100 μl of 2.5% trypsin, Invitrogen, Paisley, UK and 100 μl of 1.33% collagenase, ICN Pharmaceuticals, UK). Enzymatic activity was stopped by the addition of 1 ml of a solution (SD) containing 0.52 mg/ml soyabean trypsin inhibitor, 3.0 mg/ml bovine serum albumin and 0.04 mg/ml DNase (Sigma, UK) to prevent cell clumping. The cells were triturated through a glass pipette and spun at 800 rpm and the pellet resuspended in 5 ml of plating medium [PM: 50% low glucose DMEM, 25% horse serum, 25% HBSS (Hanks balanced salt solution without Ca2+ and Mg2+) and 2 mM L-glutamine (Invitrogen). The dissociated mixed spinal cord cells were plated onto PLL-coated coverslips at a density of 150,000 cells/100 μl and left to attach for 2 hr in the incubator, after which 300 μl of PM and 500 μl of differentiation medium (DM) was added, which contained DMEM 4,500 mg/mL glucose, 10 ng/ml biotin, 0.5% hormone mixture (1 mg/mL apotransferrin, 20 mM putrescine, 4 μM progesterone, and 6 μM selenium) (formulation based on N2 mix ), 50 nM hydrocortisone, and 0.5 μg/ml insulin (all reagents from Sigma) was added. Cultures were maintained by replacing half of the medium with fresh medium three times a week. After 12 days in culture, insulin was excluded from the DM. The cultures were maintained for up to 28 days, in a humidified atmosphere of 7% CO2/93% air, at 37°C. For time-lapse imaging in some cases, ascorbic acid (0.5 μl/1 ml) was added to the medium to enhance cell survival.
Laminectomy and transplantation of neurospheres in vivo
The method of Edgar et al., 2004 was followed . Briefly a short vertical incision was made with a scalpel blade over the thoraco-lumbar region of the spine, from anterior to posterior. With the aid of an operating microscope, the transverse laminae of a vertebra were broken (laminectomy) to reveal the spinal cord. A small opening was made in the dura with a sterile needle and a cell suspension of cGFP or fGFP neurospheres which had been actively growing in neurospheres medium containing EGF, was injected using a CellTram Oil manual micromanipulator (Eppendorf Ltd, Cambridge, UK) at a rate of 1 μl/min, using a glass microelectrode that was inserted into the exposed spinal cord. One injection was made into the dorsal spinal cord, avoiding the midline dorsal vein. A total volume of 3-5 μl of cell suspension containing, approximately 5 × 104 cells/μl was injected, at the depth of < 1 mm. The microelectrode was left in place for an additional 2 min to minimise back-flow of cells. No immunosuppressant treatment was used. Mice were sacrificed at various time points 3 days, 7-10 days, 14 days and 4 weeks (±3 days) after transplantation.
Tissue processing and immunocytochemistry
Animals were euthanized with CO2 and perfused transcardially with 10 ml of saline, followed by 20 ml of 4% paraformaldehyde. The spinal cords were dissected and placed in 0.1 M glycine in PBS, then cryoprotected in 20% sucrose overnight at 4°C followed by embedding in OCT and rapid freezing in liquid nitrogen cooled-isopentane. The rostro-caudal segment at the lumbar thoracic junction of the spinal cord encompassing the transplant site was embedded and cryosectioned dorso-ventrally. Cryostat serial sections (10 μm) were cut, mounted onto APES-coated slides and stored at -20°C. Before immunostaining, sections were air dried at room temperature (RT) for 10-20 min and then rehydrated in PBS for 10 min. The sections were permeabilised with methanol at -20°C for 10 min and blocked with 10% normal goat serum in PBS for 1 hour at RT followed by incubation in primary antibody in the same blocking solution overnight at 4°C. Antibodies used were: rabbit anti-GFP (1:1000, Abcam), mouse anti-GFP (1:250, Abcam) for GFP transplanted cells, rat anti-MBP (1:500, Serotec) for mature oligodendrocytes, the O4 antibody for oligodendrocytes ; 1:1 hybridoma, IgM), mouse anti-GFAP (1:1000, Sigma) for astrocytes and mouse anti-phosphorylated neurofilament SMI-31 for neurites. The slides were washed three times with PBS and then incubated with appropriate fluorescent-conjugated secondary antibodies (Cambridge Biosciences) for 1 hour at RT. The slides were washed three times with PBS and then mounted in Citifluor antifade (Citifluor Ltd, UK) mounting medium.
Quantification of transplanted cells in the shiverermouse
10x 10 micron thick sections were collected over the transplant site for each experiment and immuolabelled with cell type markers. Using Image J and a macro (generated in the lab) that created homocentric circles 150 microns apart, allocated from the epicentre of the injection area, the distance of oligodendroglial cells (MBP), astrocytes (GFAP) and other cells (GFAP/MBP–ve) from the transplantation site was measured. Any cells found within the borders of two circles were counted and plotted against distance from the injection site.
Maintenance of the explants for ex vivoimaging
The protocol was based on studies by Kerschensteiner and colleagues . All procedures were performed in a combination of F12 + L-glutamine, CO2 independent medium and DMEM + Glutamax (4 g/L D-glucose) or Neurobasal A medium (Invitrogen) that had been bubbled with 95% O2 and 5% CO2 for at least 15 min before imaging. During dissection, the mouse was placed on aluminium foil with ice underneath to protect the tissue from hypoxia. During imaging the explant was superfused with pre-warmed O2-bubbled medium. The temperature of the explant was maintained at 35-37°C using a heating stage, superfused with pre-warmed medium.
Multiphoton microscopy was performed using a LaVision BioTec 2-photon TRIM scope, and a Zeiss 7MP. The LaVision system consisted of a Nikon Eclipse TE2000 inverted stand, Olympus long working distance 20x 0.95 NA water immersion objective and Coherent Chameleon II laser tuned to 830 nm. Fluorescence was detected using non-descanned detectors (NDD, Hamamatsu H6780-01-LV 1 M for < 500 nm detection and H6780-20-LV 1 M for > 500 nm detection). A dichroic filter (Chroma 475 DCXR) was used to separate spectrally the second harmonic signal, when present, from the GFP emission of the transplanted cells. Band pass filters (Semrock 435/40 and Chroma 525/50) were used to further filter the emission for the SHG and GFP channels respectively. The Zeiss system consisted of an Axio Imager upright stand, 20x, 1.0 NA water immersion objective, and Chameleon II laser. To keep tissue stationary during ex vivo imaging it was glued, dorsal part side up to a plastic cover slip with cyanoacrylate (Vetbond, 3M Health Care Ltd, Leicestershire, UK). The cover slip was cut to fit in a perfusion chamber (Harvard Apparatus ltd, Kent, UK) where it was held in place by grease. The perfusate was equilibrated with oxygen and thermo-stated to 35-37°C. The chamber was placed on the motorized stage of the upright microscope and fluorescent cells located using epifluorescence with (blue) excitation. The wavelength of the laser was set to either 840 nm (to excite predominantly CFP) or 940 nm (to excite predominantly GFP). Detection channels selected light with wavelength < 485 nm (for CFP) and 500 - 550 nm (for GFP). For in vivo imaging a laminectomy was performed and the mouse placed dorsally in a custom made V-shaped plinth with adjustable legs and a small opening to allow the microscope lens (inverted) to access the spinal cord. Repetitive administration of anaesthetic drugs every 2 hours, kept the heart beat and breathing relatively low to facilitate the acquisition of images during time-lapse.
The Nikon time-lapse microscope TE2000 is fitted with a Nikon perfect focus system (PFS) to maintain focus over the imaging. The PSF only works with glass bottom dishes. So 35 mm glass bottom microwell Petri dishes, with 14 mm diameter of microwell (MakTek Corporation, MA, USA) were used for all the time-lapse experiments. The system has a temperature-controlled 37°C chamber, provided with an oxygen supply and images were acquired using a 40x short distance 0.75 NA air objective. Analysis was performed with MetaMorph (Version 5) imaging software, which compensate for stage shift, vibration or similar small whole field movement that can occur during time-lapse acquisition.