Inflammation stimulates remyelination in areas of chronic demyelinatio
http://www.100md.com
《大脑学杂志》
Department of Veterinary Medicine, University of Cambridge, UK
Summary
A major challenge in multiple sclerosis research is to understand the cause or causes of remyelination failure and to devise ways of ameliorating its consequences. This requires appropriate experimental models. Although there are many models of acute demyelination, at present there are few suitable models of chronic demyelination. The taiep rat is a myelin mutant that shows progressive myelin loss and, by 1 year of age, its CNS tissue has many features of chronic areas of demyelination in multiple sclerosis: chronically demyelinated axons present in an astrocytic environment in the absence of acute inflammation. Using the taiep rat and a combination of X-irradiation and cell transplantation, it has been possible to address a number of questions concerning remyelination failure in chronic multiple sclerosis lesions, such as whether chronically demyelinated axons have undergone changes that render them refractory to remyelination and why remyelination is absent when oligodendrocyte progenitor cells (OPCs) are present. Our experiments show that (i) transplanted OPCs will not populate OPC-containing areas of chronic demyelination; (ii) myelination competent OPCs can repopulate OPC-depleted chronically demyelinated astrocytosed tissue, but this repopulation does not result in remyelination—closely resembling the situation found in some multiple sclerosis plaques; and (iii) the induction of acute inflammation in this non-remyelinating situation results in remyelination. Thus, we can conclude that axonal changes induced by chronic demyelination are unlikely to contribute to remyelination failure in multiple sclerosis. Rather, remyelination fails either because OPCs fail to repopulate areas of demyelination or because if OPCs are present they are unable to generate remyelinating oligodendrocytes owing to the presence of inhibitory factors and/or a lack of the stimuli required to activate these cells to generate remyelinating oligodendrocytes. This non-remyelinating situation can be transformed to a remyelinating one by the induction of acute inflammation.
Key Words: multiple sclerosis; oligodendrocyte progenitor cell; remyelination; taiep rat; transplantation
Abbreviations: FBS = fetal bovine serum; OPC = oligodendrocyte progenitor cell; PDL = poly-D-lysine
Introduction
Remyelination failure is a feature of multiple sclerosis that probably contributes to the relentless progression of this disease. A major challenge in multiple sclerosis research is to understand the cause, or causes, of remyelination failure and devise ways of ameliorating its consequences. Amelioration could be achieved by the development of remyelination enhancing therapies. These might take the form of stimulating the brain's inherent remyelinating potential or transplanting cells that have remyelinating potential. Evaluation of both approaches requires appropriate animal models and an understanding of why remyelination fails in multiple sclerosis.
It has been demonstrated that transplantation of myelin forming cells can result in extensive myelin sheath formation when undertaken during development or in association with acute demyelination (reviewed, Duncan et al., 1997; Blakemore and Franklin, 2000; Baron-Van Evercooren and Blakemore, 2004). Such environments can be regarded as physiological recipient environments since either myelination or remyelination is occurring, and thus the introduced cells are exposed to an environment that is conducive to myelin formation. Deciding how to approach a pathological environment such as pertains in areas of chronic demyelination, where oligodendrocyte progenitor cells (OPCs) with the potential for remyelination may be present but fail to achieve remyelination (Wolswijk, 1998; Chang et al., 2000, 2002), has been hampered by a lack of suitable models. To be suitable, the experimental model should resemble chronic multiple sclerosis lesions in the following respects: (i) it should have axons that have remained without myelin sheaths for some considerable time; (ii) it should contain OPCs and an absence of remyelination; (iii) the demyelinated axons should be present in an area that contains astrocytes; and (iv) there should be an absence of acute inflammation. Results from experiments using such a model would have relevance for understanding how to enhance endogenous remyelination in areas of chronic demyelination, as well as potential application to cell therapy. Our study shows that such a model can be achieved using the taiep rat. The taiep rat has an autosomal recessive genetic defect which leads to an accumulation of microtubules in oligodendrocytes and altered intracellular transport (Song et al., 2003). This results in hypomyelination and progressive demyelination such that, by 1 year of age, all small diameter and many medium diameter CNS axons are without myelin sheaths (Duncan et al., 1992; Lunn et al., 1997; Foote and Blakemore, 2005). The loss of myelin sheaths is accompanied by an astrocytosis (Duncan et al., 1992; Foote and Blakemore, 2005) and an ongoing acute inflammatory response is absent. Therefore, the tissue has the features of chronic areas of demyelination in multiple sclerosis, with chronically demyelinated axons present in a non-inflammatory astrocytic environment.
By using the taiep rat in combination with X-irradiation to deplete the endogenous OPC population and transplantation of remyelination competent OPCs, it has been possible to examine a number of suggested reasons for remyelination failure in multiple sclerosis and to predict the likely outcome of transplanting OPCs into areas of chronic demyelination. In particular, our studies show that (i) transplanted OPCs will not populate OPC-containing areas of chronic demyelination; (ii) OPCs can repopulate OPC-depleted chronically demyelinated astrocytosed tissue, but this repopulation by myelination competent cells does not result in remyelination, closely resembling the situation found in some multiple sclerosis plaques; and (iii) the induction of acute inflammation in this non-remyelinating situation stimulates remyelination. Thus, axonal changes induced by chronic demyelination are unlikely to contribute to remyelination failure in multiple sclerosis. Rather, remyelination fails either because OPCs fail to repopulate areas of demyelination or because if they are present they are dysfunctional because of the presence of inhibitory factors and/or a lack of the stimuli required to activate these cells to generate remyelinating oligodendrocytes. It is therefore significant that a non-remyelinating situation can be transformed to a remyelinating one by the induction of acute inflammation.
Methods
Animals
Taiep rats were obtained from Dr M. Roncagliolo (University of Valparaiso, Chile) and bred by mating either affected to affected animals or affected to non-affected heterozygotes. Control animals used for transplantation were heterozygous littermates of the taiep rats. Sprague–Dawley rats carrying the autosomal dominant GFP gene on an actin promoter were obtained from Professor M. Okabe (Osaka University, Japan) and heterozygous males were mated to normal Sprague–Dawley females. All animal experiments were performed under a Home Office licence and with ethical approval from the University of Cambridge.
Progression of pathology
To study the progression of pathology in the taiep rats, animals were perfused with 4% glutaraldehyde at 3, 7, 10, 14 and 18 months of age (at least two animals for each time point) and tissue from the spinal cord was processed into resin. Sections 1 μm thick were cut and examined using light microscopy following toluidine blue staining. For electron microscopy, 90 nm sections were cut, stained with uranyl acetate and lead citrate and examined using a Hitachi Model H600 transmission electron microscope.
Transplantation
Using halothane anaesthesia and aseptic surgical techniques, a laminectomy was performed at the level of T13 to expose the dorsal spinal cord for cell transplantation. With the aid of a three-way micromanipulator, the tip of a glass micropipette attached to a 10 μl Hamilton syringe was inserted into the dorsal funiculus and 1 μl of cell suspension (105 cells) was injected. Some animals (as indicated in Table 1) were immunosuppressed with cyclosporin A (Sandoz, 15 mg/kg s/c daily) from 1 day before transplantation. All animals used as transplant recipients were between 11 and 14 months of age.
Cell preparations
Unlabelled OPC enriched preparation
Mixed glial cell cultures were prepared from the cerebral hemispheres of 2–3 day old Sprague–Dawley rat pups. Briefly, cerebral hemispheres were removed, stripped of their meninges and mechanically dissociated in MEM containing 20 mM HEPES (Invitrogen). Cells were harvested by filtration through a 20 μm nylon mesh followed by centrifugation (500 x g for 5 min) and plated on poly-D-lysine (PDL) coated 80 cm2 tissue culture flasks containing DMEM (Invitrogen) supplemented with glucose (6 mg/ml) and 10% heat inactivated fetal bovine serum (FBS; Invitrogen). Each flask contained cells from two animals. Every 3–4 days 60% of the culture medium was replaced. After 10 days in vitro, cultures were shaken on an orbital shaker (200 r.p.m. for 2 h) to remove loosely adhered microglia; the medium was replaced and then the cultures were shaken again (375 r.p.m. overnight) to harvest OPCs. The detached cells were collected, the more adhesive contaminating cells were allowed to adhere on to non-tissue culture coated petri dishes for 30 min at 37°C and the less adherent OPCs were collected, centrifuged and resuspended in MEM–HEPES at a concentration of 105 cells per microlitre for transplantation. The mixed glial cultures were maintained in culture for a further 7 days before shaking for a third time to harvest, in the same way as before, additional OPCs for transplantation.
GFP labelled OPC enriched preparation
OPC enriched preparations were prepared as described for unlabelled cells using 2–3 day old rat pups heterozygous for GFP, as indicated by green fluorescence under UV light.
LacZ labelled OPCs
OPC enriched cell preparations were prepared as described for unlabelled cells, using 2–3 day old rat pups. Before transplantation, the OPC enriched preparation was plated onto PDL coated 80 cm2 tissue culture flasks containing 70% DMEM/F12 (with L-glutamine; Invitrogen) and 30% medium conditioned by the B104 neuronal cell line, prepared using methods described by Fok-Seang (1995). The medium was supplemented with N2 (10 μl/ml; Invitrogen), BSA (0.66 mg/ml; Sigma), glucose (2.25 mg/ml) and the growth factors FGF-2 (20 ng/ml; Sigma) and PDGF-AA (10 ng/ml; Sigma). Fresh growth factors were added every 24 h, and 60% of the medium was replaced with fresh medium every 48 h.
Replication deficient Maloney murine leukaemia virus, lacZ (with nuclear localization sequence), was collected from an MFG packaging cell line (a gift of Dr Richard Asher). It was concentrated by precipitation using PEG8000 (Sigma) and ultracentrifugation through a sucrose density gradient. The titre of the concentrated virus preparation was estimated by titration on the 3T3 fibroblast cell line. Approximately 106 virus particles in 10 μl DMEM were added to each flask of the OPC cell preparation daily for 5 days. After a further 5 days, the OPC cell preparation was detached from the flask using a brief (<10 min) incubation with trypsin and EDTA (ATV, Sigma). Soya bean trypsin inhibitor (Sigma) was added and the detached cells were collected, centrifuged and resuspended in MEM–HEPES for transplantation at a concentration of 50 000 cells per microlitre.
X-irradiation
To irradiate the spinal cord, animals were anaesthetized using fentanyl citrate/fluanisone (0.3 ml/kg IM, Hypnorm, Janssen Pharmaceuticals) and an intraperitoneal injection of diazepam (2.5 mg/kg, Phoenix Pharmaceuticals) and placed in lateral recumbency. A 4 cm length of the spinal cord, centred on T13, was exposed to 40 Gy X-irradiation using a Pantak 255-kV radiotherapy machine (Blakemore and Crang, 1992).
Experiments and tissue processing
Four transplantation experiments were performed (summarized in Table 1). (i) Ten adult taiep rats were transplanted with unlabelled cells. After 4 weeks these animals were perfused with 4% glutaraldehyde and the tissue examined in resin sections for evidence of remyelination. (ii) Five taiep and three irradiated taiep rat spinal cords were transplanted with lacZ labelled cells and perfused after 4 weeks with 3% paraformaldehyde/1% gluteraldehyde. The cord was cut into 1.5 mm transverse blocks and the tissue was reacted with X-gal. The blocks were processed for light and electron microscopy to examine remyelination and the distribution of transplanted cells. (iii) GFP labelled cells were transplanted into control spinal cord (n = 3), taiep spinal cord (n = 3), irradiated control spinal cord (n = 3) and irradiated taiep spinal cord (n = 4). The tissue from these animals was analysed in longitudinal cryostat sections after 4 weeks to examine the distribution and immunophenotype of transplanted cells. (iv) Ten irradiated taiep spinal cords were transplanted with GFP cells, and five of them received two unilateral injections of saline and sterile charcoal into the lateral funiculus 3 weeks after implantation of cells. These animals were perfused with 4% paraformaldehyde 6 weeks after transplantation. The spinal cord of each animal was cut into 1.5 mm transverse blocks and alternate blocks were processed for either resin (postfixed in 4% gluteraldehyde followed by 2% osmium tetroxide and then resin embedded) or cryostat sectioning. The animals in experiments 1 and 4 were not immunosuppressed, whereas animals in experiments 2 and 3 were given immunosuppressant doses of cyclosporin.
In addition, in order to examine the effect of irradiation on the endogenous OPC population, the spinal cords of 4 adult taiep rats were irradiated over a 7 mm length. Animals were perfused 4 weeks after irradiation and in situ hybridization for PDGFR was performed on cryostat sections. A saline/charcoal injection was also made into the dorsolateral funiculus of three irradiated taiep spinal cords (which had not received transplanted OPCs) and the effect of this injection on the tissue was examined in cryostat and resin sections 3 days post injection.
As well as transplantation into taiep spinal cord, all cell preparations were injected into X-irradiated ethidium bromide lesions in normal adult rat spinal cord 3 days after lesion induction, and the animals were killed 1 month later to evaluate the extent of remyelination (Blakemore and Crang, 1992).
Immunostaining and ISH
Samples of the cell preparation used for transplantation were plated onto CC2-coated chamber slides (Labtec) for 12 h in DMEM/F12 containing 1% FBS, then fixed with 4% paraformaldehyde and characterized by immunocytochemistry using antibodies to A2B5 (mouse IgM, Roche, 1:200), O4 (mouse IgM hybridoma supernatant, a gift from Dr Mark Noble), GalC (mouse IgG, Chemicon 1:100), GFAP (polyclonal rabbit IgG, Dako, 1:150) and CD11b (OX42, mouse IgG, Serotec, 1:200). Cell nuclei were labelled with Hoescht (Calbiochem, 1:500).
For immunohistochemistry, cryostat sections were stained using antibodies to NG2 proteoglycan (rabbit IgG, Chemicon, 1:150), GFAP (rabbit IgG, Dako, 1:150), APC (CC2, mouse IgG, Oncogene, 1:100), CD45 (LCA, mouse IgG, Chemicon, 1:100), rat B cells (mouse IgG, Serotec, 1:100), CD4 (mouse IgG, Serotec, 1:100), CD8 (mouse IgG) and GFP (rabbit IgG or mouse IgG, Molecular Probes, 1:300). Appropriate FITC, TRITC or Cy3 conjugated secondary antibodies were used (Jackson Laboratories). PDGFR in situ hybridization was performed on cryostat sections using a DIG-labelled riboprobe (cDNA, a gift from Professor W. D. Richardson).
Analysis
The repopulation of non-irradiated and irradiated tissue by transplanted cells after 4 weeks was observed in transverse resin sections for lacZ labelled cells and in parasagittal sections for GFP labelled cells. The total length of repopulation was measured from the latter, as was the length of the area of increased GFP cell density containing APC+GFP+ cells (APC being a marker of mature oligodendrocytes). The lengths were measured from a minimum of three sections for each animal, using sections that passed through the centre of the transplant site.
In resin sections, transplant mediated remyelination was assessed by the detection of areas in which all axons were myelinated. For lacZ labelled cells, remyelination was confirmed as being transplant derived by its association with labelled cells in closely adjacent sections. For GFP labelled cells, regions of high cell density containing GFP+APC+cells could be detected in the cryostat blocks either side of resin blocks showing complete remyelination. For quantification, the cross-sectional area of white matter showing evidence of complete remyelination in the resin sections was outlined and the area measured using a live capture digital imaging system (MCID). By maintaining the craniocaudal order of the resin blocks in relation to the transplant site, the area of remyelination was compared at different levels of the cord and a volume of remyelination estimated for each animal.
Results
Examination of the spinal cord of taiep rats of different ages confirmed that there was progressive loss of myelin with age, as previously observed (Lunn et al., 1997; Foote and Blakemore, 2005). By 10 months of age, all small diameter axons and most medium sized axons lacked myelin sheaths, while large diameter axons were surrounded by thinner than normal myelin sheaths. In particular, almost all the axons in the corticospinal tract, gracile tract and the tracts in the dorsolateral funiculi were without detectable myelin sheaths in toluidine blue stained resin sections and at the ultrastructural level (Fig. 1).
Only small focal areas of remyelination are found when OPCs are injected into areas of chronic demyelination
To determine whether axons that had been chronically demyelinated were receptive to remyelination we transplanted oligodendrocyte precursors into the dorsal funiculus of the thoracolumbar spinal cord in adult (11–14 month old) taiep rats. Experiments were undertaken using cells from normal animals, cells transfected with the lacZ gene and cells derived from GFP expressing animals. The cell preparations used contained a similar proportion of OPCs (Table 2) and each showed the expected remyelinating capacity when injected into X-irradiated ethidium bromide lesions (data not shown). Four weeks after transplantation into the taiep rat spinal cord only small areas of remyelination could be detected in resin-embedded tissue (Fig. 2A–D). Using cells prepared from GFP expressing rats or lacZ labelled cells it was apparent that all the transplanted cells were restricted to a small area near to the point of cell implantation (Figs. 2E and 3E). The length of this region was consistent with the passive spread of cells that results from the injection of the cell suspension (Franklin et al., 1996; O'Leary and Blakemore, 1997). Therefore, repopulation of the chronically demyelinated tissue by the transplanted cells had not occurred, indicating that some property of the host tissue was inhibitory to widespread integration of the transplanted cells. However, although very restricted in its distribution, the presence of remyelination indicates that chronically demyelinated axons of the taiep rat can be remyelinated by myelination competent cells.
Depletion of OPCs results in widespread repopulation of OPC-depleted tissue by transplanted OPCs but does not result in extensive remyelination
The probable explanation for the lack of migration away from the site of implantation is the presence of endogenous OPCs, since it has been shown in previous studies that transplanted OPCs fail to colonize normal tissue in adult animals, but will colonize OPC-depleted tissue (Franklin et al., 1996; O'Leary and Blakemore, 1997). To confirm this, the spinal cord of taiep and normal rats was exposed to 40 Gy of X-irradiation—the dose known to deplete rat tissue of its OPC population in normal (Hinks et al., 2001; Chari and Blakemore, 2002a) and taiep rats (Fig. 3A and B) (Foote and Blakemore, 2005). Transplantation of cells into taiep spinal cord exposed to X-irradiation resulted in extensive repopulation of the OPC-depleted tissue (Fig. 3C and E). A greater area of remyelination was also seen around the point of cell implantation in irradiated animals than in non-irradiated animals, detected either on resin sections or, indirectly, as an area of increased GFP cell density containing many oligodendrocytes (APC positive cells) (Fig. 3D and F). However, in contrast to these areas of remyelination close to the injection site, transplanted cells repopulated the whole cross-sectional area of the spinal cord at the level of the implantation site and extended longitudinally cranially and caudally for some considerable distance (Fig. 3E). The length of repopulated tissue was similar in X-irradiated normal and taiep rats (Fig. 3E), and thus the astrocytosis and the chronically demyelinated axons present in the taiep rats were not significantly influencing the rate of repopulation. Nearly all the transplanted cells that established themselves away from the transplant site co-labelled with antibodies to the oligodendrocyte progenitor marker NG2 (Fig. 3K). It was noteworthy that the density of transplant derived OPCs in areas distant from the transplant site was similar to that for OPCs in normal tissue (50–60 cells/mm2) and there was no evidence of remyelination or generation of transplant derived oligodendrocytes (Fig. 3G–I, L) except in the area of remyelination associated with the transplant site (Fig. 3D, M–O). We had therefore achieved a situation in which OPCs were present adjacent to demyelinated axons without evidence of oligodendrocyte differentiation and remyelination. Thus, failure of remyelination at a distance from the implantation point cannot be attributed to a failure of OPCs to interact with demyelinated axons.
Induction of inflammation stimulates remyelination in areas of chronic demyelination
Since the widespread repopulation of the chronically demyelinated taiep spinal cord by OPCs resulted in remyelination only near to the transplant site, we hypothesized that the act of implantation of cells was inducing changes in the tissue that rendered it conducive to remyelination, whereas the mere repopulation of tissue by myelination competent OPCs was insufficient to bring about remyelination. To test this hypothesis, we conducted an experiment in which we first allowed repopulation of the irradiated taiep tissue by transplanted OPCs and then created, at a site distant from the cell transplant site, a situation that would mimic the injection of cells (Fig. 4A). This was achieved by injecting a small volume of saline (containing sterile charcoal to mark the site). The site chosen for this injection was the dorsal part of the lateral funiculus where there was an almost complete absence of myelin (Fig. 4D) and where remyelination had not been observed following transplantation of OPCs into the dorsal funiculus. In this experiment, unlike the previous experiments with genetically labelled cells, the recipient animals were not given cyclosporin, since we did not want to interfere with any inflammatory changes that may be associated with the saline injection. When animals were examined 3 weeks after the saline injection, remyelination was associated with the site of saline/charcoal injection in all five animals injected (Fig. 4C) and there was no evidence of remyelination in the same area on the contralateral side of the spinal cord (Fig. 4D). Since both areas contained transplanted OPCs, this result indicates that the injection of saline had stimulated the transplant established OPCs to remyelinate the tissue.
To examine the consequences of injecting saline and charcoal into the irradiated spinal cord, resin embedded and frozen sections of the injection site were examined 3 days after this injection. The injection resulted in a marked increase the number of macrophages and OX42 positive cells (Fig. 4B). Lymphocytes (CD4, CD8, B cell) were not detected in these areas using immunohistochemistry.
In some of the transplant repopulated animals subjected to the saline/charcoal injection, as well as in some animals allowed to survive for 6 weeks without saline/charcoal injection, we also found extensive remyelination around the periphery of the spinal cord (Fig. 4E). It was noticeable that, in the animals showing the most extensive remyelination, inflammatory cells were present in the meninges in toluidine blue stained resin sections, and CD4 and CD8 positive cells could be detected by immunostaining within the CNS parenchyma and within the meninges (Fig. 4F). When the extent of complete remyelination was measured and compared between the animals with and without evidence of meningeal infiltration (as detected in resin sections), it was clear that the extent of remyelination in the animals with cell infiltration was significantly greater than in those without evidence of meningitis (Fig. 4G and H); in the sections showing the most extensive remyelination, up to 80% of the white matter was remyelinated. Our results therefore show that induction of inflammation in non-remyelinating OPC-populated areas of chronic demyelination stimulates remyelination.
Discussion
Our studies represent the first in which extensive remyelination has been achieved using a model of chronic demyelination in adult animals. As a consequence, they establish a number of principles that have relevance both to understanding the reasons for remyelination failure in multiple sclerosis and to evaluating remyelination enhancing strategies for this disease. First, by transplanting OPCs into this model of chronic demyelination, we have confirmed the recent observation on the retina that axons that have been without myelin for a considerable time can be myelinated (Setzu et al., 2004) and thus axonal changes resulting from demyelination are unlikely to play a role in remyelination failure (Chang et al., 2002). Second, by replacing taiep OPCs with normal OPCs we have demonstrated that cells capable of undertaking remyelination can be present in tissue containing receptive demyelinated axons without consequent remyelination. Thus, the OPCs found in some chronic multiple sclerosis lesions and considered by some to be non-functional (Greenwood and Butt, 2003) may not in fact be so, since they, like the transplanted OPCs in this study, may simply require induction of inflammation to enable their proliferation and subsequent differentiation into remyelinating cells. Third, by showing that transplanted OPCs established away from the implantation site have a similar density to that found in normal tissue, our study extends to pathological tissue a conclusion drawn from studies on normal tissue that, in the absence of inflammation, OPC numbers are tightly regulated (Dawson et al., 2003; Woodruff et al., 2004). This has implication for attempts to introduce transplanted OPCs into areas of chronic demyelination as well as normal tissue, since it indicates that transplanted OPCs can be established only by either depleting tissue of endogenous OPCs, or changing the environment so that it can support these supernummary cells. Finally, our studies demonstrate that although transplanted OPCs can be established in OPC-depleted areas of chronic demyelination, remyelination is dramatically stimulated only when the environment is made conducive to OPC expansion and differentiation; this can be achieved by establishing acute inflammation.
In agreement with a number of previous studies using longer-lived mature myelin mutants (Archer et al., 1997; Milward et al., 2000; Zhang et al., 2003), we found very limited remyelination following transplantation of myelination competent cells into the spinal cord of taiep rats. In the shaking pup the extent of remyelination decreased with the age of the recipient and was enhanced when late embryonic rather than neonatal cells were transplanted in adult animals (Archer et al., 1997), indicating that donor age influences the extent of remyelination. The poor results in the Long Evans rat were attributed to the presence of activated microglial cells, as down-regulation of this response by minocycline enhanced cell survival and remyelination (Zhang et al., 2003). Our study clearly shows that transplanted OPCs, as in normal tissue (Franklin et al., 1996; O'Leary and Blakemore, 1997), do not enter tissue containing endogenous OPCs. Since there can be no remyelination without repopulation of tissue by myelination competent cells, a major reason for the failure to achieve significant remyelination following transplantation into these long-lived myelin mutants is the presence of endogenous OPCs.
The rate of OPC repopulation of OPC-depleted tissue by transplanted OPCs was not significantly different in taiep and normal rats, indicating that the chronically demyelinated environment found in the taiep rat does not significantly inhibit or enhance the OPC repopulation rate. The repopulation rate of the transplanted OPCs was slightly slower in the taiep than in normal animals, which may suggest an inhibitory role for astrocytes in vivo, as has been demonstrated in vitro (Fok-Seang et al., 1995). In a previous study, we found no difference in the rate of endogenous OPC repopulation between normal and taiep rats of similar age (Foote and Blakemore, 2005). However, failure to detect a difference in that study may have been a consequence of the very slow OPC repopulation rate in old animals. Despite a possible slight negative influence on the rate of repopulation, it is clear that transplanted OPCs colonize OPC-depleted chronically demyelinated tissue at a significantly faster rate than endogenous cells [1.5 mm/week (present data) compared with 0.2 mm/week (data from Foote and Blakemore, 2005)]. Thus, where tissue is depleted of OPCs, transplanted neonatal OPCs can repopulate normal and chronically demyelinated tissue more rapidly than endogenous cells.
More important than the identification of how to achieve OPC repopulation in areas of chronic demyelination was the finding that this widespread repopulation resulted in only a modest increase in remyelination. The increased remyelination was restricted to the transplant area, and away from this area there was no evidence of remyelination despite repopulation of the tissue by the transplanted myelination competent OPCs. Thus, repopulation and remyelination are distinct processes, with the former not necessarily resulting in the latter. The non-remyelinating OPC-repopulated chronically demyelinated tissue that we have created in the taiep rat is comparable to the situation observed in non-remyelinating OPC-populated multiple sclerosis plaques (Chang et al., 2002). Since we were able to demonstrate that the chronically demyelinated axons of the taiep rat could be remyelinated under some circumstances, we can conclude that chronically demyelinated tissue has properties that either inhibit or fail to stimulate remyelination. Our studies show a clear relationship between acute inflammation and successful remyelination. This adds to the growing body of evidence that acute inflammation provides the stimulus that initiates remyelination (reviewed, Franklin, 2002) and provides proof of principle for the ‘temporal mismatch’ hypothesis as an explanation for remyelination failure in multiple sclerosis (Chari and Blakemore, 2002b). This hypothesis proposed that if OPCs were destroyed in large areas of demyelination, remyelination would fail because the slow rate of repopulation shown by endogenous OPCs in old individuals would result in a temporal mismatch between the acute inflammatory environment associated with myelin sheath breakdown and the interaction of OPCs with demyelinated axons; that is, in the absence of an acute inflammatory environment to provide the signals required to drive remyelination, the process will fail. In a previous study, we obtained some support for this concept by delaying the interaction of OPCs with demyelinated axons by transplanting cells at a distance from a focus of acute demyelination induced in OPC-depleted tissue, and showing that the extent of remyelination was significantly reduced compared with the situation when OPCs were present at the time of, or soon after, demyelination was initiated (Blakemore et al., 2002). Having established that OPC repopulation rates are extremely slow in old animals (Chari et al., 2003; Foote and Blakemore, 2005), our current results strengthen the temporal mismatch hypothesis by demonstrating for the first time in an experimental model the three further situations required for this hypothesis to be sustainable: (i) OPCs can repopulate OPC-depleted chronically demyelinated tissue; (ii) in the absence of acute inflammation, these repopulating cells persist without remyelination occurring; and (iii) when an acute inflammatory environment is created, remyelination occurs efficiently.
The finding that areas of chronic demyelination can contain both axons that can be remyelinated and a normal density of remyelination competent OPCs without evidence of remyelination would indicate that, unlike in myelination during development, the axon does not provide the stimulus that initiates OPC proliferation (Barres and Raff, 1999). It is well established that OPCs become activated, increase in number and change their morphology in association with acute damage to the central nervous system (Levine et al., 2001). It is not known whether this activation occurs as a consequence of changes in properties of the environment that regulate the activation status of OPCs or are a consequence of direct stimulation of the OPC by factors associated with acute inflammation. In this context it is known that certain molecules, such as Il-1, associated with acute inflammation can stimulate OPC division directly (Vela et al., 2002) and could influence OPC division and survival via their effects on astrocytes (Aloisi et al., 1995; Silberstein et al., 1996; Mason et al., 2001). In adult rats, demyelination is associated with rapid activation of OPCs, as indicated by up-regulation of the transcription factor Nkx2.2 and olig2 (Fancy et al., 2004), and, in terms of remyelination, the two consequences of OPC activation are cell division and differentiation into remyelinating oligodendrocytes. The OPCs which repopulated the OPC-depleted chronically demyelinated taiep spinal cord established themselves at a normal density and showed a normal morphology and no evidence of oligodendrocyte differentiation, suggesting that they are unstimulated and under the control of the factors that regulate OPC behaviour in normal tissue. However, once activated by acute inflammation they proliferate and differentiate into remyelinating oligodendrocytes. Investigation is needed of whether inflammation also induces changes on the axon that would facilitate myelination, such as removal of the polysialylated neural cell adhesion molecule shown to inhibit myelination (Charles et al., 2000).
In the present experiments two forms of inflammation were associated with stimulation of remyelination. The saline/charcoal injections stimulated a predominantly macrophage (moncyte/macrophage/microglial) response, whereas lymphocyte infiltration was a feature of the more widespread remyelination found in the animals with meningitis. Macrophages have been shown to have a significant role in facilitating remyelination (Kotter et al., 2001), as have both lymphocytes (Bieber et al., 2003) and the inflammatory responses associated with rejection of xenogeneic cells (Blakemore et al., 1995). It was not possible to establish the cause of the inflammatory reaction in the animals with meningitis and extensive remyelination around the periphery of the spinal cord. The most probable explanation is that it represents an immune response to the transplanted cells. This group of animals was not immunosuppressed, the transplanted cells were allografts and expressed a foreign protein and, moreover, the tissue had been subjected to a high dose of X-irradiation. The dose of X-irradiation used in our studies is known to render tissue more susceptible to immune-mediated inflammation (Love et al., 1987) and GFP can be a powerful immunogen (Stripecke et al., 1999; Follenzi et al., 2004). Although it is clear that acute inflammation provides a stimulus that initiates remyelination and that macrophages and acutely reactive astrocytes have a positive effect on the success of remyelination (Franklin et al., 1991; Hinks and Franklin, 1999), there is also evidence that activated microglia and astrocytes can have a negative effect. The reactive microglia present in the Long Evans rat (Zhang et al., 2003), which are also present in the taiep rat (Goetz et al., 2000), have been shown to have a negative effect on remyelination and, unlike during the acute situation, astrocytes have been implicated as having a deleterious effect on remyelination when the inflammatory response initiated by demyelination is waning (Blakemore et al., 2002; 2003). One therefore has to consider that astrocytes and microglia/macrophages can have both positive and negative effects on remyelination. The duality of influence of a single cell type is also apparent in vitro, where macrophages can either enhance or impede remyelination in co-culture systems (Diemel et al., 2004). Similarly, astrocytes impede OPC migration under some culture conditions but can be altered by exposure to cytokines to negate this effect (Fok-Seang et al., 1998). It is likely, therefore, that it is the nature of the response of these cells to acute inflammation that creates an environment conducive to remyelination. In this context it is possible that the immune modulating therapies currently in use for multiple sclerosis could be having a deleterious effect on remyelination. However, OPC repopulation was not affected by the immunosuppressant regimen used in the present study, indicating once again that OPC repopulation of OPC-depleted tissue and remyelination are distinct processes.
In conclusion, our experiments have relevance for both interpreting the pathology present in areas of chronic demyelination in multiple sclerosis and developing OPC transplantation as a therapy for chronic demyelinating diseases. They demonstrate that (i) the failure of remyelination observed in OPC-containing areas of chronic demyelination cannot be taken as evidence that chronically demyelinated axons have undergone changes that preclude their remyelination; (ii) as with normal tissue, the presence of endogenous OPCs in areas of chronic demyelination prevents the establishment of transplanted OPCs and (iii) acute inflammation provides the stimulus for remyelination.
Acknowledgements
A.F. was in receipt of a BBRSC studentship. The taiep rats were made available by Dr M. Roncagliolo, with funds from the Myelin Project. The GFP rats were made available by Professor M. Okabe. Funds from the Multiple Sclerosis Society and the Medical Research Council, UK, were used to support this study. The technical assistance of Jennifer Gilson and Mike Peacock is gratefully acknowledged.
References
Aloisi F, Borsellino G, Care A, Testa U, Gallo P, Russo G, et al. Cytokine regulation of astrocyte function: in-vitro studies using cells from the human brain. Int J Dev Neurosci 1995; 13: 265–74.
Archer DR, Cuddon PA, Lipsitz D, Duncan ID. Myelination of the canine central nervous system by glial cell transplantation: a model for repair of human myelin disease. Nat Med 1997; 3: 54–9.
Baron-Van Evercooren A, Blakemore WF. Remyelination through engraftment. In: Lazzarini RA, editor. Myelin biology and disorders. New York: Elsevier Science; 2004. p. 143–72.
Barres BA, Raff MC. Axonal control of oligodendrocyte development. J Cell Biol 1999; 147: 1123–8.
Bieber AJ, Kerr S, Rodriguez M. Efficient central nervous system remyelination requires T cells. Ann Neurol 2003; 53: 680–4.
Blakemore WF, Crang AJ. Transplantation of glial cells into areas of demyelination in the adult rat spinal cord. In: Dunnett S, Bjorklund A, editors. Neural transplantation: a practical approach. Oxford: Oxford University Press; 1992. p. 105–22.
Blakemore WF, Franklin RJM. Transplantation options for therapeutic CNS remyelination. Cell Transplant 2000; 9: 289–94.
Blakemore WF, Crang AJ, Franklin RJM, Tang K, Ryder S. Glial cell transplants that are subsequently rejected can be used to influence regeneration of glial cell environments in the CNS. Glia 1995; 13: 79–91.
Blakemore WF, Chari DM, Gilson JM, Crang AJ. Modelling large areas of demyelination in the rat reveals the potential and possible limitations of transplanted glial cells for remyelination in the CNS. Glia 2002; 38: 155–68.
Blakemore WF, Gilson JM, Crang AJ. The presence of astrocytes in areas of demyelination influences remyelination following transplantation of oligodendrocyte progenitors. Exp Neurol 2003; 184: 955–63.
Chang A, Nishiyama A, Peterson J, Prineas J, Trapp BD. NG2-positive oligodendrocyte progenitor cells in adult human brain and multiple sclerosis lesions. J Neurosci 2000; 20: 6404–12.
Chang A, Tourtellotte WW, Rudick R, Trapp BD. Premyelinating oligodendrocytes in chronic lesions of multiple sclerosis. N Engl J Med 2002; 346: 165–73.
Chari DM, Blakemore WF. Efficient recolonisation of progenitor-depleted areas of the CNS by adult oligodendrocyte progenitor cells. Glia 2002a; 37: 307–13.
Chari DM, Blakemore WF. New insights into remyelination failure in MS: implications for glial cell transplantation. Mult Scler 2002b; 8: 271–7.
Chari DM, Crang AJ, Blakemore WF. Rate of repopulation of progenitor-depleted areas of the adult CNS by adult oligodendrocyte progenitors (OPCs) declines with age. J Neuropath Exp Neurol 2003; 62: 908–16.
Charles P, Hernandez MP, Stankoff B, Aigrot MS, Colin C, Rougon G, et al. Negative regulation of central nervous system myelination by polysialylated-neural cell adhesion molecule. Proc Natl Acad Sci USA 2000; 97: 7585–90.
Dawson MR, Polito A, Levine JM, Reynolds R. NG2-expressing glial progenitor cells: an abundant and widespread population of cycling cells in the adult rat CNS. Mol Cell Neurosci 2003; 24: 476–88.
Diemel LT, Wolswijk G, Jackson SJ, Cuzner ML. Remyelination of cytokine- or antibody-demyelinated CNS aggregate cultures is inhibited by macrophage supplementation. Glia 2004; 45: 278–86.
Duncan ID, Lunn KF, Holmgren B, Urba-Holmgren R, Brignolo-Holmes L. The taiep rat: a myelin mutant with an associated oligodendrocyte microtubular defect. J Neurocytol 1992; 21: 870–84.
Duncan ID, Grever WE, Zhang S-C. Repair of myelin disease: strategies and progress in animal models. Mol Med Today 1997; 3: 554–61.
Fancy SPJ, Zhao C, Franklin RJM. Increased expression of Nkx2.2 and Olig2 identifies reactive oligodendrocyte progenitor cells responding to demyelination in the adult CNS. Mol Cell Neurosci 2004; 27: 247–54.
Fok-Seang J, Mathews GA, ffrench-Constant C, Trotter J, Fawcett JW. Migration of oligodendrocyte precursors on astrocytes and meningeal cells. Dev Biol 1995; 171: 1–15.
Fok-Seang J, DiProspero NA, Meiners S, Muir E, Fawcett JW. Cytokine-induced changes in the ability of astrocytes to support migration of oligodendrocyte precursors and axon growth. Eur J Neurosci 1998; 10: 2400–15.
Follenzi A, Battaglia M, Lombardo A, Annoni A, Roncarolo MG, Naldini L. Targeting lentiviral vector expression to hepatocytes limits transgene-specific immune response and establishes long-term expression of human antihemophilic factor IX in mice. Blood 2004; 103: 3700–9.
Foote AK, Blakemore WF. Repopulation of oligodendrocyte progenitor cell depleted tissue in a model of chronic demyelination. Neuropathol Appl Neurobiol 2005; Published online 02/12/04; doi:10.1111dp.1365-2990. 2004.00634.x.
Franklin RJM. Why does remyelination fail in multiple sclerosis. Nat Rev Neurosci 2002; 3: 1–12.
Franklin RJM, Crang AJ, Blakemore WF. Transplanted type-1 astrocytes facilitate repair of demyelinating lesions by host oligodendrocytes in adult rat spinal cord. J Neurocytol 1991; 20: 420–30.
Franklin RJM, Bayley SA, Blakemore WF. Transplanted CG4 cells (an oligodendrocyte progenitor cell line) survive, migrate and contribute to repair of areas of demyelination in X-irradiated and damaged spinal cord, but not in normal spinal cord. Exp Neurol 1996; 137: 263–76.
Goetz BD, Zhang S-C, Song J, Duncan ID. Microglial response to progressive demyelination in the taiep mutant rat. Soc Neurosci Abst 2000; 26: 515.7.2000.
Greenwood K, Butt AM. Evidence that perinatal and adult NG2-glia are not conventional oligodendrocyte progenitors and do not depend on axons for their survival. Mol Cell Neurosci 2003; 23: 544–58.
Hinks GL, Franklin RJM. Distinctive patterns of PDGF-A, FGF-2, IGF-1, and TGF-beta1 gene expression during remyelination of experimentally-induced spinal cord demyelination. Mol Cell Neurosci 1999; 14: 153–68.
Hinks GL, Chari DM, O'Leary MT, Zhao C, Keirstead HS, Blakemore WF, et al. Depletion of endogenous oligodendrocyte progenitors rather than increased availability of survival factors is a likely explanation for the enhanced survival of transplanted oligodendrocyte progenitors in X-irradiated compared to normal CNS. Neuropathol Appl Neurobiol 2001; 27: 59–67.
Kotter MR, Setzu A, Sim FJ, Van Rooijen N, Franklin RJM. Macrophage depletion impairs oligodendrocyte remyelination following lysolecithin-induced demyelination. Glia 2001; 35: 204–12.
Levine JM, Reynolds R, Fawcett JW. The oligodendrocyte precursor cell in health and disease. Trends Neurosci 2001; 24: 39–47.
Love S, Wiley CA, Fujinami RS, Lampert PW. Effects of regional spinal X-irradiation on demyelinating disease caused by Theiler's virus, mouse hepatitis virus or experimental allergic encephalomyelitis. J Neuroimmunol 1987; 14: 19–33.
Lunn KF, Clayton MK, Duncan ID. The temporal progression of the myelination defect in the taiep rat. J Neurocytol 1997; 26: 267–81.
Mason JL, Suzuki K, Chaplin DD, Matsushima GK. Interleukin-1 promotes repair of the CNS. J Neurosci 2001; 21: 7046–52.
Milward EA, Zhang SC, Zhao M, Lundberg C, Ge B, Goetz BD, et al. Enhanced proliferation and directed migration of oligodendroglial progenitors co-grafted with growth factor-secreting cells. Glia 2000; 32: 264–70.
O'Leary MT, Blakemore WF. Oligodendrocyte precursors survive poorly and do not migrate following transplantation into the normal adult central nervous system. J Neurosci Res 1997; 48: 159–67.
Setzu A, ffrench-Constant C, Franklin RJ. CNS axons retain their competence for myelination throughout life. Glia 2004; 45: 307–11.
Silberstein FC, De Simone R, Levi G, Aloisi F. Cytokine-regulated expression of platelet-derived growth factor gene and protein in cultured human astrocytes. J Neurochem 1996; 66: 1409–17.
Song J, Carson JH, Barbarese E, Li FY, Duncan ID. RNA transport in oligodendrocytes from the taiep mutant rat. Mol Cell Neurosci 2003; 24: 926–38.
Stripecke R, Carmen VM, Skelton D, Satake N, Halene S, Kohn D. Immune response to green fluorescent protein: implications for gene therapy. Gene Ther 1999; 6: 1305–12.
Vela JM, Molina-Holgado E, Arevalo-Martin A, Almazan G, Guaza C. Interleukin-1 regulates proliferation and differentiation of oligodendrocyte progenitor cells. Mol Cell Neurosci 2002; 20: 489–502.
Wolswijk G. Chronic stage multiple sclerosis lesions contain a relatively quiescent population of oligodendrocyte precursor cells. J Neurosci 1998; 18: 601–9.
Woodruff RH, Fruttiger M, Richardson WD, Franklin RJM. Platelet-derived growth factor regulates oligodendrocyte progenitor numbers in adult CNS and their response following CNS demyelination. Mol Cell Neurosci 2004; 25: 252–62.
Zhang S-C, Goetz BD, Duncan ID. Suppression of activated microglia promotes survival and function of transplanted oligodendroglial progenitors. Glia 2003; 41: 191–98.(A. K. Foote and W. F. Blakemore)
Summary
A major challenge in multiple sclerosis research is to understand the cause or causes of remyelination failure and to devise ways of ameliorating its consequences. This requires appropriate experimental models. Although there are many models of acute demyelination, at present there are few suitable models of chronic demyelination. The taiep rat is a myelin mutant that shows progressive myelin loss and, by 1 year of age, its CNS tissue has many features of chronic areas of demyelination in multiple sclerosis: chronically demyelinated axons present in an astrocytic environment in the absence of acute inflammation. Using the taiep rat and a combination of X-irradiation and cell transplantation, it has been possible to address a number of questions concerning remyelination failure in chronic multiple sclerosis lesions, such as whether chronically demyelinated axons have undergone changes that render them refractory to remyelination and why remyelination is absent when oligodendrocyte progenitor cells (OPCs) are present. Our experiments show that (i) transplanted OPCs will not populate OPC-containing areas of chronic demyelination; (ii) myelination competent OPCs can repopulate OPC-depleted chronically demyelinated astrocytosed tissue, but this repopulation does not result in remyelination—closely resembling the situation found in some multiple sclerosis plaques; and (iii) the induction of acute inflammation in this non-remyelinating situation results in remyelination. Thus, we can conclude that axonal changes induced by chronic demyelination are unlikely to contribute to remyelination failure in multiple sclerosis. Rather, remyelination fails either because OPCs fail to repopulate areas of demyelination or because if OPCs are present they are unable to generate remyelinating oligodendrocytes owing to the presence of inhibitory factors and/or a lack of the stimuli required to activate these cells to generate remyelinating oligodendrocytes. This non-remyelinating situation can be transformed to a remyelinating one by the induction of acute inflammation.
Key Words: multiple sclerosis; oligodendrocyte progenitor cell; remyelination; taiep rat; transplantation
Abbreviations: FBS = fetal bovine serum; OPC = oligodendrocyte progenitor cell; PDL = poly-D-lysine
Introduction
Remyelination failure is a feature of multiple sclerosis that probably contributes to the relentless progression of this disease. A major challenge in multiple sclerosis research is to understand the cause, or causes, of remyelination failure and devise ways of ameliorating its consequences. Amelioration could be achieved by the development of remyelination enhancing therapies. These might take the form of stimulating the brain's inherent remyelinating potential or transplanting cells that have remyelinating potential. Evaluation of both approaches requires appropriate animal models and an understanding of why remyelination fails in multiple sclerosis.
It has been demonstrated that transplantation of myelin forming cells can result in extensive myelin sheath formation when undertaken during development or in association with acute demyelination (reviewed, Duncan et al., 1997; Blakemore and Franklin, 2000; Baron-Van Evercooren and Blakemore, 2004). Such environments can be regarded as physiological recipient environments since either myelination or remyelination is occurring, and thus the introduced cells are exposed to an environment that is conducive to myelin formation. Deciding how to approach a pathological environment such as pertains in areas of chronic demyelination, where oligodendrocyte progenitor cells (OPCs) with the potential for remyelination may be present but fail to achieve remyelination (Wolswijk, 1998; Chang et al., 2000, 2002), has been hampered by a lack of suitable models. To be suitable, the experimental model should resemble chronic multiple sclerosis lesions in the following respects: (i) it should have axons that have remained without myelin sheaths for some considerable time; (ii) it should contain OPCs and an absence of remyelination; (iii) the demyelinated axons should be present in an area that contains astrocytes; and (iv) there should be an absence of acute inflammation. Results from experiments using such a model would have relevance for understanding how to enhance endogenous remyelination in areas of chronic demyelination, as well as potential application to cell therapy. Our study shows that such a model can be achieved using the taiep rat. The taiep rat has an autosomal recessive genetic defect which leads to an accumulation of microtubules in oligodendrocytes and altered intracellular transport (Song et al., 2003). This results in hypomyelination and progressive demyelination such that, by 1 year of age, all small diameter and many medium diameter CNS axons are without myelin sheaths (Duncan et al., 1992; Lunn et al., 1997; Foote and Blakemore, 2005). The loss of myelin sheaths is accompanied by an astrocytosis (Duncan et al., 1992; Foote and Blakemore, 2005) and an ongoing acute inflammatory response is absent. Therefore, the tissue has the features of chronic areas of demyelination in multiple sclerosis, with chronically demyelinated axons present in a non-inflammatory astrocytic environment.
By using the taiep rat in combination with X-irradiation to deplete the endogenous OPC population and transplantation of remyelination competent OPCs, it has been possible to examine a number of suggested reasons for remyelination failure in multiple sclerosis and to predict the likely outcome of transplanting OPCs into areas of chronic demyelination. In particular, our studies show that (i) transplanted OPCs will not populate OPC-containing areas of chronic demyelination; (ii) OPCs can repopulate OPC-depleted chronically demyelinated astrocytosed tissue, but this repopulation by myelination competent cells does not result in remyelination, closely resembling the situation found in some multiple sclerosis plaques; and (iii) the induction of acute inflammation in this non-remyelinating situation stimulates remyelination. Thus, axonal changes induced by chronic demyelination are unlikely to contribute to remyelination failure in multiple sclerosis. Rather, remyelination fails either because OPCs fail to repopulate areas of demyelination or because if they are present they are dysfunctional because of the presence of inhibitory factors and/or a lack of the stimuli required to activate these cells to generate remyelinating oligodendrocytes. It is therefore significant that a non-remyelinating situation can be transformed to a remyelinating one by the induction of acute inflammation.
Methods
Animals
Taiep rats were obtained from Dr M. Roncagliolo (University of Valparaiso, Chile) and bred by mating either affected to affected animals or affected to non-affected heterozygotes. Control animals used for transplantation were heterozygous littermates of the taiep rats. Sprague–Dawley rats carrying the autosomal dominant GFP gene on an actin promoter were obtained from Professor M. Okabe (Osaka University, Japan) and heterozygous males were mated to normal Sprague–Dawley females. All animal experiments were performed under a Home Office licence and with ethical approval from the University of Cambridge.
Progression of pathology
To study the progression of pathology in the taiep rats, animals were perfused with 4% glutaraldehyde at 3, 7, 10, 14 and 18 months of age (at least two animals for each time point) and tissue from the spinal cord was processed into resin. Sections 1 μm thick were cut and examined using light microscopy following toluidine blue staining. For electron microscopy, 90 nm sections were cut, stained with uranyl acetate and lead citrate and examined using a Hitachi Model H600 transmission electron microscope.
Transplantation
Using halothane anaesthesia and aseptic surgical techniques, a laminectomy was performed at the level of T13 to expose the dorsal spinal cord for cell transplantation. With the aid of a three-way micromanipulator, the tip of a glass micropipette attached to a 10 μl Hamilton syringe was inserted into the dorsal funiculus and 1 μl of cell suspension (105 cells) was injected. Some animals (as indicated in Table 1) were immunosuppressed with cyclosporin A (Sandoz, 15 mg/kg s/c daily) from 1 day before transplantation. All animals used as transplant recipients were between 11 and 14 months of age.
Cell preparations
Unlabelled OPC enriched preparation
Mixed glial cell cultures were prepared from the cerebral hemispheres of 2–3 day old Sprague–Dawley rat pups. Briefly, cerebral hemispheres were removed, stripped of their meninges and mechanically dissociated in MEM containing 20 mM HEPES (Invitrogen). Cells were harvested by filtration through a 20 μm nylon mesh followed by centrifugation (500 x g for 5 min) and plated on poly-D-lysine (PDL) coated 80 cm2 tissue culture flasks containing DMEM (Invitrogen) supplemented with glucose (6 mg/ml) and 10% heat inactivated fetal bovine serum (FBS; Invitrogen). Each flask contained cells from two animals. Every 3–4 days 60% of the culture medium was replaced. After 10 days in vitro, cultures were shaken on an orbital shaker (200 r.p.m. for 2 h) to remove loosely adhered microglia; the medium was replaced and then the cultures were shaken again (375 r.p.m. overnight) to harvest OPCs. The detached cells were collected, the more adhesive contaminating cells were allowed to adhere on to non-tissue culture coated petri dishes for 30 min at 37°C and the less adherent OPCs were collected, centrifuged and resuspended in MEM–HEPES at a concentration of 105 cells per microlitre for transplantation. The mixed glial cultures were maintained in culture for a further 7 days before shaking for a third time to harvest, in the same way as before, additional OPCs for transplantation.
GFP labelled OPC enriched preparation
OPC enriched preparations were prepared as described for unlabelled cells using 2–3 day old rat pups heterozygous for GFP, as indicated by green fluorescence under UV light.
LacZ labelled OPCs
OPC enriched cell preparations were prepared as described for unlabelled cells, using 2–3 day old rat pups. Before transplantation, the OPC enriched preparation was plated onto PDL coated 80 cm2 tissue culture flasks containing 70% DMEM/F12 (with L-glutamine; Invitrogen) and 30% medium conditioned by the B104 neuronal cell line, prepared using methods described by Fok-Seang (1995). The medium was supplemented with N2 (10 μl/ml; Invitrogen), BSA (0.66 mg/ml; Sigma), glucose (2.25 mg/ml) and the growth factors FGF-2 (20 ng/ml; Sigma) and PDGF-AA (10 ng/ml; Sigma). Fresh growth factors were added every 24 h, and 60% of the medium was replaced with fresh medium every 48 h.
Replication deficient Maloney murine leukaemia virus, lacZ (with nuclear localization sequence), was collected from an MFG packaging cell line (a gift of Dr Richard Asher). It was concentrated by precipitation using PEG8000 (Sigma) and ultracentrifugation through a sucrose density gradient. The titre of the concentrated virus preparation was estimated by titration on the 3T3 fibroblast cell line. Approximately 106 virus particles in 10 μl DMEM were added to each flask of the OPC cell preparation daily for 5 days. After a further 5 days, the OPC cell preparation was detached from the flask using a brief (<10 min) incubation with trypsin and EDTA (ATV, Sigma). Soya bean trypsin inhibitor (Sigma) was added and the detached cells were collected, centrifuged and resuspended in MEM–HEPES for transplantation at a concentration of 50 000 cells per microlitre.
X-irradiation
To irradiate the spinal cord, animals were anaesthetized using fentanyl citrate/fluanisone (0.3 ml/kg IM, Hypnorm, Janssen Pharmaceuticals) and an intraperitoneal injection of diazepam (2.5 mg/kg, Phoenix Pharmaceuticals) and placed in lateral recumbency. A 4 cm length of the spinal cord, centred on T13, was exposed to 40 Gy X-irradiation using a Pantak 255-kV radiotherapy machine (Blakemore and Crang, 1992).
Experiments and tissue processing
Four transplantation experiments were performed (summarized in Table 1). (i) Ten adult taiep rats were transplanted with unlabelled cells. After 4 weeks these animals were perfused with 4% glutaraldehyde and the tissue examined in resin sections for evidence of remyelination. (ii) Five taiep and three irradiated taiep rat spinal cords were transplanted with lacZ labelled cells and perfused after 4 weeks with 3% paraformaldehyde/1% gluteraldehyde. The cord was cut into 1.5 mm transverse blocks and the tissue was reacted with X-gal. The blocks were processed for light and electron microscopy to examine remyelination and the distribution of transplanted cells. (iii) GFP labelled cells were transplanted into control spinal cord (n = 3), taiep spinal cord (n = 3), irradiated control spinal cord (n = 3) and irradiated taiep spinal cord (n = 4). The tissue from these animals was analysed in longitudinal cryostat sections after 4 weeks to examine the distribution and immunophenotype of transplanted cells. (iv) Ten irradiated taiep spinal cords were transplanted with GFP cells, and five of them received two unilateral injections of saline and sterile charcoal into the lateral funiculus 3 weeks after implantation of cells. These animals were perfused with 4% paraformaldehyde 6 weeks after transplantation. The spinal cord of each animal was cut into 1.5 mm transverse blocks and alternate blocks were processed for either resin (postfixed in 4% gluteraldehyde followed by 2% osmium tetroxide and then resin embedded) or cryostat sectioning. The animals in experiments 1 and 4 were not immunosuppressed, whereas animals in experiments 2 and 3 were given immunosuppressant doses of cyclosporin.
In addition, in order to examine the effect of irradiation on the endogenous OPC population, the spinal cords of 4 adult taiep rats were irradiated over a 7 mm length. Animals were perfused 4 weeks after irradiation and in situ hybridization for PDGFR was performed on cryostat sections. A saline/charcoal injection was also made into the dorsolateral funiculus of three irradiated taiep spinal cords (which had not received transplanted OPCs) and the effect of this injection on the tissue was examined in cryostat and resin sections 3 days post injection.
As well as transplantation into taiep spinal cord, all cell preparations were injected into X-irradiated ethidium bromide lesions in normal adult rat spinal cord 3 days after lesion induction, and the animals were killed 1 month later to evaluate the extent of remyelination (Blakemore and Crang, 1992).
Immunostaining and ISH
Samples of the cell preparation used for transplantation were plated onto CC2-coated chamber slides (Labtec) for 12 h in DMEM/F12 containing 1% FBS, then fixed with 4% paraformaldehyde and characterized by immunocytochemistry using antibodies to A2B5 (mouse IgM, Roche, 1:200), O4 (mouse IgM hybridoma supernatant, a gift from Dr Mark Noble), GalC (mouse IgG, Chemicon 1:100), GFAP (polyclonal rabbit IgG, Dako, 1:150) and CD11b (OX42, mouse IgG, Serotec, 1:200). Cell nuclei were labelled with Hoescht (Calbiochem, 1:500).
For immunohistochemistry, cryostat sections were stained using antibodies to NG2 proteoglycan (rabbit IgG, Chemicon, 1:150), GFAP (rabbit IgG, Dako, 1:150), APC (CC2, mouse IgG, Oncogene, 1:100), CD45 (LCA, mouse IgG, Chemicon, 1:100), rat B cells (mouse IgG, Serotec, 1:100), CD4 (mouse IgG, Serotec, 1:100), CD8 (mouse IgG) and GFP (rabbit IgG or mouse IgG, Molecular Probes, 1:300). Appropriate FITC, TRITC or Cy3 conjugated secondary antibodies were used (Jackson Laboratories). PDGFR in situ hybridization was performed on cryostat sections using a DIG-labelled riboprobe (cDNA, a gift from Professor W. D. Richardson).
Analysis
The repopulation of non-irradiated and irradiated tissue by transplanted cells after 4 weeks was observed in transverse resin sections for lacZ labelled cells and in parasagittal sections for GFP labelled cells. The total length of repopulation was measured from the latter, as was the length of the area of increased GFP cell density containing APC+GFP+ cells (APC being a marker of mature oligodendrocytes). The lengths were measured from a minimum of three sections for each animal, using sections that passed through the centre of the transplant site.
In resin sections, transplant mediated remyelination was assessed by the detection of areas in which all axons were myelinated. For lacZ labelled cells, remyelination was confirmed as being transplant derived by its association with labelled cells in closely adjacent sections. For GFP labelled cells, regions of high cell density containing GFP+APC+cells could be detected in the cryostat blocks either side of resin blocks showing complete remyelination. For quantification, the cross-sectional area of white matter showing evidence of complete remyelination in the resin sections was outlined and the area measured using a live capture digital imaging system (MCID). By maintaining the craniocaudal order of the resin blocks in relation to the transplant site, the area of remyelination was compared at different levels of the cord and a volume of remyelination estimated for each animal.
Results
Examination of the spinal cord of taiep rats of different ages confirmed that there was progressive loss of myelin with age, as previously observed (Lunn et al., 1997; Foote and Blakemore, 2005). By 10 months of age, all small diameter axons and most medium sized axons lacked myelin sheaths, while large diameter axons were surrounded by thinner than normal myelin sheaths. In particular, almost all the axons in the corticospinal tract, gracile tract and the tracts in the dorsolateral funiculi were without detectable myelin sheaths in toluidine blue stained resin sections and at the ultrastructural level (Fig. 1).
Only small focal areas of remyelination are found when OPCs are injected into areas of chronic demyelination
To determine whether axons that had been chronically demyelinated were receptive to remyelination we transplanted oligodendrocyte precursors into the dorsal funiculus of the thoracolumbar spinal cord in adult (11–14 month old) taiep rats. Experiments were undertaken using cells from normal animals, cells transfected with the lacZ gene and cells derived from GFP expressing animals. The cell preparations used contained a similar proportion of OPCs (Table 2) and each showed the expected remyelinating capacity when injected into X-irradiated ethidium bromide lesions (data not shown). Four weeks after transplantation into the taiep rat spinal cord only small areas of remyelination could be detected in resin-embedded tissue (Fig. 2A–D). Using cells prepared from GFP expressing rats or lacZ labelled cells it was apparent that all the transplanted cells were restricted to a small area near to the point of cell implantation (Figs. 2E and 3E). The length of this region was consistent with the passive spread of cells that results from the injection of the cell suspension (Franklin et al., 1996; O'Leary and Blakemore, 1997). Therefore, repopulation of the chronically demyelinated tissue by the transplanted cells had not occurred, indicating that some property of the host tissue was inhibitory to widespread integration of the transplanted cells. However, although very restricted in its distribution, the presence of remyelination indicates that chronically demyelinated axons of the taiep rat can be remyelinated by myelination competent cells.
Depletion of OPCs results in widespread repopulation of OPC-depleted tissue by transplanted OPCs but does not result in extensive remyelination
The probable explanation for the lack of migration away from the site of implantation is the presence of endogenous OPCs, since it has been shown in previous studies that transplanted OPCs fail to colonize normal tissue in adult animals, but will colonize OPC-depleted tissue (Franklin et al., 1996; O'Leary and Blakemore, 1997). To confirm this, the spinal cord of taiep and normal rats was exposed to 40 Gy of X-irradiation—the dose known to deplete rat tissue of its OPC population in normal (Hinks et al., 2001; Chari and Blakemore, 2002a) and taiep rats (Fig. 3A and B) (Foote and Blakemore, 2005). Transplantation of cells into taiep spinal cord exposed to X-irradiation resulted in extensive repopulation of the OPC-depleted tissue (Fig. 3C and E). A greater area of remyelination was also seen around the point of cell implantation in irradiated animals than in non-irradiated animals, detected either on resin sections or, indirectly, as an area of increased GFP cell density containing many oligodendrocytes (APC positive cells) (Fig. 3D and F). However, in contrast to these areas of remyelination close to the injection site, transplanted cells repopulated the whole cross-sectional area of the spinal cord at the level of the implantation site and extended longitudinally cranially and caudally for some considerable distance (Fig. 3E). The length of repopulated tissue was similar in X-irradiated normal and taiep rats (Fig. 3E), and thus the astrocytosis and the chronically demyelinated axons present in the taiep rats were not significantly influencing the rate of repopulation. Nearly all the transplanted cells that established themselves away from the transplant site co-labelled with antibodies to the oligodendrocyte progenitor marker NG2 (Fig. 3K). It was noteworthy that the density of transplant derived OPCs in areas distant from the transplant site was similar to that for OPCs in normal tissue (50–60 cells/mm2) and there was no evidence of remyelination or generation of transplant derived oligodendrocytes (Fig. 3G–I, L) except in the area of remyelination associated with the transplant site (Fig. 3D, M–O). We had therefore achieved a situation in which OPCs were present adjacent to demyelinated axons without evidence of oligodendrocyte differentiation and remyelination. Thus, failure of remyelination at a distance from the implantation point cannot be attributed to a failure of OPCs to interact with demyelinated axons.
Induction of inflammation stimulates remyelination in areas of chronic demyelination
Since the widespread repopulation of the chronically demyelinated taiep spinal cord by OPCs resulted in remyelination only near to the transplant site, we hypothesized that the act of implantation of cells was inducing changes in the tissue that rendered it conducive to remyelination, whereas the mere repopulation of tissue by myelination competent OPCs was insufficient to bring about remyelination. To test this hypothesis, we conducted an experiment in which we first allowed repopulation of the irradiated taiep tissue by transplanted OPCs and then created, at a site distant from the cell transplant site, a situation that would mimic the injection of cells (Fig. 4A). This was achieved by injecting a small volume of saline (containing sterile charcoal to mark the site). The site chosen for this injection was the dorsal part of the lateral funiculus where there was an almost complete absence of myelin (Fig. 4D) and where remyelination had not been observed following transplantation of OPCs into the dorsal funiculus. In this experiment, unlike the previous experiments with genetically labelled cells, the recipient animals were not given cyclosporin, since we did not want to interfere with any inflammatory changes that may be associated with the saline injection. When animals were examined 3 weeks after the saline injection, remyelination was associated with the site of saline/charcoal injection in all five animals injected (Fig. 4C) and there was no evidence of remyelination in the same area on the contralateral side of the spinal cord (Fig. 4D). Since both areas contained transplanted OPCs, this result indicates that the injection of saline had stimulated the transplant established OPCs to remyelinate the tissue.
To examine the consequences of injecting saline and charcoal into the irradiated spinal cord, resin embedded and frozen sections of the injection site were examined 3 days after this injection. The injection resulted in a marked increase the number of macrophages and OX42 positive cells (Fig. 4B). Lymphocytes (CD4, CD8, B cell) were not detected in these areas using immunohistochemistry.
In some of the transplant repopulated animals subjected to the saline/charcoal injection, as well as in some animals allowed to survive for 6 weeks without saline/charcoal injection, we also found extensive remyelination around the periphery of the spinal cord (Fig. 4E). It was noticeable that, in the animals showing the most extensive remyelination, inflammatory cells were present in the meninges in toluidine blue stained resin sections, and CD4 and CD8 positive cells could be detected by immunostaining within the CNS parenchyma and within the meninges (Fig. 4F). When the extent of complete remyelination was measured and compared between the animals with and without evidence of meningeal infiltration (as detected in resin sections), it was clear that the extent of remyelination in the animals with cell infiltration was significantly greater than in those without evidence of meningitis (Fig. 4G and H); in the sections showing the most extensive remyelination, up to 80% of the white matter was remyelinated. Our results therefore show that induction of inflammation in non-remyelinating OPC-populated areas of chronic demyelination stimulates remyelination.
Discussion
Our studies represent the first in which extensive remyelination has been achieved using a model of chronic demyelination in adult animals. As a consequence, they establish a number of principles that have relevance both to understanding the reasons for remyelination failure in multiple sclerosis and to evaluating remyelination enhancing strategies for this disease. First, by transplanting OPCs into this model of chronic demyelination, we have confirmed the recent observation on the retina that axons that have been without myelin for a considerable time can be myelinated (Setzu et al., 2004) and thus axonal changes resulting from demyelination are unlikely to play a role in remyelination failure (Chang et al., 2002). Second, by replacing taiep OPCs with normal OPCs we have demonstrated that cells capable of undertaking remyelination can be present in tissue containing receptive demyelinated axons without consequent remyelination. Thus, the OPCs found in some chronic multiple sclerosis lesions and considered by some to be non-functional (Greenwood and Butt, 2003) may not in fact be so, since they, like the transplanted OPCs in this study, may simply require induction of inflammation to enable their proliferation and subsequent differentiation into remyelinating cells. Third, by showing that transplanted OPCs established away from the implantation site have a similar density to that found in normal tissue, our study extends to pathological tissue a conclusion drawn from studies on normal tissue that, in the absence of inflammation, OPC numbers are tightly regulated (Dawson et al., 2003; Woodruff et al., 2004). This has implication for attempts to introduce transplanted OPCs into areas of chronic demyelination as well as normal tissue, since it indicates that transplanted OPCs can be established only by either depleting tissue of endogenous OPCs, or changing the environment so that it can support these supernummary cells. Finally, our studies demonstrate that although transplanted OPCs can be established in OPC-depleted areas of chronic demyelination, remyelination is dramatically stimulated only when the environment is made conducive to OPC expansion and differentiation; this can be achieved by establishing acute inflammation.
In agreement with a number of previous studies using longer-lived mature myelin mutants (Archer et al., 1997; Milward et al., 2000; Zhang et al., 2003), we found very limited remyelination following transplantation of myelination competent cells into the spinal cord of taiep rats. In the shaking pup the extent of remyelination decreased with the age of the recipient and was enhanced when late embryonic rather than neonatal cells were transplanted in adult animals (Archer et al., 1997), indicating that donor age influences the extent of remyelination. The poor results in the Long Evans rat were attributed to the presence of activated microglial cells, as down-regulation of this response by minocycline enhanced cell survival and remyelination (Zhang et al., 2003). Our study clearly shows that transplanted OPCs, as in normal tissue (Franklin et al., 1996; O'Leary and Blakemore, 1997), do not enter tissue containing endogenous OPCs. Since there can be no remyelination without repopulation of tissue by myelination competent cells, a major reason for the failure to achieve significant remyelination following transplantation into these long-lived myelin mutants is the presence of endogenous OPCs.
The rate of OPC repopulation of OPC-depleted tissue by transplanted OPCs was not significantly different in taiep and normal rats, indicating that the chronically demyelinated environment found in the taiep rat does not significantly inhibit or enhance the OPC repopulation rate. The repopulation rate of the transplanted OPCs was slightly slower in the taiep than in normal animals, which may suggest an inhibitory role for astrocytes in vivo, as has been demonstrated in vitro (Fok-Seang et al., 1995). In a previous study, we found no difference in the rate of endogenous OPC repopulation between normal and taiep rats of similar age (Foote and Blakemore, 2005). However, failure to detect a difference in that study may have been a consequence of the very slow OPC repopulation rate in old animals. Despite a possible slight negative influence on the rate of repopulation, it is clear that transplanted OPCs colonize OPC-depleted chronically demyelinated tissue at a significantly faster rate than endogenous cells [1.5 mm/week (present data) compared with 0.2 mm/week (data from Foote and Blakemore, 2005)]. Thus, where tissue is depleted of OPCs, transplanted neonatal OPCs can repopulate normal and chronically demyelinated tissue more rapidly than endogenous cells.
More important than the identification of how to achieve OPC repopulation in areas of chronic demyelination was the finding that this widespread repopulation resulted in only a modest increase in remyelination. The increased remyelination was restricted to the transplant area, and away from this area there was no evidence of remyelination despite repopulation of the tissue by the transplanted myelination competent OPCs. Thus, repopulation and remyelination are distinct processes, with the former not necessarily resulting in the latter. The non-remyelinating OPC-repopulated chronically demyelinated tissue that we have created in the taiep rat is comparable to the situation observed in non-remyelinating OPC-populated multiple sclerosis plaques (Chang et al., 2002). Since we were able to demonstrate that the chronically demyelinated axons of the taiep rat could be remyelinated under some circumstances, we can conclude that chronically demyelinated tissue has properties that either inhibit or fail to stimulate remyelination. Our studies show a clear relationship between acute inflammation and successful remyelination. This adds to the growing body of evidence that acute inflammation provides the stimulus that initiates remyelination (reviewed, Franklin, 2002) and provides proof of principle for the ‘temporal mismatch’ hypothesis as an explanation for remyelination failure in multiple sclerosis (Chari and Blakemore, 2002b). This hypothesis proposed that if OPCs were destroyed in large areas of demyelination, remyelination would fail because the slow rate of repopulation shown by endogenous OPCs in old individuals would result in a temporal mismatch between the acute inflammatory environment associated with myelin sheath breakdown and the interaction of OPCs with demyelinated axons; that is, in the absence of an acute inflammatory environment to provide the signals required to drive remyelination, the process will fail. In a previous study, we obtained some support for this concept by delaying the interaction of OPCs with demyelinated axons by transplanting cells at a distance from a focus of acute demyelination induced in OPC-depleted tissue, and showing that the extent of remyelination was significantly reduced compared with the situation when OPCs were present at the time of, or soon after, demyelination was initiated (Blakemore et al., 2002). Having established that OPC repopulation rates are extremely slow in old animals (Chari et al., 2003; Foote and Blakemore, 2005), our current results strengthen the temporal mismatch hypothesis by demonstrating for the first time in an experimental model the three further situations required for this hypothesis to be sustainable: (i) OPCs can repopulate OPC-depleted chronically demyelinated tissue; (ii) in the absence of acute inflammation, these repopulating cells persist without remyelination occurring; and (iii) when an acute inflammatory environment is created, remyelination occurs efficiently.
The finding that areas of chronic demyelination can contain both axons that can be remyelinated and a normal density of remyelination competent OPCs without evidence of remyelination would indicate that, unlike in myelination during development, the axon does not provide the stimulus that initiates OPC proliferation (Barres and Raff, 1999). It is well established that OPCs become activated, increase in number and change their morphology in association with acute damage to the central nervous system (Levine et al., 2001). It is not known whether this activation occurs as a consequence of changes in properties of the environment that regulate the activation status of OPCs or are a consequence of direct stimulation of the OPC by factors associated with acute inflammation. In this context it is known that certain molecules, such as Il-1, associated with acute inflammation can stimulate OPC division directly (Vela et al., 2002) and could influence OPC division and survival via their effects on astrocytes (Aloisi et al., 1995; Silberstein et al., 1996; Mason et al., 2001). In adult rats, demyelination is associated with rapid activation of OPCs, as indicated by up-regulation of the transcription factor Nkx2.2 and olig2 (Fancy et al., 2004), and, in terms of remyelination, the two consequences of OPC activation are cell division and differentiation into remyelinating oligodendrocytes. The OPCs which repopulated the OPC-depleted chronically demyelinated taiep spinal cord established themselves at a normal density and showed a normal morphology and no evidence of oligodendrocyte differentiation, suggesting that they are unstimulated and under the control of the factors that regulate OPC behaviour in normal tissue. However, once activated by acute inflammation they proliferate and differentiate into remyelinating oligodendrocytes. Investigation is needed of whether inflammation also induces changes on the axon that would facilitate myelination, such as removal of the polysialylated neural cell adhesion molecule shown to inhibit myelination (Charles et al., 2000).
In the present experiments two forms of inflammation were associated with stimulation of remyelination. The saline/charcoal injections stimulated a predominantly macrophage (moncyte/macrophage/microglial) response, whereas lymphocyte infiltration was a feature of the more widespread remyelination found in the animals with meningitis. Macrophages have been shown to have a significant role in facilitating remyelination (Kotter et al., 2001), as have both lymphocytes (Bieber et al., 2003) and the inflammatory responses associated with rejection of xenogeneic cells (Blakemore et al., 1995). It was not possible to establish the cause of the inflammatory reaction in the animals with meningitis and extensive remyelination around the periphery of the spinal cord. The most probable explanation is that it represents an immune response to the transplanted cells. This group of animals was not immunosuppressed, the transplanted cells were allografts and expressed a foreign protein and, moreover, the tissue had been subjected to a high dose of X-irradiation. The dose of X-irradiation used in our studies is known to render tissue more susceptible to immune-mediated inflammation (Love et al., 1987) and GFP can be a powerful immunogen (Stripecke et al., 1999; Follenzi et al., 2004). Although it is clear that acute inflammation provides a stimulus that initiates remyelination and that macrophages and acutely reactive astrocytes have a positive effect on the success of remyelination (Franklin et al., 1991; Hinks and Franklin, 1999), there is also evidence that activated microglia and astrocytes can have a negative effect. The reactive microglia present in the Long Evans rat (Zhang et al., 2003), which are also present in the taiep rat (Goetz et al., 2000), have been shown to have a negative effect on remyelination and, unlike during the acute situation, astrocytes have been implicated as having a deleterious effect on remyelination when the inflammatory response initiated by demyelination is waning (Blakemore et al., 2002; 2003). One therefore has to consider that astrocytes and microglia/macrophages can have both positive and negative effects on remyelination. The duality of influence of a single cell type is also apparent in vitro, where macrophages can either enhance or impede remyelination in co-culture systems (Diemel et al., 2004). Similarly, astrocytes impede OPC migration under some culture conditions but can be altered by exposure to cytokines to negate this effect (Fok-Seang et al., 1998). It is likely, therefore, that it is the nature of the response of these cells to acute inflammation that creates an environment conducive to remyelination. In this context it is possible that the immune modulating therapies currently in use for multiple sclerosis could be having a deleterious effect on remyelination. However, OPC repopulation was not affected by the immunosuppressant regimen used in the present study, indicating once again that OPC repopulation of OPC-depleted tissue and remyelination are distinct processes.
In conclusion, our experiments have relevance for both interpreting the pathology present in areas of chronic demyelination in multiple sclerosis and developing OPC transplantation as a therapy for chronic demyelinating diseases. They demonstrate that (i) the failure of remyelination observed in OPC-containing areas of chronic demyelination cannot be taken as evidence that chronically demyelinated axons have undergone changes that preclude their remyelination; (ii) as with normal tissue, the presence of endogenous OPCs in areas of chronic demyelination prevents the establishment of transplanted OPCs and (iii) acute inflammation provides the stimulus for remyelination.
Acknowledgements
A.F. was in receipt of a BBRSC studentship. The taiep rats were made available by Dr M. Roncagliolo, with funds from the Myelin Project. The GFP rats were made available by Professor M. Okabe. Funds from the Multiple Sclerosis Society and the Medical Research Council, UK, were used to support this study. The technical assistance of Jennifer Gilson and Mike Peacock is gratefully acknowledged.
References
Aloisi F, Borsellino G, Care A, Testa U, Gallo P, Russo G, et al. Cytokine regulation of astrocyte function: in-vitro studies using cells from the human brain. Int J Dev Neurosci 1995; 13: 265–74.
Archer DR, Cuddon PA, Lipsitz D, Duncan ID. Myelination of the canine central nervous system by glial cell transplantation: a model for repair of human myelin disease. Nat Med 1997; 3: 54–9.
Baron-Van Evercooren A, Blakemore WF. Remyelination through engraftment. In: Lazzarini RA, editor. Myelin biology and disorders. New York: Elsevier Science; 2004. p. 143–72.
Barres BA, Raff MC. Axonal control of oligodendrocyte development. J Cell Biol 1999; 147: 1123–8.
Bieber AJ, Kerr S, Rodriguez M. Efficient central nervous system remyelination requires T cells. Ann Neurol 2003; 53: 680–4.
Blakemore WF, Crang AJ. Transplantation of glial cells into areas of demyelination in the adult rat spinal cord. In: Dunnett S, Bjorklund A, editors. Neural transplantation: a practical approach. Oxford: Oxford University Press; 1992. p. 105–22.
Blakemore WF, Franklin RJM. Transplantation options for therapeutic CNS remyelination. Cell Transplant 2000; 9: 289–94.
Blakemore WF, Crang AJ, Franklin RJM, Tang K, Ryder S. Glial cell transplants that are subsequently rejected can be used to influence regeneration of glial cell environments in the CNS. Glia 1995; 13: 79–91.
Blakemore WF, Chari DM, Gilson JM, Crang AJ. Modelling large areas of demyelination in the rat reveals the potential and possible limitations of transplanted glial cells for remyelination in the CNS. Glia 2002; 38: 155–68.
Blakemore WF, Gilson JM, Crang AJ. The presence of astrocytes in areas of demyelination influences remyelination following transplantation of oligodendrocyte progenitors. Exp Neurol 2003; 184: 955–63.
Chang A, Nishiyama A, Peterson J, Prineas J, Trapp BD. NG2-positive oligodendrocyte progenitor cells in adult human brain and multiple sclerosis lesions. J Neurosci 2000; 20: 6404–12.
Chang A, Tourtellotte WW, Rudick R, Trapp BD. Premyelinating oligodendrocytes in chronic lesions of multiple sclerosis. N Engl J Med 2002; 346: 165–73.
Chari DM, Blakemore WF. Efficient recolonisation of progenitor-depleted areas of the CNS by adult oligodendrocyte progenitor cells. Glia 2002a; 37: 307–13.
Chari DM, Blakemore WF. New insights into remyelination failure in MS: implications for glial cell transplantation. Mult Scler 2002b; 8: 271–7.
Chari DM, Crang AJ, Blakemore WF. Rate of repopulation of progenitor-depleted areas of the adult CNS by adult oligodendrocyte progenitors (OPCs) declines with age. J Neuropath Exp Neurol 2003; 62: 908–16.
Charles P, Hernandez MP, Stankoff B, Aigrot MS, Colin C, Rougon G, et al. Negative regulation of central nervous system myelination by polysialylated-neural cell adhesion molecule. Proc Natl Acad Sci USA 2000; 97: 7585–90.
Dawson MR, Polito A, Levine JM, Reynolds R. NG2-expressing glial progenitor cells: an abundant and widespread population of cycling cells in the adult rat CNS. Mol Cell Neurosci 2003; 24: 476–88.
Diemel LT, Wolswijk G, Jackson SJ, Cuzner ML. Remyelination of cytokine- or antibody-demyelinated CNS aggregate cultures is inhibited by macrophage supplementation. Glia 2004; 45: 278–86.
Duncan ID, Lunn KF, Holmgren B, Urba-Holmgren R, Brignolo-Holmes L. The taiep rat: a myelin mutant with an associated oligodendrocyte microtubular defect. J Neurocytol 1992; 21: 870–84.
Duncan ID, Grever WE, Zhang S-C. Repair of myelin disease: strategies and progress in animal models. Mol Med Today 1997; 3: 554–61.
Fancy SPJ, Zhao C, Franklin RJM. Increased expression of Nkx2.2 and Olig2 identifies reactive oligodendrocyte progenitor cells responding to demyelination in the adult CNS. Mol Cell Neurosci 2004; 27: 247–54.
Fok-Seang J, Mathews GA, ffrench-Constant C, Trotter J, Fawcett JW. Migration of oligodendrocyte precursors on astrocytes and meningeal cells. Dev Biol 1995; 171: 1–15.
Fok-Seang J, DiProspero NA, Meiners S, Muir E, Fawcett JW. Cytokine-induced changes in the ability of astrocytes to support migration of oligodendrocyte precursors and axon growth. Eur J Neurosci 1998; 10: 2400–15.
Follenzi A, Battaglia M, Lombardo A, Annoni A, Roncarolo MG, Naldini L. Targeting lentiviral vector expression to hepatocytes limits transgene-specific immune response and establishes long-term expression of human antihemophilic factor IX in mice. Blood 2004; 103: 3700–9.
Foote AK, Blakemore WF. Repopulation of oligodendrocyte progenitor cell depleted tissue in a model of chronic demyelination. Neuropathol Appl Neurobiol 2005; Published online 02/12/04; doi:10.1111dp.1365-2990. 2004.00634.x.
Franklin RJM. Why does remyelination fail in multiple sclerosis. Nat Rev Neurosci 2002; 3: 1–12.
Franklin RJM, Crang AJ, Blakemore WF. Transplanted type-1 astrocytes facilitate repair of demyelinating lesions by host oligodendrocytes in adult rat spinal cord. J Neurocytol 1991; 20: 420–30.
Franklin RJM, Bayley SA, Blakemore WF. Transplanted CG4 cells (an oligodendrocyte progenitor cell line) survive, migrate and contribute to repair of areas of demyelination in X-irradiated and damaged spinal cord, but not in normal spinal cord. Exp Neurol 1996; 137: 263–76.
Goetz BD, Zhang S-C, Song J, Duncan ID. Microglial response to progressive demyelination in the taiep mutant rat. Soc Neurosci Abst 2000; 26: 515.7.2000.
Greenwood K, Butt AM. Evidence that perinatal and adult NG2-glia are not conventional oligodendrocyte progenitors and do not depend on axons for their survival. Mol Cell Neurosci 2003; 23: 544–58.
Hinks GL, Franklin RJM. Distinctive patterns of PDGF-A, FGF-2, IGF-1, and TGF-beta1 gene expression during remyelination of experimentally-induced spinal cord demyelination. Mol Cell Neurosci 1999; 14: 153–68.
Hinks GL, Chari DM, O'Leary MT, Zhao C, Keirstead HS, Blakemore WF, et al. Depletion of endogenous oligodendrocyte progenitors rather than increased availability of survival factors is a likely explanation for the enhanced survival of transplanted oligodendrocyte progenitors in X-irradiated compared to normal CNS. Neuropathol Appl Neurobiol 2001; 27: 59–67.
Kotter MR, Setzu A, Sim FJ, Van Rooijen N, Franklin RJM. Macrophage depletion impairs oligodendrocyte remyelination following lysolecithin-induced demyelination. Glia 2001; 35: 204–12.
Levine JM, Reynolds R, Fawcett JW. The oligodendrocyte precursor cell in health and disease. Trends Neurosci 2001; 24: 39–47.
Love S, Wiley CA, Fujinami RS, Lampert PW. Effects of regional spinal X-irradiation on demyelinating disease caused by Theiler's virus, mouse hepatitis virus or experimental allergic encephalomyelitis. J Neuroimmunol 1987; 14: 19–33.
Lunn KF, Clayton MK, Duncan ID. The temporal progression of the myelination defect in the taiep rat. J Neurocytol 1997; 26: 267–81.
Mason JL, Suzuki K, Chaplin DD, Matsushima GK. Interleukin-1 promotes repair of the CNS. J Neurosci 2001; 21: 7046–52.
Milward EA, Zhang SC, Zhao M, Lundberg C, Ge B, Goetz BD, et al. Enhanced proliferation and directed migration of oligodendroglial progenitors co-grafted with growth factor-secreting cells. Glia 2000; 32: 264–70.
O'Leary MT, Blakemore WF. Oligodendrocyte precursors survive poorly and do not migrate following transplantation into the normal adult central nervous system. J Neurosci Res 1997; 48: 159–67.
Setzu A, ffrench-Constant C, Franklin RJ. CNS axons retain their competence for myelination throughout life. Glia 2004; 45: 307–11.
Silberstein FC, De Simone R, Levi G, Aloisi F. Cytokine-regulated expression of platelet-derived growth factor gene and protein in cultured human astrocytes. J Neurochem 1996; 66: 1409–17.
Song J, Carson JH, Barbarese E, Li FY, Duncan ID. RNA transport in oligodendrocytes from the taiep mutant rat. Mol Cell Neurosci 2003; 24: 926–38.
Stripecke R, Carmen VM, Skelton D, Satake N, Halene S, Kohn D. Immune response to green fluorescent protein: implications for gene therapy. Gene Ther 1999; 6: 1305–12.
Vela JM, Molina-Holgado E, Arevalo-Martin A, Almazan G, Guaza C. Interleukin-1 regulates proliferation and differentiation of oligodendrocyte progenitor cells. Mol Cell Neurosci 2002; 20: 489–502.
Wolswijk G. Chronic stage multiple sclerosis lesions contain a relatively quiescent population of oligodendrocyte precursor cells. J Neurosci 1998; 18: 601–9.
Woodruff RH, Fruttiger M, Richardson WD, Franklin RJM. Platelet-derived growth factor regulates oligodendrocyte progenitor numbers in adult CNS and their response following CNS demyelination. Mol Cell Neurosci 2004; 25: 252–62.
Zhang S-C, Goetz BD, Duncan ID. Suppression of activated microglia promotes survival and function of transplanted oligodendroglial progenitors. Glia 2003; 41: 191–98.(A. K. Foote and W. F. Blakemore)