Structure of autoinhibited Akt1 reveals mechanism of PIP3-mediated activation

Abstract

The protein kinase Akt is one of the primary effectors of growth factor signaling in the cell. Akt responds specifically to the lipid second messengers phosphatidylinositol-3,4,5-trisphosphate [PI(3,4,5)P₃] and phosphatidylinositol-3,4-bisphosphate [PI(3,4)P₂] via its PH domain, leading to phosphorylation of its activation loop and the hydrophobic motif of its kinase domain, which are critical for activity. We have now determined the crystal structure of Akt1, revealing an autoinhibitory interface between the PH and kinase domains that is often mutated in cancer and overgrowth disorders. This interface persists even after stoichiometric phosphorylation, thereby restricting maximum Akt activity to PI(3,4,5)P₃- or PI(3,4)P₂-containing membranes. Our work helps to resolve the roles of lipids and phosphorylation in the activation of Akt and has wide implications for the spatiotemporal control of Akt and potentially lipid-activated kinase signaling in general.

Keywords: Akt; PIP3; kinase; lipid; signaling.

Fig. 1.

The PIP₃ binding site is sequestered in autoinhibited Akt. (A) Structure of autoinhibited Akt1 1 to 445 in complex with a nanobody. Cartoon schematic illustrates domain architecture of Akt1. Color scheme: PH domain, orange; N-lobe of kinase domain, yellow; C-lobe of kinase domain, magenta; C-tail of kinase domain, cyan. Dashed boxes indicate regions of disorder in the structure. (B) Superposition of Akt1 PH domain in complex with Ins(1,3,4,5)P₄ (PDB 1UNQ) with autoinhibited Akt1. Ins(1,3,4,5)P₄ shown in sticks. PIP₃-coordinating residues of PH domain shown in blue sticks (3′ phosphate, K14 and R25; 4’ phosphate, R86). (C) Interaction map of PH-kinase domain interface. Figure produced using Arpeggio (63). Dashed red lines, hydrogen bonds; red springs, cation-pi; blue springs, donor-pi; gray springs carbon-pi; dotted gray lines, hydrophobic van der Waals. (D) Map of disease-associated mutations (red, bold) and mutations that drive growth factor-independent cell survival in vitro (black). Mutations in PH domain shown in orange sticks; mutations in kinase domain shown in magenta sticks. (E) Superposition of structure of Akt1 in complex with inhibitor VIII (3O96) on autoinhibited Akt1. Inward rotation of PH domain indicated by 23° rotation of α1. Rmsd of PH domain over all atoms is 9 Å. (F) Superposition of active Akt1 kinase domain (4EKK) on autoinhibited Akt1. APE-αF loop of autoinhibited Akt1 shown as magenta mesh. Activation loop and APE-αF loop of active Akt1 shown in blue. Zoom: conformation of phosphorylated T308 in activation loop (red sticks) and network of stabilizing interactions.

Fig. 2.

Akt1 prepared by protein semisynthesis lacks a phosphorylated turn motif. (A) Phosphorylation state analysis of Akt1 after expressed protein ligation and in vitro phosphorylation with PDK1. Tandem mass spectrometry of pepsin digest. Missing turn motif phosphorylation highlighted in red. (B) Thermal stability analysis of Akt1 by differential scanning fluorimetry. Black curves, monophosphorylated Akt1^1P; red curves, diphosphorylated Akt1^2P. Solid lines, +1 mM ATP; dashed lines, no ATP. EPL, expressed protein ligation. (C) Liposome pelleting assay of Akt1^2P and Akt1^1P in the presence of 0% and 5% PIP₃-containing liposomes. (D) Akt1 kinase assay in the presence of liposomes containing 0 or 5 mol % PI(3,4,5)P₃, ± 10 μM MK-2206 (added postliposome binding). Diphosphorylated (T308/S473) Akt1^2P, red bars; Akt1^1P, black bars.

Fig. 3.

Phosphorylation does not override the requirement for PIP₃. (A) Phosphorylation state analysis of Akt1^3P after coculture with A-443654 and okadaic acid. Tandem mass spectrometry of GluC digest. Additional substoichiometric phosphorylation of S477 in 4P species indicated in red. (B) Liposome pelleting assay for 0% and 5% PIP₃ liposomes indicating the binding of Akt1^3P and Akt1^1P to PIP₃ in the kinase assay shown in F. (C) Akt1 kinase assay ± PIP₃ liposomes, ± 10 μM MK-2206. Akt1^3P (37.5 nM), black bars; Akt1^1P (750 nM), gray bars. (D) Kinase assay of Akt1^3P in the presence of liposomes containing increasing concentrations of PIP₃. Left axis, and blue lines correspond to kinase activity. Right axis and black lines correspond to % PIP3 binding (determined by a liposome pelleting assay). Squares, Akt1^3P; circles, Akt1^3P preincubated for 10 min with 10 μM MK-2206. PIP₃ binding and PIP₃-dependent increase in kinase activity were fit to one-site binding models, taking into account basal Akt1^3P activity and nonspecific binding in the presence of 0 mol % PIP₃ liposomes respectively. (E) Kinase assay of Akt1^3P and Akt1^3P R144A with increasing substrate concentration in the presence of liposomes containing 0 or 5 mol% PIP₃. Error bars indicate the SD of three independent measurements.

Fig. 4.

Phosphorylation alone does not drive Akt into an active conformation. (A) Size-exclusion profile of Akt1^3P from in-line SEC-SAXS data collection. Gray bar indicates the region of the chromatogram evaluated in the SAXS data processing. (B) SAXS curve of Akt1^3P. Radius of gyration (R_g) derived from Guinier analysis of the low-angle scattering regime. (C) Guinier plot of the low-angle SAXS regime for Akt1^3P. (D) Pair distribution function for Akt1^3P, indicating the radius of gyration (R_g) and maximum dimension of the particle (D_max). (E) Hydrogen-deuterium exchange mass spectrometry analysis of Akt1^3P in the presence of liposomes containing 0 or 5 mol % PI(3,4,5)P₃. Regions of Akt1^3P that showed significant increases or decreases in exchange (meeting the three criteria: ≥6% change in exchange, ≥0.4 Da difference in exchange, and a P value <0.01 using a two-tailed Student’s t test) upon PIP₃ binding are mapped on the structures of the PH domain and the active kinase domain (PDB ID 4EKK) (16) with the corresponding color scheme. (F) Plot of differences in deuterium incorporation upon PIP₃ binding. Changes in deuterium incorporation are plotted against the center of each peptide. Regions of protection and deprotection are indicated above the plot and correspond to those mapped in E. Error bars indicate the SD of three independent replicates. Red data points indicate increases or decreases in exchange that passed the three significance criteria. (G) Plot of changes in deuterium incorporation upon ATPγS binding. Changes in deuterium incorporation are plotted against the center of each peptide. Regions of protection and deprotection are indicated above the plot. Error bars indicate the SD of three independent replicates. No changes were deemed significant according to the three significance criteria.

Fig. 5.

PIP₃ binding promotes Akt hydrophobic motif exposure. (A) SAXS scattering curve for Akt1^ΔC. (B) Pair-distribution function (PDF) for Akt1^ΔC indicating the radius of gyration (R_g) and maximum dimension of the particle (D_max). (C) HDX-MS of Akt1^1P ± PIP₃ liposomes. Exposure (deprotection) of N-lobe peptide 218 to 225 indicated in red on structure of kinase domain. Plot: deuterium incorporation as a function of time for Akt1^1P ± PIP₃ liposomes. Deuterium incorporation plots were reproduced with permission. Adapted from ref. , which is licensed under CC BY-NC-ND 4.0. (D) HDX-MS of Akt1^1P versus Akt1^ΔC. Two regions showed significant increases in exchange (meeting the three criteria: ≥6% change in exchange, ≥0.4 Da difference in exchange, and a P value <0.01 using a two-tailed Student's t test). Regions 171 to 183 (YAMKILKKEVIVA) in the N-lobe and 260 to 274 (HSEKNVVYRDLKLEN) in the C-lobe are indicated in red on the structure of the kinase domain. Deuterium incorporation plots for these peptides as a function of time for Akt1^1P and Akt1^ΔC are shown to the right. (E) Plot of changes in deuterium incorporation between Akt1^1P and Akt1^ΔC. Changes in deuterium incorporation are plotted against the center of each peptide. Regions of protection and deprotection are indicated above the plot and correspond to those mapped in D. Error bars indicate the SD of three independent replicates. Red data points indicate increases or decreases in exchange that passed the three significance criteria. (F) Fluorescence anisotropy binding assay for C-terminal tail peptide (FITC-SMEAVDSERRPHFPQFSYSASGTA) to Akt1^ΔC. The K_D was estimated from three independent titrations. Each data point is the mean of 50 technical replicates with an integration time of 1 s. Error bars indicate the SD from the mean. Data were fit to a one-site binding model. (G) Composite model of full-length Akt1. The model comprises autoinhibited Akt1 (PH domain, PH-kinase linker, and kinase domain C-lobe), the N-lobe of active Akt1 (4EKK), and the phosphorylated C-terminal regulatory tail of PKCι (4DC2). The inactive conformation of the activation loop (unknown) is indicated with dashed magenta lines. (H) Stepwise activation of Akt by PIP₃ and phosphorylation. Activating steps are indicated with green arrows. Inactivating steps are indicated with red arrows. Phosphorylation state of each species in the activation and inactivation cycle is indicated in the blue boxes for each of the three regulatory residues: T308, T450, and S473.