Phylogenetic analysis of CDK and cyclin proteins in premetazoan lineages

Abstract

Background: The molecular history of animal evolution from single-celled ancestors remains a major question in biology, and little is known regarding the evolution of cell cycle regulation during animal emergence. In this study, we conducted a comprehensive evolutionary analysis of CDK and cyclin proteins in metazoans and their unicellular relatives.

Results: Our analysis divided the CDK family into eight subfamilies. Seven subfamilies (CDK1/2/3, CDK5, CDK7, CDK 20, CDK8/19, CDK9, and CDK10/11) are conserved in metazoans and fungi, with the remaining subfamily, CDK4/6, found only in eumetazoans. With respect to cyclins, cyclin C, H, L, Y subfamilies, and cyclin K and T as a whole subfamily, are generally conserved in animal, fungi, and amoeba Dictyostelium discoideum. In contrast, cyclin subfamilies B, A, E, and D, which are cell cycle-related, have distinct evolutionary histories. The cyclin B subfamily is generally conserved in D. discoideum, fungi, and animals, whereas cyclin A and E subfamilies are both present in animals and their unicellular relatives such as choanoflagellate Monosiga brevicollis and filasterean Capsaspora owczarzaki, but are absent in fungi and D. discoideum. Although absent in fungi and D. discoideum, cyclin D subfamily orthologs can be found in the early-emerging, non-opisthokont apusozoan Thecamonas trahens. Within opisthokonta, the cyclin D subfamily is conserved only in eumetazoans, and is absent in fungi, choanoflagellates, and the basal metazoan Amphimedon queenslandica.

Conclusions: Our data indicate that the CDK4/6 subfamily and eumetazoans emerged simultaneously, with the evolutionary conservation of the cyclin D subfamily also tightly linked with eumetazoan appearance. Establishment of the CDK4/6-cyclin D complex may have been the key step in the evolution of cell cycle control during eumetazoan emergence.

Figure 1

Phylogenetic tree from analysis of CDK family proteins in Homo sapiens, Nematostella vectensis, Thecamonas adhaerens, Amphimedon queenslandica, Monosiga brevicollis, and Salpingoeca rosetta. Maximum likelihood (ML) and Bayesian analyses were conducted using RAxML and PHYLOBAYES 3.3, respectively. Both methods produced trees with nearly identical topologies. The first numbers above branches indicate Bayesian posterior probabilities (only key branches are labeled), and the second numbers above branches indicate ML bootstrap percentages. The scale bar shows the number of substitutions per site. Sequences of Hsa-GSK3alpha, Hsa-MAK, and Hsa-HCDKL1 were used as outgroups. All proteins are labeled with species names followed by accession numbers. Species abbreviations are as follows: Hsa, H. sapiens; Nve, N. vectensis; Tad, T. adhaerens; Aqe, A. queenslandica; Mbr, M. brevicollis. The alignment used for this analysis is found in Additional file 1: File S1.

Figure 2

Schematic representation of the distribution of different CDK subfamilies in eukaryotic organisms. The results of phylogenetic analyses of CDK family proteins in different organisms are summarized. A black dot indicates the presence of clear homologs of CDK subfamilies or clades (see text for further explanations). Phylogenetic relationships of these organisms are based on recent reports [43,61,62] and the results of the Origins of Multicellularity project [10]. Detailed information regarding this figure, including CDK protein accession numbers, is given in Additional file 6: Table S1.

Figure 3

Tree derived from phylogenetic analysis of cyclin family proteins in H. sapiens, N. vectensis, T. adhaerens, A. queenslandica, M. brevicollis, and S. rosetta. Maximum likelihood (ML) and Bayesian analyses were conducted using RAxML and PHYLOBAYES 3.3, respectively. Both methods produced trees with nearly identical topologies. The first numbers above branches indicate Bayesian posterior probabilities (only key branches are labeled), and the second numbers above branches indicate ML bootstrap percentages. The scale bar shows the number of substitutions per site. Sequences of Hsa-Cables1 and Hsa-Cables2 were used as outgroups. All proteins are labeled with their accession numbers preceded by their species names. Species abbreviations are as follows: Hsa, H. sapiens; Nve, N. vectensis; Tad, T. adhaerens; Aqe, A. queenslandica; MBr, M. brevicollis. The alignment used for this analysis is found in Additional file 1: File S3.

Figure 4

Schematic representation of the distribution of different cyclin subfamilies in eukaryotic organisms. The results of phylogenetic analyses of cyclin family proteins in different organisms are summarized. A black dot indicates the presence of clear homologs of cyclin subfamily members (see text for further explanations). Phylogenetic relationships illustrated for these organisms are derived form a proteome-based phylogeny [43,61,62] and the results of the Origins of Multicellularity project [10]. The names of cyclin B-like (cyclins B, A, D, E, J, F, G, I, O, CLB, and CLN); cyclin Y-like (cyclins Y and PCL), and cyclin C-like (cyclins C, H, L, K, T, and Fam58) group members are indicated by different colors. Detailed information regarding this figure, including cyclin protein accession numbers, may be found in Additional file 12: Table S2.

Figure 5

Alignments of cyclin B and cyclin A proteins. A. Alignment of cyclin B proteins from representative organisms. The region of Cyclin_N domain was underlined. Protein accession numbers are as follows: Hsa-cyclin B1: gi:14327896 from H. sapiens; Tad-cyclin B: gi:196002535 from T. adhaerens; Sro-cyclin B: gi:326428978 from S. rosetta; Cow-cyclin B: gi:320166256 from C. owczarzaki; Ttr-cyclin B: AMSG_03352 from T. trahens; Ddi-cyclin B: gi:66819865 from D. discoideum. B. Alignment of cyclin A proteins from representative organisms. The region of Cyclin_N domain was underlined. Protein accession numbers are as follows: Hsa-cyclin A1:gi:4502611 from H. sapiens; Tad-cyclin A: gi:196005765 from T. adhaerens; Mbr-cyclin A: gi:167517989 from M. brevicollis; Sro-cyclin A: gi:326426811 from S. rosetta; Cow-cyclin A: gi:320169862 from C. owczarzaki.

Figure 6

Alignments of cyclin E and cyclin D proteins. A. Alignment of cyclin E proteins from representative organisms. The region of Cyclin_N domain was underlined. Cyclin accession numbers are as follows: Hsa-cyclin E1: gi:17318559 from H. sapiens; Tad-cyclin E:gi:196003236 from T. adhaerens; Sro-cyclin E: gi:326437558 from S. rosetta; Cow-cyclin E:gi:320167008 from C.owczarzaki; Ttr-cyclin E: AMSG_07694 from T. trahens. B. Alignment of cyclin D proteins from representative organisms. The region of Cyclin_N domain was underlined. Cyclin accession numbers are as follows: Hsa-cyclin D1: gi|16950655 from H. sapiens; Nve-cyclin D: gi:156350442 from N. vectensis; Tad-cyclin D: gi:196001479 from T. adhaerens; Ttr-cyclin D: AMSG_02061 from T. trahens.

Figure 7

Schematic scenarios of CDK and cyclin protein function in cell cycle regulation of different representative organisms. Schemes for organisms S. cerevisiae and H. sapiens were drawn based on previous reports [6,37,67,70], and schemes for M. brevicollis, A. queenslandica, and T. adhaerens were drawn based on inferences derived from our evolutionary analysis (see text for further explanations). Accession numbers of CDK and cyclin proteins in the figure are as follows: Sce-CDK1: gi:6319636; Sce-Cln1/2: gi:6323855, gi:6324999; Sce-Clb5/6: gi:6325377, gi:6321546; Sce-Clb1/2/3/4: gi:6321545, gi:6325376, gi:6320046, gi:6323239; Mbr-CdK1: gi:167517533; Mbr-cyclin B: gi:167523717, gi:167524669; Mbr-cyclin E: gi:167519314; Mbr-cyclin A: gi:167517989; Aqe-CDK1: gi:340381019, Aqe-CdK2: gi:340379293; Aqe-cyclin B: gi:340376468, gi:340374274; Aqe-cyclin E: gi:340379787; Tad- CDK1: gi:196003954; Tad-CDK2: 196013348; Tad-CDK4: gi:195999760; Tad-cyclin B: gi:196002535; gi:196003740; Tad-cyclin E: gi:196003236; Tad-cyclin A: gi:196005765; Tad-cyclin D: gi:196001479.