Skip to main content


NIH Symposium: Challenges & Promise of Cell-Based Therapies
May 6, 2008
Natcher Conference Center
NIH Main Campus, Bethesda, Maryland

Session 1: Neurological Disorders

Moderator: Ronald McKay, Ph.D.; National Institute of Neurological Disorders and Stroke, NIH

Combining Stem Cell and Gene Therapy for Neurological Disorders
Clive N. Svendsen, Ph.D.; University of Wisconsin-Madison

Dr. Svendsen discussed the potential uses of combined stem-cell and gene therapies for neurologic disorders such as Parkinson's disease (PD) and amyotrophic lateral sclerosis (ALS, also known as Lou Gehrig's disease). Human neural stem cells have the potential to differentiate into neurons, astrocytes, or oligodendrocytes. Of these pathways, neuronal replacement is the most challenging because it requires integration to the neural circuit and a long-distance molecular "connection." Oligodendrocyte replacement represents a less difficult treatment option, although remyelination is more practical for treating diseases such as MS. The astrocyte, however, is a relatively simple cell by comparison. Astrocytes modulate the transfer of components from the blood to the brain and thus regulate many disease processes. Moreover, transplantation of stem cells that give rise to new astrocytes is comparatively easy, suggesting that they could be used to deliver drugs that prevent cell death and enhance plasticity. New astrocytes can also be modified to release growth factors such as glial cell line-derived neurotrophic factor (GDNF).

In the treatment of PD, stem cells could theoretically be used to replace lost or damaged dopamine neurons in the substantia nigra. However, transplant procedures to date have yet to enabled circuit reconstruction or reconnection. Can dopamine neurons be protected from death through a combination of astrocyte replacement and the release of growth factors such as GDNF? GDNF was originally isolated from a glial cell line and found to have survival and differentiation effects on dopamine neurons (Lin LF, Science 1993;260:1130-1132). It is now known to protect many different types of neurons in the brain following damage, although it does not penetrate the brain and therefore cannot be administered orally (Airaksinen MS and Saarma M. Nat Rev Neurosci 2002;3:383-394). Several options have been investigated for delivering GDNF into the brain, including direct infusion into the striatum (Gill SS, Nat Med 2003;9:589-595) and in vivo gene therapy using viral particles. However, a large clinical trial using the former method failed (Lang AE, Ann Neurol 2006;59:459-466), suggesting that the field had reached a crossroads (Sherer TB, Mov Disord 2006;21:136-141; Salvatore MF, Exp Neurol 2006;202:497-505).

While one clinical trial to assess the applicability of in vivo gene therapy using viral particles is underway, stem cells that produce GDNF may represent an alternate strategy. GDNF delivery ex vivo has several advantages; namely, it obviates the need for introducing viral particles, and there is a known release of GDNF prior to transplant. Moreover, astrocytes alone may be therapeutic, thereby eliminating the need to infect host neurons. In addition, human neural progenitor cells can be modified to produce GDNF (Capowski EE, J Neurosci Methods 2007;163:338-349). Because these stem cells are derived from neural tissue, they create only neural tissue—tumor formation is greatly reduced. GDNF increases the survival of dopamine neurons, and human neural progenitor cells have been shown to migrate and mature into astrocytes following transplantation into a rat model of PD (Svendsen CN, Exp Neurol 1997;148:135-146). Moreover, there is no evidence of tumor formation from these cells (Ostenfeld T, Exp Neurol 2000;164:215-226), and they release GDNF after approximately ten weeks post-transplant (Behrstock S, Gene Ther 2006;13:379-388). Following release of GDNF and insulin-like growth factor-1, these cells restored function in a rat model of PD (Ebert AD, Exp Neurol 2008;209:213-223). Studies are now underway to assess the success of this approach in monkey models of PD.

These experiments support a general strategy to regenerate neural tissue; many growth factors may have different uses to address the issues in PD. Rat and monkey model studies thereby offer the following conclusions:

  • Human neural progenitor cells modified to release GDNF survive transplantation into the lesioned rat and monkey striatum.
  • These cells integrate and form glial fibrillary acidic protein (GFAP)-positive astrocytes at late time points.
  • GDNF released from these cells protects dopamine neurons and induces dopamine neuron sprouting and functional effects.
  • Human trials now await regulated vectors to allow appropriate dosing of GDNF over time.

Dr. Svendsen then discussed the combination of stem cell- and gene therapies to treat ALS. A lack of muscle nourishment seen with this disorder ultimately leads to the death of motor neurons. Persons with ALS usually survive less than four years after diagnosis, and the cause of the disease is not known. No treatment is currently available. ALS affects upper and lower motor neurons, and astrocytes play a role in modulating disease (Clement AM, Science 2003;302:113-117). As it does with dopamine neurons, GDNF represents a potent survival factor for motor neurons present in peripheral nerve and muscle (Henderson CE, Science 1994;266:1062-1064). Human neural progenitor cells transplanted into a rat model of ALS release GDNF and survive to integrate within the rat spinal cord (Klein SM, Hum Gene Ther 2005;16:509-521) and protect motor neurons (Suzuki M, PLoS ONE 2007;2:e689). However, despite the evidence of surviving cells, functional improvement is not observed in these rats, suggesting that contact of the motor neuron with the muscle is not maintained by GDNF delivered to the spinal cord.

However, could the protection be administered at the point of the muscle, rather than the spinal cord? Retrograde viral delivery of IGF-1 has been shown to prolong survival in a mouse model of ALS (Kaspar BK, Science 2003;301:839-842). In addition, human mesencyhmal stem cells (hMSCs) survive injection into muscle and release GDNF in a rat model of ALS, protecting motor neurons and increasing survival. Can ALS therefore be treated with a combination of growth factors and stem cells? In humans, the cells would have to linger in order to mature into astrocytes, and GDNF may therefore have more functional effects in sporadic ALS. The main hurdle is delivery of the cells into the spinal cord.

Dr. Svendsen concluded by noting that neuronal replacement in neurological disease remains challenging due to the difficulty of establishing new connections in the brain, although it exhibits great potential as a therapeutic strategy. Combining astrocyte replacement with growth factor release is a powerful strategy that is both practical and immediately applicable. Other areas of active investigation for this approach include retinal diseases, stroke, and Huntington’s disease.


One attendee asked about the need for immunosuppression for astrocyte grating. Dr. Svendsen noted that immunosuppression was not necessary for treatment of PD after six months.