On gene dosage balance in protein complexes: a comment on Semple JI, Vavouri T, Lehner B. A simple principle concerning the robustness of protein complex activity to changes in gene expression.

I have read with interest the paper by Semple et al. (2008) dealing with the phenotypic effects of protein under- and overexpression as a function of their belonging to protein complexes [1]. Semple et al. 2008 confirmed that, in yeast, genes that inhibit growth when underexpressed often encode subunits of protein complexes. This finding was valid for both core and peripheral subunits. They also reanalyzed overexpression data from an array of yeast strains each one overexpressing a single gene. This array covered 85% of all yeast genes [2]. Semple et al. (2008) find that genes leading to growth defects when over-expressed are not enriched amongst the core or peripheral subunits of protein complexes [1]. Accordingly, they propose a simple principle: "the overall activity of a protein complex is in general robust to an increase, but not to a decrease in the expression of its subunits". The verification of such a simple principle would be more than welcome: at least something simple in biology. With this note, I merely seek to point out places where existing incomplete data leaves untested some hypotheses. Thus, the comment that follows is not a rebuttal of Semple's work.The first point that attracts the attention in the paper by Semple et al. is the fact that complete gene deletions (when dealing with essential genes or those whose absence leads to a growth defect) are taken as underexpression and not as plain absence of expression. This can be considered as a matter of words but only 'haploinsufficient genes' should qualify as truly underexpressed in their analysis.In the context of macromolecular complexes an alteration of the stoichiometric balance (i.e. relative amounts) between/among the subunits has been suggested, and often experimentally confirmed, to be harmful and to lead to fitness defects. These alterations result from underexpression (i.e. haploinsufficiency) or overexpression (i.e. in the case of trisomy) of a subunit [3,4]. The proposal of the existence