This is Part 2 of a two-part meeting report from the 37th annual meeting of the Society for Neuroscience, held 3-7 November, in San Diego. See Part 1.
15 November 2007. Cell-based therapies have the potential for treating a variety of diseases including diabetes, leukemia, and even neurodegenerative diseases such as Parkinson and Alzheimer disease. But while replacing worn out pancreatic or hematopoietic cells may be relatively straightforward, getting new neurons to seamlessly slip into pre-existing circuitry may be a bit like sending an understudy into the last act of a play without one ever having read the lines. Might there be a better way to keep the show on the road? At this year’s Society for Neuroscience annual meeting, Clive Svendsen, University of Wisconsin, Madison, suggested that for neurodegenerative diseases such as ALS and Parkinson disease (PD), it may be unnecessary to replace the lost neurons if the neurons that are still there can be protected. Svendsen was chair of a mini-symposium entitled “The Role of Glial Cells in Brain Injury and Disease.”
There have been several attempts at shoring up neurons in humans, but none were successful. A recent trial for GDNF in Parkinson disease was perhaps the most disappointing because initial signs had been promising (see ARF related news story). Svendsen avoids the complications that plague some delivery systems, such as pumps and live viruses, by using ex-vivo techniques to turn glial cells into mini-protein factories. Other advantages of this approach are that the glial cells alone may be therapeutic and the delivery system can be tested in vitro before implantation. With Patrick Aebischer, at the Swiss Federal Institute of Technology in Lausanne, Svendsen has developed a way to differentiate human neural precursor cells into predominantly astroglia and engineer them to release GDNF. The researchers tried this system in rodent and primate models of PD and have seen some improvement in function (see Behrstock et al., 2006). They have also tried the same approach in ALS models but were less successful, Svendsen said. Even though the cell transplants survived, spread into areas of degeneration, and almost completely protected against neuronal loss, the animals’ function did not improve (see Suzuki et al., 2007). The outcome was poor probably because the treatment failed to protect the neuromuscular junction. Svendsen showed how α-bungarotoxin staining revealed loss of the muscle endplate, and he is now working on strategies to deliver a cocktail of GDNF, IGF-1, VEGF, and BDNF via mesenchymal cells to the neuromuscular junctions. So far this has protected about 50 percent of the junctions, and improved function. Svendsen noted that cell-based strategies may work better in a slowly progressing disease, such as PD, than in more rapidly deteriorating diseases such as ALS. This is because the transplanted glia take time to mature. Even though cells injected into the brain diffused into the striatum in the PD models, it took 120 days before GFAP showed up there. In the case of ALS, targeting both the spinal cord and the muscle may achieve the best results.
One other potential use of cell therapy lies in spinal cord injury repair. Mary Bunge, University of Miami, Florida, described how transplanted Schwann cells, under varying conditions, can support spinal cord repair. The idea is that when axons die back after injury, they leave behind a tunnel of Schwann cells. If the axons can be coaxed to grow back, they may re-establish their previous connections. Since Schwann cells provide not only myelin to ensheath axons, but also a variety of neurotrophic factors, matrix proteins, and cell adhesion molecules that promote axon growth, transplanting these cells to sites of injury could prove beneficial. Strategies based on a single therapeutic approach, i.e., transplanting Schwann cells alone, have proven unsuccessful. At present, the scientists are experimenting with combinations of multiple approaches to get better results. These include providing Schwann cells in a growth-supportive matrix, or matrigel, limiting the formation of scar tissue by delivering chondroitinase to the site of injury, coaxing axons to travel through the site of injury by adding olfactory ensheathing glia (OEG) to the distal side (these cells promote axon elongation), and activating neurons by transfecting with transgenes or Schwann cell-derived neurotrophins.
Bunge reported that the combination of Schwann cells, OEG, and chondroitinase led to an increase in myelinated nerve fibers within grafts at the site of spinal cord transection, and to significant improvement in locomotion. Her lab also found that axon growth into the injury site can be stimulated by transfecting the spinal cord with AAV virus carrying MEK and ERK kinase constructs for expression in neurons.
Spinal contusion is a more common form of injury in humans than complete transection. Contusions in animals are usually followed about 12 weeks later by formation of a cyst at the site of injury, but it can be reduced, and function improved by transplanting Schwann cells 1 week after injury. To build on this, Bunge’s lab has genetically modified Schwann cells to produce a bi-functional neurotrophin called D15A. This is a mutant form of the human neurotrophin NT-3. Substituting aspartic acid 15 with alanine confers both TrkB and TrkC receptor binding, so the protein effectively mimics both NT-3 and BDNF. Six weeks after transplantation of D15A-expressing Schwann cells into rat spinal cord, the graft volume was about fivefold bigger than that elicited by untransformed cells, and there were more myelinated axons and the serotonergic and sensory fibers were longer. Unfortunately, the strategy did not improve locomotion, probably because the axons failed to migrate through the graft to the distal side. One strategy to overcome this problem might be as simple as inducing cAMP signaling, Bunge said. Infusing cAMP and the phosphodiesterase inhibitor rolipram into the site of damage along with Schwann cells was better than cell therapy alone. The numbers of myelinated fibers increased, serotonergic fiber grew into and beyond the graft, and that correlated with improved function.
Glia are not the first things that come to mind when one thinks of Alzheimer disease; nevertheless, understanding their contributions to both the healthy and diseased central nervous systems could lead to a better grasp of pathogenesis and, potentially, novel therapeutic approaches. The November issue of Nature Neuroscience features additional review articles on the role of glial cells in demyelination, ischemic toxicity, regulation of the microvasculature, and neuropathology—all areas that impinge in some way on age-related dementia. So do microglia and the brain’s oligodendroglia, but these cell types unfortunately got little attention at this symposium.—Tom Fagan.