The dictyostelium kinome--analysis of the protein kinases from a simple model organism

Abstract

Dictyostelium discoideum is a widely studied model organism with both unicellular and multicellular forms in its developmental cycle. The Dictyostelium genome encodes 285 predicted protein kinases, similar to the count of the much more advanced Drosophila. It contains members of most kinase classes shared by fungi and metazoans, as well as many previously thought to be metazoan specific, indicating that they have been secondarily lost from the fungal lineage. This includes the entire tyrosine kinase-like (TKL) group, which is expanded in Dictyostelium and includes several novel receptor kinases. Dictyostelium lacks tyrosine kinase group kinases, and most tyrosine phosphorylation appears to be mediated by TKL kinases. About half of Dictyostelium kinases occur in subfamilies not present in yeast or metazoa, suggesting that protein kinases have played key roles in the adaptation of Dictyostelium to its habitat. This study offers insights into kinase evolution and provides a focus for signaling analysis in this system.

Figure 1. Tree of the Dictyostelium Kinome

A tree of 248 Dictyostelium ePK domains is presented. Pseudogenes, Chromosome 2 duplications, and sequences with very divergent ePK domains were omitted. The N- and C-terminus domains of dual-domain kinases, respectively designated by an “-a” or “-b” extension, were analyzed independently. Group and family names are shown in bold type, subfamily names in plain type, and Dictyostelium-specific families and subfamilies in italic. Selected protein names are shown in red. Branch lengths reflect relative distances between ePK domains. A branching order could not be assigned in the region indicated by the small gray circle because of the diversity of the sequences.

Figure 2. Group-Level Comparison of the Dictyostelium and Other Kinomes

Pie charts depict the proportion of the kinome that is devoted to the major groups in Dictyostelium, yeast, worm, flies, and humans. The size of the pie is proportional to the number of kinase domains in each organism, and the total number of kinases in each group is shown in the slice. Data for other organisms are from KinBase (http://kinase.com). The portion of each group that is in families or subfamilies not found in the other four kinomes is indicated with a black arc drawn outside the slice for that group.

Figure 3. Summary of Kinase Subfamily Invention and Loss

Comparison of Dictyostelium with four other kinomes suggests that 75 distinct subfamilies existed in their common ancestor, and that new subfamilies were born (positive numbers) and lost (negative numbers) in most lineages. Numbers in parenthesis indicate “unique” kinases within each lineage that may be classified as novel subfamilies when more kinomes are analyzed. Most notably, S. cerevisiae has lost 24 subfamilies present in the common ancestor, while metazoans invented an additional 80 conserved subfamilies.

Figure 4. Receptor Kinases

The 17 Dictyostelium kinases that are strongly predicted to have TMDs are depicted. The drawings are approximately to scale.

Figure 5. Dual-Domain Kinases

The Dictyostelium dual-domain kinases are drawn to scale. The kinase domains are shown in blue, broken by a dashed line in instances where there is a large insert in the kinase domains. The classification (group/family/subfamily) of each kinase domain is indicated above each domain. An X through the kinase domain indicates that it is predicted to be catalytically inactive.

Figure 6. Protein Kinase Pseudogenes

The genomic loci for putative pseudogenes, together with their most likely parental genes, are shown. Pseudogene exons are shaded yellow, and parental gene exons are shaded orange. Stop codons are indicated with an asterisk, and frameshifts by “fs.” The percentage identity in amino acid sequence between pairs of sequences is shown to the right of the drawings. The number of ESTs, as reported at http://www.dictybase.org, is shown. The results of RT-PCR experiments are shown. +, a product was obtained; −, no product was obtained; ND, not determined. All are drawn to the scale shown in the bottom right of the figure, except Gdt3_ps, which has a condensed scale. For DDB0187382_ps and DDB0204419_ps, no putative parental gene could be identified.