The new results help account for previous observations that SOD1 mutants do not have to be expressed directly in neurons to cause cell death. Originally, researchers thought the toxic effects of SOD1 mutations might be due to the propensity of the misfolded protein to aggregate in neurons, as in several other neurodegenerative diseases. But work by Donald Cleveland and colleagues at the University of California at San Diego pointed to a different mechanism (see ARF related news story). Two years ago, those researchers showed that expression of SOD1 mutants in the non-neuronal cells of mouse spinal cord was sufficient to cause motor neuron death, even if the neurons themselves contained normal SOD1. These elegant studies established the importance of the neuron milieu for cell death. Now, Julien and coworkers have stepped up to provide a possible explanation for how toxicity can spread from cell to cell.

In the current work, first author Makoto Urushitani and coworkers identified the chromogranins A and B as binding partners for the mutant, but not wild-type, SOD1 protein. Chromogranin B, a major constituent of some secretory vesicles in neurons and endocrine cells, bound the mutant via an internal HSP70-like domain that presumably recognizes misfolded SOD1 protein. The proteins were found to be associated in spinal cord extracts from transgenic mice expressing mutated SOD1. In an extensive series of experiments using confocal microscopy, cell fractionation, and immunoelectron microscopy, the investigators established that mutant SOD1, but not wild type, colocalized with chromogranins in the secretory trans-Golgi network in cells.

The functional effect of the chromogranin-SOD1 association seems to be enhanced secretion of the misfolded proteins. Expression of SOD1 proteins in COS cells, which do not normally express chromogranins, resulted in a constitutive secretion of both SOD1 mutant and wild-type proteins. Coexpression of chromogranins increased the output of mutant, but not wild-type protein. In a more physiological setting, embryonic spinal cord cultures from SOD1 transgenic mice showed constitutive and stimulated secretion of SOD1 proteins.

What happens when SOD1 gets outside the cell? In-vitro experiments showed that recombinant mutant SOD1 could activate a microglia cell line to express immune mediators including TNA-α, COX2, and iNOS. Recombinant wild-type SOD1 had the opposite effect, causing suppression of microglial activation. Treatment of embryonic spinal cord cultures with mutant SOD1 (2 microgram/ml) not only increased the number of active microglia, but also killed nearly all the motor neurons. Motor neuron toxicity was not mediated solely by the microglia in those cultures, since specifically poisoning the microglia did not rescue the motor neurons.

The function of chromogranins in neurons is unknown, but chromogranin A has been linked previously to microglial activation, via production of its N-terminal bioactive peptide, vasostatin. Consistent with this suggestion, Urushitani et al. showed an elevated expression of chromogranin A in reactive astrocytes in the SOD1 mutant mice. This is intriguing in light of previous work showing an altered staining pattern for chromogranin A in people with sporadic ALS (Schiffer et al., 1995).

The roles of chromogranins and mutant SOD1 secretion in stimulating gliosis and motor neuron death are an endorsement of the idea that the ALS disease process is not autonomous to motor neurons. Some of the Julien group’s data suggest that interneurons could be a major source of extracellular SOD1, but further work will be needed to determine the roles of different cell types and inflammatory mediators in disease progression, as well as the exact mechanism of SOD1 toxicity. Given that chromogranin A has been found in neuritic plaques in Alzheimer disease (Marksteiner et al., 2000) and in prion deposits of Creutzfeldt-Jacob disease (Rangon et al., 2003), one question going forward is whether chromogranins mediate secretion of abnormal proteins in these and other neurodegenerative diseases.—Pat McCaffrey.

Reference:
Urushitani M, Sik A, Sakurai T, Nukina N, Takahashi R, Julien J. Chromogranin-mediated secretion of mutant superoxide dismutase proteins linked to amyotrophic lateral sclerosis. Nat Neurosci. 2006 Jan;9(1):108-18. Epub 2005 Dec 20. Abstract