Novel Molecular Signatures in the PIP4K/PIP5K Family of Proteins Specific for Different Isozymes and Subfamilies Provide Important Insights into the Evolutionary Divergence of this Protein Family

Abstract

Members of the PIP4K/PIP5K family of proteins, which generate the highly important secondary messenger phosphatidylinositol-4,5-bisphosphate, play central roles in regulating diverse signaling pathways. In eukaryotic organisms, multiple isozymes and subfamilies of PIP4K/PIP5K proteins are found and it is of much interest to understand their evolution and species distribution and what unique molecular and biochemical characteristics distinguish specific isozymes and subfamilies of proteins. We report here the species distribution of different PIP4K/PIP5K family of proteins in eukaryotic organisms and phylogenetic analysis based on their protein sequences. Our results indicate that the distinct homologs of both PIP4K and PIP5K are found in different organisms belonging to the Holozoa clade of eukaryotes, which comprises of various metazoan phyla as well as their close unicellular relatives Choanoflagellates and Filasterea. In contrast, the deeper-branching eukaryotic lineages, as well as plants and fungi, contain only a single homolog of the PIP4K/PIP5K proteins. In parallel, our comparative analyses of PIP4K/PIP5K protein sequences have identified six highly-specific molecular markers consisting of conserved signature indels (CSIs) that are uniquely shared by either the PIP4K or PIP5K proteins, or both, or specific subfamilies of these proteins. Of these molecular markers, 2 CSIs are distinctive characteristics of all PIP4K homologs, 1 CSI distinguishes the PIP4K and PIP5K homologs from the Holozoa clade of species from the ancestral form of PIP4K/PIP5K found in deeper-branching eukaryotic lineages. The remaining three CSIs are specific for the PIP5Kα, PIP5Kβ, and PIP4Kγ subfamilies of proteins from vertebrate species. These molecular markers provide important means for distinguishing different PIP4K/PIP5K isozymes as well as some of their subfamilies. In addition, the distribution patterns of these markers in different isozymes provide important insights into the evolutionary divergence of PIP4K/PIP5K proteins. Our results support the view that the Holozoa clade of eukaryotic organisms shared a common ancestor exclusive of the other eukaryotic lineages and that the initial gene duplication event leading to the divergence of distinct types of PIP4K and PIP5K homologs occurred in a common ancestor of this clade. Based on the results gleaned from different studies presented here, a model for the evolutionary divergence of the PIP4K/PIP5K family of proteins is presented.

Keywords: conserved signature indels; evolution of the PIP4K/PIP5K family of proteins; holozoa clade of eukaryotic organisms; molecular signatures for the pip4k/pip5k isozymes and isoforms; phosphatidylinositol phosphate kinases; phylogenetic analysis; protein evolution; species distribution of pip4k/pip5k proteins.

Figure 1

A maximum-likelihood phylogenetic tree of the PIP4K/PIP5K family of proteins based on the core conserved kinase domain region of the protein sequences from representative species. The accession numbers of the protein sequences that were utilized are provided in Table S1. Bootstrap support values > 50% are shown at the nodes.

Figure 2

Excerpts from the sequence alignment of PIP4K and PIP5K homologs showing a 1 aa insert (boxed) in a conserved region that is uniquely shared by all PIP4K homologs. This insert is commonly shared by all PIP4K homologs from metazoan phyla including the Choanoflagellates and Filasterea but it is not found in any PIP5K homologs or the single orthologs of PIP4K/PIP5K found in the deeper-branching eukaryotic organisms. Detailed information regarding the species distribution of this conserved signature indels provided in Figure S2. The dashes (-) in the alignment indicates identity with the amino acids on the top line. Numbers on the top indicate the location of this sequence region in the protein from Saccharomyces cerevisiae. The accession numbers of various sequences are given in the second column. Secondary structure information for this protein region is presented on top of the sequence.

Figure 3

Partial sequence alignment of the PIP4K/PIP5K family of proteins showing a 2 aa deletion in a conserved region (boxed) that is uniquely shared by all PIP4K homologs. The boxed CSI is not present in any of the PIP5K homologs as well as the PIP4K/PIP5K orthologs from plants and other deep-branching eukaryotic lineages. The PIP4K/PIP5Korthologs from fungi contain a shorter 1 aa deletion in this position. More detailed sequence information for this CSI is provided in Figure S3. Other details are the same as in Figure 2 legend. Numbers on the top indicate the position of this sequence in the human PIP5Kα.

Figure 4

Partial sequence alignment of PIP4K/PIP5K family of proteins showing 1 aa deletion (boxed) in a conserved region that is commonly shared by different PIP4K and PIP5K homologs but lacking in the single copy PIP4K/PIP5K orthologs from deeper branching eukaryotic phyla including plants and fungi. This 1 aa CSI distinguishes the distinct PIP4K and PIP5K homologs found in Holozoa species (i.e., all multicellular metazoan species and their unicellular relatives Choanoflagellates and Filasterea) from early branching eukaryotic organisms harboring only a single ortholog of the PIP4K/PIP5K protein. The PIP4K from some nematode species lack this deletion or contain a longer insertion (indicated by *) in this position and its possible significance is unclear. Numbers on the top indicate the position in human PIP5Kα. More detailed information for this CSI is provided in Figure S4.

Figure 5

Partial sequence alignment of different subfamilies of the PIP5K protein showing 1 aa CSI (boxed) that is uniquely shared by the PIP5Kα subfamily of proteins. More detailed information regarding species distribution of this CSI is provided in Fig. S5. The predicted secondary structure of this sequence region is shown on top of the sequence alignment. Other details are the same as in the Figure 2 legend.

Figure 6

Excerpts from the multiple sequence alignment of PIP5K homologs showing a 2 aa insert in a conserved region (boxed) that is uniquely shared by the PIP5Kβ homologs of mammals, birds, and reptiles, but absent from all other PIP4K and PIP5K homologs. Sequence information is shown here for only representative species from different vertebrate classes. More detailed information for this CSI is provided in Figure S6. Other details are as in Figure 2 legend. Numbers on the top indicate the position in the human PIP5Kβ.

Figure 7

Partial sequence alignment of different isoforms of the PIP4K proteins showing a conserved region where 1–4 aa deletions (boxed) are uniquely found in the PIP4Kγ subfamily of proteins from different classes of vertebrates. The PIP4Kγ protein mammals and reptiles have a 3 aa deletion in this position, whereas those from the birds contain a 4 aa deletion in the same position. The fish and amphibians are found to contain multiple copies of PIP4Kγ with 1–2 aa deletion in this position. More detailed information for this CSI is provided in Figure S7. Other details are the same as in Figure 2 legend. Numbers on the top indicate the position of the sequence in the human PIP4Kγ.

Figure 8

Surface representation of the identified CSIs in a structural model of the human PIP5Kβ protein. For mapping of the CSIs in protein structures, structural information for a number of solved/modeled structures for the PIP4K/PIP5K family of proteins (see Methods section) was utilized. The CSIs which constitute inserts are marked in red on the surface, while for the CSIs that are deletions, the protein regions where these deletions are found are marked in blue on the surface. The location of the 1 aa deletion (Figure 4) that is commonly shared by different homologs of PIP4K and PIP5K is shown in magenta based on structural comparison with the Saccharomyces cerevisiae PIP4K/PIP5K homolog. The close-up views of the locations in the protein structure for different identified CSIs are shown in cartoon representation. The structure model of PIP4Kγ isoform from H. sapiens is shown in yellow and crystal structure of PIP4Kβ isoform is shown in green.

Figure 9

A summary diagram showing the evolutionary divergence of different members of the PIP4K/PIP5K family of proteins in eukaryotic organisms. The model presented here is based on the species distribution of different proteins as well as the species/isozyme specificities of different CSIs in these proteins that were identified in the present work. The arrows mark the evolutionary stages where the rare genetic changes leading to the specific CSI(s) are postulated to have occurred. The red stars mark the evolutionary stages where gene duplication events have occurred during the divergence of this protein family. Holozoa clade comprises of the multicellular metazoan phyla as well as phyla consisting of their unicellular metazoan common ancestor (UMCA).