Fig 2: Reconstitution of RanBP2/SUM01-RanGAP1/Ubc9/Crm1/RanQ69L(GTP) complexes.A Components and domain organization: full length RanBP2 and fragments, Crm1, Ran GTPase, RanGAPI conjugated to SUM01, and the E2 Ubc9. RBD, FG, and IR domains indicated within RanBP2.B Schematic indicating multivalent protein-protein interactions in complexes assembled with RanBP2 variants containing RBD3 and RBD4 (Complex I), RBD3 (Complex II), or RBD4 (Complex III). Dashed lines indicate possible interactions.C Profiles obtained by size exclusion chromatography for complexes I, II and III shown in B. See Extended Data Fig. 1 for SDS-PAGE gels and fractions.

Fig 3: engineered cells expressing the RanGAP1ΔNES variant do not exhibit overt defects in nucleocytoplasmic transport – related to Figure 6.A Construct encoding 2xGFP-NES-IRES-2xRFP-NLS61 used to generate stable RPE1 cell lines in cell lines expressing wild type RanGAP1 or RanGAP1ΔNES.B Expected localization for GFP and RFP in cells exhibiting normal NCT.C Representative GFP and RFP average confocal z-stack images of live wild type RPE1 and RanGAP1ΔNES clone 2 cells in the absence (−) or presence (+) of 50 nM LMB. Brightness/contrast adjusted for individual channels.D Quantification of N/C ratios for GFP for wild type and RanGAP1ΔNES clone 2 cells in the absence (−) or presence (+) of 50 nM LMB. Individual points plotted and bars represent means +/− SD. Statistical significance calculated in the same way as in Fig. 6 (F and H). P-values as defined in Fig. 6 (F and H).E Quantification of N/C ratios for RFP for wild type and RanGAP1ΔNES clone 2 cells in the absence (−) or presence (+) of 50 nM LMB. Individual points plotted and bars represent means +/− SD. Statistical significance calculated in the same way as in Fig. 6 (F and H). P-values as defined in Fig. 6 (F and H).F Schematic of construct and experimental design for the LEXY assay.63 mCherry fused to a constitutive NLS signal and an NES bound to and masked by the AsLov2 domain localizes primarily to the nucleus in the dark (OFF). Upon illumination with 458 nm light (ON), the NES is exposed by disengaging from the AsLov2 domain resulting in the export of mCherry into the cytoplasm.G Example images of mCherry in live wild type RPE1, RanGAP1ΔNES clone 2, and RanGAP1ΔNES clone 10 cell lines stably expressing the LEXY construct before exposure (time = 0) and 5 minutes after pulsatile exposure (time = 300 seconds) to 458 nm light.H Representative data generated by quantifying nuclear mCherry fluorescence in LEXY cells at 2.57 second intervals during 458 nm light exposure (see Methods for details).I Plots of half-lives vs. initial nuclear fluorescence calculated from curves (fit to first order exponential models in Prism) such as shown in (H). Note that although the half-life for mCherry export increases with increasing initial mCherry nuclear fluorescence, no difference is observed for wild type and RanGAP1ΔNES cells across a wide range of initial nuclear mCherry fluorescence levels.J Schematic of construct and experimental design for the LiNUS assay.62 mCherry fused to a constitutive NES signal and an NLS masked by the AsLov2 domain localizes primarily to the cytoplasm in the dark (OFF). Upon illumination with 458 nm light, the NLS is exposed resulting in nuclear translocation of mCherry (ON).K Example images of mCherry in wild type RPE1, RanGAP1ΔNES clone 2, and RanGAP1ΔNES clone 10 cell lines stably expressing the LiNUS construct before exposure (time = 0) and 5 minutes after pulsatile exposure (time = 300 seconds) to 458 nm light.L Representative data generated by quantifying nuclear mCherry fluorescence in LiNUS cells at 2.57 second intervals during 458 nm light exposure (see Methods for details).M Plots of half-lives vs. final nuclear fluorescence calculated from curves (fit to first order exponential models in Prism) such as shown in (L). No difference is observed for wild type and RanGAP1ΔNES cells.

Fig 4: Cells expressing the RanGAP1ΔNES variant exhibit the same mitotic localization of RanGAP1, Crm1, and Ran as wild type cells and do not show overt cell cycle defects. Related to Figure 6.A Confocal microscopy images of fixed metaphase and anaphase wild type RPE1 (top) and RanGAP1ΔNES clone 2 cells (bottom) co-stained with α-RanGAP1 and α-Crm1 antibodies and counterstained by DAPI. RanGAP1 detected with rabbit α-RanGAP1 polyclonal primary antibody (Bethyl, A302–027A) and alpaca α-rabbit IgG-Alexa488 secondary nanobody (Thermo, SA5–10323), both at 1:1000 dilution. Crm1 detected with mouse monoclonal α-Crm1 antibody directly labelled with Alexa647 (Santa Cruz, SC74454) used at 1:40 dilution. Representative 2D images are shown; brightness/contrast adjusted for individual channels.B Confocal microscopy images of fixed metaphase and anaphase wild type RPE1 (top) and RanGAP1ΔNES clone 2 cells (bottom) co-stained with α-Ran and α-Crm1 antibodies and counterstained by DAPI. Ran detected with rabbit polyclonal α-Ran antibody (Proteintech, 10469–1-AP) at a 1:100 dilution and alpaca α-rabbit IgG-Alexa488 secondary nanobody (Thermo, SA5–10323) at 1:1000 dilution. Crm1 detected as in (A) but antibody used at 1:300 dilution. Representative z-stack average projections are shown; brightness/contrast adjusted for individual channels.C Representative FACS plots for wild type RPE1, RanGAP1ΔNES clone 2, and RanGAP1ΔNES clone 10 cell lines. Distribution of G0/G1, S, G2/M phases shown in BrdU vs. 7-AAD plots (top); cells in mitosis shown in BrdU vs. phosphohistone H3 plots (bottom). Boxed regions assigned to indicated cell cycle phases and cells within were quantified as indicated. Note that the indicated percentage of cells in mitosis is out of the G2/M population.D Quantification of FACS data from plots such as those shown in C. Data from individual biological replicates plotted. Bars represent means +/− SD. At least 3 biological replicates were analyzed for each cell line.

Fig 5: The N-terminal helix of RanGAPI encodes a nuclear export signal.A Schematics for the N-terminal fragments of RanGAP1 fused to monomeric EGFP (wild type left panel; mutant with 4 of 5 NES residues mutated to alanine right panel) shown above confocal microscopy images of stable cell lines as representative z-stack average projections to highlight cytoplasmic localization for wild type and nuclear localization for the mutant. Cells stained with DAPI; GFP visualized via native GFP fluorescence using 488 nm excitation. Brightness/contrast adjusted for individual channels.B Schematic of CRISPR-Cas9 strategy to generate an in-frame deletion of RanGAP1 residues 5–17 to remove the RanGAP1 NES.C Western blot analysis of wild type and clones (2, 3 and 10) expressing RanGAP1ΔNES to highlight expression levels and amounts of unmodified and SUMO1-modified RanGAP1. 20 μg of lysate was loaded in each lane and probed with a rabbit polyclonal α-RanGAP1 antibody (Sigma, HPA062034) at a 1:4000 dilution (top blot) and β-actin as a loading control (bottom blot) probed with a mouse monoclonal α-β-actin antibody (Santa Cruz, SC81178) at a 1:500 dilution.D Western blot analysis of wild type RanGAP1 or RanGAP1ΔNES lysates to highlight similar levels of phosphorylation after nocodazole treatment. 7.5 μg of lysate was loaded in each lane and blots were probed with rabbit polyclonal α-RanGAP1 antibody (Sigma, HPA062034) at 1:2000 dilution.E Confocal microscopy images for cells expressing wild type RanGAP1 or RanGAP1ΔNES to highlight RanGAP1 localization to the nuclear periphery in both cell lines independent of the RanGAP1 NES. White arrowheads point to prototypical nuclear ring staining for RanGAP1 in the nuclear focal plane (top panels, z-stack slice shown) and RanGAP1 puncti on the periphery of the nuclear envelope after translocating 1 μm in the z-plane (bottom panels, z-stack slice shown). RanGAP1 detected with rabbit polyclonal α-RanGAP1 primary antibody (Bethyl, A302–027A) and α-rabbit IgG-Alexa647 secondary VHH nanobody (Thermo, SA5–10327), both at 1:1000 dilution. Brightness/contrast adjusted for individual channels.F Confocal microscopy images for cells expressing wild type RanGAP1 or RanGAP1ΔNES (clones 2 and 10) in the absence or presence of leptomycin B (LMB) to highlight nuclear accumulation RanGAP1ΔNES relative to wild type RanGAP1. Cells were treated with methanol control (left panels, −) or 50 nM leptomycin B (LMB, right panels, +) for 20–21 hours, fixed, and stained for RanGAP1 using the same antibodies and dilutions as in (E). Representative average 2D projections of confocal z-stacks shown with brightness/contrast adjusted separately for each channel. Right graph depicts the ratio of nuclear to cytoplasmic RanGAP1 calculated using CellProfiler.73 Individual values and bars representing means +/− SD are shown. Kruskal-Wallis with Dunn’s multiple comparison tests were used to calculate statistical significance using Prism. n ≥ 66 cells across at least 3 biological replicates. * P < 0.05, ** P < 0.01; *** P < 0.001; **** P < 0.0001, ns – not significant.G Western blot analysis of Ran in cells expressing wild type RanGAP1 or RanGAP1ΔNES to highlight similar levels of expression. Ran (top blot) detected with a mouse monoclonal α-Ran antibody (BD, 610340) at 1:8000 dilution and β-actin (bottom blot) as loading control (bottom blot) detected with a mouse monoclonal α-β-actin antibody (Novus, NBP1–47423) at 1:300,000 dilution. 20 μg of lysate was loaded in each lane.H Confocal microscopy images with an α-Ran antibody in cells expressing wild type RanGAP1 or RanGAP1ΔNES (clones 2 and 10) in the absence or presence of leptomycin B (LMB) to highlight cytoplasmic mislocalization of Ran in cells expressing RanGAP1ΔNES with or without LMB or in wild type cells after LMB treatment. Ran detected with mouse α-Ran monoclonal primary antibody (BD, 610340) and α-mouse IgG-Alexa647 secondary antibody (H+L Thermo A32787) at 1:250 and 1:1000 dilutions, respectively. LMB treatment, imaging, and analysis as in (F). Right graph depicts results for n ≥ 73 across at least 3 biological replicates. p-values and statistical analysis as in (F). No significant difference between the means of Δ2 (−LMB), Δ10 (−LMB), WT (+LMB), Δ2 (+LMB), Δ10 (+LMB).