Inherited mutations in SOD1—of which more than 100 varieties have been reported—are responsible for approximately 20 percent of familial ALS cases, or 2 percent of ALS cases overall. Rodents overexpressing mutant human SOD1 are a common lab model for the disease, and both the rodents and people have aggregated SOD1 in their motor neurons. Scientists suspected the aggregation was a late-stage process, but the current work confirms it, Borchelt said: “What we have done is basically tied the package up.”

Borchelt, first author Celeste Karch, and colleagues followed three mouse lines, expressing mutant SOD1-G93A, G37R, or H46R/H48Q. Karch harvested spinal cords from animals at time points ranging from 30 days to end-stage, when animals are paralyzed, and measured the amount of detergent-insoluble, aggregated SOD1. Pathological features such as Golgi apparatus degradation and astrogliosis preceded the appearance of SOD1 clusters, confirming that the animals were already getting sick when SOD1 started to aggregate. The authors did not investigate other forms of the protein, such as soluble SOD1 clusters, which could still play a role in disease onset. In AD, for example, soluble, oligomeric Aβ attacks synapses (see ARF related news story and Berman et al., 2008) and is toxic to dendritic spines (see ARF related news story and Koffie et al., 2009).

The researchers also used their experiments to examine the link between aberrant disulfide bond formation and aggregation. SOD1 contains four cysteines, and in its mature, dimerized form each monomer contains an intramolecular disulfide bridge. It has been suggested that deviant, intermolecular disulfide bonds between SOD molecules can seed oligomerization (see ARF related news story and Banci et al., 2009). However, human SOD1 mutations that remove one or the other cysteine involved in the intermolecular bonding have been associated with familial ALS, and in previous work, Karch and Borchelt found that cysteine-free mutants could still aggregate (Karch and Borchelt, 2008). In the current study, the authors found that even when they used the reducing agent β-mercaptoethanol, which destroys disulfide bridges, when extracting tissue from G93A mice, the SOD1 aggregates remained. Borchelt suggests that aberrant disulfide bonding is concomitant with aggregation, but not required. “We cannot convince ourselves that there is cause and effect,” he said.

Karch did find that the majority of insoluble SOD1 was reduced, implying that it never formed, or lost, the normal disulfide bond. This was the case both in end-stage mice and in cultured human embryonic kidney cells overexpressing A4V, G37R, and G93A mutant SOD1. The mutant protein may be less likely to form the correct disulfide bridge, the authors suggest, remaining in an immature, aggregation-prone state.

The work rules out a role for SOD1 aggregates in causing disease, but does not clarify what they actually do. The aggregates could merely be markers for diseased cells that cannot maintain protein homeostasis, or they could accelerate progression of the illness when they appear. “I’m guessing they make it worse,” said Joan Valentine of UCLA, who has collaborated with Borchelt and communicated the current paper to PNAS.

Understanding how ALS progresses could be more important that comprehending how it starts, Valentine said, because disease progression is a better clinical target. Armed with the details of progression, “we could stop it in its tracks,” she said.

Borchelt’s next step will be to discover what, if anything, the aggregates do to damage cells. “We won’t know what aggregation does in disease until we are able to stop it and see if it makes a difference,” he said. To that end, Borchelt is searching for compounds that will block aggregation.—Amber Dance.

Reference:
Karch CM, Prudencio M, Winkler DD, Hart PJ, Borchelt DR. Role of mutant SOD1 disulfide oxidation and aggregation in the pathogenesis of familial ALS. PNAS Early Edition. Abstract