Referee 1's review:

This paper - would be of interest to the SOD1-ALS community as it provides further evidence for the role of apo-sod (wt and mutants) in the formation of aggregates and adds new data. On the other hand, it does not say much fundamentally new that we don't already know or believe from other reports in the last few years. The paper does primarily indicate that it is the apo-SOD1 structure (whether mutant in the case of fALS or native in the case of sporadic ALS) that seems to cause the aggregation problems. This is an interesting hypothesis but it is not entirely new. Interestingly the cited reference 16 is a PNAS paper from this year by more or less the same group which already talks about the same issue/mechanism (based on the native apo-SOD1). Having said that the paper is an important extension.

The importance of the metal-depleted enzyme has been discussed for some time as being a common factor in the role of SOD1 in ALS (e.g. Stathopolous et al PNAS 2003, 7021-26; Tiwari et al, J. Biol. Chem. 2005, 280, 29771-29779). Questions remain about the mechanism, e.g. whether mutations make the enzyme more prone to metal loss (or less prone to metal incorporation), and these are raised but not answered by the present study (page 9). By the way Figure 4 in both papers (the PLoSONE manuscript and ref.16 PNAS paper) look very much alike. Concluding from this finding it would mean that ALS is not at all related to the normal metal-loaded and properly functioning protein but to the immature protein that is still lacking metals i.e. it would not be a gain of function but an aberration/dysfunction on the way to maturity (although the authors mention that it may be one of several possible pathways to aggregation).

A recent contrary view (Roderiguez et al, 2005, 102, 10516-21) claims that SOD1-linked ALS cannot be simply explained by reduced global stability of the mutant apo-proteins (three exceptions mentioned were E100K, D101N, and N139K). These mutations were not part of the present study, but can this view be reconciled with the present study?

Also, were the de-metallated samples prepared at 37degC (the text implies this)? Would the rates of aggregation be any different if during sample preparation the samples were treated at a lower temperature than 37degC (for example, it was reported that apo-A4V is >95% soluble at 30degC but rapidly forms aggregates at 37degC,Leinweber et al Free Rad Biol & Med, 2004, 36, 911-910)?

I would have liked to see some comments on relationship between ease of aggregation and severity of disease.

The manuscript is clearly written, methods well described and referenced and could be understood by a general interest reader. It would have been easier to highlight the mutations in a ribbon model rather than in a table (it seems to be one of the few papers without any structure or ribbon model being shown).

**********
N.B. These are the comments made by the referee when reviewing an earlier version of this paper. Prior to publication the manuscript has been revised in light of these comments and to address other editorial requirements.