Figure 3. Identification and Analyses of Cdk1 Substrates in ES Cells.

(A) Schematic representation of the direct labeling and inhibition approaches.

(B) Summary of the direct labeling approach. A total of 1181 unique thiophosphorylated peptides were detected; 92% of these were ‘Cdk1-specific’, since they were never detected in Cdk1^+/+ cells. Only ‘Cdk1-specific’ peptides were used for subsequent analyses. 64% of ‘Cdk1-specific’ peptides contained thiophosphorylated serine or threonine residue followed by a proline.

(C) Summary of the inhibition approach. In total, we quantified 11842 unique phosphopeptides. 19% of them showed altered phosphorylation status (p<0.05 using the t-test with Benjamini-Hochberg correction) in Cdk1-inhibited cells. 97% of these peptides displayed reduction in phosphorylation upon Cdk1 inhibition. In 53% of peptides showing decreased phosphorylation upon Cdk1 inhibition, the residue displaying decreased phosphorylation corresponded to serine or threonine residue followed by a proline.

(D)Waterfall plot depicting changes in phosphorylation of peptides upon Cdk1 inhibition. Shown are log2 values of the ratio between abundance of each phosphopeptide in 3-MB-PP1- versus in vehicle-treated cells. Negative values correspond to decreased phosphorylation, positive to increased phosphorylation.

(E) Venn diagram summarizing results of the inhibition approach. The blue square depicts all phosphoproteins quantified by mass spectrometry (the number of unique phosphopeptides is in brackets). The yellow rectangle corresponds to quantified proteins (peptides) containing S/T-P sequence. The red rectangle depicts proteins (peptides) showing decreased phosphorylation upon Cdk1-inhibition (Benjamini-Hochberg p-value < 0.05). We considered the overlap between the yellow and red rectangles as putative Cdk1 substrates (622 proteins, 1177 peptides).

(F) Venn diagram illustrating the overlap between proteins identified by us as Cdk1 substrates in the direct labeling (green circle) and inhibition (orange circle) approaches. Proteins were considered as substrates only if the phosphorylation site matched the consensus S/T-P sequence. The number (and percentage) of known Cdk1 substrates (as reported by PhosphoSite Plus substrate page for Cdk1) identified using each approach, and by both approaches (intersection) is shown.

(G) Graphic representation of the top ten significantly enriched GO categories among Cdk1 substrates identified using the inhibition approach. p-values were adjusted using the Benjamini-Hochberg procedure.

See also Figures S2, S3, Tables S2, S3, S5.