Fig 2: Phospholipid, phosphatidylinositol 3,4,5-triphosphate (PIP3) binding affinities and PP2A-mediated phosphatase removal of pT308 among several full-length Akt mutant forms.(A) PIP3 binding assays with several isolated Akt PH domain forms. WT: black, R86A: blue, E17K: yellow, Y18A: red, E17K Y18A: green. Fluorescent anisotropy spectra were obtained from the mixture of 50 nM fluorescein-labeled soluble (C8) PIP3 and varying amount of Akt PH domain (0–50 µM). The KD values were determined from fluorescence anisotropy versus PH domain protein concentration plots (n=3, SD shown). (B) Full-length Akt mutants have different sensitivities toward the dephosphorylation of pT308 by PP2A phosphatase. Dephosphorylation assays were performed with Akt mutants containing pT308 and lacking C-tail phosphorylation (A6–A9). The dephosphorylation rates were monitored by western blots with anti-pT308 antibody and T1/2 values shown were determined from plots for fraction of pT308 versus time (n=3, SEM shown). Figure 5—source data 1.Fluorescence anisotropy binding experiments to measure the phospholipid, phosphatidylinositol 3,4,5-triphosphate (PIP3) binding affinities of the PH domain mutants. Figure 5—source data 2.Dephosphorylation assays using PP2A phosphatase on full-length Akt mutants containing pT308 and a non-phosphorylated C-tail with raw western blot images.

Fig 3: Heat-map analysis of PP1c/PP2Ac amino acid preference.Amino acid preference surrounding pSer/pThr for PP1 and PP2A on the protein level from the phosphoproteomic data (Supplementary Data 2). Amino acids with <25 raw counts in a given position were excluded and are marked in gray. a Color coding represents fold-change of the relative abundance of a given amino acid in a given position between phosphatase-sensitive/insensitive phosphorylation sites. b Direct comparison of PP1c and PP2Ac fold changes displayed in (a). Blue highlights amino acids selectively preferred by PP1c, and residues statistically different between PP1c and PP2Ac (adjusted p-value <0.01) according to a two-sided Fisher’s exact test are highlighted in bold. Please see Methods and Source Data for details and exact p-values. Source data are provided as a Source Data file.

Fig 4: Analysis of PP1c/PP2Ac-affected p-sites.The data are from the phosphoproteomic experiments (Supplementary Data 2; Supplementary Fig. 4b, c). a Motif analysis of PP1c/PP2Ac (in)sensitive class I sites. Color of single letter code represents biochemical properties with red = basic, purple = acidic. The enrichment of R in PP1c-sensitive p-sites over PP2Ac-sensitive sites was tested by a two-sided Fisher’s exact test and was found to be statistically different (p-value 0.019). Source data are provided as a Source Data file. b Hierarchical clustering of p-sites after one-way ANOVA analysis using Perseus software. Color scale represents log2-fold change after z-transformation. The three largest clusters are depicted in detail on the right. Each line corresponds to one p-site. These data are also the basis for PP1c/PP2Ac-specific sites presented in Table 1.

Fig 5: Analysis of PP1c/PP2Ac amino acid preference using the PLDMS approach.a PP1c and PP2Ac preferably dephosphorylate pThr over pSer. For statistical analysis, a two-sided Fisher’s exact test was used. b Heat-map analysis of the sequence context surrounding pSer/pThr dephosphorylated by PP1c and PP2Ac. Color coding: Phosphorylation-fold-change of all amino acids compared to the average library dephosphorylation. At 1 the dephosphorylation rate of an amino acid is equal to the average dephosphorylation rate. For a third heat map comparing PP1c and PP2Ac, dephosphorylated peptides for each amino acid were divided and Nterm/Cterm heat-maps were normalized by the median. To allow better visualization by color coding, Pro in+1 was excluded. Fold-changes >1.2 with an adjusted p-value <0.01 according to Fisher’s exact test in the comparison of PP1c/PP2Ac are highlighted in bold. Please see the methods section and Source Data for details on statistics. c Analysis of the additive influence of positive and negative charges on dephosphorylation by PP1c/PP2Ac of Nterm/Cterm libraries. p-Values were derived by carrying out a two-sided Fisher’s exact test. Exact p-values are provided in the Source Data. d, e Kinetic analysis of the dephosphorylation rate of four synthetic peptides upon PP2Ac or PP1c treatment. Error bars represent s.e.m. of three independent repeats with technical duplicates. kcat/Km was calculated by comparison to a phosphate standard curve. Source data underlying Fig. 2a–e are provided as a Source Data file.