J Neuropathol Exp Neurol Copyright Ó 2009 by the American Association of Neuropathologists, Inc.

Vol. 68, No. 6 June 2009 pp. 595Y604

REVIEW ARTICLE

Sending Mixed Signals: Bone Morphogenetic Protein in Myelination and Demyelination Jill M. See, PhD and Judith B. Grinspan, PhD

Abstract Developing oligodendrocytes undergo a well-characterized maturation process that is controlled by extrinsic factors that promote specification, proliferation, and differentiation. Inhibitory factors also influence oligodendrocyte development and may regulate the location and number of oligodendrocytes available for myelination. These factors may also repress regeneration and remyelination after injury. Bone morphogenetic proteins (BMPs) comprise a family of factors that inhibit oligodendrocyte development in vitro and when they are overexpressed in vivo. These effects seem to be mediated by the actions of inhibitors of DNA-binding protein on transcription factors that promote myelination. Bone morphogenetic protein signaling deletion studies have generated a complex picture in which the main effect of BMPs is on oligodendrocyte differentiation and depends on the level of signaling. Bone morphogenetic proteins are significantly upregulated in demyelinated areas in models of myelin injury and disease, and blocking of BMP signaling aids recovery. It is not yet known, however, whether this occurs by promoting differentiation of oligodendrocyte precursors or by inhibiting astrogliosis because BMPs also promote astrogliogenesis. Understanding the actions of BMPs will be important for promoting recovery in patients with demyelinating diseases and other types of CNS injury. Key Words: Astrocytes, Bone morphogenetic proteins, Demyelination, Inhibitors of DNA-binding proteins, Myelin, Oligodendrocytes, Remyelination.

INTRODUCTION Specification, proliferation, and differentiation of the various cell types in the developing CNS require interactions between internal molecular mechanisms and external signaling factors. Oligodendrocyte development provides an excellent model in which to study these interactions because the stages and the internal cellular mechanisms involved in oligodendrocyte lineage progression are distinct and well

characterized (Fig. 1). Oligodendrocyte specification and later maturation are promoted internally by transcription factors that include Olig1, Olig2, Nkx2.2, and Sox10 (1Y4), and externally by factors such as sonic hedgehog (Shh) for specification and thyroid hormone and insulin-like growth factor 1 for differentiation (5Y9). Inhibitory factors may also regulate the location and extent of oligodendrogliogenesis, restricting both oligodendrocyte precursor cell (OPC) specification and oligodendrocyte differentiation; the identity of these factors is presently unknown. One potential group of repressive factors for oligodendrocyte development is the bone morphogenetic proteins (BMPs). These signaling proteins are concentrated in the roof plate of the developing nervous system and act as dorsalizing factors, specifying the patterning of interneurons in the dorsal neural tube (10). Because most OPCs arise in ventral regions of the neural tube and because BMPs inhibit oligodendrocyte differentiation in vitro, BMPs are thought to limit dorsal oligodendrogliogenesis, at least early in development (11Y13). Recently published in vivo studies from several laboratories, however, suggest that the regulation of oligodendrocyte development by BMPs may be more complex (14, 15). This review will discuss in vitro and in vivo studies of the role of BMPs in the development of the oligodendrocyte lineage. We will also review the mounting body of evidence suggesting that BMPs are upregulated during nervous system injury and disease involving demyelination and may serve to inhibit regeneration. Thus, understanding the role of BMPs in oligodendrogliogenesis may lead to development of strategies to repair demyelination in the CNS.

Most Oligodendrocytes Are Ventrally Derived

From the Department of Neurobiology, Drexel University College of Medicine (JMS); and Department of Neurology, Children’s Hospital of Philadelphia and University of Pennsylvania (JBG), Philadelphia, Pennsylvania. Send correspondence and reprint requests to: Judith B. Grinspan, PhD, 516D Abramson Center, 3615 Civic Center Blvd, Children’s Hospital of Philadelphia, Philadelphia, PA 19104; E-mail: [email protected] This work was supported by NMSS RG3662 (Judith B. Grinspan).

Cells of the oligodendrocyte lineage first appear in the germinal zones of the mouse neural tube at E12.5, shortly after motor neurons are generated (13). In culture, the first recognizable stage in the oligodendrocyte lineage is the oligodendrocyte preprogenitors that are specifically recognized by their expression of polysialic acid portion of neural cell adhesion molecule (PSAYNCAM) on the cell surface (16). In the absence of other cues, these cells will become oligodendrocyte precursors (i.e. OPCs), which are bipolar proliferative cells that migrate out of the ventricular zone to their final locations (17). The OPCs can be identified in vitro by the A2B5 antibody that identifies a surface ganglioside (18) and in vivo by the expression of platelet-derived growth factor receptor-> and possibly chondroitin sulfate proteoglycans both in vitro and in vivo (19, 20). Both oligodendrocyte

J Neuropathol Exp Neurol  Volume 68, Number 6, June 2009

595

Copyright @ 2009 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.

See and Grinspan

J Neuropathol Exp Neurol  Volume 68, Number 6, June 2009

preprogenitors and OPCs are multipotential in vitro and possibly in vivo, and certain external signaling factors induce their expression of astrocytic and even neuronal markers (12, 21Y23). In the spinal cord, most OPCs are specified in the ventral ventricular zone in the Olig2-expressing domain (1, 4, 13). These ventrally arising OPCs appear in a tightly restricted area of the preYmotor neuron domain, induced by Shh signaling from the notochord and floor plate. Experiments in mice in which the notochord has been removed or Shh signaling has been disrupted have resulted in almost no oligodendrocytes, suggesting that Shh signaling is necessary for generation of ventral oligodendrocytes (8, 9). Furthermore, addition of notochord explants or Shh to developing chick neural tube has resulted in increased and ectopic oligodendrogliogenesis, suggesting that Shh signaling may be sufficient for OPC generation (8, 9, 24). There is recent in vitro evidence that Shh signaling may not be necessary for generation of all oligodendrocytes because some may be derived from a dorsal population of OPCs (25). Sonic hedgehog signaling from the notochord seems to be responsible for the ventral origin of most OPCs, however. Inhibitory factors that restrict OPC generation to the ventral neural tube may also exist and share responsibility for the largely ventral origin of OPCs. Using spinal cord explant cultures, Wada et al (26) demonstrated inhibition of oligodendrogliogenesis in ventral spinal cord segments by placing pieces of dorsal spinal cord adjacent to the ventral sections. Conversely, removing dorsal areas entirely resulted in increased and ectopic O4+ oligo-

dendrocytes, indicating the presence of a dorsal inhibitory factor that restricts most oligodendrocyte development to ventral areas. The BMPs are highly expressed in the roof plate of the developing neural tube and seem to be the most likely candidates for dorsal inhibition of oligodendrocytes for several reasons. First, a growing body of evidence supports the idea that exogenous BMP treatment of OPCs inhibits oligodendrocyte maturation both in vitro and in vivo. Second, BMPs oppose the actions of Shh in the development of motor neuron subclasses and other dorsal structures (27). This is not simply a question of the ventral versus dorsal source of these proteins because BMP is also expressed in some ventral areas (28). The influence of the BMPs ventrally is inhibited by expression of the specific BMP inhibitor noggin; in the absence of noggin, dorsal cell fates predominate even in the presence of Shh, suggesting that BMPs can override Shh signaling (28).

The Bone Morphogenetic Proteins Bone morphogenetic proteins are a family of secreted transforming growth factor-A signaling factors that were first identified through their involvement in bone formation but are also important for development of numerous tissues. There are at least 20 structurally distinct BMPs. The diverse cellular processes controlled by BMPs include proliferation, differentiation, and apoptosis. In the developing CNS, BMP expression in the roof plate exerts a dorsalizing effect on neural stem cells and developing neurons (27, 29). Disruption

FIGURE 1. Progression of the oligodendrocyte lineage from preprogenitor to mature oligodendrocyte. Upper portions diagram the stages of the oligodendrocyte lineage showing increased size and process complexity as the cells differentiate and mature. Surface or internal markers found at specific stages of the lineage are indicated in the blue cytoplasm of the cell. Transcription factors important for specification and later differentiation are indicated in the pink nucleus of the cell. Lower portions show the corresponding morphology of typical oligodendrocyte lineage cells at each stage. These are neonatal rat cells grown in purified culture and labeled at appropriate stages with antibodies to polysialic acid portion of neural cell adhesion molecule (PSAYNCAM) for preprogenitors, ganglioside marker A2B5 for precursors, galactocerebroside ([GalC] R-mAb) for immature oligodendrocytes, and myelin proteolipid protein (PLP) for mature oligodendrocytes. CNP, 2¶,3¶-cyclic nucleotide phosphodiesterase; MAG, myelinassociated glycoprotein; MBP, myelin basic protein; MOG, myelin oligodendrocyte glycoprotein; NG2, NG2 chondroitin sulfate proteoglycan; PDGFR, platelet-derived growth factor receptor.

596

Ó 2009 American Association of Neuropathologists, Inc.

Copyright @ 2009 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.

J Neuropathol Exp Neurol  Volume 68, Number 6, June 2009

BMPs in Myelination and Demyelination

of BMP signaling in the neural tube expands ventral domains such that DI1 interneurons are eliminated, numbers of DI2 interneurons are reduced, and DI3 and DI4 interneuron domains are expanded (30). The BMPs are expressed in both dorsal and ventral areas of the spinal cord during embryonic development, although expression is much higher in dorsal areas (31). In human embryos, BMP is expressed in the diencephalic floor and may have a role in telencephalic regionalization and patterning; BMP mutations have been found in humans with brain anomalies and developmental delays (32). Although BMPs are secreted molecules, their actions are local and their diffusion is limited, possibly by interaction with extracellular matrix proteins such as heparin sulfate proteoglycans (33). The BMPs can, however, control patterning at a distance through local effects on other families of dorsal growth factors such as the Wnts (30). The BMPs signal through a serine-threonine receptor dimer consisting of the BMP receptors BMPRI and BMPRII, of which there are multiple subunits. The BMP dimer binds to the high-affinity Type II receptor, which then associates with and phosphorylates the Type I receptor (34). The Type I receptor then phosphorylates a receptor Smad (R-Smad) 1, 5, or 8, which then forms a complex with the co-Smad, Smad4, and translocates to the nucleus to initiate transcription (35, 36). Bone morphogenetic protein signaling can also activate an alternative signaling pathway in cells through activation of STAT proteins mediated by the serine-threonine kinase FKBP12/rapamycin-associated protein and the cyclic AMP response binding protein (CBP/p300) (37). The BMPs that are most relevant to oligodendrocyte development use BMP receptors BMPR-IA and BMPR-IB, which are encoded by the genes Bmpr1a and Bmpr1b,

respectively (Fig. 2) (10, 38). The two receptors can mediate different responses even within the same cells in some developmental situations (39). In the specification of dorsal spinal cord patterning, however, the two Type I BMP receptors have recently been shown to be functionally redundant (30). Both the BMPR-IA and BMPR Type II receptors are present in OPCs and mature oligodendrocytes, but the BMPR-IB receptor has so far been identified only in OPCs (11, 40). The OPCs and mature oligodendrocytes synthesize BMP-4, but possible paracrine effects of BMP signaling are not known (40). Bone morphogenetic protein signaling is tightly regulated by cell intrinsic and extrinsic factors. Extracellular antagonists chordin, noggin, gremlin, and dan bind to BMP itself and prevent receptor binding (41). Intracellular proteins Smad6, Smad7, and Lim compete with receptor-activated Smads for receptor binding (41Y43). This allows precise regulation of BMP signaling levels throughout the CNS, thereby allowing high levels of BMP signaling in dorsal tissues, whereas ventral tissues are free from dorsalizing effects.

BMPs in Oligodendrocyte Development: Effect of BMP Addition During Early Stages Of Oligodendrocyte Development In Vitro Bone morphogenetic protein addition to neurospheres and to cultures of oligodendrocyte preprogenitors and precursors (the early stages of the oligodendrocyte lineage) first demonstrated that BMPs inhibit oligodendrogliogenesis while promoting astrogliogenesis (Fig. 3). Gross et al (44) showed that neurospheres composed of undifferentiated neural stem cells generated astrocytes at the expense of neurons and oligodendrocytes when they are treated with BMPs. In addition, BMP treatment has been shown to direct early stages of the oligodendrocyte lineage to an astrocyte phenotype. For

FIGURE 2. Bone morphogenetic protein (BMP) signaling can inhibit oligodendrocyte differentiation and increase astrogliogenesis in neural cells. (Left) The BMP ligand is bound by specific inhibitors and cannot contact its receptors. Smad1 remains unphosphorylated in the cytoplasm. Olig1/2, Sox10, and Nkx2.2 cooperate to promote differentiation and myelin protein formation. (Right) Bone morphogenetic protein signaling to its receptors phosphorylates Smad1, signaling to inhibitors of DNAbinding proteins (Id) proteins sequesters Olig 1/2 in the cytoplasm, and differentiation and myelin protein synthesis do not occur. Astrogliogenesis is promoted possibly through the serine-threonine kinase FKBP12/rapamycin-associated protein (FRAP)/signal transducers and activators of transcription (STAT) pathway. BMPR, bone morphogenetic protein receptor. Ó 2009 American Association of Neuropathologists, Inc.

597

Copyright @ 2009 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.

See and Grinspan

J Neuropathol Exp Neurol  Volume 68, Number 6, June 2009

example, BMP treatment of PSA-NCAM+ oligodendrocyte preprogenitors generated cells that retained PSA-NCAM expression but also expressed glial fibrillary acidic protein (GFAP), a marker of mature astrocytes and radial glia and inhibited progression to the next stage of the oligodendrocyte lineage, the OPC. Furthermore, treatment of OPCs with BMP2 or BMP-4 inhibited oligodendrocyte differentiation and instead diverted the cells to a Type 2 astrocyte phenotype, characterized by coexpression of GFAP and A2B5, an antibody marker that labels surface gangliosides on OPCs (11, 12, 45). These studies suggest that BMPs prevent multipotential progenitors from entering the oligodendrocyte lineage while promoting astrogliogenesis. These processes may be separately controlled (Fig. 2). Although BMP signaling to oligodendrocyte precursors upregulates Smad labeling in the nucleus, which may activate mechanisms to inhibit oligodendrocyte differentiation (15), the generation of astrocytes may require activation of the STAT/FRAP pathway. Inhibition of this pathway in neural stem cells prevents differentiation of astrocytes (37). This area awaits further study.

progression from an OPC to an immature phenotype is expression of the proligodendroblast antigen, which is recognized by the O4 antibody, marking the proligodendroblast stage (46). This shortly precedes differentiation into immature oligodendrocytes, characterized by expression of markers galactocerebroside (GalC) and 2¶,3¶-cyclic nucleotide phosphodiesterase (CNP), which signifies commitment to an oligodendrocyte fate (47, 48). Finally, mature oligodendrocytes express myelin proteins such as proteolipid protein, myelin basic protein, myelin-associated glycoprotein, myelin-oligodendrocyte protein, in addition to GalC and CNP (47, 49). Bone morphogenetic proteins seem to have a role in regulating myelin protein expression, independent of their function in differentiation. When BMP-4 was added to OPC cultures that had been allowed to differentiate for 2 days, achieving an immature oligodendrocyte phenotype, the cells expressed the differentiation marker GalC but not myelin proteins (Fig. 3) (50). These immature oligodendrocytes were unable to express GFAP in the presence of BMP because they were no longer multipotential. Furthermore, when beads soaked in recombinant BMP-4 proteins were placed in the ventral midline of spinal cord explant cultures, myelin protein expresion was almost completely absent in a local area surrounding the beads, whereas immature markers such as O4 and GalC were also reduced, but to a much lesser extent (50). These data suggest that exogenous expression of BMP has a specific inhibitor effect on myelin protein expression. Interestingly, Wada et al (26) were unable to show an effect of BMP-4 on similar explant cultures when BMPs were simply added to the

Effect of BMPs on Oligodendrocyte Development: BMPs and Myelin Protein Expression In Vitro After migration, OPCs cease proliferation and begin to differentiate, activating internal machinery involved in differentiation, expressing markers of maturing oligodendrocytes and beginning formation of the myelin sheath. The first marker of

FIGURE 3. Effects of exogenous bone morphogenetic protein (BMP) addition in vitro. Addition of BMP-2 or BMP-4 inhibits the progression through the oligodendrocyte lineage and generates cells with an astrocyte phenotype that expresses glial fibrillary acidic protein (GFAP) and oligodendrocyte precursor markers platelet-derived growth factor receptor-> (PDGF-R>) or ganglioside marker A2B5. Treatment of immature oligodendrocytes that are committed to the oligodendrocyte lineage with BMP-2 or BMP-4 inhibits myelin protein expression. CNP, 2¶,3¶- cyclic nucleotide phosphodiesterase; GalC, galactocerebroside; MAG, myelinassociated glycoprotein; MBP, myelin basic protein; MOG, myelin oligodendrocyte glycoprotein; NG2, NG2 chondroitin sulfate proteoglycan; PDGFR, platelet-derived growth factor receptor; PLP, myelin proteolipid protein; PSA-NCAM, polysialic acid portion of neural cell adhesion molecule (CD56).

598

Ó 2009 American Association of Neuropathologists, Inc.

Copyright @ 2009 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.

J Neuropathol Exp Neurol  Volume 68, Number 6, June 2009

BMPs in Myelination and Demyelination

explant medium. Oligodendrocyte differentiation was suppressed, however, when BMP was delivered in a locally concentrated manner, illustrating the locally limiting effects of BMPs (26, 50). This restricted effect was also shown in Xenopus spinal cord using BMP-4Ycoated beads or beads coated with function-blocking antibody to BMP-4 (31).

Because BMP signaling is involved in numerous and fundamental aspects of development, simple genetic deletions of key BMPs or their receptors have resulted in embryonic lethality. However, a conditional knockout was created by the Cre/loxP recombination system using proximal cis-active transcriptional regulatory elements of the POU-domain gene Brn4/Pou3f4 (53). Bmpr1a conditional knockouts were crossed with traditional Bmpr1b knockouts to generate double knockouts. These animals show dorsal/ventral neural tube patterning defects, hindlimb deformities and neonatal lethality. When the animals were analyzed at P0, BMP receptor signaling was clearly missing, as indicated by the lack of response to BMP in vitro and the lack of nuclear localization of the phosphorylated form of Smad1, the downstream signaling partner of BMP-4 (15). Spinal cord sections from the animals had a 50% decrease in astrocytes, as expected, but the number of OPCs was the same as the controls and the numbers of cells expressing the oligodendrocyte proteins myelin proteolipid protein and myelin basic protein were decreased by more than 50%, indicating a significant lack of oligodendrocytes (Fig. 4) (15). This suggested that the lack of BMP signaling did not affect OPCs and that some amount of BMP signaling, either directly to OPCs or indirectly through astrocytes, was actually required for timely myelination. A second model used the disruption of BMPR1a only using an Olig1-Cre that deletes expression of Olig1 from the neural tube by E13.5 (14). These animals retained some nuclear phospho-Smad labeling probably caused by the continued presence of the BMPR1B receptor. The OPC numbers and distribution were unchanged, as were numbers of astrocytes in the brains of these animals. but there were increased numbers of mature oligodendrocytes and calbindinpositive interneurons at P21. The latter was ostensibly because the Olig1 Cre targets some motor neurons as well as oligodendrocytes (14). Both of the genetic studies previously described had the common conclusion that the deletion of BMP signaling at a time point before or at the start of the appearance of OPCs did not alter the number or distribution of precursors. This suggests that BMP signaling is not involved in the specification of oligodendrocytes and runs counter to the general notion in the field. Most of the analysis of the effect of exogenous BMPs on oligodendrocyte development has been performed using markers of immature or mature oligodendrocytes,

BMP in Oligodendrocyte Development: Overexpression In Vivo Several lines of evidence suggest that addition of BMPs to developing neural tube also inhibits oligodendrogenesis in vivo. For example, in chick embryos, addition of BMP-soaked beads to developing neural tube resulted in decreased numbers of maturing oligodendrocytes but no increase in astrogliogenesis (51). To investigate the effects of genetic BMP overexpression, Gomes et al (52) generated mice in which overexpression of BMP-4, controlled by the neuron-specific enolase promoter, resulted in an 11% to 19% decrease in oligodendrocyte maturation in multiple brain regions, depending on the area examined, coupled with increased astrogliogenesis. Together, these results support an inhibitory effect of excess BMP signaling, but they do not address the effects of endogenous BMPs on gliogenesis (Fig. 4).

BMP in Oligodendrocyte Development: Disruption of Signaling Recent studies have shown that endogenous BMP signaling does play a role in gliogenesis. In chick neural tube, the disruption of BMP signaling by adding nogginexpressing cells resulted in ectopic dorsal oligodendrogenesis, suggesting that the presence of BMP was the primary factor preventing dorsal oligodendrocytes from developing (51). In a related study, depletion of endogenous noggin, an inhibitor of BMP signaling, from rat optic nerve decreased oligodendrogenesis and increased the number of astrocytes (40). This effect was confirmed in Xenopus embryos (31). Although these studies show the effect of a diminution in BMP signaling, the most rigorous test of the role of BMPs comes from genetic deletion experiments.

BMPs in Oligodendrocyte Development: Genetic Deletion Models Based on conditional deletion of Type I receptors, 2 models of BMP expression abrogation have been developed.

FIGURE 4. Effects of bone morphogenetic protein (BMP) levels in vivo on the generation of mature oligodendrocytes. Four scenarios of BMP signaling that have been tested in vivo by manipulating levels of BMP both experimentally and genetically are shown. Excess BMP signaling decreases the numbers of mature oligodendrocytes, whereas a decrease in signaling generates a surplus of mature oligodendrocytes. Complete deletion of signaling through both BMP Type 1 receptors seems to also decrease the number of oligodendrocytes achieving maturity. Shh, sonic hedgehog. Ó 2009 American Association of Neuropathologists, Inc.

599

Copyright @ 2009 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.

See and Grinspan

J Neuropathol Exp Neurol  Volume 68, Number 6, June 2009

not OPCs; however, the O4 antibody and antibody to GalC (11, 12) have been used most often. O4 identifies the proligodendroblast antigen, which in the mouse appears in the later stages of precursor development and remains expressed in mature cells but is not detected in early precursors (46), whereas GalC is present in both immature and mature oligodendrocytes. Chick embryo studies in which BMPs were overexpressed also used the O4 antibody and antibody against galactocerebroside as outcome measures (51). Although O4 is expressed earlier in chick oligodendrocytes than in rodents, its expression remains present throughout development, thereby confounding the determination of a lack of precursors versus mature cells. Analyses showing decreases in oligodendrogenesis in a BMP-overexpressing mouse was performed using only mature markers, leaving in doubt whether OPCs were also decreased in number (52). Only Miller et al (31) showed a decrease in OPCs upon BMP treatment using the A2B5 antibody, which specifically labels OPCs. They also observed that the effect of BMP was less strong on A2B5+ cells and speculated that OPCs become increasingly sensitive to BMP during development. Based on these data, we hypothesize that BMPs do not play a major role in OPC specification in vivo but do regulate the differentiation of OPCs to mature oligodendrocytes.

implicated. Downregulation of cell proliferation through factors such as P27kip1, CDK2, and p53 are necessary but not sufficient to induce differentiation (54, 55). External factors such as thyroid hormones and insulin-like growth factor 1 potentiate differentiation, and the presence of Notch receptors and Jagged signaling inhibits differentiation (5, 56, 57). The BMP-mediated inhibition of differentiation may be related to any of the factors previously listed, but the cell-intrinsic pathway by which BMPs inhibit oligodendrocyte differentiation is probably mediated by inhibitors of DNA binding (Id) proteins and inhibition of transcription factors that promote oligodendrogenesis. The 4 mammalian Id proteins identified bind to bHLH transcription factors, thereby preventing DNA binding; they are upregulated in the presence of BMP signaling (58Y60). Disruption of BMP signaling through BMPR1a and BMPR1b receptors decreases Id protein expression, suggesting that BMP signaling regulates Id protein expression (30). The Id2 and Id4 overexpression in cultured OPCs prevents differentiation and induces an astrocyte phenotype, mimicking the effects of BMP treatment (61Y63). The mechanism by which BMP signaling inhibits oligodendrogenesis is likely to include the regulation of the Olig proteins, Olig1 and Olig2, either through Id proteins or through direct interactions with components of the BMP pathway. In neurospheres treated with BMP-4, Id2 and Id4 bound to Olig1 and Olig2 proteins and prevented their entrance into the nucleus and their actions in differentiation (62). Conversely, in adult OPCs, overexpression of Olig1 and Olig 2 inhibited oligodendrocyte maturation after BMP treatment (63). Direct interactions between Olig2 and the BMP pathway have been demonstrated at the level of the coSmad, Smad4, which in dorsal spinal cord cultures can bind directly to the Olig2 promoter and dissociate upon differentiation (64). In vivo, both a genetic deletion of Smad4 and overexpression of noggin increased the number of Olig2+ progenitors in the adult subependymal zone. The neurogenic phenotype in these Smad4-/- mice could be rescued by suppression of Olig2 function, thus providing in vivo evidence of the relationship between the BMP pathway and Olig protein regulation (65).

Generation of Astrocytes From Multipotential Precursors: Is There a Role for BMP? The role of BMPs in astrogliogenesis remains incompletely understood. The application of BMPs to neural stem cells in vitro favors the appearance of GFAP+ astrocytes over neurons and oligodendrocytes (44). The same treatment of cultured OPCs or oligodendrocyte preprogenitors from the brain generates cells that label with OPC or preprogenitor markers and express GFAP (11, 12). These cells also assume a stellate morphology clearly distinct from the large elongated astrocytes typically seen in CNS cell cultures (12). Modifications of BMP signaling in vivo have generated mixed results, however. The addition of recombinant BMP-4 to chick ventral neural tube explants did not result in a change in astrocyte numbers (51) nor did the addition of BMP-soaked beads to mouse spinal cord explants (50). Rat spinal cord cultures treated with BMP-4 showed a morphological change in astrocytes, but a numerical change was not noted (31). By contrast, genetic overexpression of BMP-4 in neuron-specific enolaseYexpressing cells significantly increased astrocyte numbers in the brain (52). It would be tempting to ascribe the differences seen here as a difference in astrocyte development in the more rostral parts of the CNS versus the spinal cord, but deletion of both BMP Type 1 receptors results in mice with 50% fewer spinal cord astrocytes (15). Thus, the reason why spinal cord explant models lacked an astrocyte effect and the genetic models had one remains uncertain, although genetic ablation studies favor an in vivo role.

Downstream of BMPs: Mediation by Inhibitors of DNA-Binding Proteins The mechanisms of oligodendrocyte differentiation are complex, and many extrinsic and intrinsic factors have been

600

Dorsal Oligodendrocytes In light of recent evidence that some portions of oligodendrocytes arise dorsal to the subventricular zone of the spinal cord, the inhibitory effects of BMPs on dorsal oligodendrogenesis seem contradictory. Until relatively recently, there was little evidence that a significant proportion of OPCs could be generated outside the ventral subventricular zone. Chickquail chimera studies generated conflicting results regarding the existence of a dorsal source of oligodendrocytes, suggesting that there may be a small population of dorsal oligodendrocytes (66). It is not known, however, whether the differing results of these studies were caused by technical details or a dorsally derived oligodendrocyte population. Studies investigating disruption of ventralizing signal Shh in ventral spinal cord grafts did not detect a significant number of dorsally derived oligodendrocytes. Recent evidence suggests, however, that a later dorsal population of OPCs exists, although its contribution to the final population of mature myelinating Ó 2009 American Association of Neuropathologists, Inc.

Copyright @ 2009 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.

J Neuropathol Exp Neurol  Volume 68, Number 6, June 2009

BMPs in Myelination and Demyelination

oligodendrocytes is unclear (24, 67Y70). Chandran et al (25) showed that oligodendrocytes can be generated in the absence of Shh signaling in vitro, suggesting that a dorsal Shhindependent population of oligodendrocytes could exist. Two lines of evidence using genetic experiments have revealed this dorsal population of oligodendrocytes. First, mice null for Nkx6.1 and Nkx6.2, which normally activate Olig2 and promote oligodendrogenesis, exhibit a loss of ventrally derived OPCs in the spinal cord while a later-emerging dorsal population of Olig2+/Pax6+ OPCs is able to develop (68Y70). As confirmation of the existence of dorsally derived oligodendrocytes, Fogarty et al (69) used Dbx1-Cre mice crossed with a reporter strain to fate-map Dbx1-expressing cells and found that a small population of spinal cord oligodendrocytes expressed Dbx-Cre, suggesting that oligodendrocytes can arise from Dbx-expressing cells, which are found dorsal to the subventricular zone. It is unclear what external signaling factors are responsible for the generation of this population of oligodendrocytes or whether they respond to the same signals as ventrally derived oligodendrocytes. In dorsal neural stem cells, fibroblast growth factor 2 signaling through mitogenactivated protein kinase promotes oligodendrogliogenesis by sequestering Smad1 in the cytoplasm, thus directly opposing the effects of BMPs (64). Because fibroblast growth factor 2 generally serves to inhibit oligodendrocyte differentiation from OPCs (71, 72), however, the in vivo relevance of this observation for at least the maturation of OPCs is not known. How then can mature oligodendrocytes develop in areas where BMPs are expressed? There are 3 possible explanations. First, BMPs are downregulated in the dorsal spinal cord during late embryogenesis and dorsally derived oligodendrocytes appear only after BMP downregulation (31). Second, it seems that this dorsal population of oligodendrocytes expresses a different complement of transcription factors than the ventral population (68), which could result in a differing response to BMP signaling. Third, based on in vivo data from the deletion of both the BMPR1a and the BMPR1b, it is possible that BMPs do not simply inhibit oligodendrocyte development but may play a dose-dependent role; some level of BMP signaling could be required for oligodendrocyte development while excess BMP signaling is inhibitory (15).

minutes after ischemic injury (75) to hours after the compression injury (74, 77) to weeks for the others (63, 78, 79). The neural cell types that exhibit BMP upregulation seem to differ among these studies, although in general, the increased expression was mostly present in astrocytes, particularly those that are reactive and hypertrophic (63, 77). Increased BMP expression, however, has also been associated with neurons (74, 77) and macrophages (76). Mature oligodendrocytes express some amount of BMP, at least in culture (40, 63), but BMP upregulation in oligodendrocytes was only noted in a few of these studies (78). It is not known whether this is a rare observation or was simply not assessed in some of these studies. Two reports also showed upregulation of BMP receptors in neurons upon injury (80, 81). Bone morphogenetic protein upregulation has also followed several other experimental paradigms with demyelination. Lesions caused by ethidium bromide injections into the caudal cerebellar peduncle resulted in upregulation of BMP-4 in oligodendrocyte precursors (82). Lesions formed by lysolethicin injection in the spinal cord resulted in upregulation of BMP-4 that was not colocalized with astrocytes (83). In chronic experimental autoimmune encephalomyelitis, an animal model of multiple sclerosis, BMP-4, -6, and -7 were upregulated, and BMP-4, by far the most abundant of the BMPs tested, was found on macrophages and some astrocytes and oligodendrocytes in the area of the lesion in the lumbar spinal cord (76). Finally, BMP-4 and BMP-5 were detected in human tissue from a multiple sclerosis plaque tissue and in a Creutzfeldt-Jakob disease lesion (84). Whether the lesion was in the spinal cord or other parts of the CNS does not seem to have made a difference to the upregulation of the BMPs in these studies. What function does BMP upregulation play in demyelination/remyelination? Based on the role of BMP overexpression during development, there are several possibilities. One is that the abundance of BMPs inhibits the adult oligodendrocyte precursors remaining in the CNS from differentiating to mature oligodendrocytes and making myelin. Another is that BMPs promote the astrocytic response and gliosis. Evidence for the latter comes from 2 studies. First, lesioning spinal cord by lysolethicin injection upregulated BMPs in nonastrocytic cells and upregulated phospho-Smad1/5/8 in GFAP+ astrocytes, suggesting a heightened sensitivity of astrocytes to the BMPs (83). Second, BMP-4 treatment in vitro and in vivo increased expression of chondroitin sulfate proteoglycan, an important component of the glial scar made by astrocytes and a molecule known to inhibit axonal growth (83). In a stab injury model, infusion of an anti-noggin antibody increased the number of GFAP+ cells and NG2 chondroitin sulfate proteoglycanY positive cells in the lesion area (79). Thus, BMP upregulation could inhibit remyelination by increasing the glial response. If BMPs act to inhibit remyelination, will inhibition of BMP signaling promote recovery? In lesioned areas formed by contusion or compression, addition of BMP inhibitor noggin significantly enhanced recovery of locomotion, whether the noggin was expressed by transplanted neural precursor cells engineered to express it or by infusion via osmotic minipumps (73, 78). No change in reactive gliosis or the size of the lesion was noted, but there was significant regrowth of the spinal

The Role of BMP in Demyelination/Remyelination Recent studies show that BMPs are upregulated in a variety of CNS injury and disease paradigms that result in demyelination and poor remyelination, thereby implicating BMP signaling pathways as possible targets for myelin repair. Although the highest expression of BMPs is observed during development, low levels of BMP-2, -4, -6, and -7 expression are found in many brain and spinal cord regions in the adult. Bone morphogenetic protein expression is mostly associated with neurons, although astrocytes have also been mentioned, as well as one report of BMP expression in oligodendrocytes (73Y76). Upregulation of BMP-2, -4, -6, and -7 has been noted in many injury paradigms, including cerebellar cortical ischemia (75), spinal cord compression injury (74, 77), spinal cord contusion (63, 78), and stab injury to both spinal cord and brain (79). The time frame of this upregulation varies from Ó 2009 American Association of Neuropathologists, Inc.

601

Copyright @ 2009 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.

See and Grinspan

J Neuropathol Exp Neurol  Volume 68, Number 6, June 2009

tract (78). These results suggest that inhibiting BMP signaling could improve regeneration, although no mechanism is suggested and the effect on remyelination and other recovery processes has not been examined. These reports also contrast with studies in which noggin-expressing stem cells neither altered the resulting number of astrocytes nor increased oligodendrocyte differentiation, but instead increased the size of the lesion in a contused spinal cord model (85). Differences between the type of lesions and method of delivery of the noggin cells might explain this contradiction. Although the evidence previously described points to BMPs playing an inhibitory role in remyelination, BMPs have been found to be neuroprotective in models of stroke, ischemia, and Parkinson disease (86Y88), possibly by the same mechanism of increasing astrogliogenesis. One can hypothesize that increasing astrogliogenesis might enhance the survival of new neurons (89) or that generation of Type 2 astrocytes themselves might serve as a supply of new neural cells for cells damaged in the CNS (79). These models may not be relevant to the effect of BMPs on remyelination, however, because an additional complicating factor is the upregulation of the BMP inhibitor noggin, along with BMPs in some disease models (79).

The upregulation of BMPs has now been demonstrated in many models of myelin injury and disease, and there is evidence that BMP upregulation increases gliosis. That excess BMP directly inhibits remyelination is possible, but this has not yet been tested. Limited but provocative evidence exists that inhibition of BMP signaling can ameliorate the recovery from injury, but whether the mechanism involves remyelination, inhibition of gliosis, and/or possible other mechanisms is not yet known. It is exciting to speculate that these regulators of gliogenesis could be manipulated to promote maximal neural regeneration and remyelination in demyelinating diseases or after different types of CNS injury. Thus, it is important to understand their function in glial development and disease.

SUMMARY We have attempted to understand how a family of signaling molecules expressed in specific locations and specific times during development participates in the generation of a mature cell type from multipotential stem cells. A complex picture of regulation has emerged. Based on in vitro experiments in which BMPs are overexpressed, they clearly seem inhibitory to oligodendrocyte maturation and myelin protein expression. This is corroborated by overexpression in a mouse model in which modest decreases in oligodendrocytes are seen. Likewise, reduction of the BMP signals via treatment with inhibitors, antibodies, or ablation results in increased numbers of mature oligodendrocytes. Thus, it would be tempting to call BMPs simply inhibitory to oligodendrogliogenesis in general and postulate that the decrease in BMP signaling seen around the time of birth permits the generation of oligodendrocytes in areas where BMP was previously strongly expressed or in areas not counterbalanced by the expression of the BMP inhibitor noggin. Based on many studies of BMPs and oligodendrocyte development, it would also be tempting to define roles for BMPs in both critical junctures in the generation of oligodendrocytes, that is, specification from neural stem cells and differentiation from OPCs. Studies of the endogenous role of BMP signaling using 2 somewhat different genetic models, however, indicate a more complex picture in which BMPs may not affect the specification of OPCs and might indirectly play a positive role in the generation of oligodendrocytes. Intertwined in these observations is the role of BMP in the generation of astrocytes, possibly from the same stem cells that would generate oligodendrocytes. The stage is now set for additional studies to provide clarification on, among other things, the direct versus indirect roles of BMP in oligodendrocyte and astrocyte generation during CNS development.

602

REFERENCES 1. Lu QR, Yuk D-I, Alberta JA, et al. Sonic hedgehogYregulated oligodendrocyte lineage genes encoding bHLH proteins in the mammalian central nervous system. Neuron 2000;25:317Y29 2. Qi Y, Cai J, Wu Y, et al. Control of oligodendrocyte differentiation by the Nkx2.2 homeodomain transcription factor. Development 2001;128: 2723Y33 3. Stolt CC, Rehberg S, Ader M, et al. Terminal differentiation of myelin-forming oligodendrocytes depends on the transcription factor Sox 10. Genes Dev 2002;16:165Y70 4. Zhou Q, Wang S, Anderson DJ. Identification of a novel family of oligodendrocyte lineageYspecific basic helix-loop-helix transcription factors. Neuron 2000;25:331Y43 5. Barres BA, Lazar MA, Raff MC. A novel role for thyroid hormone, glucocorticoids and retinoic acid in timing oligodendrocyte development. Development 1994;120:1097Y108 6. Billon N, Jolicoeur C, Tokumoto Y, et al. Normal timing of oligodendrocyte development depends on thyroid hormone receptor alpha 1 (TRalpha1). EMBO J 2002;21:6452Y60 7. Hsieh J, Aimone JB, Kaspar BK, et al. IGF-1 instructs multipotent adult neural progenitor cells to become oligodendrocytes. J Cell Biol 2004; 164:111Y22 8. Orentas DM, Hayes JE, Dyer KL, et al. Sonic hedgehog signaling is required during the appearance of spinal cord oligodendrocyte precursors. Development 1999;126:2419Y29 9. Pringle NP, Yu WP, Guthrie S, et al. Determination of neuroepithelial cell fate: Induction of the oligodendrocyte lineage by ventral midline cells and sonic hedgehog. Dev Biol 1996;177:30Y42 10. Liu A, Niswander LA. Bone morphogenetic protein signalling and vertebrate nervous system development. Nat Rev Neurosci 2005;6: 945Y54 11. Mabie PC, Mehler MF, Marmur R, et al. Bone morphogenetic proteins induce astroglial differentiation of oligodendroglial-astroglial progenitor cells. J Neurosci 1997;17:4112Y20 12. Grinspan JB, Edell E, Carpio DF, et al. Stage-specific effects of bone morphogenetic proteins on the oligodendrocyte lineage. J Neurobiol 2000;43:1Y17 13. Pringle NP, Richardson WD. A singularity of PDGFalpha receptor expression in the dorsoventral axis of the neural tube may define the origin of the oligodendrocyte lineage. Development 1993;117: 525Y33 14. Samanta J, Burke GM, McGuire T, et al. BMPR1a signaling determines numbers of oligodendrocytes and calbindin-expressing interneurons in the cortex. J Neurosci 2007;27:7397Y407 15. See J, Mamontov P, Ahn K, et al. BMP signaling mutant mice exhibit glial cell maturation defects. Mol Cell Neurosci 2007;35:171Y82 16. Grinspan JB, Franceschini B. PDGF is a survival factor for PSANCAM + oligodendroglial pre-progenitor cells. J Neurosci Res 1995; 41:540Y51 17. LeVine SM, Goldman J. Embryonic divergence of oligodendrocyte and astrocyte lineages in the developing rat cerebrum. J Neurosci 1988;8: 3992Y4006

Ó 2009 American Association of Neuropathologists, Inc.

Copyright @ 2009 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.

J Neuropathol Exp Neurol  Volume 68, Number 6, June 2009

BMPs in Myelination and Demyelination

18. Raff MC, Abney ER, Cohen J, et al. Two types of astrocytes in cultures of developing rat white matter: Differences in morphology, surface gangliosides, and growth characteristics. J Neurosci 1983;3: 1289Y300 19. Hart IK, Richardson WD, Bolsover SR, et al. PDGF and intracellular signalling in the timing of oligodendrocyte differentiation. J Cell Biol 1989;109:3411Y17 20. Nishiyama A, Lin X-H, Giese N, et al. Interaction between NG2 proteoglycan and PDGF a-receptor on O2A progenitor cells is required for optimal response to PDGF. J Neurosci Res 1996;43:315Y30 21. Ben-Hur T, Rogister B, Murray K, et al. Growth and fate of PSA-NCAM + precursors of the postnatal brain. J Neurosci 1998;18: 5777Y88 22. Kondo T, Raff M. Oligodendrocyte precursor cells reprogrammed to become multipotential CNS stem cells. Science 2000;289:1754Y57 23. Raff MC, Miller RH, Noble M. A glial progenitor cell that develops in vitro into an astrocyte or an oligodendrocyte depending on the culture medium. Nature 1983;303:390Y96 24. Poncet C, Soula C, Trousse F, et al. Induction of oligodendrocyte progenitors in the trunk neural tube by ventralizing signals: Effects of notochord and floor plate grafts, and of sonic hedgehog. Mech Dev 1996;60:13Y32 25. Chandran S, Kato H, Gerreli D, et al. FGF-dependent generation of oligodendrocytes by a hedgehog-independent pathway. Development 2003;130:6599Y609 26. Wada T, Kagawa T, Ivanova A, et al. Dorsal spinal cord inhibits oligodendrocyte development. Dev Biol 2000;227:42Y55 27. Liem KF Jr, Jessell TM, Briscoe J. Regulation of the neural patterning activity of sonic hedgehog by secreted BMP inhibitors expressed by notochord and somites. Development 2000;127:4855Y66 28. McMahon JA, Takada S, Zimmerman LB, et al. Noggin-mediated antagonism of BMP signaling is required for growth and patterning of the neural tube and somite. Genes Dev 1998;12:1438Y52 29. Liem KF Jr, Tremml G, Roelink H, et al. Dorsal differentiation of neural plate cells induced by BMP-mediated signals from epidermal ectoderm. Cell 1995;82:969Y79 30. Wine-Lee L, Ahn KJ, Richardson RD, et al. Signaling through BMP type 1 receptors is required for development of interneuron cell types in the dorsal spinal cord. Development 2004;131:5393Y403 31. Miller RH, Dinsio K, Wang R, et al. Patterning of spinal cord oligodendrocyte development by dorsally derived BMP4. J Neurosci Res 2004;76:9Y19 32. Bakrania P, Efthymiou M, Klein JC, et al. Mutations in BMP4 cause eye, brain, and digit developmental anomalies: Overlap between the BMP4 and hedgehog signaling pathways. Am J Hum Genet 2008;82: 304Y19 33. Ohkawara B, Iemura S, ten Dijke P, et al. Action range of BMP is defined by its N-terminal basic amino acid core. Curr Biol 2002;12: 205Y9 34. Wrana JL, Attisano L, Wieser R, et al. Mechanism of activation of the TGF-beta receptor. Nature 1994;370:341Y47 35. Kretzschmar M, Doody J, Massague J. Opposing BMP and EGF signalling pathways converge on the TGF-beta family mediator Smad1. Nature 1997;389:618Y22 36. Zhang H, Miller RH. Density-dependent feedback inhibition of oligodendrocyte precursor expansion. J Neurosci 1996;16:6886Y95 37. Rajan P, Panchision DM, Newell LF, et al. BMPs signal alternately through a SMAD or FRAP-STAT pathway to regulate fate choice in CNS stem cells. J Cell Biol 2003;161:911Y21 38. Mehler MF, Mabie PC, Zhang D, et al. Bone morphogenetic protein in the nervous system. Trends Neurosci 1997;20:309Y17 39. Zou H, Wieser R, Massague J, et al. Distinct roles of type I bone morphogenetic protein receptors in the formation and differentiation of cartilage. Genes Dev 1997;11:2191Y203 40. Kondo T, Raff MC. A role for Noggin in the development of oligodendrocyte precursor cells. Dev Biol 2004;267:242Y51 41. Zimmerman LB, De Jesus-Escobar JM, Harland RM. The Spemann organizer signal noggin binds and inactivates bone morphogenetic protein 4. Cell 1996;86:599Y606 42. Hata A, Lagna G, Massague J, et al. Smad6 inhibits BMP/Smad1 signaling by specifically competing with the Smad4 tumor suppressor. Genes Dev 1998;12:186Y97

43. Onichtchouk D, Chen YG, Dosch R, et al. Silencing of TGF-beta signalling by the pseudoreceptor BAMBI. Nature 1999;401:480Y85 44. Gross RE, Mehler MF, Mabie PC, et al. Bone morphogenetic proteins promote astroglial lineage commitment by mammalian subventricular zone progenitor cells. Neuron 1996;17:595Y606 45. Mabie PC, Mehler MF, Kessler JA. Multiple roles of bone morphogenetic protein signaling in the regulation of cortical cell number and phenotype. J Neurosci 1999;19:7077Y88 46. Bansal R, Stefansson K, Pfeiffer SE. Proligodendroblast antigen (POA), a developmental antigen expressed by A007/O4-positive oligodendrocyte progenitors prior to the appearance of sulfatide and galactocerebroside. J Neurochem 1992;58:2221Y29 47. Raff MC, Mirsky R, Fields KL, et al. Galactocerebroside is a specific cell-surface antigenic marker for oligodendrocytes in culture. Nature 1978;274:813Y16 48. Scherer S, Braun P, Grinspan J, et al. Differential regulation of the 2¶3¶-cyclic nucleotide 3¶-phosphodiesterase gene in oligodendrocyte development. Neuron 1994;12:1363Y75 49. Grinspan J, Wrabetz L, Kamholz J. Oligodendrocyte maturation and myelin gene expression in PDGF-treated cultures from rat cerebral white matter. J Neurocytol 1993;22:322Y33 50. See J, Zhang X, Eraydin N, et al. Oligodendrocyte maturation is inhibited by bone morphogenetic protein. Mol Cell Neurosci 2004;26: 481Y92 51. Mekki-Dauriac S, Agius E, Kan P, et al. Bone morphogenetic proteins negatively control oligodendrocyte precursor specification in the chick spinal cord. Development 2002;129:5117Y30 52. Gomes WA, Mehler MF, Kessler JA. Transgenic overexpression of BMP4 increases astroglial and decreases oligodendroglial lineage commitment. Develop Biol 2003;255:164Y77 53. Ahn K, Mishina Y, Hanks MC, et al. BMPR-IA signaling is required for the formation of the apical ectodermal ridge and dorsal-ventral patterning of the limb. Development 2001;128:4449Y61 54. Tang X-M, Beesley JS, Grinspan JB, et al. Cell cycle arrest induced by ectopic expression of p27 is not sufficient to promote oligodendrocyte differentiation. J Cell Biochem 1999;76:270Y79 55. Casaccia-Bonnefil P, Hardy RJ, Teng KK, et al. Loss of p27Kip1 function results in increased proliferative capacity of oligodendrocyte progenitors but unaltered timing of differentiation. Development 1999; 126:4027Y37 56. Wang S, Sdrulla AD, diSibio G, et al. Notch receptor activation inhibits oligodendrocyte differentiation. Neuron 1998;21:63Y75 57. McMorris FA, Dubois-Dalcq M. Insulin-like growth factor 1 promotes cell proliferation and oligodendroglial commitment in rat glial progenitor cells developing in vitro. J Neurosci Res 1988;21:199Y209 58. Hollnagel A, Oehlmann V, Heymer J, et al. Id genes are direct targets of bone morphogenetic protein induction in embryonic stem cells. J Biol Chem 1999;274:19838Y45 59. Nakashima K, Takizawa T, Ochiai W, et al. BMP2-mediated alteration in the developmental pathway of fetal mouse brain cells from neurogenesis to astrocytogenesis. Proc Natl Acad Sci U S A 2001;98: 5868Y73 60. Norton JD, Deed RW, Craggs G, et al. Id helix-loop-helix proteins in cell growth and differentiation. Trends Cell Biol 1998;8:58Y65 61. Wang SL, Sdrulla A, Johnson JE, et al. A role for the helix-loop-helix protein Id2 in the control of oligodendrocyte development. Neuron 2001; 29:603Y14 62. Samanta J, Kessler JA. Interactions between ID and OLIG proteins mediate the inhibitory effects of BMP4 on oligodendroglial differentiation. Development 2004;131:4131Y42 63. Cheng X, Wang Y, He Q, et al. Bone morphogenetic protein signaling and olig1/2 interact to regulate the differentiation and maturation of adult oligodendrocyte precursor cells. Stem Cells 2007;25:3204Y14 64. Bilican B, Fiore-Heriche C, Compston A, et al. Induction of Olig2 precursors by FGF involves BMP signalling blockade at the Smad level. PLoS ONE 2008;3:e2863 65. Colak D, Mori T, Brill MS, et al. Adult neurogenesis requires Smad4-mediated bone morphogenic protein signaling in stem cells. J Neurosci 2008;28:434Y46 66. Cameron-Curry P, LeDouarin NM. Oligodendrocyte precursors originate from both the dorsal and ventral parts of the spinal cord. Neuron 1995; 15:1299Y310

Ó 2009 American Association of Neuropathologists, Inc.

603

Copyright @ 2009 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.

See and Grinspan

J Neuropathol Exp Neurol  Volume 68, Number 6, June 2009

67. Orentas DM, Miller RH. The origin of spinal cord oligodendrocyte is dependent on local influences from the spinal cord. Dev Biol 1996;177: 43Y53 68. Cai J, Qi Y, Hu X, et al. Generation of oligodendrocyte precursor cells from mouse dorsal spinal cord independent of Nkx6 regulation and Shh signaling. Neuron 2005;45:41Y53 69. Fogarty M, Richardson WD, Kessaris N. A subset of oligodendrocytes generated from radial glia in the dorsal spinal cord. Development 2005; 132:1951Y59 70. Vallstedt A, Klos JM, Ericson J. Multiple dorsoventral origins of oligodendrocyte generation in the spinal cord and hindbrain. Neuron 2005;45:55Y67 71. Bansal R, Pfeiffer SE. Regulation of oligodendrocyte differentiation by fibroblast growth factors. Adv Exp Med Biol 1997;429:69Y77 72. Grinspan JB, Reeves MF, Coulaloglou MJ, et al. Re-entry in to the cell cycle is required for bFGF-induced oligodendroglial dedifferentiation and survival. J Neurosci Res 1996;46:456Y64 73. Setoguchi T, Nakashima K, Takizawa T, et al. Treatment of spinal cord injury by transplantation of fetal neural precursors cells engineered to express BMP inhibitor. Exp Neurol 2004;189:33Y44 74. Setoguchi T, Yone K, Matsuoka E, et al. Traumatic injuryYinduced BMP 7 expression in the adult rat spinal cord. Brain Res 2001;921:219Y25 75. Martinez G, Carnazza M, Di Giacomo C, et al. Expression of bone morphogenetic protein 6 and transforming growth factor b1 in the rat brain after a mild and reversible ischemic damage. Brain Res 2001;894:1Y11 76. Ara J, See J, Mamontov P, et al. Bone morphogenetic proteins 4, 6, and 7 are up-regulated in mouse spinal cord during experimental autoimmune encephalomyelitis. J Neurosci Res 2008;86:125Y35 77. Chen J, Leong SY, Schachner M. Differential expression of cell fate determinants in neurons and glial cells of adult mouse spinal cord after compression injury. Eur J Neurosci 2005;22:1895Y906 78. Matsuura I, Taniguchi J, Hata K, et al. BMP inhibition enhances axonal growth and functional recovery after spinal cord injury. J Neurochem 2008;105:1471Y79

79. Hampton DW, Asher RA, Kondo T, et al. A potential role for bone morphogenetic protein signalling in glial cell fate determination following adult central nervous system injury in vivo. Eur J Neurosci 2007;26:3024Y35 80. Lewen A, Soderstrom S, Hillered L, et al. Expression of serine/threonine kinase receptors in traumatic brain injury. Neuroreport 1997;8:475Y79 81. Lai M, Gluckman P, Dragunow M, et al. Focal brain injury increases activin betaA mRNA expression in hippocampal neurons. Neuroreport 1997;8:2691Y94 82. Zhao C, Fancy SPJ, Magy L, et al. Stem cells, progenitors and myelin repair. J Anat 2005;207:251Y58 83. Fuller ML, Dechant AK, Rothstein B, et al. Bone morphogenetic proteins promote gliosis in demyelinating spinal cord lesions. Ann Neurol 2007;62:288Y300 84. Deininger M, Meyermann R, Schluesener H. Detection of two transforming-growth-factor-betaYrelated morphogens, bone morphogenetic proteins-4 and -5, in RNA of multiple sclerosis and Creutzfeldt-Jakob disease lesion. Acta Neuropathol 1995;90:76Y79 85. Enzmann GU, Benton RL, Woock JP, et al. Consequences of noggin expression by neural stem, glial, and neuronal precursor cells engrafted into the injured spinal cord. Exp Neurol 2005;195:293Y304 86. Shen H, Luo Y, Kuo CC, et al. 9-Cis-retinoic acid reduces ischemic brain injury in rodents via bone morphogenetic protein. J Neurosci Res 2009;87:545Y55 87. Harvey BK, Hoffer BJ, Wang Y. Stroke and TGF-beta proteins: Glial cell lineYderived neurotrophic factor and bone morphogenetic protein. Pharmacol Ther 2005;105:113Y25 88. Zuch CL, David D, Ujhelyi L, et al. Beneficial effects of intraventricularly administered BMP-7 following a striatal 6-hydroxydopamine lesion. Brain Res 2004;1010:10Y16 89. Xin H, Li Y, Chen X, et al. Bone marrow stromal cells induce BMP2/4 production in oxygen-glucoseYdeprived astrocytes, which promotes an astrocytic phenotype in adult subventricular progenitor cells. J Neurosci Res 2006;83:1485Y93

604

Ó 2009 American Association of Neuropathologists, Inc.

Copyright @ 2009 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.