Due to its key role in dysregulated cellular signaling in cancer, AKT has been subject to intense drug discovery efforts leading to small molecule inhibitors investigated in clinical trials, notably breast cancer. To shed more light on how these drugs exert their therapeutic effects, we integrated chemoproteomic target affinity profiling using the kinobeads approach and phosphoproteomics for the five designated AKT inhibitors AZD5363, GSK2110183, GSK690693, Ipatasertib and MK-2206 in breast cancer cells. Kinobead analysis identified between four and 29 nanomolar targets for these compounds and revealed that AKT1 and AKT2 were the only common targets. Similarly, measuring the response of the phosphoproteome to the same inhibitors identified ~1,700 regulated phosphorylation sites among which 276 were regulated by all drugs. This analysis expanded the AKT signaling network by 119 phosphoproteins that may represent direct or indirect targets of AKT. Within this network, we found 41 regulated phosphorylation sites bearing the AKT substrate motif and recombinant kinase assays validated 16 as novel AKT substrates. These included CEP170 and FAM83H, which suggest a direct regulatory function of AKT in mitosis and cytoskeleton organization. In addition, a specific phosphorylation pattern on ULK1, FIP200, ATG13 and VAPB was found to represent an active state of ULK1, leading to elevated autophagy in response to AKT inhibition. And last, secretome analysis of AKT inhibitor-treated cells identified STC2 as a putative blood-based drug response marker that may be tested in patients in the future.

In order to obtain the spectral library, respective raw files were searched in MaxQuant with carbamidomethylated cysteine as a fixed modification; and phosphorylation of serine, threonine, and tyrosine, oxidation of methionine, and N-terminal protein acetylation as variable modifications. Trypsin/P was specified as the proteolytic enzyme and up to two missed cleavage sites were allowed. Precursor and fragment ion tolerances were 10 ppm and 4.5 ppm, respectively. The PRM raw data files were imported into the Skyline software package (version 4.1.0.11714). Confident phosphopeptide identification was carried out based on iRT information using spiked-in retention time peptides, matching relative transition intensities between the PRM peak and the library MS2 spectrum (if available) and requiring site determining ions for each phosphorylation site. For accurate phosphopeptide quantification, low‐quality or interfered transitions were discarded manually. Peaks were integrated using the automatic peak finding function followed by the manual curation of all peak boundaries. By calculating the sums of all transition areas associated with the phosphopeptide, quantification of each phosphopeptide was accomplished. Peak areas were normalized based on the total MS1 intensity across the experiments. Log2 fold changes for each inhibitor against the vehicle control were calculated per phosphopeptide. The median log2 fold change of the 13 quantified p-peptides for each inhibitor were computed to obtain a simple AKT perturbation score.

After treatment with designated AKT inhibitors, BT-474 cells were washed twice with PBS and lysed in 40mM Tris-HCl pH 7.6, 8 M urea, EDTA-free protease inhibitor (Roche) and phosphatase inhibitors (Roche). Lysate was centrifuged for 1 h at 21,000 x g und the supernatant was subjected to sample preparation. 2 mg protein of BT-474 lysate was reduced with 10 mM DTT for 40 min at 56 °C and alkylated with 25 mM CAA at room temperature in the dark for 20 min. After dilution of the urea concentration from 6 M to 1.5 M with 40 mM Tris-HCl pH 7.6, the proteins were digested in a 1:100 trypsin/substrate weight ratio for 4 h at 37 °C and 700 rpm. The second digestion step was performed overnight again in a 1:100 trypsin/substrate weight ratio in presence of 1 mM CaCl2. Desalting of tryptic peptides was performed on Sep-Pak C18 50 mg columns (Waters) as described elsewhere in 0.07 % TFA in 50 % ACN. Phosphopeptides were enriched using Fe-IMAC as previously described. Phosphopeptides were further desalted by using 0.07 % TFA in 50 % ACN in the micro-column format (three discs, Ø 1.5 mm, C18 material, 3M Empore per micro-column were used) as described. After adding an equivalent of 500 fmol Pierce Retention Time Calibration Mixture (Thermo Fisher Scientific) for each MS injection to each PRM sample, sample preparation for label-free PRM assays has been complete at this point.