NMR experiments

If not stated otherwise, NMR experiments were conducted on a Bruker spectrometer operating at 750 MHz, equipped with a triple-channel ¹H, ¹³C, ¹⁵N TCI cryoprobe, and z-shielded gradients. All data were processed using Topspin (Bruker) or NmrPipe (Delaglio et al., 1995) and analyzed with CCPNmr (Vranken et al., 2005). For NMR resonance assignments, samples of approximately 600 μM isotopically labeled R86A Akt PH domain (aa 1–121) in 25 mM HEPES pH 6.5, 500 mM NaCl, 1 mM TCEP, and 5% v/v ²H₂O were measured at 13°C. Due to the high ionic strength, rectangular-shaped sample tubes were used and effectively reduced the ¹H 90° pulse length and increased sensitivity significantly (de Swiet, 2005). Eighty-one percent of non-proline backbone resonances were assigned, significantly more than for the WT counterpart due to the apparent higher stability and rigidity of the R86A PH domain (Chu et al., 2020). Chemical shift assignments were deposited in the BMRB under accession code 51419. HSQC spectra of 100–120 μM segmentally isotopically labeled full-length R86A Akt were measured in 25 mM HEPES pH 6.5, 500 mM NaCl, 1 mM TCEP, and 5% v/v ²H₂O using TROSY-HSQC. Recycle delays of 1 s were used and approximately 768 transients (and 256 increments) were accumulated, leading to measurement times of 54 hr. Chemical shifts were calibrated using internal water calibration in NmrPipe. Combined CSPs and associated standard deviations were calculated according to the protocol from Schumann et al., 2007.

Nuclear spin relaxation (R₁, R₂, Eta_xy) were measured on a Bruker 800 MHz using standard pulse sequences and recycle delays of 5 s. R₂ relaxation employed Carr-Purcell-Meiboom-Gill (CPMG) pulse trains with varied delays and corresponding temperature compensation blocks. All experiments were acquired using single scan interleaving. R₂ relaxation delays were set to 10, 30, 50, 70, 90, 130, 150, 170 ms (50 ms was repeated twice). R₁ relaxation delays were set to 0.01, 0.05, 0.1, 0.2, 0.3, 0.5, 0.8, 1, 1.2, 1.5, 1.8 s (0.5 s was repeated twice). Cross-correlated (DD/CSA) relaxation rates were measured at a relaxation delay of 50 ms. For steady-state hetNOE, ¹H saturation was achieved with a train of 180° pulses at 12 kHz power for a delay of 6 s (>4⋅T₁^max), recycle delays of 10 s were employed. CPMG relaxation dispersion experiments were conducted with a T₂ relaxation delay of 40 ms. CPMG frequencies were modulated in the following order (in Hz): 0, 1000, 50, 850, 100, 650, 500, 250, 400, 100, 50, 850, 400, 1000. Dispersion curves were analyzed with the program Relax.

¹⁹F experiments were measured on a Bruker spectrometer operating at 600 MHz, equipped with a QCIF triple-channel ¹H, ¹⁹F, ¹³C, ¹⁵N TCI cryoprobe, and z-shielded gradients. Samples of approximately 25–50 μM R86A Akt PH domain modified with 3-¹⁹F-Tyr18 were measured, either isolated or after EPL with Akt kinase domain. Two other Tyr residues of the PH domain, Tyr26, and Tyr38, were mutated to Phe, hence the ¹⁹F signal can be assigned to Tyr18. ¹⁹F 1D spectra were measured with relaxation delays of 1 s, 1024 points in the direct dimension for a spectral width of 200 ppm (4.5 ms acquisition time), and 64k scans for a total experimental time of approximately 10 hr.