Abstract #1

Modeling the potential for complex genetic interactions: Actin displays a rich spectrum of complex haploinsufficiencies. Brian Haarer¹, Susan Viggiano¹, Olga Troyanskaya², David C. Amberg¹. 1) Dept Biochemistry & Molec Biol, SUNY Upstate Medical Univ, Syracuse, NY; 2) Department of Computer Science, Princeton University, Princeton, NJ.
   The yeast actin cytoskeleton was used to determine the potential for deleterious binary complex genetic interactions in a eukaryotic genome. Using an act1? allele, we screened for complex haploinsufficient interactions (CHI) with ~4800 non-essential genes and found that at least 200 show a deleterious CHI interaction. This set is enriched for genes with GO terms co-annotated with actin (P value=10-5) and contains genes for several known actin-binding proteins, including ANC1 the only gene previously known to be CHI with actin. This collection also includes many genes not previously connected to actin and many of unknown function. To determine whether these genes influence the actin cytoskeleton we stained null haploids with rhodamine-phalloidin and found that nearly 1/2 have defects in actin organization. The most common phenotype in these strains was excessive actin assembly suggesting that a CHI interaction with actin can result from a monomer crisis, a theory that was confirmed using the act1-159 allele that express an F-actin stabilizing mutant. CHI interactions with actin are frequently seen for multiple genes with shared functions. For example, many of the genes for the large ribosomal subunit are CHI with actin. However, nearly all of the RPL genes have paralogs and in almost all cases only one of the pair is CHI with actin. Invariably only the null strain for the CHI paralog has severe actin defects suggesting that these genes are not functionally equivalent. Using the alanine scan alleles of actin, we asked if CHI interactions can be recapitulated by reductions in subsets of actin functions. In many cases they can and in fact there is spatial congruence to this data suggesting that the loss of binding of specific actin binding proteins causes subsets of CHI interactions. Based on these data we will present the beginnings of the first genetically derived structure/function map of actin.

Return to YGM 2006 Home at SGD