- Research article
- Open Access
Phylogenetic diversification of glycogen synthase kinase 3/SHAGGY-like kinase genes in plants
BMC Plant Biology volume 6, Article number: 3 (2006)
The glycogen synthase kinase 3 (GSK3)/SHAGGY-like kinases (GSKs) are non-receptor serine/threonine protein kinases that are involved in a variety of biological processes. In contrast to the two members of the GSK3 family in mammals, plants appear to have a much larger set of divergent GSK genes. Plant GSKs are encoded by a multigene family; analysis of the Arabidopsis genome revealed the existence of 10 GSK genes that fall into four major groups. Here we characterized the structure of Arabidopsis and rice GSK genes and conducted the first broad phylogenetic analysis of the plant GSK gene family, covering a taxonomically diverse array of algal and land plant sequences.
We found that the structure of GSK genes is generally conserved in Arabidopsis and rice, although we documented examples of exon expansion and intron loss. Our phylogenetic analyses of 139 sequences revealed four major clades of GSK genes that correspond to the four subgroups initially recognized in Arabidopsis. ESTs from basal angiosperms were represented in all four major clades; GSK homologs from the basal angiosperm Persea americana (avocado) appeared in all four clades. Gymnosperm sequences occurred in clades I, III, and IV, and a sequence of the red alga Porphyra was sister to all green plant sequences.
Our results indicate that (1) the plant-specific GSK gene lineage was established early in the history of green plants, (2) plant GSKs began to diversify prior to the origin of extant seed plants, (3) three of the four major clades of GSKs present in Arabidopsis and rice were established early in the evolutionary history of extant seed plants, and (4) diversification into four major clades (as initially reported in Arabidopsis) occurred either just prior to the origin of the angiosperms or very early in angiosperm history.
The glycogen synthase kinase 3 (GSK3)/SHAGGY-like kinases are non-receptor serine/threonine protein kinases that are involved in a variety of signal transduction pathways . In animals, they are involved in cell fate determination, in metazoan pattern formation, and in tumorigenesis [2–6]. In mammals, two enzymes, GSK3α and GSK3β, are involved in the regulation of glycogen metabolism , in stability of the cytoskeleton , and in numerous processes related to oncogenesis . In Saccharomyces cerevisiae, the GSK3 homologs MCK1 and MDS1 play a role in chromosomal segregation , and in Schizosaccharomyces pombe the GSK3 homolog Skp1 regulates cytokinesis .
In contrast to the two members of the GSK3 family found in mammals, plants appear to have a much larger set of divergent GSK3/SHAGGY-like kinase genes [12–28], with functions as numerous as in animals. Genetic and biochemical approaches indicate that different plant GSKs are involved in diverse processes, including signaling, development, and stress response. For example, the Arabidopsis SHAGGY-like protein kinase AtGSK1 complements the salt-sensitive phenotype of yeast calcineurin mutants . In Medicago sativa, GSK3 (WIG) is activated by wounding . Arabidopsis AtSK11 and AtSK12 participate in the regulation of flower patterning at several developmental stages ; both genes are expressed during perianth and gynoecium development. Cloning of the BIN2 (brassinosteroid-insensitive 2) locus, which is identical to UCU1 (ULTRACURVATA1) and DWF12 (DWARF12), revealed that ASKη (AtSK21) is involved in brassinosteroid signaling [25–28]. However, in contrast to the known functions of GSK in animals, much less is known about the specific functions of these genes in plants.
Arabidopsis has ten different GSK genes [13, 15–17, 20, 21, 23]. The protein sequences of family members are highly conserved throughout the kinase domain. In contrast, the N- and C-terminal regions of the plant GSK genes are highly variable, consistent with observations that the various plant genes are involved in divergent biological processes. However, because the functional analyses of the plant GSK genes are based on mutant phenotypes or transcript expression levels [12–28], more precise analyses of mutant phenotypes without the N- and/or C-terminal regions are needed to determine whether the variable N- and C-terminal regions are related to the functional differences of plant GSK genes. Based on phylogenetic analyses of amino acid and cDNA sequences, Arabidopsis GSK genes have been grouped into four classes (I-IV) [13, 15–17, 21].
Besides Arabidopsis GSKs, GSK3/SHAGGY-like kinase genes have been reported from the angiosperms Oryza sativa, Brassica napus, Medicago sativa, Petunia hybrida, Nicotiana tabacum, and Ricinus communis [14, 15, 18, 19, 22, 23, 29, 30], all of which are highly derived monocot or eudicot species. No basal eudicots or basal angiosperm lineages, representing phylogenetically ancient groups, were included in any previous analyses. Furthermore, no phylogenetic analyses of plant GSK genes have included sequences from diverse green plant lineages. Thus, it is not clear when plant-specific GSK3/SHAGGY-like kinases diverged or what complement of GSK genes is present in basal angiosperms or indeed other land plants. Recently, the Floral Genome Project (FGP) research consortium  provided expressed sequence tag (EST) sequences of GSK genes for a number of basal angiosperms, including Amborella trichopoda and the water lily Nuphar advena . These taxa are phylogenetically important because they represent the earliest-diverging lineages of extant flowering plants [e.g., [33–42]].
In this study we examined the diversification of the GSK3/SHAGGY-like kinase genes in plants. Specifically, we (1) compared the structure of GSK3/SHAGGY-like kinase genes in Arabidopsis and rice, and (2) addressed whether the diversity of GSK genes in Arabidopsis is unique to Arabidopsis or is more generally true of all angiosperms and all land plants. For example, if the diversification of the gene family predated or coincided with the origin of the angiosperms, then ESTs from basal angiosperm taxa should appear in all major clades identified in Arabidopsis. Likewise, if GSK gene diversity in plants is ancient, basal lineages of land plants, such as mosses, should also contain orthologs to the Arabidopsis genes. Alternatively, some gene lineages may have diversified since the origin of the angiosperms, or land plants, and will not contain sequences from all basal lineages.
Results and discussion
Gene structure and patterns of sequence evolution
The structure of five Arabidopsis GSK3/SHAGGY-like kinase genes was reported by Dornelas et al. . We sought to obtain a more comprehensive view of the structure of these genes. To accomplish this, we used the complete genome sequences now available for Arabidopsis and rice [43, 44]; we describe the gene structure of additional GSKs from Arabidopsis, as well as the structure of GSKs reported from rice. We followed the numbering scheme of Dornelas et al.  for numbering exons and introns.
The structure of GSK genes in Arabidopsis and rice is highly conserved (Figure 1). This conservation of gene structure is also apparent by inspection of the aligned sequences across a diverse array of plants, including angiosperms, gymnosperms, a fern, a moss, and green and red algae .
Most of the GSK genes have 12 exons interrupted by 11 introns, but there are some exceptions. AtSK12 does not contain intron 6, and AtSK21 does not possess introns 3 and 11. As a result, these two genes have the smallest number of exons among the GSKs we examined. In addition, AtSK31 and AtSK32 have one additional exon (located between exons 1 and 2) compared to most other members of the GSK gene family. In our phylogenetic analyses, these two genes from Arabidopsis appear together in a clade with a sequence from Oryza (Os10g37740), which also has one additional exon similarly located between exons 1 and 2. These results suggest that the presence of this additional exon in Arabidopsis and rice was inherited from a common ancestor, prior to the divergence of monocots and eudicots, suggesting that the addition of this exon was an ancient event that occurred early in the diversification of flowering plants or possibly prior to the origin of flowering plants. It would be interesting to determine whether other sequences from clade III (see phylogenetic results below) similarly have an extra exon. Tichtinsky et al.  reported that PSK6.2 and PSK7 from Petunia hybrida also have an additional exon between exons 1 and 2. However, genomic sequences are not available for other members of clade III. Recent studies demonstrate that the structure of three GSK genes from the moss Physcomitrella patens is very similar to that of Arabidopsis and rice .
The structurally variable 5' region of plant GSKs is composed of exons 1 and 2, and the catalytic domain is encoded by exons 3–10 . The structurally variable 3' region typically comprises exons 11 and 12 (Figure 1).
The length of the GSK genes in Arabidopsis ranges from 2135 bp (AtSK12) to 3558 bp (AtSK22), whereas the length ranges from 2341 bp (Os05g04340) to 6186 bp (Os06g35530) in rice. The large variation of gene length in rice is due to the presence of long introns (up to 2173 bp in Os06g35530) in some genes.
We investigated the patterns of nucleotide substitution across 116 plant-specific GSK homologs. This comparison provides a minimum estimate of change in a 4-position window. The substitution pattern of plant GSK homologs varied across the nucleotide sequences (Figure 2). The most variable 4-nucleotide window occurs at positions 945–948, with 70 substitutions in this interval. The substitution pattern of plant-specific GSK homologs when analyzed across amino acid sequences revealed a pattern similar to that found for nucleotide sequences. Variable regions are spread across the protein, but the most highly variable regions occur at amino acid positions 121–124 and 317–320 (Figure 3). The latter region (corresponding to the variable region of exon 12) accumulated 147 amino acid substitutions over an 8-aa interval, in a region that underwent 325 nucleotide substitutions. This high ratio of amino acid to nucleotide substitutions implies that many amino acid substitutions are tolerated in the 3' region outside of the catalytic domain (Figure 3). In contrast, amino acid positions 29–32, 37–44, and 161–180 were conserved, although these regions were not conserved at the nucleotide level, suggesting that selection and/or functional constraints may be important in this part of the protein.
Changes at the first, second, and third codon positions varied substantially. Substitutions in third positions were much more frequent than those at first and second positions (Figure 4). The ratio of base substitutions by codon position is 2.0: 1.0: 7.6. A similar pattern was observed in each clade analyzed: green plants, mosses, clade I, II, III, and IV. Substitutions also vary similarly among organismal groups, regardless of gene clade, for example, among all angiosperm sequences and among all monocot sequences (Figure 4). This result implies a similar pattern of base substitution in diverse gene lineages and organismal lineages.
Phylogeny of GSK3/SHAGGY-like kinase genes
A total of 842 variable sites was found in the nucleotide sequences, 735 of which were parsimony-informative. Seventeen most parsimonious trees with a length of 11641 steps were obtained from the maximum parsimony (MP) analysis. The consistency index (CI) was 0.1522, and the retention index (RI) was 0.5789. In the amino acid analysis, 288 variable sites were detected, with 234 parsimony-informative; 77 most parsimonious trees of 2156 steps were obtained (CI = 0.4935; RI = 0.7532).
The clades identified in the support-weighted tree based on nucleotide sequences (SW; Figure 5) are very similar to those of the maximum parsimony tree based on the same data set (MP-N; Figure 6), although relationships among basal nodes are not resolved in the support-weighted tree. Furthermore, the clades found in the trees based on nucleotide sequences (both MP and SW) are very similar to those found in the MP trees based on translated amino acid sequences (MP-AA; Figure 7). Therefore, AT content, codon usage, and other molecular evolutionary biases do not appear to have compromised the reliability of the nucleotide-based results. In fact, the nucleotide data are more informative than the amino acid sequences, yielding greater support for most clades (see Figures 5, 6, 7). However, support for most clades is quite low in all analyses.
The clades found in the Bayesian phylogenetic analysis based on nucleotide sequences are almost identical to those of the maximum parsimony tree based on the same data set. Therefore, the posterior probabilities are indicated on the maximum parsimony strict consensus tree (MP-N) (Figure 6).
In all four phylogenetic analyses, all of the land plant GSK sequences formed a clade distinct from non-plant sequences with high values of internal support as measured by bootstrap, posterior probabilities, and jackknife resamplings (Figures 5, 6, 7). In all four analyses, the Porphyra sequence is sister to all green plant sequences (0.97 posterior probability, support values of 59%, <50%, and 82% from parsimony jackknifing mapped onto the SW tree, MP-N, and MP-AA, respectively), and the Chlamydomonas sequence is sister to all other green plant GSKs (0.99 posterior probability, support values of 81%, 75%, and 64% from parsimony jackknifing mapped onto the SW tree, MP-N, and MP-AA, respectively).
The trees from all four analyses recovered five major clades of sequences within land plants. One clade is composed only of sequences from the moss Physcomitrella (1.0 posterior probability, support values of 100%, 99%, and 72% from parsimony jackknifing mapped onto the SW tree, MP-N, and MP-AA, respectively), and the remaining four clades (I, II, III, and IV) correspond to the GSK subgroups recognized in Arabidopsis [13, 15–17, 21]. Relationships among these five clades varied among the analyses, but internal support was weak except in the Bayesian analysis. A large clade containing clades I, II, and III received a posterior probability of 0.90, and a clade including clades I and II had a posterior probability of 1.0 (Figure 6).
The MP-N tree (Figure 6) shows the moss clade as sister to the remaining four clades, whereas the MP-AA tree places the moss clade as sister to clades I, II, and III, with clade IV sister to this entire clade of moss + clades I, II, and III (Figure 7). The SW analysis also placed the moss clade as sister to the remaining four clades, and clade I was split into two separate clades (Figure 5). The fact that several taxa bear multiple GSKs that fall into separate subclades within clade I suggests that "clade I" may actually represent the products of an additional ancient duplication. However, the non-monophyly of clade I in the SW tree, lack of bootstrap support >50% in the MP trees, and the low posterior probability in the Bayesian analysis suggest that these two subclades may not be each other's closest relatives.
Although we recovered four major clades that correspond to the four groups recognized in Arabidopsis by Dornelas et al. , relationships among and within these clades are generally not well supported based on analyses of either nucleotide or amino acid sequences (Figure 5, 6, 7), apparently due to the conflict among characters. Low support was not due to the choice of outgroups. We repeated the phylogenetic analyses using only Chlamydomonas as an outgroup and obtained the same topology and similar levels of support.
Clade IV was supported most strongly, with 98% jackknife support (on the SW tree; Figure 5), 1.0 posterior probability, and 81% and 78% bootstrap support from the MP-N and MP-AA analyses. Clade III received jackknife support of 100% (SW tree), 0.98 posterior probability, and bootstrap support less than 50% in both MP analyses. Clade II was supported by a jackknife value of 89% (on the SW tree; Figure 5), 0.99 posterior probability, and bootstrap values of 85% and 72%, respectively, in the MP-N and MP-AA analyses. Clade I received less than 50% bootstrap support in both MP analyses and <0.90 posterior probability in the Bayesian analysis, and the SW analysis split this clade into two parts, with jackknife values of 57% and 95%, respectively, mapped onto the SW tree (Figure 5).
Oryza sequences were included in the same four major clades with the Arabidopsis GSKs (Figures 5, 6, 7). Clade I contains three rice sequences, Os01g19150, Os01g14860, and Os05g04340, and clade II includes Os01g10840, Os05g11730, Os06g35530, and Os02g14130 in all trees. The presence of duplicate Oryza sequences within individual clades raises the possibility that some rice GSK genes may have resulted from relatively recent gene duplication, as reported in Arabidopsis . Recently reported evidence of genome duplication in rice  may explain, at least in part, the multiple Oryza sequences within clades I and II. Sequences of several other plant genera are found in three of the four clades. Sequences of the grasses Triticum and Zea are found in clades I, II, and IV, and GSKs from the eudicots Medicago and Lycopersicon are found in clades I, III, and IV.
Clade IV includes AtSK41 and AtSK42 from Arabidopsis, plus sequences from other eudicots, monocots, and the basal angiosperms Persea americana and Nuphar advena. Nuphar advena 4 and 5 form a clade with 83% bootstrap support, appearing well separated from Nuphar advena 3 (Figure 7). These data for Nuphar advena suggest at least one gene duplication in clade IV and indicate a diversity of GSK genes within some basal angiosperm species, comparable to that observed within the eudicot Arabidopsis. Finally, Pinus taeda 2 grouped with eudicot sequences in the MP-AA analysis (Figure 7), but in other analyses it failed to form a subclade.
In clade III, two Pinus ESTs (Pinus taeda 3 and 4) were sister to all other sequences in both MP trees, but this relationship was weakly supported (<50%) even though the posterior probability was high (0.98). In addition, in the SW tree, these two Pinus sequences failed to form a clade (Figure 5). Also within clade III, one Persea EST sequence is sister to a eudicot-specific clade that contains sequences from Nicotiana, Petunia, Lycopersicon, Medicago, Brassica, and Arabidopsis (AtSK31 and AtSK32).
Clade II contains the Arabidopsis sequences AtSK21, AtSK22, and AtSK23. The sequences from rice, wheat, and maize formed a clade with 77% bootstrap support in the MP-N analysis, 1.0 posterior probability, and 100% jackknife support mapped on the SW tree; this clade was not recovered in the MP-AA analysis. This clade also includes sequences from the basal angiosperms Persea and Nuphar in all trees and from Amborella in the MP-AA tree. Sequences from the eudicots Eschscholzia, Ricinus, and Cucumis are also included in clade II.
Clade I contains the Arabidopsis sequences AtSK11 and AtSK12, which formed a sister pair in all analyses (Figures 5, 6, 7); AtSK13 appeared in a separate subclade near the base of clade I, well removed from AtSK11 and AtSK12 in the MP-N and MP-AA trees. In the SW tree, clade I is not monophyletic, and these sequences fall into two clades (I-A and I-B), with 52% and 93% jackknife support, respectively, mapped on the SW tree. AtSK11 and AtSK12 occur in I-A, and AtSK13 occurs in I-B (Figure 5). Expression studies have demonstrated that both AtSK11 and AtSK12 seem to be involved in flower development [13, 16]. In contrast, AtSK13 plays a role in the response to saline treatment and osmotic pressure. It is therefore not surprising that AtSK11 and AtSK12 are not closely related to AtSK13, although phylogenetic position and function are not always coupled. Clade I also contains multiple copies of GSKs from the basal monocot Acorus, two in I-A and one in I-B in the SW tree. The relationship between the two sequences in I-A is not resolved, and their positions in the two MP consensus trees did not receive bootstrap support >50%; it is therefore possible that these two sequences are in fact sisters and represent the product of a gene duplication within the Acorus lineage. The functions of these divergent copies remain to be investigated.
From an evolutionary standpoint, it is significant that ESTs from basal angiosperms were represented in all four major clades in all analyses (Figures 5, 6, 7). ESTs of Nuphar (Nymphaeaceae) occur in three of the four clades (Figures 5, 6, 7). ESTs of Amborella, the sister to all other living flowering plants (either alone or with Nymphaeales; reviewed in ), are found in clades I and II. ESTs of Persea, the avocado (Lauraceae), occur in clades I, II, III, and IV, and an EST of Liriodendron (tulip poplar; Magnoliaceae) is in clade I. ESTs of Eschscholzia (poppy; Paveraceae), a basal eudicot, are in clades I and II. Sequences from the basal angiosperm lineages typically attach at, or near, the base of the clades in which they appear. For example, a sequence of Nuphar is sister to other sequences in clade IV in the MP-AA tree. A sequence of Amborella attaches near the base of clade I in the MP-N tree and clade II in the MP-AA tree, and a sequence of Persea attaches very close to the base of clades II and III in the MP-N and MP-AA tree.
There is a distinct monocot subclade in both clades II and IV, and most of the monocots form two or three subclades in clade I. These monocot-specific subclades are particularly evident in the MP-N tree (Figure 6). Within most clades, the eudicot sequences form a distinct subclade, for example, the subclade of Nicotiana, Petunia, Lycopersicon, and Medicago sequences within clade III. In clade II the GSK homologs of the eudicots Arabidopsis and Ricinus form a subclade. The other eudicot member of clade II, Cucumis, does not appear with the Arabidopsis and Ricinus sequences. However, the Cucumis sequence is a partial sequence (only 72 amino acid residues), which could affect its phylogenetic placement. Recently, Wiens [50, 51] reviewed the effect of missing data in phylogenetic analyses, and his simulations showed that incomplete sequences can be accurately placed in phylogenies; furthermore, they typically do not impact the overall tree, in agreement with empirical studies [e.g., [39, 40]]. We analyzed our data set with and without the partial Cucumis sequence, but removal of this sequence did not alter the topology of the remaining sequences.
Sequences of GSK3/SHAGGY-like kinases are also available for a fern and for several gymnosperms. An EST of the fern Ceratopteris appeared within clade I, as sister to a subclade that includes AtSK11 and AtSK12 in the MP-AA tree (Figure 7). Sequences from Zamia attached near the base of clade I, a sequence of Welwitschia was sister to clades I, II, and III, and the four EST sequences of Pinus taeda appeared in clades III and IV (MP-AA), although these positions varied in the SW and MP-N trees (Figures 5, 6). The placement of gymnosperm sequences in clades I, III, and IV in the MP-AA tree suggests that GSKs diversified to some extent prior to the origin of seed plants, over 300 million years ago [e.g., [52, 53]]. In addition, the presence of a GSK sequence in Porphyra and its phylogenetic placement as sister to all green plant sequences (at least in the two MP analyses) indicates that the plant-specific GSKs were already established before the origin of green plants, the oldest fossils of which are unicellular and filamentous green algae from the Neoproterozoic of Australia (900 mya; [54, 55]) and Spitzbergen (700–800 mya; [56, 57]; reviewed in ). Taken together, our structural and phylogenetic analyses indicate that plant GSK3/shaggy-like kinases were established prior to, or at least early in, the diversification of green plants and that the common ancestor of seed plants already had a diverse tool kit of GSK3/shaggy-like kinase genes that could be used for various signaling-related processes. Future comparative studies of gene function, based on orthologous genes, may be informative about patterns of functional diversification of GSK genes.
The structure of GSK genes in Arabidopsis and rice is highly conserved, and most GSK genes have 12 exons interrupted by 11 introns. Genes included in the same clade based on parsimony analyses share similar structural characteristics. Our phylogenetic results indicate that the plant-specific GSK gene lineage was established prior to, or early in, the history of green plants, and plant GSKs began to diversify prior to the origin of extant seed plants. In addition, at least three of the four major clades of GSKs (I, III, IV) present in Arabidopsis and rice were established early in the history of extant seed plants. Sequences of basal angiosperms are present in all four of the major GSK clades, indicating that the fourth major subgroup of these genes (II) was established either early in angiosperm evolution or prior to the origin of the angiosperms (but after their last common ancestor with extant gymnosperms), if the absence of Clade II sequences from gymnosperms is real and not an artifact of limited sampling. In addition, our data indicate that GSK gene duplication events may have occurred in several of the basal angiosperms investigated, most notably Nuphar. Thus, duplication of GSK genes, which is prevalent in both Arabidopsis and rice, has also occurred in basal angiosperms. This phylogenetic analysis of numerous plant GSK sequences provides a framework for the investigation of the functional genetics of GSKs in signaling, development, and stress response.
A search for GSK3/SHAGGY-like kinase homologs was performed using BLAST [58, 59] at the websites of NCBI , TIGR , PlantGDB , Kazusa DNA Research Institute , and the FGP . We started our search with 10 Arabidopsis and nine rice sequences, and then continued with various published GSK3/SHAGGY-like kinase homologs from human, yeast, Drosophila, Brassica, Medicago, Petunia, Nicotiana, and Ricinus to identify as many GSK homologs as possible from protists, fungi, animals, and plants. Putative GSK homology was defined initially by sequence similarity when the sequences were retrieved and then confirmed by phylogenetic analysis (see below). A total of 139 GSK homologs was collected, of which 73 sequences were ESTs: 26 ESTs from 10 taxa at the FGP web site, 40 ESTs from 17 taxa at the PlantGDB web site, 5 ESTs from the NCBI web site (Ceratopteris and Pinus), and two ESTs from the Kazusa DNA Research Institute database (Clamydomonas and Porphyra). Some ESTs were integrated into a contig, which was constructed using the CAP3 Sequence Assembly Program , and therefore some gene designations have several accession numbers (Additional File 1). Of the remaining 66 sequences, 43 were previously reported land plant sequences, and 23 were sequences from protists, fungi, and animals (Additional File 1).
All sequences were translated into amino acid sequences using Se-Al . The sequences corresponding to the catalytic domain (as defined by Hanks ; 285 amino acid residues corresponding to exon 3 to exon 10 in Arabidopsis; see Figure 1) and part of the 3' region (corresponding to 78 amino acid residues; exons 11 and 12 in Arabidopsis) were aligned manually in a stepwise manner using Se-Al; other regions were too variable to align. The aligned matrix therefore comprised exons 3 to 12 and was 348 amino acid residues in length; the average length of all included sequences was 293 amino acid residues, and the average length of the translated EST sequences was 193 amino acid residues. The aligned sequences were exported for phylogenetic analyses as separate data matrices of nucleotide sequences and amino acid sequences, and all data matrices and trees were deposited in TreeBASE (Study accession S1459, matrix accessions M2623-M2624) . For Arabidopsis and rice, the genomic sequences were aligned and compared with cDNA sequences to investigate gene structure.
A series of analyses was conducted to explore the pattern of sequence evolution in GSK homologs. We investigated patterns of substitution across both nucleotide and protein sequences using the CHART option of MacClade 4.05 , using 116 plant-specific GSK homologs and Tree 1, selected arbitrarily from the phylogenetic analysis. This approach provides a minimum estimate of change for each site. Plotting of substitutions was conducted across a 4-bp or 4-amino acid interval on the x-axis. The analyses were conducted across the entire aligned sequences. We tested for variation in mean substitution rate among codon positions using the CHART option of MacClade 4.05, across the entire data set, within all green plants, within mosses, and within each of the four major clades of seed plant sequences identified by phylogenetic analyses.
Maximum parsimony analyses were conducted with (i) equally weighted characters and character states and (ii) support weighting . Equally weighted parsimony analyses for matrices of nucleotides and amino acids were conducted using PAUP* 4.0b10 . The search strategy involved 100 random addition replicates with TBR branch swapping, saving all optimal trees. Gaps were treated as missing data. To assess support for each node, bootstrap analyses  were performed using 100 replicate heuristic searches, each with 10 random addition replicates and TBR branch swapping, saving all optimal trees.
The support weighting method  provides an alternative approach to assessing internal support for phylogenetic results, by measuring the degree to which changes in a character (site) are concentrated in the supported branches of a tree. Jackknife resampling was used to generate randomly selected suites of initial weights for successive support weighting, providing a means of assessing the stability of branches supported in a standard parsimony jackknife tree [67, 70]. We applied the support weighting method to the nucleotide data matrix. Support values mapped onto the support-weighted tree topology were generated by standard parsimony jackknifing  of the original data matrix using 1000 replicates with SPR branch swapping on each of 10 random data entry orders.
A Bayesian phylogenetic analysis was performed using MrBayes 3.1.1  to compare the tree topology and support values to those obtained from maximum parsimony analyses. The GTR + I + Γ model was selected by the Akaike information criterion (AIC) in ModelTest v.3.6 [72, 73] and applied for the Bayesian analysis. Default parameter values were used for the priors. The analysis was run for 20 million generations, sampling trees every 1000 generations. The first 3000 trees produced during 3 million generations were discarded as burn-in, and the 50% majority-rule consensus of the remaining trees was used to obtain posterior probabilities. Two chains were run, and results from both chains were combined as convergence diagnostics indicated they had converged on similar results (the average standard deviation of split frequencies at 20 million generations was 0.062054).
In previous phylogenetic analyses , mitogen activated protein kinase (MAPK) and cyclin-dependent kinase (CDK) sequences were shown to be the sister group to a clade of all GSK homologs. We analyzed plant MAPK/CDK/Casein kinase II/GSK sequences because these four kinases are included in the same group . In an unrooted tree, GSK sequences formed a clade in which non-plant GSK homologs were sister to plant GSKs (tree not shown). As a result, we used non-plant GSKs as outgroups for analysis of all plant-specific GSK homologs.
Kim L, Kimmel AR: GSK3, a master switch regulating cell-fate specification and tumorigenesis. Curr Opin Genet Dev. 2000, 10: 508-514. 10.1016/S0959-437X(00)00120-9.
Perrimon N, Smouse D: Multiple functions of a Drosophila homeotic gene, zeste-white 3, during segmentation and neurogenesis. Dev Biol. 1989, 135: 287-305. 10.1016/0012-1606(89)90180-2.
Siegfried E, Chou TB, Perrimon N: wingless signaling acts through zeste-white 3, the Drosophila homolog of glycogen synthase kinase-3, to regulate engrailed and establish cell fate. Cell. 1992, 71: 1167-1179. 10.1016/S0092-8674(05)80065-0.
He X, Saint-Jeannet J-P, Woodgett JR, Varmus HE, Dawid IB: Glycogen synthase kinase-3 and dorsoventral patterning in Xenopus embryos. Nature. 1995, 374: 617-622. 10.1038/374617a0.
Emily-Fenouil F, Ghiglione C, Lhomond G, Lepage T, Gache C: GSK3beta/shaggy mediates patterning along the animal-vegetal axes of the sea urchin embryo. Development. 1998, 125: 2489-2498.
Simpson P, El Messal M, Moscoso del Prodo J, Ripoll P: Stripes of positional homologies across the wing blade of Drosophila melanogaster . Development. 1998, 103: 391-401.
Oreña SJ, Torchia AJ, Garofalo RS: Inhibition of glycogen-synthase kinase 3 stimulates glycogen synthase and glucose transport by distinct mechanisms in 3T3-L1 adipocytes. J Biol Chem. 2000, 275: 15765-15772. 10.1074/jbc.M910002199.
Zumbrunn J, Kinoshita K, Hyman AA, Näthke IS: Binding of the adenomatous polyposis coli protein to microtubules increases microtubule stability and is regulated by GSK3 beta phosphorylation. Curr Biol. 2001, 11: 44-49. 10.1016/S0960-9822(01)00002-1.
Webster MT, Rozycka M, Sara E, Davis E, Smalley M, Young N, Dale TC, Wooster R: Sequence variants of the axin gene in breast, colon, and other cancers: An analysis of mutations that interfere with GSK3 binding. Genes Chromosomomes Cancer. 2000, 28: 443-453. 10.1002/1098-2264(200008)28:4<443::AID-GCC10>3.0.CO;2-D.
Puziss JW, Hardy TA, Johnson RB, Roach PJ, Hieter P: MDS1, a dosage suppressor of an mck1 mutant, encodes a putative yeast homolog of glycogen synthase kinase 3. Mol Cell Biol. 1994, 14: 831-839.
Plyte SE, Feoktistova A, Burke JD, Woodgett JR, Gould KL: Schizosaccharomyces pombe skp1+ encodes a protein kinase related to mammalian glycogen synthase kinase 3 and complements a cdc14 cytokinesis mutant. Mol Cell Biol. 1996, 16: 179-191.
Bianchi MW, Guivarc'h D, Thomas M, Woodgett JR, Kreis M: Arabidopsis homologs of the shaggy and GSK-3 protein kinases: molecular cloning and functional expression in Escherichia coli. Mol Gen Genet. 1994, 242: 337-345. 10.1007/BF00280424.
Charrier B, Champion A, Henry Y, Kreis M: Expression profiling of the whole Arabidopsis Shaggy-like kinase multigene family by real-time reverse transcriptase-polymerase chain reaction. Plant Physiol. 2002, 130: 577-590. 10.1104/pp.009175.
Decroocq-Ferrant V, Van Went J, Bianchi MW, de Vries SC, Kreis M: Petunia hybrida homologues of shaggy/zeste-white 3 expressed in female and male reproductive organs. Plant J. 1995, 7: 897-911. 10.1046/j.1365-313X.1995.07060897.x.
Dornelas MC, Lejeune B, Dron M, Kreis M: The Arabidopsis SHAGGY -related protein kinase (ASK) gene family: structure, organization and evolution. Gene. 1998, 212: 249-257. 10.1016/S0378-1119(98)00147-4.
Dornelas MC, van Lammeren AAM, Kreis M: Arabidopsis thaliana SHAGGY-related protein kinases (AtSK11 and 12) function in perianth and gynoecium development. Plant J. 2000, 21: 419-429. 10.1046/j.1365-313x.2000.00691.x.
Dornelas MC, Wittich P, von Recklinghausen I, van Lammeren A, Kreis M: Characterization of three novel members of the Arabidopsis SHAGGY-related protein kinase (ASK) multigene family. Plant Mol Biol. 1999, 39: 137-147. 10.1023/A:1006102812280.
Einzenberger E, Eller N, Heberle-Bors E, Vicente O: Isolation and expression during pollen development of a tobacco cDNA clone encoding a protein kinase homologous to shaggy/glycogen synthase kinase-3. Biochim Biophys Acta. 1995, 1260: 315-319.
Jonak C, Beisteiner D, Beyerly J, Hirt H: Wound-Induced Expression and Activation of WIG, a Novel Glycogen Synthase Kinase 3. Plant Cell. 2000, 12: 1467-1475. 10.1105/tpc.12.8.1467.
Jonak C, Heberle-Bors E, Hirt H: Inflorescence-specific expression of AtK-1, a novel Arabidopsis thaliana homologue of shaggy/glycogen synthase kinase-3. Plant Mol Biol. 1995, 27 (1): 217-221. 10.1007/BF00019194.
Jonak C, Hirt H: Glycogen synthase kinase 3/SHAGGY-like kinases in plants: an emerging family with novel functions. Trends Plant Sci. 2002, 7: 457-461. 10.1016/S1360-1385(02)02331-2.
Pay A, Jonak C, Bogre L, Meskiene I, Mairinger T, Szalay A, Heberle-Bors E, Hirt H: The MsK family of alfalfa protein kinase genes encodes homologues of shaggy/glycogen synthase kinase-3 and shows differential expression patterns in plant organs and development. Plant J. 1993, 3: 847-856. 10.1111/j.1365-313X.1993.00847.x.
Tichtinsky G, Tavares R, Takvorian A, Schwebel-Dugue N, Twell D, Kreis M: An evolutionary conserved group of plant GSK-3/shaggy-like protein kinase genes preferentially expressed in developing pollen. Biochim Biophys Acta. 1998, 1442: 261-273.
Piao HL, Pih KT, Lim JH, Kang SG, Jin JB, Kim SH, Hwang I: An Arabidopsis GSK3/shaggy-like gene that complements yeast salt stress-sensitive mutants is induced by NaCl and abscisic acid. Plant Physiol. 1999, 119: 1527-1534. 10.1104/pp.119.4.1527.
Li J, Nam KH, Vafeados D, Chory J: BIN2, a new brassinosteroid-insensitive locus in Arabidopsis. Plant Physiol. 2001, 127: 14-22. 10.1104/pp.127.1.14.
Li J, Nam KH: Regulation of brassinosteroid signaling by a GSK3/SHAGGY-like kinase. Science. 2002, 295: 1299-1301.
Pérez-Pérez JM, Ponce MR, Micol JL: The UCU1 Arabidopsis gene encodes a SHAGGY/GSK3-like kinase required for cell expansion along the proximodistal axis. Dev Biol. 2002, 242: 161-173. 10.1006/dbio.2001.0543.
Choe S, Schmitz RJ, Fujioka S, Takatsuto S, Lee M-O, Yoshida S, Feldmann KA, Tax FE: Arabidopsis brassinosteroid-insensitive dwarf12 mutants are semidominant and defective in a glycogen synthase kinase 3beta-like kinase. Plant Physiol. 2002, 130: 1506-1515. 10.1104/pp.010496.
Sasaki T, Matsumoto T, Yamamoto K, Sakata K, Baba T, Katayose Y, Wu J, Niimura Y, Cheng Z, Nagamura Y, Antonio BA, Kanamori H, Hosokawa S, Masukawa M, Arikawa K, Chiden Y, Hayashi M, Okamoto M, Ando T, Aoki H, Arita K, Hamada M, Harada C, Hijishita S, Honda M, Ichikawa Y, Idonuma A, Iijima M, Ikeda M, Ikeno M, Ito S, Ito T, Ito Y, Ito Y, Iwabuchi A, Kamiya K, Karasawa W, Katagiri S, Kikuta A, Kobayashi N, Kono I, Machita K, Maehara T, Mizuno H, Mizubayashi T, Mukai Y, Nagasaki H, Nakashima M, Nakama Y, Nakamichi Y, Nakamura M, Namiki M, Negishi M, Ohta I, Ono N, Saji S, Sakai K, Shibata M, Shimokawa T, Shomura A, Song J, Takazaki Y, Terasawa K, Tsuji K, Waki K, Yamagata H, Yamane H, Yoshiki S, Yoshihara R, Yukawa K, Zhong H, Iwama H, Endo T, Ito H, Hahn JH, Kim HI, Eun MY, Yano M, Jiang J, Gojobori T: The genome sequence and structure of rice chromosome 1. Nature. 2002, 420: 312-316. 10.1038/nature01184.
The Rice Full-Length cDNA Consortium: Collection, mapping, and annotation of over 28,000 cDNA clones from japonica rice. Science. 2003, 301: 376-379. 10.1126/science.1081288.
The Floral Genome Project. [http://fgp.bio.psu.edu/fgp/]
Albert VA, Soltis DE, Carlson JE, Farmerie WG, Wall PK, Ilut DC, Solow TM, Mueller LA, Landherr LL, Hu Y, Buzgo M, Kim S, Yoo MJ, Frohlich MW, Perl-Treves R, Schlarbaum SE, Bliss BJ, Zhang X, Tanksley SD, Oppenheimer DG, Soltis PS, Ma H, dePamphilis CW, Leebens-Mack JH: Floral gene resources from basal angiosperms for comparative genomics research. BMC Plant Biol. 2005, 5: 5-10.1186/1471-2229-5-5.
Mathews S, Donoghue MJ: The root of angiosperm phylogeny inferred from duplicate phytochrome genes. Science. 1999, 286: 947-950. 10.1126/science.286.5441.947.
Qiu YL, Lee J, Bernasconi-Quadroni F, Soltis DE, Soltis PS, Zanis M, Zimmer EA, Chen Z, Savolainen V, Chase MW: The earliest angiosperms: evidence from mitochondrial, plastid and nuclear genomes. Nature. 1999, 402: 404-407. 10.1038/46536.
Soltis PS, Soltis DE, Chase MW: Angiosperm phylogeny inferred from multiple genes as a tool for comparative biology. Nature. 1999, 402: 402-404. 10.1038/46528.
Parkinson CL, Adams KL, Palmer JD: Multigene analyses identify the three earliest lineages of extant flowering plants. Curr Biol. 1999, 9: 1485-1488. 10.1016/S0960-9822(00)80119-0.
Barkman TJ, Chenery G, McNeal JR, Lyons-Weiler J, Ellisens W, Moore G, Wolfe AD, dePamphilis CW: Independent and combined analyses of sequences from all three genomic compartments converge on the root of flowering plant phylogeny. Proc Natl Acad Sci USA. 2000, 97: 13166-13171. 10.1073/pnas.220427497.
Graham SW, Olmstead RG: Utility of 17 chloroplast genes for inferring the phylogeny of the basal angiosperms. Am J Bot. 2000, 87: 1712-1730.
Soltis DE, Soltis PS, Chase MW, Mort ME, Albach DC, Zanis M, Savolainen V, Hahn WH, Hoot SB, Fay MF, Axtell M, Swensen SM, Prince LM, Kress WJ, Nixon KC, Farris JS: Angiosperm phylogeny inferred from 18S rDNA, rbcL, and atpB sequences. Bot J Linean Soc. 2000, 133: 381-461. 10.1006/bojl.2000.0380.
Zanis MJ, Soltis DE, Soltis PS, Mathews S, Donoghue MJ: The root of the angiosperms revisited. Proc Natl Acad Sci USA. 2002, 99: 6848-6853. 10.1073/pnas.092136399.
Borsch T, Hilu KW, Quandt D, Wilde V, Neinhuis C, Barthlott W: Noncoding plastid trnT-trnF sequences reveal a well resolved phylogeny of basal angiosperms. J Evol. 2003, 16: 558-576. 10.1046/j.1420-9101.2003.00577.x.
Hilu KW, Borsch T, Muller K, Soltis DE, Soltis PS, Savolainen V, Chase MW, Powell MP, Alice LA, Evans R, Sauquet H, Neinhuis C, Slotta TAB, Rohwer JG, Campbell CS, Chatrou LW: Angiosperm phylogeny based on matK sequence information. Am J Bot. 2003, 90: 1758-1776.
The Arabidopsis Genome Initiative: Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature. 2000, 408: 796-815. 10.1038/35048692.
Goff SA, Ricke D, Lan TH, Presting G, Wang RL, Dunn M, Glazebrook J, Sessions A, Oeller P, Varma H, Hadley D, Hutchison D, Martin C, Katagiri F, Lange BM, Moughamer T, Xia Y, Budworth P, Zhong J, Miguel T, Paszkowski U, Zhang S, Colbert M, Sun WL, Chen L, Cooper B, Park S, Wood TC, Mao L, Quail P, Wing R, Dean R, Yu Y, Zharkikh A, Shen R, Sahasrabudhe S, Thomas A, Cannings R, Gutin A, Pruss D, Reid J, Tavtigian S, Mitchell J, Eldredge G, Scholl T, Miller RM, Bhatnagar S, Adey N, Rubano T, Tusneem N, Robinson R, Feldhaus J, Macalma T, Oliphant A, Briggs S: A draft sequence of the rice genome (Oryza sativa L. ssp japonica). Science. 2002, 296: 92-100. 10.1126/science.1068275.
Richard O, Paquet N, Haudecoeur E, Charrier B: Organization and expression of the GSK3/shaggy kinase gene family in the moss Physcomitrella patens suggest early gene multiplication in land plants and an ancestral response to osmotic stress. J Mol Evol. 2005, 61: 99-113. 10.1007/s00239-004-0302-6.
Hanks SK: Eukaryotic protein kinases. Curr Opin Struct Biol. 1991, 1: 369-383. 10.1016/0959-440X(91)90035-R.
Paterson AH, Bowers JE, Chapman BA: Ancient polyploidization predating divergence of the cereals, and its consequences for comparative genomics. Proc Natl Acad Sci USA. 2004, 101: 9903-9908. 10.1073/pnas.0307901101.
Soltis PS, Soltis DE: The origin and diversification of angiosperms. Am J Bot. 2004, 91: 1614-1626.
Wiens JJ: Can incomplete taxa rescue phylogenetic analyses from long-branch attraction?. Syst Biol. 2005, 54: 731-742.
Wiens JJ: Missing data and the design of phylogenetic analyses. J Biomed Inform. 2006, 39: 34-42. 10.1016/j.jbi.2005.04.001.
Kenrick P, Crane PR: The origin and early diversification of land plants: a cladistic study. 1997, Washington, DC: Smithsonian Institution Press
Niklas KJ: The evolutionary biology of plants. 1997, Chicago: The University of Chicago Press
Schopf JW: Microflora of the Bitter Springs Formation, late Precambrian, central Australia. J Paleontol. 1968, 42: 651-688.
Schopf JW, Blacic JM: New microorganisms from the Bitter Springs Formation (late Precambrian) of the north-central Amadeus Basin, Australia. J Paleontol. 1971, 45: 925-960.
Butterfield NJ, Knoll AH, Swett K: Exceptional preservation of fossils in an Upper Proterozoic shale. Nature. 1988, 334: 424-427. 10.1038/334424a0.
Knoll AH: The early evolution of eukaryotes: a geological perspective. Science. 1992, 256: 622-627.
Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ: Basic local alignment search tool. J Mol Biol. 1990, 215: 403-410. 10.1006/jmbi.1990.9999.
Altschul SF, Madden TL, Schaffer AA, Zhang JH, Zhang Z, Miller W, Lipman DJ: Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 1997, 25: 3389-3402. 10.1093/nar/25.17.3389.
National Center for Biotechnology Information. [http://0-www.ncbi.nlm.nih.gov.brum.beds.ac.uk/]
TIGR-The Institute for Genomic Research. [http://www.tigr.org/]
PlantGDB-Resources for Plant Comparative Genomics. [http://www.plantgdb.org/]
Kazusa DNA Research Institute database. [http://www.kazusa.or.jp/en/plant/database.html]
Huang X, Madan A: CAP3: a DNA Sequence Assembly Program. Genome Res. 1999, 9: 868-877. 10.1101/gr.9.9.868.
Rambaut A: Se-Al: Sequence Alignment Editor. 1996, [http://evolve.zoo.ox.ac.uk/]
Maddison DR, Maddison WP: MacClade, version 4.05. 2002, Sunderland: Sinauer Associates
Farris JS: Support weighting. Cladistics. 2001, 17: 389-394. 10.1111/j.1096-0031.2001.tb00133.x.
Swofford DL: PAUP* 4.0b1:phylogenetic analysis using parsimony (*and other methods). 2001, Sunderland: Sinauer Associates
Felsenstein J: Confidence limits on phylogenies: an approach using the bootstrap. Evolution. 1985, 39: 783-791.
Farris JS, Albert VA, Kallersjo M, Lipscomb D, Kluge AG: Parsimony jackknifing outperforms neighbor-joining. Cladistics. 1996, 12: 99-124. 10.1111/j.1096-0031.1996.tb00196.x.
Ronquist F, Huelsenbeck JP: MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics. 2003, 19: 1572-1574. 10.1093/bioinformatics/btg180.
Posada D, Crandall KA: Modeltest: testing the model of DNA substitution. Bioinformatics. 1998, 14: 817-818. 10.1093/bioinformatics/14.9.817.
Posada D, Buckley TR: Model selection and model averaging in phylogenetics: advantages of the AIC and Bayesian approaches over likelihood ratio tests. Syst Biol. 2004, 53: 793-808. 10.1080/10635150490522304.
PlantsP Kinase Classification web site. [http://plantsp.sdsc.edu/plantsp/family/class.html]
This research was supported by the Floral Genome Project (NSF PGR-0115684). We thank those members of the Floral Genome Project who contributed to tissue collection, library construction, and EST sequencing, especially Bill Farmerie and Kevin Holland of the UF Genome Sequencing Service Laboratory. We also thank David G. Oppenheimer, Matyas Buzgo, Sangtae Kim, Jin Koh, André Chanderbali, and Samuel Brockington for helpful comments and discussion, and Matt Gitzendanner for assistance with Bayesian analyses, and James S.Farris for access to the support weighting program.
M-JY carried out the sequence analysis, the equally weighted maximum parsimony analysis, and the Bayesian analysis, and with PSS and DES wrote the manuscript. VAA performed the support weighting analysis. PSS and DES supervised the project. All authors read and approved the final submission.
Authors’ original submitted files for images
Below are the links to the authors’ original submitted files for images.
About this article
Cite this article
Yoo, M., Albert, V.A., Soltis, P.S. et al. Phylogenetic diversification of glycogen synthase kinase 3/SHAGGY-like kinase genes in plants. BMC Plant Biol 6, 3 (2006) doi:10.1186/1471-2229-6-3
- Major Clade
- Basal Angiosperm
- Jackknife Support
- Extant Seed Plant
- Maximum Parsimony Strict Consensus Tree