However, only 12 of these 85 have an NRSE motif in the human homolog

will refer to this amino acid as the DFGx residue to avoid confusion with the +1 amino acid position of the substrate peptide. Since the DFGx residue communicates with the phosphoacceptor, and the phosphoacceptor in turn influences +1 specificity, we hypothesized that the identity of the DFGx residue may be a determinant for +1P vs +1R specificity. This hypothesis is consistent with X-ray crystal structures of kinase peptide complexes, in which the +1 residue in the substrate is in close proximity to the DFGx residue . Our ancestral reconstructions indicate that the DFGx residue was a leucine in AncCMGI and then mutated to serine multiple times in the CMGC family. Further, the presence of leucine vs serine at DFGx in present-day kinases correlates with specificity for +1P vs +1R. Taken together, this suggests that Leu vs Ser at DFGx could affect kinase specificity at the +1 site. To test this hypothesis, we examined the effect of mutating Leu to Ser in mammalian ICK. This single mutation at DFGx made ICK approximately threefold more proline specific, such that its specificity resembled that of MOK. Conversely, MedChemExpress JW 55 Howard et al. eLife 2014;3:e04126. DOI: 10.7554/eLife.04126 9 of 22 Research Article Biochemistry Genomics and Evolutionary Biology Howard et al. eLife 2014;3:e04126. DOI: 10.7554/eLife.04126 10 of 22 Research Article Biochemistry Genomics and Evolutionary Biology mutating the DFGx residue in the opposite direction in MOK had the reverse effect of converting MOK from +1P preference to a non-selective kinase. We also mutated the DFGx residue in S. cerevisae Ime2 and in the ancestors AncLF4, AncICK, and AncCMGI. In all cases, we observed that mutation from leucine to serine at DFGx shifted kinase function towards +1P specificity, while mutation from serine to leucine at DFGx shifted the kinases towards +1R specificity. We note that though DFGx mutation was sufficient to substantially shift the +1 residue preference, other residues must also be important for +1 specificity, since proline to arginine selectivity shifts occur in the evolution of CMGC kinases without a DFGx mutation. In addition to affecting kinase selectivity at the +1 position, as anticipated mutation of the DFGx residue also impacted phosphoacceptor preference. However, the identity of the DFGx residue appeared to only modestly affect the phosphoacceptor specificity in comparison to specificity at the +1 position, and did not follow a clear pattern. These results are in keeping with previous observations that CMGC kinases are generally less specific than other kinase groups for the phosphoacceptor residue. We also examined phosphoacceptor preference in our resurrected CMGC ancestors. A general trend is observed in which AncCMGI has a slight preference for serine and this shifts towards PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19825579 a slight threonine preference in the lineage leading to Ime2. The S. cerevisiae meiotic phosphoregulatory network tolerates expansion of kinase specificity In S. cerevisiae, Ime2 is expressed exclusively during meiosis and is required for all stages of this process, including meiotic initiation, S-phase, the meiotic divisions, and gametogenesis. This meiotic exclusive expression allows us to engineer allelic replacements of IME2 without any impact on vegetative cells, and then assess the ability of strains to complete various aspects of the meiotic program. The Ime2 DFGx mutant shifts from a strong +1R preference to an expanded specificity that tolerates both proline and arginine at th