Vitrobot

For cryo-EM observation, GMPCPP-MT was polymerized by incubating 0.18 mg/mL GMPCPP-tubulin in PEM/NP40 buffer (80 mM PIPES pH 6.8, 1 mM EGTA, 1 mM MgCl₂, 0.01% NP40) for 30 min at 37 °C. For MTBD-High-MT complex, 3 μL of the GMPCPP-MT solution was applied to a glow-discharged, carbon-coated grid (Quantifoil R 1.2/1.3) inside a Vitrobot (Thermo Fisher Scientific) and incubated for 30 s at 22 °C and 100% humidity. The grid was washed once with 3 μL of 1.1 mg/mL MTBD solution and once with 3 μL of PEM/NP40 buffer (30-s incubation each), followed by blotting and vitrifying in liquid ethane. To observe the reduced form of MTBD in complex with MTs, GMPCPP-MT was polymerized by incubating 1.0 mg/mL GMPCPP-tubulin in PEM buffer (80 mM PIPES pH 6.8, 1 mM EGTA and 1 mM MgCl₂) for 30 min at 37 °C. MTBD-High, NP40 and DTT were added to the GMPCPP-MT solution, to the final concentration of 0.8 mg/mL MTBD-High, 0.4 mg/mL GMPCPP-MT, 0.01% NP40 and 1 mM DTT, and the mixtures were maintained for 2 h at room temperature. Disulfide bond reduction of MTBD-High was confirmed by SDS-PAGE analysis (Supplementary Fig. 1f). An aliquot (2 μL) of the specimen was applied to a glow-discharged, carbon-coated grid (Quantifoil R 1.2/1.3) inside a Vitrobot set at 6 °C and 100% humidity and incubated for 10 s before blotting and vitrifying in liquid ethane.

For imaging apo-RE-CmeB, a 20 μM RE-CmeB-nanodisc sample was directly applied to glow-discharged holey carbon grids (Quantifoil Cu R1.2/1.3, 300 mesh), blotted for 18 s, and then plunge-frozen in liquid ethane using a Vitrobot (Thermo Fisher). For imaging the RE-CmeB-Cip, RE-CmeB-Ery, or RE-CmeB-Chl, a 20 μM RE-CmeB-nanodisc sample was incubated with 1 mM Cip, Ery, or Chl for 2 h on ice to form the corresponding RE-CmeB-drug complexes. The samples were then applied to glow-discharged holey carbon grids (Quantifoil Cu R1.2/1.3, 300 mesh), blotted for 18 s, and then plunge-frozen in liquid ethane using a Vitrobot. All of the grids were then transferred into cartridges prior to data collection.

Purified hOCT1 in DDM (13 mg/mL) or LMNG (10 mg/mL) micelles were incubated with substrate metformin at 100 mM or the inhibitor spironolactone at 0.5× saturated concentration for 1 h at 4 °C. For hOCT1-DDM samples, 4 μL of purified protein was applied to glow-discharged (25 mA for 45 s) cryo-EM grids (Quantifoil NiTi 1.2/1.3 300 mesh). The blotted grids (5 s, 8 °C, 100% humidity) were rapidly frozen in liquid ethane cooled by liquid nitrogen with Vitrobot (FEI Mark IV, Thermo Fisher Scientific). For hOCT1-LMNG samples, a 4 μL aliquot of purified protein was applied to glow-discharged (25 mA for 30 s) cryo-EM grids (Quantifoil Au 1.2/1.3 200 mesh) and plunge-frozen with Vitrobot (4 s, 8 °C, 100% humidity). The grids were transferred to a 300 kV Titan Krios equipped with Gatan K3 Summit detector or Falcon 4 electron detector with GIF Quantum energy filter (slit width 20 eV). Micrographs of hOCT1-Nb5660-metformin were recorded with a defocus range of –1.1 to –1.5 μm and a pixel size of 0.86 Å. The total dose was about 50 e^–/Ų for each stack. For the other samples, data collection was done using EPU data acquisition software in super resolution mode with a calibrated pixel size of 0.54 Å and a defocus range of –1.1 to –1.5 μm. All 32 frames in each stack were aligned and summed using the MotionCor2 and binned to a pixel size of 1.08 Å^{48 (link)}.