The Berlin group, joint first authors Felix Meissner and Kaaweh Molawi, and senior author Arturo Zychlinsky, were following up on their previous results that wild-type SOD1, in response to pathogenic infection, activates caspase-1, which in turn ramps up production of IL-1β, inducing inflammation (Meissner et al., 2008). They wondered whether mutant SOD1 might also drive the same inflammatory pathway in an ALS model. In fact, they found out that mSOD1’s enzymatic activity is not required for inflammation in ALS model systems, but that the mSOD1 aggregates do lead to caspase-1 and IL-1β activation. “We implicate interleukin-1β directly in the pathogenesis of the disease,” Zychlinsky said.

The researchers cultured microglia from neonatal mouse brains and added purified SOD1. Wild-type SOD1 had little effect, but the G93A mutant protein induced caspase activity as well as cleavage and secretion of IL-1β. Zychlinsky and colleagues suspected, then, that ALS mice unable to activate this inflammatory pathway would have a less severe disease than mSOD1 mice usually do. To test this, they crossed SOD1-G93A mice with animals that were deficient for either caspase-1 or IL-1β. The double-mutant progeny suffered motor neuron symptoms, which appeared at the same age as single mutant SOD1 animals, but their disease progressed more slowly. SOD1-G93A mice survived a maximum of 160 days, but the double-mutants lived for up to 175 days, a significant lifespan extension in this particular model.

The result is surprising, according to Serge Rivest of McGill University in Montréal, Québec. Researchers in his laboratory found that knocking out IL-1β had no effect on disease progression in mSOD1/ mice (Nguyen et al., 2001; Kang and Rivest, 2007). Both Rivest and Zychlinksy noted that the two labs used different mSOD1 models, which may account for the discrepancy. Rivest’s group used SOD1-G87R mice, which have a slower progression, less inflammation, and more neurodegeneration, and are more similar to human ALS cases, Rivest wrote in an e-mail to ARF.

Zychlinsky and colleagues also examined how cells deal with mSOD1 aggregates. They collected bone marrow macrophages from mice and treated them with fluorescently tagged SOD1 as well as fluorescent dextran, which labels the endolysosomes where autophagy occurs. Mutant SOD1 colocalized with the dextran, so they suspected the misfolded dismutase might wind up in an autophagosome. When they treated the cells with the drug 3MA, which blocks the lysosomal degradation pathway, they found that more mSOD1 appeared in the cytosol and cells secreted more IL-1β, suggesting autophagy normally protects cells from inflammation by cleaning up SOD1 aggregates.

These results dovetail with those of the other research group, from the Università degli Studi di Milano in Italy, reported in Human Molecular Genetics. These scientists, led by first author Valeria Crippa and senior author Angelo Poletti, were interested in how the cell deals with aggregated SOD1. Mutant SOD1 inhibits the proteasome, so cells likely need another mechanism to destroy this damaging protein. The researchers knew that the small heat shock protein HspB8 promoted autophagy of mutant huntingtin in a model of Huntington disease (Carra et al., 2005) and suspected it might similarly dispatch mSOD1 in ALS.

Crippa and colleagues also experimented with SOD1-G93A mice, and using RT-PCR, they found that HspB8 is overexpressed in the spinal cord motor neurons of these animals compared to those carrying wild-type SOD1. And when the researchers overexpressed HspB8 in cultured NSC34 motor neurons also expressing mutant SOD1, they found that it increased clearance of the mutant dismutase.

From these data, the group suspected that HspB8 induced autophagy to eradicate mSOD1 aggregates. The key proof, Poletti said, came from co-immunoprecipitation (co-IP) experiments showing that mSOD1 and HspB8 bind each other and complex with known autophagy proteins. In cells expressing mSOD1 under normal conditions, the mutant protein did not co-immunoprecipitate with the others. But the scientists reasoned that if HspB8 was binding SOD1 to destroy it, the interaction might not last long enough to see the complex. Therefore, they treated their cultures with 3MA to halt completion of autophagy. Under these conditions, mSOD1 and HspB8 did co-immunoprecipitate.

“I think this heat shock protein may recognize many different proteins that are misfolded,” Poletti said, so it might be relevant to other neurodegenerative diseases as well. However, since HspB8 cannot ultimately save motor neurons from mSOD1 in animals or people with SOD1 mutations, the autophagy might come too late, he suggested. Therapeutics that increase HspB8 expression early in disease might be useful, the authors suggested.

Therapies aimed at blocking IL-1β could also be an avenue worth exploring, Zychlinsky suggested. However, he noted that there is as of yet no data from humans suggesting such a treatment would be worthwhile. Rivest was skeptical of this approach. “All clinical trials using any kind of anti-inflammatory treatments have failed terribly so far,” he wrote. “Clearly, this is not the direction to go.”—Amber Dance.

References:
Crippa V, Sau D, Rusmini P, Boncoraglio A, Onesto E, Bolzoni E, Galbiati M, Fontana E, Marino M, Carra S, Bendotti C, De Biasi S, Poletti A. The Small Heat Shock Protein B8 (HspB8) promotes autophagic removal of misfolded proteins involved in Amyotrophic Lateral Sclerosis (ALS). Hum Mol Genet. 2010 Jun 22. Abstract

Meissner F, Molawi K, Zychlinsky A. Mutant superoxide dismutase 1-induced IL-1beta accelerates ALS pathogenesis. Proc Natl Acad Sci U S A. 2010 Jul 20;107(29):13046-50. Abstract