Ran ^GTP interaction with Kap114, Nap1 ₂ , and H2A-H2B mixtures. (A) The full gel of one of the two binding assays shown in Fig. 1 B, which is also similar to assays in Fig. S1, D and E. MBP-Kap114 (1 µM) was immobilized with equimolar Nap1₂ FL ± Sc or Xl H2A-H2B ± Ran^GTP. Quantification of the Kap114-bound Nap1 band intensities from triplicate experiments is shown on the right. ** indicate P value <0.01. Student’s t test was performed two-tailed and unpaired. Data distribution was assumed to be normal but it was not formally tested. MBP-Kap114 pulled down more Nap1 when both H2A-H2B and Ran^GTP are present. This result is consistent with SEC-MALS data showing a smaller shoulder at the ∼12.2 ml peak of the Nap1₂•H2A-H2B•Kap114•Ran^GTP trace (green; likely due to Kap114•H2A-H2B) than in the Nap1₂•H2A-H2B•Kap114 trace (orange) in Fig. 1 C. (B) The same pull-down assay as in A, except 3 µM Ran^GTP was added after pre-assembly of the MBP-Kap114/Nap1₂/H2A-H2B (K:N₂:H) complex and immobilization on beads. Individual steps of the binding assay are visualized by SDS-PAGE. The quaternary Kap114/Nap1₂/H2A-H2B/Ran^GTP complex formed regardless of the order of protein addition, and no Nap1 or H2A-H2B was dissociated by Ran^GTP. (C) AUC analysis of Kap114-Nap1 interaction in the presence of Ran^GTP. Plots of the c(s) distributions of unliganded proteins, Kap114 (K, 5.2 S), Nap1₂ core (N₂, 4.9 S) and Ran^GTP (R, 2.1 S), and mixtures of the proteins with indicated molar ratios. The species corresponding to the individual peaks are labeled. Ran^GTP binding increased sedimentation coefficient similarly for unliganded Kap114 (K, 5.2 → K•R, 6.3S) and for Kap114•Nap1₂ (K•N₂, 7.3 → N₂•K•R, 8.3S). The Kap114•Ran^GTP and Nap1₂•Kap114•Ran^GTP complexes had estimated molecular weights of 138 and 206 kDa, respectively, consistent with equimolar complexes. (D) Left: SEC chromatogram of the Kap114/Nap1₂/H2A-H2B/Ran^GTP sample. Kap114, Nap1₂, H2A-H2B, and Ran^GTP were mixed in equimolar ratio and dialyzed overnight before mild crosslinking. Fractions 1–6 of the SEC chromatogram are colored red to purple, and the typical elution volume of a non-crosslinked complex that contains all four proteins is marked with a dotted line. Right: Mass photometry traces of each of the six SEC fractions, with mean masses (kDa) and relative populations (%) indicated above the fitted gaussian peaks, along with the likely protein or complex that correspond to the approximate masses. Fraction 3, most enriched with the complex containing Kap114, Nap1₂, H2A-H2B and Ran^GTP, was used for cryo-EM grid preparation. The largest (∼380–410 kDa) species in each of the six SEC fractions may contain crosslinked complexes of K/R/N₂/H with K•R; such a large assembly was not observed in AUC or SEC-MALS analyses where the proteins were not crosslinked. (E) Particle distribution of cryo-EM data obtained for the quaternary Kap114/Ran^GTP/Nap1₂/H2A-H2B complex. Blob picking was used first and then the particles containing Kap114 and Ran^GTP were used for Topaz training. Topaz-picked particles were cleaned up and submitted for 3D reconstruction to obtain four maps. The population with density for H2A-H2B and Nap1₂ (albeit poor density) was used for non-uniform (NU; gray map; EMD-44151) and local refinement (cyan map; EMD-44150) to obtain the final maps. The maps were overlayed onto the final Nap1₂•H2A-H2B•Kap114•Ran^GTP structure (9B3I; in cyan•yellow-red•gray•green). Below is the consensus map colored by local resolution. (F) Phenix map-to-model FSC curves for the composite map (EMD-44141). (G) 3D angular distribution of the particles used for reconstruction. Left orientation is the same as in E. (H) Directional FSCs unmasked and masked by cryoSPARC refine mask, for the consensus map. Source data are available for this figure: SourceData FS2.