Stimulation of oligodendroglial proliferation and maturation by interleukin Kiernan, B. Tenascin-C inhibits oligodendrocyte precursor cell migration by both adhesion-dependent and adhesion-independent mechanisms. Blaschuk, K. Development , — Garcion, E. Knockout mice reveal a contribution of the extracellular matrix molecule tenascin-C to neural precursor proliferation and migration.
Cell Biol. Shi, J. Purification and characterization of adult oligodendrocyte precursor cells from the rat optic nerve. Jiang, F. Environmental enhancement of growth factor-mediated oligodendrocyte precursor proliferation. Baron, W. EMBO J. This paper reveals how different signalling systems can be intimately linked, highlighting the potential complexity of the signals that regulate OPC behaviour during remyelination.
Tourbah, A. Endogenous aFGF expression and cellular changes after a demyelinating lesion in the spinal cord of adult normal mice: immunohistochemical study. Hinks, G. Messersmith, D. Fibroblast growth factor 2 FGF2 and FGF receptor expression in an experimental demyelinating disease with extensive remyelination. Delayed changes in growth factor gene expression during slow remyelination in the CNS of aged rats.
Expression of the highly polysialylated neural cell adhesion molecule during postnatal myelination and following chemically induced demyelination of the adult mouse spinal cord. Calver, A. Oligodendrocyte population dynamics and the role of PDGF in vivo. Neuron 20 , — Fruttiger, M. Ye, P. Goddard, D. In vivo actions of fibroblast growth factor-2 and insulin-like growth factor-I on oligodendrocyte development and myelination in the central nervous system.
Carson, M. Insulin-like growth factor I increases brain growth and central nervous system myelination in transgenic mice. Neuron 10 , — Komoly, S. Insulin-like growth factor I gene expression is induced in astrocytes during experimental demyelination. USA 89 , — Gveric, D. Insulin-like growth factors and binding proteins in multiple sclerosis plaques. Rosenthal, S. Opposing early and late effects of insulin-like growth factor I on differentiation and the cell cycle regulatory retinoblastoma protein in skeletal myoblasts.
Barres, B. Control of oligodendrocyte number in the developing rat optic nerve. Neuron 12 , — Mckinnon, R. O'Leary, M. Neuron 5 , — Fibroblast growth factor-2 inhibits myelin production by oligodendrocytes in vivo. Canoll, P. Neuron 17 , — Park, S. The erbB2 gene is required for the development of terminally differentiated spinal cord oligodendrocytes.
Viehover, A. Neuregulin: an oligodendrocyte growth factor absent in active multiple sclerosis lesions. Gallo, V. Ghiani, C. Neurotransmitter receptor activation triggers p27 Kip1 and p21 CIP1 accumulation and G1 cell cycle arrest in oligodendrocyte progenitors.
Wang, S. Notch receptor activation inhibits oligodendrocyte differentiation. Neuron 21 , 63—75 Givogri, M. Central nervous system myelination in mice with deficient expression of Notch1 receptor. Charles, P. Negative regulation of central nervous system myelination by polysialylated-neural cell adhesion molecule.
Brain in the press. References 94 and 95 provide a potential mechanism whereby demyelinated axons might have a crucial role in determining whether OPCs associated with them can differentiate into remyelinating cells. Demerens, C. Induction of myelination in the central nervous system by electrical activity.
USA 93 , — Buttery, P. Relvas, J. Sobel, R. Vitronectin and integrin vitronectin receptor localization in multiple sclerosis lesions. Gutowski, N. Tenascin-R and C in multiple sclerosis lesions: relevance to extracellular matrix remodelling. Durand, B. Gao, F. Cell-intrinsic timers and thyroid hormone regulate the probability of cell-cycle withdrawal and differentiation of oligodendrocyte precursor cells.
Stolt, C. Terminal differentiation of myelin-forming oligodendrocytes depends on the transcription factor Sox Genes Dev. A role for the helix—loop—helix protein Id2 in the control of oligodendrocyte differentiation.
Neuron 29 , — Kondo, T. The Id4 HLH protein and the timing of oligodendrocyte differentiation. Basic helix—loop—helix proteins and the timing of oligodendrocyte differentiation. Armstrong, R. Glia 14 , — FitzGerald, U. AP-1 activity during the growth, differentiation, and death of O-2A lineage cells. Awatramani, R. Evidence that the homeodomain protein Gtx is involved in the regulation of oligodendrocyte myelination.
Saluja, I. Glia 33 , — CAS Google Scholar. Lu, Q. Cell , 75—86 Multiple sclerosis: oligodendrocyte proliferation and differentiation in fresh lesions. Multiple sclerosis: remyelination of nascent lesions.
Raine, C. Multiple sclerosis: remyelination in acute lesions. Oligodendrocyte precursor cells in the demyelinated multiple sclerosis spinal cord. Brusa, A. Long-term remyelination after optic neuritis: A 2-year visual evoked potential and psychophysical serial study. Graca, D. Delayed remyelination in the rat spinal cord following ethidium bromide injection. Kotter, M. Macrophage depletion impairs oligodendrocyte remyelination following lysolecithin-induced demyelination.
Glia 35 , — Loughlin, A. Myelination and remyelination of aggregate rat brain cell cultures enriched with macrophages. Arnett, H. Nature Neurosci. References and provide experimental evidence of the importance of an appropriately regulated inflammatory response for efficient remyelination to occur. Contact with central nervous system myelin inhibits oligodendrocyte progenitor maturation. Shamash, S. It is, therefore, crucial to understand all the elements at work in this regenerative process fully, to facilitate the identification of new targets and the development of therapeutic strategies.
The complex nature of MS makes it difficult to mimic the disease faithfully in animal models, but these models have nevertheless been the source of most of our knowledge on the cell biology of myelin repair. The most widely used models of demyelination are as follows Burrows et al. EAE has proved very useful for studies of the pathogenesis of the disease and the role of immune cells, but the demyelination lesions generated are highly variable in size and unpredictable and occur at different stages of development.
Furthermore, axonal integrity is compromised in this model, making it difficult to study remyelination. For these reasons, the LPC and cuprizone models in which extensive demyelination is followed by robust remyelination are preferred for studies of the cellular mechanisms of demyelination and remyelination. However, these models do not encompass the complexity of MS pathogenesis due to the absence of adaptive immune system involvement.
Progress in mouse genetic techniques opened up opportunities for lineage cell tracing, making it possible to identify the different cell types contributing to OLG replacement and remyelination Figure 1.
The data obtained in rodents concerning the cells involved in myelin regeneration are summarized below. Figure 1. Cell sources for myelin regeneration.
Following a demyelination insult, oligodendrocyte progenitor cells OPCs are mobilized: they proliferate, migrate toward the injury, and finally differentiate into new myelinating oligodendrocytes OLGs. Surviving OLG can also produce new myelin segments thus contributing to remyelination. Most of the OPCs present in adult rodent brain are generated from the neuroepithelium during embryonic development; then they proliferate and disseminate throughout the brain parenchyma. They constitute the major population of dividing cells in the healthy adult brain.
In physiological conditions, they have a very long cell cycle, with a prolonged G1 phase, and only a minority of these cells differentiate into OLGs Dimou et al. Early studies reported a rapid increase in OPC density following demyelination Franklin et al.
These observations suggested that the parenchymal OPC pool was responsible for spontaneous remyelination. These studies provided direct evidence for the generation of remyelinating oligodendrocytes from OPCs Zawadzka et al. There is growing evidence to suggest that OPCs do not constitute a homogeneous cell population for a review, see Werkman et al. However, it is difficult to differentiate between real cell diversity and different lineage stages within the same cell population, because only a few studies have shown phenotypic differences to be intrinsic and associated with functional specificity Foerster et al.
The first evidence for OPC heterogeneity was provided by the difference in cell cycle kinetics between OPCs residing in the white and gray matter. The OPC cell cycle in the corpus callosum CC of young adult mice takes about 7 days, whereas that in the cortex takes between 21 and 50 days Dimou et al. The OPCs in the white and gray matter also differ in terms of their differentiation potential. Cultured rat gray matter OPCs have a less mature phenotype in terms of morphology and gene expression , higher rates of proliferation, and slower differentiation than white matter OPCs.
In the context of demyelination, these characteristics may confer advantages on advantages to gray matter OPCs. The heterogeneity of OPCs is also revealed by their electrophysiological properties. OPCs are sensitive to neuronal activity due to the expression of ion channels and neurotransmitter receptors. However, not all OPCs are excitable, and white and gray matter OPCs have different electrophysiological signatures, often linked to their differentiation potential Spitzer et al.
Recent studies have revealed not all OPCs contribute equally to myelin repair. During embryonic development, OPCs arise from different neuroepithelium domains. In response to demyelination in the adult brain, OPCs from dorsal origin are more strongly mobilized than ventral OPCs, but these cells are more sensitive to the age-related decline in differentiation potential Crawford et al.
This is one of the rare examples of the functional diversity of OPCs. A subset of OPCs evenly distributed throughout the brain and characterized by the expression of the G protein-coupled receptor GPR17 has been identified as a reserve pool for repair purposes Lecca et al. In physiological conditions, GPR17 is required to initiate OPC differentiation, but must be downregulated to allow cells to undergo terminal maturation.
GPR17 is activated by purines and leukotrienes, the levels of which increase after lesion formation. GPR17 reactivity is not observed in the cortex of cuprizone-fed mice Nyamoya et al. These results suggest that GPR17 may be a suitable molecular target for the promotion of endogenous myelin repair.
These differences probably account for remyelination occurring later in the cortex than in the CC: after acute cuprizone-induced demyelination, OPC repopulation and differentiation is much faster in the CC than in the cortex, resulting in later, incomplete remyelination in the cortex Gudi et al.
However, in a chronic model based on cuprizone treatment associated with rapamycin to inhibit OPC differentiation so as to improve reproduction of the characteristics of chronic MS, remyelination was faster and more robust in the cortex than in the CC Bai et al. In both cases cuprizone model and patients , these differences in remyelination efficiency were associated with differential astrocyte reactivity in the cortex and CC see below for the role of astrocytes in myelin repair.
In the adult brain, the wall of the lateral ventricle is a stem cell niche with a unique cellular and extracellular organization Mirzadeh et al. The neural stem cells present in the subventricular zone SVZ persist throughout life; they are quiescent but can be activated and re-enter the cell cycle when needed.
These low levels of oligodendrogenesis increase by a factor of four after LPC-induced demyelination Menn et al. Substantial levels of progenitor cell emigration from the SVZ to demyelinated lesion sites are also consistently observed in MS models Nait-Oumesmar et al. The migration of these cells from their niche to the lesion site is regulated by guidance cues, such as Slit1 Deboux et al. Here again, genetic lineage tracing appears to be a useful tool for unequivocally demonstrating the contribution of SVZ-derived progenitors to myelin repair, particularly in periventricular areas, such as the rostrolateral CC Xing et al.
Despite these demonstrations, the role of SVZ-derived progenitors in myelin repair has been called into question in a few studies. Guglielmetti et al. Another study reported that SVZ progenitors responded rapidly to focal demyelination but failed to produce myelinating OLGs in the lesion Kazanis et al.
The stem cells in the SVZ are not homogeneous. They are organized into domains according to the combination of transcription factors that confer them functional specificities Merkle et al. Thus, particular subpopulations of SVZ progenitors may be mobilized after demyelination.
A subset of progenitors expressing GLI1, a Sonic Hedgehog effector, has been shown to be particularly responsive after the formation of lesions; surprisingly, the mobilization of these cells is further increased by GLI1 inhibition Samanta et al.
Tracing experiments with NestinCreERT2 mice cannot determine the type of progenitors mobilized after demyelination. Neuroblasts have been reported to exit the SVZ and rostral migratory stream following experimentally induced demyelination Picard-Riera et al. This lineage plasticity was further documented in recent studies showing spontaneous neuroblast conversion into OLGs in cuprizone-fed mice El Waly et al. The respective roles of each of these cell populations in the contribution of the SVZ to myelin repair remain unknown.
Bioinformatics and in silico genomic analyses have identified a catalog of small molecules as potential regulators of SVZ microdomain-specific lineages Azim et al. This study provides proof of concept that the pharmacological stimulation of SVZ neural stem cells to produce new OLGs is a potentially valuable strategy for myelin repair.
Martino's laboratory investigated the role of SVZ-derived progenitors in myelin repair, using nestin-thymidine kinase-transgenic mice to kill neural progenitors in a specific manner in cuprizone-fed mice Butti et al. They concluded that SVZ-derived progenitors were dispensable for remyelination but provided partial protection against greater axonal loss Butti et al.
Along the same lines, our laboratory showed that some SVZ-derived progenitors mobilized to the demyelinated CC remain undifferentiated and produce factors capable of modulating microglial activation, thereby playing a protective role Brousse et al.
Schwann cells are responsible for peripheral nervous system PNS myelination. They are derived from the neural crest during embryonic development. Schwann cells are remarkably plastic and respond to lesions by dedifferentiation and re-entry into the cell cycle, facilitating rapid PNS myelin repair.
It was originally assumed that Schwann cells invaded the CNS following a breach of the glia limitans, but another mode of Schwann cell contribution to CNS remyelination is now recognized Blakemore, After spinal cord injury, extensive Schwann cell-mediated remyelination occurs, much of which is driven by OPC-derived Schwann cells Assinck et al.
Interestingly, recent studies have shown that the specific deletion of Fbxw7 a E3 ubiquitin ligase component in Schwann cells is sufficient to induce the production of thicker myelin sheaths by these cells, together with a myelination of multiple axons similar to that observed with OLGs Harty et al.
These recent findings raise questions about the relationship between PNS and CNS remyelination and suggest that the demarcation between the biology of Schwann cells and that of OLGs may be less marked than previously appreciated.
However, it remains to be demonstrated that Schwann cells can provide appropriate metabolic support to CNS axons and restore effective signal conduction Chen et al. Following demyelination, some mature oligodendrocytes are spared and survive. A possible role for these cells in myelin repair was considered but quickly ruled out when it was shown that they do not re-enter the cell cycle and proliferate after experimentally induced demyelination and that they do not migrate to the lesion or extend processes to the lesion site Keirstead and Blakemore, , ; Crawford et al.
OLGs were, thus, considered to be highly differentiated and specialized cells with a complex morphology and no postlesional plasticity.
However, recent studies have suggested that these spared OLGs are not passive witnesses of demyelination and can instead play a significant role in the repair process. OLGs located at the border of the lesion produce heparan sulfates, which, in turn, enhance remyelination via Sonic Hedgehog signaling Macchi et al. Using 3D electron microscopy in large-animal models cats and nonhuman primates , Duncan and coworkers provided the first evidence of direct remyelination by mature OLGs Duncan et al.
The development of in vivo two-photon video microscopy with longitudinal follow-up over several weeks subsequently made it possible to demonstrate unequivocally that surviving OLGs were able to generate new myelin sheaths after cuprizone-induced demyelination Bacmeister et al. However, the laboratory of Bergles failed to detect such OLG plasticity with similar techniques Orthmann-Murphy et al. OPCs, and OLGs are present throughout the central nervous system and can therefore participate in myelin repair at any lesion site.
In this respect, they occupy a privileged position. SVZ-derived progenitors have a strong migratory potential and are equipped to respond to inflammatory cues, but their contribution to myelin repair is unlikely to extend very far from periventricular structures.
Remyelination quality may also depend on the source of remyelinating cells. Interestingly, surviving mature OLGs produce fewer new myelin sheaths than newly formed OLGs, but they better preserve the pattern of myelination in the cortex, with potential implications for functional recovery Bacmeister et al. Indeed, the optimal processing capabilities of cortical circuits may be dependent on specific and stable cortical neuron myelination.
Finally, remyelination by Schwann cells leads to myelin sheaths of a different molecular composition, which are not affected in MS patients. This is an interesting and potentially advantageous feature. Conversely, peripheral myelin is less compacted, which may affect conduction efficiency in the CNS.
Furthermore, given the ratio between Schwann cells and myelin segments, extensive CNS remyelination would require an extremely large number of Schwann cells and little is currently known about the mechanisms driving the differentiation of OPCs into Schwann cells. Successful remyelination implies progenitor cell proliferation, migration to the lesion site, and differentiation into OLGs. The newly formed OLGs must then engage in dialog with axons, which they must ensheath to form compacted functional myelin sheaths.
A glitch in any one of these steps may lead to remyelination failure. We will now consider the conditions that have been shown to inhibit spontaneous myelin repair.
Like postlesional regeneration and most plasticity events, the potential for remyelination declines with age.
Remyelination is still observed in old rodents, but it is much slower than in younger animals Shields et al. Interestingly, in MS patients, the transition from relapsing—remitting to progressive MS occurs at about the same age, regardless of age at disease onset Confavreux and Vukusic, b ; Tutuncu et al.
Aging affects OPC recruitment and differentiation in a cell-autonomous or non-cell-autonomous fashion via the alteration of other cell types. OPCs from aged mice transplanted into neonatal brain recover the proliferation and differentiation rates of newborn OPCs Segel et al.
The mechanical properties of the microenvironment may be involved: a recent study revealed that tissue stiffness increases with age, impairing OPC proliferation and differentiation via the mechanoresponsive ion channel Piezzo1 Segel et al.
Consistent with cell-autonomous effects, the epigenetic control of OPC differentiation into myelinating OLGs is disrupted with aging Shen et al. Furthermore, single-cell RNA sequencing-based comparisons of OPCs obtained from young and old mice have revealed mitochondrial dysfunction and a greater activity of the inflammasome and pathways associated with nutrient signaling in aged OPCs Neumann et al. These findings were recently confirmed by a proteomic analysis revealing that the levels of proteins associated with oxidative phosphorylation and inflammatory responses increase with age, whereas those of proteins associated with cholesterol biosynthesis and the cell cycle decrease de la Fuente et al.
Several studies have suggested that, unlike neurogenesis, SVZ-derived oligodendrogenesis does not decline with age Capilla-Gonzalez et al. The SVZ may therefore be an interesting target reservoir for treatments designed to promote myelin repair in elderly patients. Finally, aging also affects microglial cells and macrophages, which play a crucial role in remyelination, by removing myelin debris that inhibits remyelination for a review, see Pinto and Fernandes, Aged microglial cells become dystrophic, and their processes become less motile Wong, ; Hefendehl et al.
These cells become more immunogenic, produce inflammatory cytokines and reactive oxygen species, and thus, have a deleterious phenotype Hammond et al. In the context of demyelination, aged microglial cells fail to take up myelin debris efficiently by phagocytosis Ritzel et al. Interestingly, the process can be reversed, as old mice exposed to a youthful systemic environment via heterochronic parabiosis recover a remyelination potential similar to that of young mice Ruckh et al.
The clearance of myelin debris and remyelination can also be restored in old mice by systemic injections of niacin, which acts by upregulating CD36 Rawji et al. Fasting and treatment with metformin a fasting mimetic drug also lead to a recovery of OPC responsiveness to differentiation signals in old mice Neumann et al. CNS injury triggers a cascade of cellular and molecular events leading to inflammation.
In MS, breakdown of the brain—blood barrier allows autoreactive T lymphocytes and macrophages to infiltrate the brain, increasing local levels of proinflammatory cytokines. Glial cells also make an active contribution to these environmental changes. Inflammation may itself cause demyelination, as in MS, in which leptomeningeal immune cell infiltration and compartmentalized inflammation within the subarachnoid space are tightly associated with the development of cortical lesions Choi et al.
Inflammatory mediators may have a negative or positive effect on progenitor cell-mediated remyelination. Acute inflammation is required for the correct remyelination of demyelinated lesions Prineas et al. Overall, these data suggest that early acute inflammation is required for the correct recruitment of OPCs to the lesion site, but that chronic inflammation may impede remyelination Figure 2.
Figure 2. Roles of astrocytes and microglia in myelin regeneration. Depending on environmental cues, microglial cells may adopt different phenotypes, from proinflammatory to regulatory.
Inflammatory microglia promotes demyelination and astrocyte reactivity, but may also be useful to the repair process, by stimulating OPC proliferation and mobilization. Regulatory microglia promotes remyelination by enhancing debris removal and OPC production. Reactive astrocytes also play multiple roles: they inhibit remyelination through inflammatory cytokine production and ECM modifications, promote myelin debris removal by recruiting microglia, and stimulate oligodendrogenesis.
Inflammation can inhibit remyelination in several ways: 1 neural progenitors contributing to remyelination may be directly attacked by inflammatory cues or immune cells Imitola et al.
Activated T lymphocytes induce the progressive collapse of the process and the apoptotic death of neural progenitors and OPCs via the secretion of semaphorin Giraudon et al. These inhibitory signals, secreted into the ECM, are generated principally by microglial cells and astrocytes activated upon injury. They have multiple complex functions in pathological conditions. Both microglia and astrocytes can present different phenotypes, behaving like Dr.
Jekyll and Mr. Hyde Figure 2. A transcriptomic analysis of astrocytes activated by different types of insults ischemia, LPS injection revealed that two types of reactive astrocytes could be distinguished: A1 astrocytes with detrimental effects on cell survival and regeneration and A2 astrocytes presenting protective effects Zamanian et al. Astrocyte reactivity upon demyelination is greater in white matter than in gray matter, probably due to the differential induction of various factors.
Astrocyte reactivity is regulated by proinflammatory cytokines and myelin debris. Myelin debris are more abundant in white matter, contributing to higher levels of astrocyte activation. In addition, microglial reactivity occurs later and is weaker in the gray matter, resulting in lower levels of cytokine production Gudi et al.
The activation state of astrocytes determines their permissive or inhibitory influence on remyelination. Astrogliosis is one of the hallmarks of MS. Reactive astrocytes have been shown to increase the recruitment of OPCs to the lesion via the secretion of chemokines, such as CXCL1, 8, and 10 Omari et al. They also promote the recruitment of microglial cells via CXCL10, thereby regulating myelin debris clearance Skripuletz et al. In this way, astrocytes contribute to the repair process.
Conversely, they also promote peripheral immune cell recruitment, thereby enhancing demyelination Brambilla et al. Astrocytes can also regulate the myelin-specific autoreactive response of effector T cells via interleukin secretion Correale and Farez, In EAE, astrocytes display impaired glutamate transporter expression, leading to deficiencies of glutamate uptake and excitotoxicity Ohgoh et al. Fibronectin accumulates in demyelinated lesions and prevents OPC differentiation in the EAE model, but not in toxin-induced demyelination, suggesting that fibronectin aggregation is mediated by inflammation.
In MS patients, fibronectin aggregates are found in chronic lesions but not in remyelinated lesions Stoffels et al. Fibronectin may therefore contribute to remyelination failure. In addition, these changes to ECM composition increase tissue stiffness, which has been shown to downregulate OPC proliferation and differentiation Jagielska et al.
Tissue stiffness is higher in chronic lesions than in actively remyelinating lesions in MS patients; similarly, tissue stiffness decreases slightly in the acute cuprizone model, but increases in the chronic model with long-term administration Urbanski et al.
Microglial cells are the resident immune cells of the CNS, and as such, they are the sensors of pathological events. They are, thus, among the first cells to be activated upon injury. Activated microglia proliferate and migrate to the lesion site, where they accumulate, with both beneficial and detrimental functions.
Activated microglia adopt different phenotypes according to the combination of cytokines and factors they express. Traditionally, microglia have been categorized into M1 proinflammatory and M2 immunomodulatory , although this classification now appears too simplistic. Indeed, the development of single-cell RNA sequencing analyses has led to the detection of microglial heterogeneity in both physiological and pathological conditions, suggesting that microglial activation is a dynamic response involving transcriptionally and spatially different subpopulations Keren-Shaul et al.
Inflammatory microglial necroptosis occurs, shutting down proinflammatory signals, and is followed by the repopulation of the lesion with regulatory microglia, a process required for efficient remyelination Lloyd et al. Regulatory microglia can also promote SVZ-derived oligodendrogenesis and remyelination via Wnt7a production Mecha et al. The astrocytes lose their ability to promote neuronal survival, become neurotoxic, block OLG differentiation, and enhance OLG death Liddelow et al.
Furthermore, like astrocytes, activated microglia secrete chemokines of the CXC and CC family, contributing to the intracerebral recruitment of T cells and antigen-presenting cells, such as macrophages and dendritic cells O'Loughlin et al. As mentioned above, microglia are sensitive to aging, becoming increasingly inflammatory and less effective at the phagocytosis of myelin debris, thereby contributing to remyelination failure. MS is an autoimmune disease leading to myelin destruction and neuronal dysfunction, finally triggering neurodegeneration.
The pathological hallmark of MS is the occurrence of focal demyelinated lesions or plaques disseminated throughout the CNS, causing diverse functional deficits according to their location.
These plaques can be classified into several categories on the basis of their inflammatory properties and the presence or absence of ongoing demyelination Kuhlmann et al.
Observations of postmortem tissues from MS patients long ago raised the possibility of spontaneous myelin repair in humans Prineas and Connell, ; Prineas et al. Little is known about the factors underlying this variability in myelin repair. The regenerative process is probably influenced by both genetic and environmental factors, although no specific genes or factors involved in this process have yet been clearly identified.
Aging obviously plays a role in decreasing the efficiency of repair in chronic MS, probably through mechanisms reminiscent of those observed in animal models see Aging section , but it cannot provide a complete explanation. Both disease duration and the location of the lesions affect remyelination potential. Many studies have suggested that remyelination is more prominent at the beginning of the disease than in chronic MS, in which the extent of remyelination is often limited Prineas et al.
Depletion of the OPC pool through repeated demyelination is another possible explanation, although sustained demyelination rather than repeated insults is required to deplete OPCs and impair remyelination in rodent models Penderis et al.
Subcortical lesions display more extensive remyelination than periventricular and cerebellar lesions, which are often poorly remyelinated Goldschmidt et al. Lesions in cortical areas are consistently more extensively remyelinated than white matter lesions Albert et al.
Cortical lesions are observed in all forms of MS, but are most prominent in long-term progressive MS and are more strongly associated with functional disability than white matter lesions. There is evidence to suggest that environmental cues, produced by astrocytes and microglia in particular, may underlie these differences in remyelination potential Chang et al.
Indeed, cortical lesions often lack the pathological hallmarks of MS white matter lesions, instead displaying only low levels of inflammation and gliosis Peterson et al. Finally, the sex of the patient may affect disease susceptibility, progression, and regeneration. Indeed, MS affects three times more women than men, but appears to follow a less aggressive course in women Confavreux and Vukusic, a. The rate of relapses decreases strongly during the last trimester of pregnancy, when estrogen and progesterone levels are at their highest, whereas there is a rebound after delivery, coinciding with the decrease in hormone levels Confavreux et al.
In vitro experiments and preclinical studies also support the hypothesis that sex hormones may improve remyelination for a review, see El-Etr et al. It therefore appears likely that several mechanisms and factors contribute to impaired remyelination, ranging from genetic background to neuropathological subtype. The identification of cells contributing to myelin repair is of course more difficult and less reliable in humans than in animal models.
However, a number of observations suggest that several cell types may be involved. However, in chronic white matter lesions, OPC differentiation into OLGs like that observed in early active lesions does not occur Kuhlmann et al. In the spinal cord of MS patients, myelin sheaths labeled with P0, a major constituent of peripheral nervous system myelin, can be detected, indicating the contribution of Schwann cells to CNS remyelination Itoyama et al.
Finally, SVZ activation is observed in MS patients, with increases in oligodendrogenesis and the ectopic migration of progenitors that are thought to be involved in remyelination Nait-Oumesmar et al. Based on these observations, it seems likely that myelin repair in the human brain can, as in rodents, proceed from different cell sources, including OPCs, Schwann cells, and SVZ-derived progenitors. Recent studies using 14 C levels to date the birth of cells concluded that, in healthy adult human white matter, only a very small number of new OLGs are generated Yeung et al.
More surprisingly, newborn OLGs were almost undetectable in the remyelinated lesions of MS patients, except in very aggressive forms of the disease Yeung et al. The absence of new OLGs in shadow plaques led the authors to conclude that myelin repair in the human brain does not stem from OPCs, instead originating from surviving OLGs that form new myelin sheaths.
This contrasts sharply with observations and demonstrations in animal models, in which remyelination from mature OLGs is restricted to very particular conditions Duncan et al.
This finding also calls into question the use of rodent models in the design of therapeutic strategies and suggests that the use of larger animal models, such as cats and nonhuman primates, should be considered Duncan et al.
Indeed, no single animal model can recapitulate the entire spectrum of heterogeneity for human MS lesions and repair mechanisms, and the best option would be to use several models. A role for mature OLGs in remyelination would have important consequences for the design of new therapeutic strategies: in this case, efforts should be made to promote OLG survival and plasticity. However, many questions remain unresolved due to several obstacles, making it impossible to draw definitive conclusions from these studies.
First, it is difficult to date the beginning of the disease and lesion initiation in humans. Second, if histology is not coupled with longitudinal in vivo imaging, it is not possible to conclude with certainty that remyelination has occurred. The histological hallmark of remyelination is the observation of shadow plaques, visible because the regenerated myelin sheath is thinner than the original one; however, this assumption is currently questioned Neumann et al.
In particular, in areas in which the axon diameter is small, as in the CC, the difference is minimal, and remyelinated areas may be very similar to healthy areas, leading to an underestimation of remyelination rates. Conversely, a shadow plaque may not necessarily be associated with remyelination: paler coloration may result from partial demyelination or a decrease in axonal density.
It is not, therefore, yet possible to rule out a contribution of OPCs to myelin repair in the human brain. MS is primarily an immune-mediated disease, and until very recently, treatments were designed to target immune cells and inflammation. These immunomodulation therapies proved efficient for limiting and reducing lesion formation and relapse rate but failed to prevent the progression of the disease toward neuron loss and irreversible disability.
As mentioned in the Remyelination Is a Major Issue for Preventing Neurodegeneration and Irreversible Losses of Function section, many studies have provided support for the idea that targeting remyelination is a sound strategy for promoting functional recovery. Here, we focus on strategies promoting myelin repair from endogenous cell sources; we do not, therefore, consider cell transplantation approaches. Bearing in mind the various factors shown to be involved in the production of new OLGs and in remyelination failure, several strategies for enhancing spontaneous repair may be considered.
Boosting OPC proliferation and differentiation appears to be a straightforward strategy for promoting remyelination.
However, the reality is more complex. Lesions at different stages coexist in MS patients, and neuropathological studies indicate that there is a high degree of heterogeneity between lesions Lucchinetti et al. It is therefore difficult to determine the most appropriate time window for the use of proliferating or differentiating agents. Indeed, promoting OPC proliferation potentially inhibiting cell differentiation would be at best useless and possibly even detrimental for lesions containing progenitors unable to differentiate into myelinating OLGs; conversely, promoting OLG differentiation in lesions with only a few rare OPCs would be highly inefficient.
The first is based on our current knowledge of factors or receptors governing oligodendrogenesis and myelination. The second approach is blind to mechanisms of action and based on the in vitro screening of compounds from libraries. This approach was made possible by the recent development of high-throughput platforms assessing myelination Mei et al.
Several promising small molecules have been identified, including miconazole, clobetazole, clemastine, and benzatropine, highlighting new pathways regulating OLG differentiation the MAP kinase, glucocorticoid receptor, and muscarinic acetylcholine receptor pathways Deshmukh et al.
Bazedoxifene BZA , a selective estrogen receptor modulator, has been identified as a promyelinating agent, with further studies validating its remyelinating effect in vivo after demyelinating insults Rankin et al.
This molecule is now being evaluated in a phase 2 clinical trial on MS patients. One recent study revealed a common mechanism of action for these diverse compounds, independent of their canonical pathways: they all interfere with the cholesterol biosynthesis pathway, leading to the accumulation of 8,9-unsaturated sterols, which stimulate the differentiation of OPCs into myelinating OLGs Hubler et al.
A small cholesterol-like compound, olesoxime, has been shown to promote oligodendrocyte maturation, remyelination, and functional recovery in rodent models of demyelination, via binding to mitochondria and the modulation of ROS levels Magalon et al. Great hopes were raised by biotin, an essential cofactor for fatty-acid synthesis and energy production that was found to have beneficial effects against progressive MS in a pilot study Sedel et al.
Recent studies in animal models have highlighted the crucial role of neuronal activity in OPC proliferation and de novo myelination for a review, see Sampaio-Baptista and Johansen-Berg, Based on these observations, two clinical trials were launched on electrical stimulation of the optic nerve in patients suffering from acute optic neuritis.
Creatine was found to be effective for this purpose in the lysolecithin model, in which it protected OLGs against caspase-dependent apoptosis by enhancing mitochondrial function Chamberlain et al. If OLGs are found to play a major role in remyelination in the human brain, they should, indeed, be considered as new targets for drug development. Blocking or removing signals that inhibit OLG differentiation and myelination should render the microenvironment more permissive for regeneration.
Preclinical studies provided strong evidence that LINGO-1 inhibition enhances repair in demyelinating disease Bourikas et al.
Clinical trials with LINGO-1 antagonists have provided encouraging preliminary results, at least for optic neuritis. MS lesions are enriched in CSPG, indicating that they may be good targets for neutralization to improve remyelination. In preclinical studies, CSPG biosynthesis inhibitors were shown to be efficient for rescuing OPC differentiation in vitro and accelerating remyelination in mice Lau et al.
A better knowledge of CSPG and the signaling cascades operating in OPCs will be required for the design of future therapeutic strategies. Similarly, low molecular weight hyaluronans impair OLG differentiation and remyelination, and their concentrations are high in chronic MS lesions Back et al. The inhibition of low molecular weight hyaluronans may, therefore, also be therapeutically relevant.
Finally, the recent discovery of OPC mechanosensitivity and of the impact of tissue stiffness on OLG maturation and myelination opens up new perspectives for treatment Makhija et al.
Myelin debris is a potent inhibitor of remyelination. The transmembrane protein EphrinB3 has been shown to be an important mediator of this inhibition of OLG maturation; the masking of EphrinB3 epitopes promotes remyelination in a focal demyelination rat model Syed et al. This discovery has been patented and may lead to the future development of new therapeutic neutralizing antibodies.
Improving myelin clearance is a major challenge. No drug is currently available for specifically preventing the aging-induced decrease in microglial phagocytosis, although several candidate molecules have been identified Pinto and Fernandes, Vitamin B 3 niacin also seems to be a promising compound for safely enhancing myelin phagocytosis by microglia Rawji et al.
TREM2 is another potentially interesting candidate target. It is expressed by microglia and is involved in the proliferation and phagocytic activity of these cells.
One very recent study of a TREM2 agonistic antibody in the cuprizone model reported enhanced myelin debris uptake and degradation, together with improved remyelination Cignarella et al. However, as microglia also take up synapses by phagocytosis, a fine balance must be achieved to promote the phagocytosis of myelin debris specifically, without eliminating synapses. Another approach could involve targeting lipid metabolism, as the uptake of too much cholesterol-rich myelin debris converts microglia into cells with a proinflammatory profile.
Molecules triggering the upregulation of ATP-binding cassette ABC transporters promote lipid efflux from human macrophages and are, thus, potentially good candidates Pinto and Fernandes, Finally, approaches fighting cellular aging may represent an interesting strategy.
Preclinical studies have shown that OPCs can be rejuvenated by alternate-day fasting or by treatment with the fasting mimetic metformin Neumann et al. Interestingly, in EAE mice, a fasting mimicking diet strongly reduces clinical severity and inflammation and promotes axonal remyelination Choi et al. Several clinical trials targeting dietary interventions have been completed, but they provided insufficient evidence of efficacy for translation into clinical practice.
However, most of these trials included limited numbers of patients and short treatment durations Evans et al. Other clinical trials are still underway, testing various specific regimens, such as a ketogenic diet and intermittent fasting. These approaches, with few deleterious side effects, may be an interesting complement to drug administration.
In recent years, the discovery of several compounds that effectively promote remyelination and provide neuroprotection in animal models has raised hopes for the development of new treatments for MS patients, particularly for preventing or treating the progressive form of the disease, for which very few options are currently available.
However, many obstacles will need to be overcome before this goal can be achieved. First, it should be remembered that no animal model fully reproduces all the characteristics of MS in humans, and young rodents are used, at ages at which regenerative potential is optimal. Drugs found to be effective in such experimental designs may be less effective in less favorable conditions. The use of larger animal models in preclinical studies may be required to overcome this problem.
It is reasonable to assume that the combination of immunomodulatory treatments with compounds alleviating inhibitory signals for remyelination, together with the use of other treatments stimulating OLG differentiation, would have the greatest effect. However, the design of such combinatorial therapies is complicated and they are difficult to test in clinical trials.
For instance, it is not immediately obvious whether the various drugs should be administered simultaneously or in a particular time window. The negative effects of permanently stimulating OPC differentiation should also be taken into account, because such treatment may ultimately lead to the depletion of the OPC pool.
Will all patients benefit from the same protocol, or do we need to characterize the clinical and neuropathological specificities of each patient more precisely to propose adapted therapy? If we are to achieve this ambitious objective, we will first need to identify valuable biological markers and to develop imaging techniques of better predictive value. MRI techniques are still being developed, to improve signal specificity, but reliable blood biomarkers are still lacking, and neurophysiological measurements will also be required, to estimate functional recovery.
MC wrote the original draft and MF generated the figures. Our data suggest cellular senescence is a disease-associated phenotype in MS. Comparative proteomic analyses of the conditioned media from control and PPMS NPCs has also identified a candidate factor we find to be responsible for impairing oligodendrocyte differentiation. Based on these new data, we hypothesize that impaired CNS remyelination in MS is a consequence of disease-acquired cellular senescence in NPCs that, through production of secreted factors, limits the endogenous regenerative potential of OPCs.
Results from this study are expected to determine whether targeting cellular senescence may be a therapeutic strategy to promote remyelination in MS patients which may have implications for understanding why remyelination failure diminished with age and in this disease.
0コメント