Solid-state NMR

Sample A (K⁺) was packed into a 1.9-mm solid-state NMR rotor (Bruker BioSpin), and 1 μl of D₂O was added to enable deuterium locking. To prevent leakage, a spacer was added between the bottom cap and the sample, and the drive cap was glued using a silicon-based glue. The washed samples [sample B (no ions) and sample C (washed to K⁺)] were packed into 1.3-mm rotors (Bruker BioSpin), and the drive caps were glued as well to prevent leakage. Experiments on sample A were recorded at 40-kHz magic angle spinning (MAS) using a four-channel (¹H, ¹³C, ¹⁵N, and ²H) 1.9-mm probe (Bruker BioSpin) on a spectrometer with an external magnetic field strength corresponding to 900 MHz (Bruker BioSpin). Four CP-based 3D experiments [(H)CANH, (H)CONH, (H)CACO(N)H, and (H)COCA(N)H] were recorded for backbone chemical shift assignments of sample A, closely following earlier work by Fricke et al. (24). The assigned chemical shifts confirm protein folding into a single dominant conformation. Spin diffusion [H(H)NH] experiments with mixing times of 5, 12, 20, 35, and 100 ms were recorded using a 3D version of the standard (H)NH experiment, with the addition of a ¹H acquisition period followed by a spin diffusion sequence on the ¹H channel (90° pulse, spin diffusion mixing time, 90° pulse) before the CP-based transfer to ¹⁵N. A recycle delay of 1 s and 90° pulses of 83.3, 50, and 35.7 kHz for ¹H, ¹³C, and ¹⁵N, respectively, were used for all experiments acquired on sample A. Experiments on samples B and C were recorded at 58- to 60-kHz MAS using a three-channel (¹H, ¹³C, and ¹⁵N) 1.3-mm probe (Bruker BioSpin) on a spectrometer with an external magnetic field strength corresponding to 600 MHz (Bruker BioSpin). For sample B, a 3D (H)CANH spectrum was recorded, and for sample C, 3D (H)CANH and (H)CONH spectra were recorded to identify new peaks compared to sample A. Spin diffusion (H(H)NH) experiments with mixing times of 20, 40, and 60 ms were recorded on sample C. A recycle delay of 1 s and 90° pulses of 100, 50, and 35.7 kHz for ¹H, ¹³C, and ¹⁵N, respectively, were used for all experiments acquired on samples B and C. The chemical shifts were referenced to DSS (4,4-dimethyl-4-silapentane-1-sulfonic acid) using an external reference. The sample temperature was kept at 25°C, as measured from the ¹H chemical shift of water. Proton detection with a 30-ms acquisition time was used for all experiments. Further experimental parameters for the 3D experiments are listed in tables S4 to S7.