Evolutionary conservation of the polyproline II conformation surrounding intrinsically disordered phosphorylation sites

Abstract

Intrinsically disordered (ID) proteins function in the absence of a unique stable structure and appear to challenge the classic structure-function paradigm. The extent to which ID proteins take advantage of subtle conformational biases to perform functions, and whether signals for such mechanism can be identified in proteome-wide studies is not well understood. Of particular interest is the polyproline II (PII) conformation, suggested to be highly populated in unfolded proteins. We experimentally determine a complete calorimetric propensity scale for the PII conformation. Projection of the scale into representative eukaryotic proteomes reveals significant PII bias in regions coding for ID proteins. Importantly, enrichment of PII in ID proteins, or protein segments, is also captured by other PII scales, indicating that this enrichment is robustly encoded and universally detectable regardless of the method of PII propensity determination. Gene ontology (GO) terms obtained using our PII scale and other scales demonstrate a consensus for molecular functions performed by high PII proteins across the proteome. Perhaps the most striking result of the GO analysis is conserved enrichment (P < 10(-8) ) of phosphorylation sites in high PII regions found by all PII scales. Subsequent conformational analysis reveals a phosphorylation-dependent modulation of PII, suggestive of a conserved "tunability" within these regions. In summary, the application of an experimentally determined polyproline II (PII) propensity scale to proteome-wide sequence analysis and gene ontology reveals an enrichment of PII bias near disordered phosphorylation sites that is conserved throughout eukaryotes.

Figure 1

Determination of a calorimetric PII scale. (A) Schematic of the binding equilibrium of SH3 (blue) and Sos peptide (green) with surface-exposed substitution site (red) (PDB code: 1SEM). (B) PII propensity scale with cis/trans isomerization correction (black), and without correction (white). Error bars are propagated error in ΔΔG + 30 cal/mol. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Figure 2

Schematic and validation of an algorithm for calculation of PII propensity within amino acid sequences. (A) A sliding window calculated the position-specific average PII propensity along a dataset of protein sequences, referencing the experimentally determined propensity at site (red) and window residues (blue). High PII regions of (B) human tau, (C) human titin, and (D) bacterial TonB, can be detected using the sliding window scheme and the calorimetrically-determined PII scale (black line). The proteome average sequence PII propensity is shown (dashed gray line) for reference. (E) A transmembrane protein (Omp85) whose family is known to have β-barrel structure shows no high PII signal. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Figure 3

PII propensity is enriched in ID segments. ID segments show significant (P < 0.05) enrichment of high PII propensity (solid line) relative to structured proteins (dashed line). Ordered segments of disordered proteins (dotted line) are shown.

Figure 4

Enrichment of PII propensity in ID sequences is detected using other PII scales. ID segments of disordered proteins show significant (P < 0.05) enrichment of high PII propensity (black line) relative to structured proteins (dashed line) using PII scales of (A) Rucker et al., (B) Tran et al. PPXPP, and (C) Grdadolnik et al. (D) Amino acids whose frequencies increase in ID sequences (red) are near the top of the average PII rank order; those with decreasing frequencies in ID sequences (blue) occur near the bottom of most PII scales. (E) Average PII rank order from all PII scales.– Error bars show standard deviation of PII rank across scales. Amino acids are nonpolar (red), polar (blue), aromatic (purple), negatively charged (orange), positively charged (green). (F) Spearman correlation of the average PII rank order to the TOP-IDP scale. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Figure 5

(A) The PII propensity distribution of folding sequences (▪) and shuffled sequences (□) are superimposable. (B) Locally enriched high PII bias (▪) is abolished upon shuffling ID sequences (□).

Figure 6

Evolutionary selection of high conformational bias. (A) Amino acid substitution by either random change or by BLOSUM62 preserves the average PII propensities of 10,000 in silico evolved “daughter” sequences generated from a “parent” set of randomly selected folding protein sequences (P > 0.25). The average PII propensity of the “parent” sequences (black dot) is well within the “daughter” distribution. (B) The average PII distributions of 10,000 “daughter” sequences deviate from the high PII “parent” sequences (•) (P < 0.05). (C) The statistical deviation of the “daughter” sequence distribution depends on the sequence identity maintained during in silico evolution. The “parent” sequence falls within the “daughter” PII distribution at all levels of sequence identity in structured proteins (squares), regardless of whether substitution was performed randomly (▪) or by BLOSUM62 (□). The log of the p-value of the “parent” sequence PII propensity relative to the “daughter” sequences sharply decreases in high PII proteins when substitution was random (•) and when substitution occurred by BLOSUM62 (○) Colored points on each line correspond to the significance of the “parent” to “daughter” difference shown in (A,B). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Figure 7

Phosphorylation is a functionally conserved feature of high PII proteins. (A) Average PII propensity distributions for six eukaryote proteomes: H. sapiens (red), M. musculus (orange), D. melanogaster (yellow), C. elegans (green), A. thaliana (light blue), and S. cerevisiae (dark blue). (B) Venn diagram of the number of features shared in the top 1% of PII proteins from H. sapiens (red), M. musculus (orange), and S. cerevisiae (blue). (C) Commonality of the top five GO terms by statistical enrichment among PII scales– and the present scale (black line) compared to ten random protein sets (gray dashed line). (D) “Phosphoprotein” GO term enrichment obtained by PII scales: (i) Rucker et al., (ii) Shi et al., (iii) Grdadolnik et al., (iv) Oh et al., (v) Brown et al., (vi) Tran et al. (PPXPP), (vii) Fleming et al. (coil library), (viii) Beck et al., (ix) Moradi et al., and (x) the present scale. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Figure 8

Phosphorylation can modulate PII and is differentially distributed in the proteome. (A) Calorimetrically-determined PII propensities for unmodified (white) and phosphorylated (red) SER, THR, and TYR. (B) PII propensity distribution of the H. sapiens proteome (black) and phosphorylation sites therein (red). (C) Bimodal enrichment of phosphorylation site density (black) observed in H. sapiens, dominated by SER (blue). THR sites (red) have enriched density in high PII contexts, while TYR (purple) is mostly in low contexts. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Figure 9

Low and high PII phosphoproteins are utilized for different cellular functions. Relative enrichment for GO terms obtained for SER (blue), THR (red), and TYR (purple) phosphoproteins correspond to the pie sizes in the above graphs. (A) Low PII phosphoprotein biological process related GO terms. (B) Low PII phosphoprotein biological process related GO terms. (C) High PII phosphoprotein biological process related GO terms. (D) High PII phosphoprotein biological process related GO terms. Comparison of the GO terms and amino acid utilization (SER, THR, or TYR) in each panel immediately shows how low PII (A,B) and high PII (C,D) phosphoproteins have different functions across the human proteome. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]