Unraveling signaling mysteries by expression profiling activated kinases

Abstract

Signal transduction is indispensible for control of cell growth, differentiation, metabolism, and migration. Key regulators in signaling are protein kinases, a major superfamily of enzymes containing a kinase domain with a high degree of conservation across eukaryotes. Improper functioning of a protein kinase is commonly observed in diseases, especially cancer. Under normal conditions, they are capable of regulating the activity of proteins by attaching a phosphate group, mostly originating from adenosine triphosphate (ATP), to an amino acid with a free hydroxyl group; serine, threonine, or tyrosine. The simplest model organism to study eukaryotic protein kinases (ePKs) in is yeast, Saccharomyces cerevisiae (S. cerevisiae). It has a relatively small number of ePKs, namely 139, and is easy to manipulate and to keep under laboratory circumstances. Expression profiling by microarray can be used to study ePKs. This technique measures the relative abundance of mRNA. The method allows for clarification of the biological process in which an ePK is involved and of the mutual relationship among ePKs via clustering. A common strategy is to compare the transcriptome of a deletion mutant of a gene of interest with mRNA levels of wild type yeast. Expression profiling has been shown to be a useful tool when it comes to structure-function analysis of a large multisubunit complex, genetic epistasis with expression profiles as phenotype, and revealing the effects of regulatory kinase activity. Single deletions could not obtain an altered expression profile compared to wild type yeast for the majority of ePKs. Other attempts, like studying essential protein kinases with DAmP mutants, were also not sufficient to obtain specific expression signatures for all studied genes. A DAmP mutant has an antibiotics cassette in the 3 prime untranslated region of a specific gene, resulting in destabilization of its mRNA. Only two out of 20 DAmP mutants gave a specific expression signature. These mutants had around fourfold decrease of mRNA level of the gene of interest. All the other mutants had about twofold decrease of mRNA level of the gene of interest. Further diminishment of mRNA levels of the essential protein kinases with an additional sequence could lead to more phenotypes. Still a group of 111 out of 139 protein kinases are without specific expression signature, due to either the essentiality of the ePK or the lack of a phenotype. A quarter of this remaining group could give a phenotype upon constitutively activating the protein kinase by mutating a specific amino acid. This is based on alignments of the reported constitutively active point mutants with the pool of 111 protein kinases. As a query sequence the residue of interest was taken and 50 amino acids on either site. It is not necessary that the mutated protein kinase is fully active. Even a slight activation can rescue a certain phenotype, which indicates that it is sufficient to obtain a specific expression signature. Combining constitutively active protein kinases by mutating a conserved residue with expression profiling will lead to more knowledge about signal transduction by ePKs.