Research ArticleMOLECULAR BIOLOGY

Allosteric modulation of nucleoporin assemblies by intrinsically disordered regions

See allHide authors and affiliations

Science Advances  27 Nov 2019:
Vol. 5, no. 11, eaax1836
DOI: 10.1126/sciadv.aax1836
  • Fig. 1 Nup53 uses distinct IDRs for binding to partner nups and Kap121.

    (A) A cross-sectional view of the NPC shows distinct nup subcomplexes linked by IDRs in adaptor nups (wavy lines). Kaps (green) facilitate NLS-mediated nuclear import. The zoomed-in section highlights the adaptor Nup157·Nup53·Nic96 complex that bridges the channel and membrane nup modules. (B) Domain architecture of Nup53, Nic96, Nup157, and Kap121 (left panel). Gray lines mark boundaries of the protein constructs used in this study. Table showing Nup53 fragments and the results of pull-down (#) and ITC (˄) binding assays (right panel). (C) Pull-downs using partially purified, GST-immobilized Nup53 fragments with Nic96, Nup157 (157N and 157C), or Kap121. Eluted GST·Nup53 complexes were visualized by SDS–polyacrylamide gel electrophoresis (PAGE) (M, molecular weight marker; GST-53C*, proteolytic truncation product). (D) ITC profiles for 53NTD-Nic96 and 53CTD-Nup157 interactions overlaid with Nic96 or 157N titrations with 53core. Integrated heat signals were analyzed by a single-site 1:1 binding model, yielding binding constants of 7.0 (± 2.2) × 105 M−1, 9.9 (± 0.9) × 105 M−1, 1.6 (± 0.2) × 106 M−1, and 1.1 (± 0.3) × 106 M−1 for 53NTD-Nic96, 53core-Nic96, 53CTD-Nup157, and 53core-Nup157 interactions, respectively. (E to G) ITC profiles for interactions of Kap121 (150 μM) with different Nup53 fragments (15 μM; blue). Monophasic binding data (E) were fitted with a single-site 1:1 binding model. The biphasic binding isotherms (F) were analyzed by either two-mode (53C) or multiple-equilibrium (RRM-53C and 53core) binding models (continuous lines). The best-fit parameters and the associated errors from global analysis are presented in Table 1. The RRM domain in Nup53 does not interact with Kap121 (G).

  • Fig. 2 Nup53 RRM and IDR domains are coupled when interacting with Kaps.

    (A) ITC profiles for 53C-Kap121 interactions at three different 53C concentrations. The isotherms were globally analyzed together with the reverse titration in fig. S2D by a two-mode binding model (continuous lines; best-fit parameters with SDs are reported in Table 1). (B) Population distribution of 1:1 and 2:1 53C·Kap121 complexes is displayed as a function of [Kap121]/[53C] molar ratio. (C) 53core monomer-dimer equilibrium at different concentrations analyzed by ITC using a single-step dissociation model (continuous lines). (D) Crystal structure of the yeast Nup53 RRM homodimer determined at a resolution of 1.75 Å. Individual monomers are shown with secondary structure elements. Key hydrophobic residues contributing to the interface are highlighted in the green monomer. Regions not observed in the electron density map are shown as dashed lines. (E) ITC profiles for 53core-Kap121 interactions at three 53core concentrations were globally analyzed by a multiple-equilibrium model (continuous lines; best-fit parameters with SDs are reported in Table 1). (F) Population distribution of the 53core·Kap121 complex is shown as a function of [Kap121]/[53core] molar ratio. (G) Negative-stain EM (NS-EM) analysis of Kap121 and its complex with 53core. Representative 2D class averages and 3D reconstitution of free Kap121 and the two distinct conformations of the 53core·Kap121 complex are displayed with fitted crystal structures of Kap121 (PDB code: 3W3T). The monomeric state of the 53core·Kap121 complex (middle panel) shows extra density that likely represents either a 53core monomer, dimer, or an average of the two conformations, as determined by ITC. To assign orientation of the two Kap molecules within the dimeric state of the complex (right panel), the experimental NS-EM density map for free Kap121 (purple and cyan) was fitted into each of the arms of the structure. The RRM domain of Nup53 (PDB code: 5UAZ) was docked into the stemlike portion of the structure.

  • Fig. 3 Kap121 allosterically destabilizes ternary 157N·53core·Nic96 complex.

    (A) SEC elution profiles of ternary 157N·53core·Nic96 complex alone (red trace) and preincubated with Kap121 (blue trace). Colored dots denote elution peak maxima, and dashed line rectangle represents the range of elution fractions visualized by SDS-PAGE on the right. (B to E) ITC profiles for interactions between (B) 53core and Nic96 in the absence or presence of 157N, (C) 53core and 157N in the absence or presence of Kap121, (D) Nic96 and 53core or equimolar 53core:Kap121 or 157N:53core:Kap121 mixtures, and (E) Nic96 and 53core or its equimolar mixture with HsKapβ1. The ITC binding profiles were analyzed by a single-site 1:1 binding model (continuous lines; best-fit parameters with SDs are reported in Table 1). (F) 3D population distribution plot of the 53core·Kap121·Nic96 complex. Increasing Kap121 concentration gradually destabilizes the 53core-Nic96 interactions—at a 1:1 ratio of 53core and Nic96, the fractional population of the 53core·Nic96 complex decreases from ~0.6 to 0.3 as the [Kap121]/[Nup53] molar ratio increases from 0 to 2. NS-EM models of 53core·Kap121 (gray) and 53core·Kap121·Nic96 (green) at a stoichiometry ratio of 2:2 are shown in top view with fitted crystal structures of Kap121 and the Nup53 RRM domain. Dashed ovals denote electron density differences between binary and ternary complex.

  • Fig. 4 Karyopherins modulate molecular architecture of nucleoporin assemblies.

    (A) Schematic representation of Nup53 hub interactions. The ternary complex containing Nup157, Nup53, and Nic96 is destabilized by Kap121 binding to the Nup53 C-terminal IDR. Kap121 allosterically reduces Nic96 affinity for the monomeric and dimeric conformations of the complex. Dissociation constants (Kds) for the different complexes were derived from Table 1. (B) Proposed model for allostery-driven Nup53 hub assembly in yeast. Kap121 binds to a cytoplasmic pool of Nup53 to destabilize its interactions with Nic96 and Nup157. In the nucleus, Ran-GTP releases Nup53. At the nuclear envelope, Nup53 sequentially recruits other adaptor nucleoporins. In the final step of NPC assembly, Nic96 recruits channel nucleoporins. (C) Classification of IDR-mediated interactions within macromolecular complexes. A hub protein consisting of a folded domain and IDR binding sites for three interacting partners is shown in the center. Different binding scenarios are considered in increasing order of complexity. By binding to chaperones (top left), IDRs target proteins to particular subcellular locations. As assemblers, IDRs independently recruit macromolecules or promote avidity-driven interactions. As effectors, IDRs are often regulated by cooperative binding and/or allostery, actively modulating biological activity of macromolecular complexes or signaling pathways. Coupling of folded and IDR domains may promote the formation of a structural ensemble or a phase-separated granule (dashed lines).

  • Table 1 Summary of ITC data.

    Detailed description of rows 1 to 5 is provided in Materials and Methods. ∆G° and TS° were determined by ∆G° = −RT lnK and TΔS = ΔH° − ΔG°, where R and T are gas constant (1.99 cal mol−1 K−1) and temperature (288 K), respectively. The errors are the SDs obtained from the global fits, pertaining to uncertainties in both the model accuracy and measurement precision (i.e., model and experimental errors). (1) 53core dissociation profiles obtained at three protein concentrations were analyzed by a monomer-dimer equilibrium model. (2) 53C·Kap121 forward and reverse titrations, performed at various protein concentrations, were globally analyzed by a two-mode binding model, yielding thermodynamic constants for 1:1 and 2:1 (53C:Kap121) interactions in rows 2a and 2b. These parameters also describe interactions between monomeric 53core and Kap121. 53core·Kap121 forward and reverse titrations, performed at various protein concentrations, were globally analyzed by a multiple-equilibrium model using constants in rows 1, 2a, and 2b. The obtained parameters in 2c and 2d describe 2:1 and 2:2 interactions between 53core dimer and one and two molecules of Kap121, respectively. A2*B and A2B complexes differ in the oligomeric state of A (two monomers versus dimer). (3 and 4) Nic96·53core and Nup157·53core binding profiles were analyzed by a single-site 1:1 binding model. (5) Nic96 titrations into stoichiometric 53core:Kap121 or 53core:HsKapβ1 mixtures were analyzed by a single-site 1:1 model to obtain apparent K’s (Kapp) for 53core-Nic96 interactions in the presence of Kaps (rows 5a and 5a′). Titrations were globally analyzed by a multiple-equilibrium model to deconvolute Nic96 binding to monomeric and dimeric 53core·Kap121 complexes (rows 5b and 5c, respectively). Parameters describing Nic96 interactions with the monomeric 53core·Kap121 complex were independently obtained using a dimerization-deficient Nup53mut fragment (table S1 and fig. S7, A and B). The subscript i in Kap121i denotes the number of bound Kap molecules (i = 1 or 2).

    NoteInteraction*KH° (kcal/mol)G° (kcal/mol)TS° (kcal/mol)
    12A ↔ A21.0 (± 0.1) × 104 M−16.5 ± 0.2−5.3 ± 0.111.8 ± 0.2
    2aA + B ↔ AB4.2 (± 0.7) × 106 M−1−2.7 ± 0.1−8.7 ± 0.16.0 ± 0.1
    2b2A + B ↔ A2*B1.9 (± 0.5) × 1011 M−24.2 ± 0.7−14.9 ± 0.219.1 ± 0.8
    2cA2 + B ↔ A2B5.6 (± 0.8) × 107 M−116.3 ± 1.4−10.2 ± 0.126.5 ± 1.4
    2dA2 + 2B ↔ A2B23.9 (± 1.0) × 1014 M−2−1.3 ± 0.4−19.2 ± 0.117.9 ± 0.4
    3A + D ↔ AD1.1 (± 0.3) × 106 M−1−10.1 ± 0.5−8.0 ± 0.2−2.1 ± 0.5
    4A + C ↔ AC9.9 (± 0.9) × 105 M−1−4.4 ± 0.1−7.9 ± 0.13.5 ± 0.1
    5a(AB)app + C ↔ (AB)appC1.3 (± 0.4) × 105 M−1−5.2 ± 0.6−6.7 ± 0.21.5 ± 0.6
    5a′(AB′)app + C ↔ (AB′)appC1.0 (± 0.1) × 105 M−1−5.1 ± 0.4−6.6 ± 0.11.5 ± 0.4
    5bAB + C ↔ ABC2.0 (± 0.9) × 105 M−1−6.7 ± 1.7−7.0 ± 0.30.3 ± 1.7
    5cA2Bi + C ↔ A2BiC1.2 (± 0.2) × 105 M−1−2.5 ± 0.3−6.7 ± 0.14.2 ± 0.3

    *53core (A), Kap121/HsKapβ1 (B/B′), Nic96 (C), and 157N (D).

    Supplementary Materials

    • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/5/11/eaax1836/DC1

      Fig. S1. Characterization of the N- and C-terminal regions of Nup53 and their interactions with Nic96, Nup157, and Kap121.

      Fig. S2. Kap121 interacts with the C-terminal region of Nup53 using the NLS and FG motif binding sites.

      Fig. S3. The RRM domain of human Nup53 forms a constitutive dimer in solution.

      Fig. S4. The RRM domain enhances biphasic profile of 53core-Kap121 interactions.

      Fig. S5. NS-EM analysis of Nup53-Kap121 interactions.

      Fig. S6. Reconstitution of Nup53 into complexes with Nic96 and Nup157.

      Fig. S7. Kap121 allosterically destabilizes Nic96 interactions with dimerization-deficient Nup53.

      Fig. S8. NS-EM analysis of the Kap121·53core·Nic96 complex.

      Table S1. Bacterial expression constructs.

      Table S2. Data collection and refinement statistics for the ScNup53 RRM domain (molecular replacement).

      References (5861)

    • Supplementary Materials

      This PDF file includes:

      • Fig. S1. Characterization of the N- and C-terminal regions of Nup53 and their interactions with Nic96, Nup157, and Kap121.
      • Fig. S2. Kap121 interacts with the C-terminal region of Nup53 using the NLS and FG motif binding sites.
      • Fig. S3. The RRM domain of human Nup53 forms a constitutive dimer in solution.
      • Fig. S4. The RRM domain enhances biphasic profile of 53core-Kap121 interactions.
      • Fig. S5. NS-EM analysis of Nup53-Kap121 interactions.
      • Fig. S6. Reconstitution of Nup53 into complexes with Nic96 and Nup157.
      • Fig. S7. Kap121 allosterically destabilizes Nic96 interactions with dimerization-deficient Nup53.
      • Fig. S8. NS-EM analysis of the Kap121·53core·Nic96 complex.
      • Table S1. Bacterial expression constructs.
      • Table S2. Data collection and refinement statistics for the ScNup53 RRM domain (molecular replacement).
      • References (5861)

      Download PDF

      Files in this Data Supplement:

    Stay Connected to Science Advances

    Navigate This Article