A mechanism for the evolution of phosphorylation sites

Abstract

Protein phosphorylation provides a mechanism for the rapid, reversible control of protein function. Phosphorylation adds negative charge to amino acid side chains, and negatively charged amino acids (Asp/Glu) can sometimes mimic the phosphorylated state of a protein. Using a comparative genomics approach, we show that nature also employs this trick in reverse by evolving serine, threonine, and tyrosine phosphorylation sites from Asp/Glu residues. Structures of three proteins where phosphosites evolved from acidic residues (DNA topoisomerase II, enolase, and C-Raf) show that the relevant acidic residues are present in salt bridges with conserved basic residues, and that phosphorylation has the potential to conditionally restore the salt bridges. The evolution of phosphorylation sites from glutamate and aspartate provides a rationale for why phosphorylation sometimes activates proteins, and helps explain the origins of this important and complex process.

Figure 1. A Possible Mechanism for the Evolution of Activating Phosphorylation Sites from Acidic Residues

(A) Phosphorylation of a surface-exposed serine residue in an active protein (yellow) would be expected to decrease the protein’s activity (gray). (B) Structures of phosphoserine, phosphothreonine, aspartic acid, and glutamic acid. Carbon atoms are represented in green, phosphorus atoms in magenta, and oxygen atoms in red. (C) Phosphorylation of a serine or threonine residue could conditionally restore an important electrostatic interaction originally mediated by an acidic amino acid, thereby activating the protein.

Figure 2. Enrichment of Asp and Glu at the Positions of Phosphoserines

(A) Construction of a multiple sequence alignment for a phosphorylated protein showing the phosphosite (red) and control serines (serines not known to be phosphorylated, black) (B) Overall incidences at which various amino acids were found at the position of phosphoserines (N=7584) and control serines (N=257,481) in the multiple sequence alignments. Error bars are standard deviations, obtained by bootstrap resampling. (C) Replacement percentages normalized to the overall abundance of each amino acid in the multiple sequence alignments. Error bars are standard deviations as calculated by the formula $\frac{A}{B}\times \sqrt{{\left(\frac{a}{A}\right)}^2+{\left(\frac{b}{B}\right)}^2}$, where A is the replacement percentage at phosphoserines, B is the replacement percentage at control serines, and a and b are the bootstrapped standard deviations of A and B. (D) Enrichment of various amino acids at phosphoserines relative to control serines. Error bars are standard deviations, calculated by bootstrapping as in (C). P-values were calculated by bootstrapping under the null hypothesis.

Figure 3. A Subset of the Phosphosites Have Very High Asp/Glu Replacement Percentages

(A) Percentage of phosphosites (red) and control serine sites (black) which had various Asp/Glu replacement percentages in the alignments. (B) Percentage of phosphosites (red) and control serine sites (black) which had various Asn/Gln replacement percentages in the alignments. Replacement percentages were calculated from the 234 multiple sequence alignments that contained at least sixty homologs. Standard deviations and p values were calculated by boostrapping. See also Table S1.

Figure 4. Evolution of Phosphorylation Sites from Acidic Residues at the Divergence of Eukaryotes from Prokaryotes

(A) Homologs of human eEF2, which is phosphorylated at S502. The tree is colored to depict the amino acids present in homologs of human eEF2 at the positions corresponding to pS502. Human eEF2 is denoted by the star. eEF2 homologs were identified using BLAST. Homologs with E-values < 10⁻¹⁶ were aligned. Trees were inferred by PhyML using maximum likelihood. The tree was rooted using E. coli EF4 as an outgroup. Leaves were colored according to the amino acid at the relevant position. Internal nodes were colored based on the maximum likelihood amino acid inferred for those ancestral sequences using FASTML. The amino acids inferred for the hypothetical ancestors of human eEF2 are shown. (B) Homologs of human topoisomerase II (Topo II) β were identified and aligned, and trees were inferred and colored as described for panel A. Human Topo II is phosphorylated at T639 and S640. The former residue is replaced by a Glu in all of the prokaryotic gyrase B proteins. A larger version of this figure, which includes species names and bootstrap values, is available as Figure 1B. The asterisks denote three viral Topo II homologs. (C, D) Amino acid sequences close to the phosphosites for a few selected eEF2 and Topo II homologs. For larger versions of the trees shown in panels A and B, and control trees, see Table S2.

Figure 5. Evolution of Phosphorylation Sites from Acidic Residues Later in Evolution

(A, B) S. cerevisiae enolase Eno1. Homologs of the S. cerevisiae enolase Eno1 were identified and aligned, and trees were inferred as described for Figure 4. The root of the tree was placed between the archaea and eukaryotes. An expanded view of thirty fungal enolases is shown in panel (B). The fungus Cryptococcus neoformans was used as an outgroup to define the root of the tree. (C) Amino acid sequences close to the phosphosite for a few selected enolase homologs. See also Table S3.

Figure 6. Evolution of Phosphorylation Sites from Acidic Residues in Particular Eukaryotic Paralogs

(A) Homologs of human C-Raf were identified and aligned, trees were inferred as described for Figure 4. The human Ksr2 protein was used as an outgroup to define the root of the tree. (B) Local alignments of various Raf proteins in the vicinity of phosphosites Y340 and Y341. (C) Homologs of bovine GRK5 were identified and aligned, and trees were inferred as described for Figure 4. The human Akt2 protein was used as an outgroup to define the root of the tree. (D) Local alignments of the seven bovine GRK proteins and two Drosophila GRK proteins in the vicinity of phosphosites S484 and T485. See also Table S4.

Figure 7. Salt Bridges in Phosphoprotein Homologs Possessing Glu Residues at the Positions of the Phosphosites

(A) Topoisomerase II. Yellow represents the E. coli gyrase A₂B₂ tetramer (PDB ID 3NUH). Gray represents the S. cerevisiae Topo II homodimer (2RGR). (B) Enolase. Yellow represents E. coli enolase (1E9I). Gray represents S. cerevisiae Eno1 protein (1ONE). (C) Raf. Yellow represents human B-Raf (1UWJ). Gray represents human C-Raf (3OMV). In all three panels, the structures were superimposed so as to minimize the overall RMS deviation of the positions of the aligned residues, using UCSF Chimera (Pettersen et al., 2004).