Glial Progenitor Cells and the Development of Glial Repair Strategies

Project Overview

One of the major goals of the field of stem cell medicine is the use of cell transplantation to repair damaged tissue. In the first stage of our work in this field, we focused our attention on repair of demyelinating damage, and were the first group to define means of promoting O-2A progenitor cell division in vitro and then of discovering means of growing sufficient numbers of progenitor cells to carry out repair by transplantation of pure populations of progenitor cells. These discoveries then enabled us, in 1993, to provide the first demonstration of CNS repair by transplantation of pure populations of ex vivo expanded progenitor cells.

In the next stage of our work on CNS repair, we have continued with our interest in repair of the damaged spinal cord to address the problem of how to optimize repair by cell transplantation. This work (carried out in collaboration with colleagues Stephen Davies and Jeannette Davies, has led to two surprising findings. First, these studies demonstrate that the degree of repair obtained in the damaged spinal cord is greatly increased by pre-differentiating progenitor cells into a specific population of astrocytes prior to repair, and indicates that injured tissues may not be able to direct undifferentiated precursor cells into the cell types optimal for repair. Second, these studies have revealed that transplantation of the wrong population of astrocytes, or even of precursor cells themselves, not only fails to promote repair but also can induce generation of neuropathic pain syndromes. The ability to cause harm by cell transplantation is a very serious problem, with grave implications for current repair strategies.

It is also important to emphasize that all of our studies on CNS repair were only possible due to our dedicated efforts in cell discovery, emphasizing the importance of basic research as the foundation for translational research.

Oligodendrocyte replacement and repair of demyelinating damage
The ability to isolate progenitor cells is a first step towards using these cells to repair damage in the CNS. After first discovering how to control division of O-2A progenitor cells in vitro (Noble and Murray, 1984; Noble et al., 1988; Raff et al., 1988), we then discovered that particular combinations of mitogens caused these cells to divide extensively without undergoing differentiation (Bögler et al., 1990). The ability to grow essentially unlimited numbers of primary O-2A progenitor cells enabled us, in 1993, to provide the first demonstration of repair of CNS damage by transplantation of purified and ex vivo expanded neural progenitor cells (Groves et al., 1993). Multiple laboratories, using isolation strategies and/or growth conditions based on the ones developed in the early stages of our research, have been pursuing repair of demyelinating damage using human-derived cells in ways that hold promise for application in the clinical setting.

Development of astrocyte transplantation therapies and the importance of pre-differentiation as a strategy for optimizing CNS repair
Our ability to study multiple neural progenitor cells also has provided the essential tools for developing replacement strategies for astrocytes, the major support cell of the CNS. Working together with Drs. Stephen Davies and Jeannette Davies (University of Colorado Health Sciences Center, Denver, Colorado), we recently found that transplantation of astrocytes derived from GRP cells treated with bone morphogenetic protein (BMP)-4 (i.e, GDAsBMP) to acute transection injuries of adult rat spinal cord promoted a ~40% efficiency of axon regeneration across sites of injury, protection of axotomized red nucleus neurons, suppression of scarring, and a degree of behavioral recovery from dorsolateral funniculus injuries that enabled rats to generate scores by 4 weeks after transplantation that were statistically indistinguishable from uninjured animals on a stringent test of volitional foot placement (Davies et al., 2006).

The impressive regeneration and behavioral recovery promoted by transplantation of GDAsBMP was not achieved by transplantation of GRP cells themselves, which did not provide any detectable benefit. These results indicate that relying upon the signals in injured tissues to produce desired cell types from transplanted stem or progenitor cells may be a deeply flawed approach to the promotion of repair. Instead, pre-differentiation to obtain a precise cell type proves more effective.

While development of astrocyte transplanation therapies is an important goal in and of itself, we also view this work in the more general context of developing strategies to prioritize among the multiple cell types available for carrying out tissue repair. For example, a variety of cell types of both non-CNS and CNS origin are being considered for clinical trials in spinal cord injury, including Schwann cells, olfactory ensheathing glia, marrow stromal cells and oligodendrocyte progenitor cells. Non-CNS cells such as Schwann cells, olfactory ensheathing cells and marrow stromal cells have been particularly attractive candidates for use in CNS repair due to their relative ease of isolation compared to cells of CNS origin. But none of these cells have had particularly dramatic effects on promoting CNS regeneration, and the results obtained with such transplants are much less than obtained by transplantation of GDAsBMP. Thus, it is critical to determine whether GDABMP transplantation provides a strategy suitable for translation to the clinic.

Causing harm by transplanting the wrong cell type
The ability to GDAsBMP to promote extensive recovery after transplantation into spinal injury lesions raises the question of whether use of this specific astrocyte population is critical or whether other astrocytes generated from GRP cells might be equally effective – and the experiments that address this question demonstrate that it is GDAsBMP that are critical in developing astrocyte transplantation therapies. We addressed this question by exposing GRP cells to ciliary neurotrophic factor (CNTF) and then transplanting the resultant astrocytes (i.e, GDAsCNTF). BMP and CNTF both have been implicated in promoting astrocyte generation in vivo, but the astrocytes they generate from GRP cells have different properties. Astrocytes induced by exposure of GRP cells to BMP have the antigenic and morphological characteristics that were originally defined as type-1 astrocytes, while those generated by exposure to CNTF have the characteristics originally considered to define a separate population of type-2 astrocytes.

Transplantation of GDAsBMP and GDAsCNTF yielded dramatically different outcomes, with GDAsCNTF providing no benefit when transplanted into acute spinal cord injuries (Davies et al., 2008). Animals receiving these transplants showed no evidence of regeneration, neuronal rescue or behavioral recovery. Thus, it is critical to generate precisely the right type of astrocytes from GRP cells in order to promote recovery by cell transplantation. These results offer the first demonstration of functional differences between astrocytes in the context of tissue repair. Moreover, as O-2A progenitor cells generate type-2 astrocytes even when exposed to BMP these results raise the possibility that glial progenitor cells other than GRP cells may not be able to generate the specific population of astrocytes most useful in promoting repair.

topic ii pic 1

Transplantation of GDAsBMP, but not of GDAsCNTF or of GRP cells, promotes extensive axonal regeneration, tissue reorganization and behavioral recovery following experimental spinal cord injury. A) Examination of the distance of regeneration of BDA labeled axons across the transected spinal cord at 8 days post injury reveals that in animals transplanted with GDAsBMP nearly two-thirds of labeled axons grew as far as the lesion center and the majority of these grew back into distal cord, often extending distances of 5mm beyond the lesion center. In contrast, only a small fraction (<5%) of BDA labeled axons reached the lesion center in animals receiving GRP cell or GDACNTF transplants. B,C) Microphotographs showing that BDA-labeled axons do not enter sites of GDACNTF transplantation (i.e, the dense red zone in the center of (B)) but readily enter and traverse the lesion site following GDABMP transplantation (C). D) Staining with anti-GFAP antibodies reveals the disorganized glial scar tissue that occurs following dorsal column transection even when GRP cells or or GDAsCNTF were transplanted (The unlabeled tissue on the left of the picture contains transplanted GRP cells). E) In contrast, transplantation of GDAsBMP is associated with extensive linear re-organization of the host astrocytes. (As the GDAsBMP down-regulate GFAP expression following transplantation, they are not visualized in this figure with anti-GFAP staining. Other stains reveal these cells (which occupy the unlabeled left to be linearly organized left side of the spinal cord) to be linearly organized with the host astrocytes. E) Transplantation of GDAsBMP, but not of GDAsCNTF, promotes extensive recovery in a volitional foot placement test following transection lesions of the ruprospinal tract. In this behavioral grid-walk (foot placement) assay, animals are first trained until they make few mistakes. Injury is associated with a sharp increase in the number of mistakes and with an absence of recovery, and transplantation of GDAsCNTF does not promote improvements in outcome. Transplantation of GRP cells themselves (data not shown) yields an outcome indistinguishable from transplantation of GDAsCNTF. In contrast, GDABMP transplantation provides initial protection and promotes behavioral recovery so extensive that by the end of 4 weeks many of the transplanted animals cannot be statistically distinguished from non-operated control animals by this assay.

Further analysis of the effects of GRP cell, GDABMP and GDACNTF transplantation demonstrated that the transplantation of GRP cells or GDAsCNTF was not just ineffective, but also caused animals to develop neuropathic pain syndromes. This is a particularly disturbing outcome, as such pain syndromes greatly limit the quality of life in individuals with spinal cord injury. The possibility that cell transplantation might actually worsen the quality of life for someone with these injuries represents a very serious concern that suggests appropriate caution is warranted in moving forward with cell transplantation to treat spinal cord injury.

topic ii pic 2

Transplantation of GDAsCNTF or of GRP cells, but not of GDAsBMP, causes neuropathic pain syndromes and sprouting of CGRP+ fibers thought to play a role in generating such pain responses. A) Animals receiving GDACNTF transplants or GRP cells transplanted showed increased sensitivity to mechanical stimuli and also showed evidence of thermal hyperalgesia (data not shown, see * for details). B) Despite the extensive regeneration promoted by transplantation of GDAsBMP, transplantation of these cells was not associated with increases in sprouting of CGRP+ neurons. In contrast, transplantation of GDAsCNTF or GRP cells was associated with a selective sprouting of CGRP+ neurons despite the absence of regeneration of other neurons examined.

This work provides an example of how our basic research in precursor cell discovery enables the pursuit of nuanced analyses of repair and functional analysis of different cell populations that would not be possible without the carefully defined cell populations that our team of investigators continues to identify.


Bögler, O., Wren, D., Barnett, S. C., Land, H., and Noble, M. (1990). Cooperation between two growth factors promotes extended self-renewal and inhibits differentiation of oligodendrocyte-type-2 astrocytes (O-2A) progenitor cells. Proc. Natl. Acad. Sci, U.S.A 87, 6368-6372. link

Davies, J. E., Huang, C., Proschel, C., Noble, M., Mayer-Proschel, M., and Davies, S. J. (2006). Astrocytes derived from glial-restricted precursors promote spinal cord repair. J Biol. 5(3):7. link

Davies, J. E., Pröschel, C., Zhang, N., Noble, M., Mayer-Proschel, M., and Davies, S. J. A. (2008). Transplanted astrocytes derived from BMP or CNTF treated glial restricted precursors have opposite effects on recovery and allodynia after spinal cord injury. J. Biol. 7:e24. link

Groves, A. K., Barnett, S. C., Franklin, R. J., Crang, A. J., Mayer, M., Blakemore, W. F., and Noble, M. (1993). Repair of demyelinated lesions by transplantation of purified O-2A progenitor cells. Nature 362, 453-455. link

Noble, M., and Murray, K. (1984). Purified astrocytes promote the in vitro division of a bipotential glial progenitor cell. EMBO-J 3, 2243-2247. link

Noble, M., Murray, K., Stroobant, P., Waterfield, M. D., and Riddle, P. (1988). Platelet-derived growth factor promotes division and motility and inhibits premature differentiation of the oligodendrocyte/type-2 astrocyte progenitor cell. Nature 333, 560-562. link

Raff, M. C., Lillien, L. E., Richardson, W. D., Burne, J. F., and Noble, M. D. (1988). Platelet-derived growth factor from astrocytes drives the clock that times oligodendrocyte development in culture. Nature 333, 562-565. link


Mark D. Noble
University of Rochester
601 Elmwood Ave., Box 633
Rochester, NY 14642
Office: MRB 2-9625
+1-585-273-1448 mark_noble@urmc.