10 October 2011. Bad neighbors are usually a mere annoyance, but in amyotrophic lateral sclerosis, they kill. According to a recent study, astrocytes, which are supposed to nurture neurons, turn deadly when they express the disease-linked mutant enzyme superoxide dismutase 1 (mSOD1). Nicholas Maragakis and colleagues at Johns Hopkins University School of Medicine in Baltimore, Maryland, proved it so when they injected normal rats with mSOD1-carrying astrocyte precursors, which then destroyed the neurons in their vicinity. The team reports their results in the October 3 Proceedings of the National Academy of Sciences online. Alzforum covered Maragakis’ preliminary results in 2010 at the Fondation André-Delambre symposium in Québec City, Canada (see ARF related news story). The evidence suggests that astrocytes may kill motor neurons, at least in part, by activating microglia and causing inflammation.
The now-published work clinches a role for astrocytes that many scientists have suspected in amyotrophic lateral sclerosis ever since Don Cleveland first published on the subject (reviewed in Ilieva et al., 2009). In cell culture, mSOD1-containing astrocytes destroy wild-type motor neurons (see ARF related news story on Nagai et al., 2007 and Di Giorgio et al., 2007; ARF related news story on Di Giorgio et al., 2008 and Marchetto et al., 2008). And in chimeric mice that carry mutant human SOD1 in all cells of the body bar astrocytes, motor neuron disease progression slows and pathology-linked microglia activation drops (Yamanaka et al., 2008; Wang et al., 2011), further implicating astrocytes in pathology. If there was any remaining doubt that astrocytes with mSOD1 can kill motor neurons, this in-vivo model ought to quell it, said Neil Cashman of the University of British Columbia in Vancouver, Canada, who was not involved in the study. “There is no ambiguity here,” he told ARF. Of course, the study does not rule out an additional toxic role for mSOD1 in motor neurons or other cell types, he added.
The Maragakis team previously showed that injecting wild-type astrocyte precursors into mSOD1 rats dampened inflammation and improved survival (see ARF related news story on Lepore et al., 2008). In the current work, first author Sophia Papadeas and Maragakis used wild-type rats as the host for injected mouse cells containing human mSOD1. They injected glial-restricted precursor (GRP) cells isolated from the developing spinal cords of mice that carried the mutation (Rao and Mayer-Proschel, 1997). Constrained to develop into glial lineages, these cells over the ensuing three months differentiated almost exclusively into astrocytes expressing the cell type-specific marker glial fibrillary acid protein (GFAP).
Injecting new cells into the spinal cord is no gentle procedure, and the researchers performed controls to ensure they were observing the effects of the mSOD1 in the transplants, not the transplantation process. For example, they injected some rats with astrocytes from animals expressing wild-type human SOD1, or astrocytes from wild-type mice with no transgene.
The GRP-derived astrocytes settled into their new surroundings, interacted with host neurons, and, only in the case of the mSOD1-toting astrocytes, decimated neurons. The researchers counted neurons within 240 micrometers of the transplant. Three months after injection, they found an average of three or four motor neurons near each transplant site compared to 11 or 12 motor neurons near control transplant sites. Motor neurons remaining near mSOD1 astrocytes contained ubiquitin inclusions, as in ALS. However, they contained no aggregates of SOD1—as appear in people who have ALS caused by SOD1 mutations—implying these accumulations are not crucial to neurodegeneration. The presence of ubiquitin inclusions extended to motor neurons beyond the central transplant site, suggesting the astrocytes influenced nearby neurons via a secreted factor or other unknown mechanism.
The injected animals did not have actual ALS, Maragakis cautioned, nor did they have as extensive a motor neuron disease as rodents carrying mSOD1 in every cell in their bodies. “It looks like a watered-down version,” commented Brett Morrison, who is also at Johns Hopkins University School of Medicine but was not involved in the study.
The rats did manifest regional signs of motor neuron disease. They lost strength in their forelimbs because the transplanted cells were in the cervical spinal cord near the front legs, and electrophysiology tests showed reduced action potentials in their diaphragms.
ALS is associated with a variety of potential mechanisms, including inflammation (see ARF related news story and ARF news story on Chiu et al., 2009), mitochondrial dysfunction (see ARF related news story on Keeney and Bennett, 2010), and apoptosis (see ARF related news story on Reyes et al., 2010). Papadeas assessed inflammation in the form of activated microglia, the nervous system’s resident immune cells. Immunohistochemistry for the microglial marker Iba-1 was twice as intense in the mSOD1-transplanted animals as in controls, although only half as strong as the signal in animals that carried mSOD1 in every cell.
The astrocytes might recruit or activate the microglia, which could then direct motor neuron degeneration. Papadeas treated the animals with minocycline, an antibiotic that inhibits microglia (Yrjänheikki et al., 1999), which reduced Iba-1 staining. The drug, which has failed in human ALS trials, was unable to save the motor neurons next to the mSOD1 astrocyte transplants; however, it did protect more distant motor neurons. The drug also prevented the loss of forelimb strength and somewhat ameliorated the reduced diaphragm action potentials. These results suggest that microglia mediate the astrocytes’ attack, at least on distant neurons. Morrison cautioned that minocycline is a “dirty drug” that has multiple effects, including blocking cell death pathways. It remains possible that the medication was preventing neuron death by apoptosis, not microgliosis. The paper leaves open the possibility that other pathological events could also be part of mSOD1 astrocytes’ toxic effects.
The rats represent a new kind of model that could allow scientists to dissect out the precise role of astrocytes in ALS. The relatively mild nature of their pathology may give scientists the opportunity to observe neurodegeneration over long periods. One disadvantage of the model, which might prevent it from catching on, is the skill required to perform the injections. “It is technically very demanding,” Cashman said, calling the work a “tour de force.”
Although approximately 2 percent of people with ALS have SOD1 mutations, the work could be relevant to non-SOD1 disease as well, Cashman said, noting that astrocytes derived from people who died of sporadic ALS are toxic to motor neurons in culture (see ARF related news story on Haidet-Phillips et al., 2011). A model like Maragakis’ may prove useful in testing astrocyte-directed therapies, Cashman said.—Amber Dance.
Papadeas ST, Kraig SE, O’Banion C, Lepore AC, Maragakis NJ. Astrocytes carrying the superoxide dismutase 1 (SOD1G93A) mutation induce wild-type motor neuron degeneration in vivo. Proc Natl Acad Sci U S A. 2011. Abstract