Conformational and oligomeric states of SPOP from small-angle X-ray scattering and molecular dynamics simulations

  1. F Emil Thomasen
  2. Matthew J Cuneo
  3. Tanja Mittag
  4. Kresten Lindorff-Larsen

  • Curated by eLife

    eLife logo

    eLife assessment

    In this important paper, the authors have developed an approach for simultaneously optimizing the conformational ensemble and degrees of oligomerization, and this has been tested by applying it to a specific protein (SPOP). Comparison of the quality of fits with different models also provides valuable insights into structural features important to the assembly of oligomers. The approach, presented with compelling experimental support, is potentially applicable to other systems as well.

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Speckle-type POZ protein (SPOP) is a substrate adaptor in the ubiquitin proteasome system, and plays important roles in cell-cycle control, development, and cancer pathogenesis. SPOP forms linear higher-order oligomers following an isodesmic self-association model. Oligomerization is essential for SPOP’s multivalent interactions with substrates, which facilitate phase separation and localization to biomolecular condensates. Structural characterization of SPOP in its oligomeric state and in solution is, however, challenging due to the inherent conformational and compositional heterogeneity of the oligomeric species. Here, we develop an approach to simultaneously and self-consistently characterize the conformational ensemble and the distribution of oligomeric states of SPOP by combining small-angle X-ray scattering (SAXS) and molecular dynamics (MD) simulations. We build initial conformational ensembles of SPOP oligomers using coarse-grained molecular dynamics simulations, and use a Bayesian/maximum entropy approach to refine the ensembles, along with the distribution of oligomeric states, against a concentration series of SAXS experiments. Our results suggest that SPOP oligomers behave as rigid, helical structures in solution, and that a flexible linker region allows SPOP’s substrate-binding domains to extend away from the core of the oligomers. Additionally, our results are in good agreement with previous characterization of the isodesmic self-association of SPOP. In the future, the approach presented here can be extended to other systems to simultaneously characterize structural heterogeneity and self-assembly.

Article activity feed

  1. Version published to 10.7554/elife.84147 on eLife
  2. eLife

    Author Response

    Reviewer 1 (Public Review):

    Protein oligomerization is essential to their in vivo function, and it is generally challenging to determine the distribution of oligomeric states and the corresponding conformational ensembles. By combining coarse-grained molecular dynamics simulations and experimental small-angle X-ray scattering profiles at different protein concentrations, the authors have established a robust approach to self-consistently determine the oligomeric state(s) and the conformational ensemble. The approach has been applied specifically to the speckle-type POZ protein (SPOP) and generated new insights into the conformational ensemble and structural features that determine the ensemble. The model was further tested by the analysis of several relevant mutants as well as models with different types of structural restraints. The results also support the isodesmic selfassociation model, with KD values comparable to those measured from independent experiments in the literature. The approach is potentially applicable to a broad set of systems.

    We thank the reviewer for taking the time to assess our work.

    Reviewer 2 (Public Review):

    This manuscript applied the SAXS data analysis of protein selfassembly by implementing the simultaneous fitting of intra- and intermolecular motions/conformations against SAXS data at a series of oligomerization states/concentrations. Despite several major assumptions hinted, a diverse pool of conformational and oligomeric candidates was generated from CG simulations, and more importantly, these candidates were fitted into these SAXS data to reach a reasonable agreement, suggesting a somewhat convergence (even if the ensemble-fitting could well be at a local minimal). This is considered a technical advance, given the fairly large numbers of both the oligomer fraction phi_i (i=1, ..., N) and the conformational weight w_k (k=1, ..., n), where N is the number of oligomers and n is the number of internal conformational states.

    We thank Prof. Yang for taking the time to assess our work.

    Central is optimizing phi_i and w_k, simultaneously. The former has been illustrated in Fig. 4 and SI-Fig. 7 for the total number of 60mers. The latter relies on an overfitting-preventing strategy, as shown in SI_Fig. 1, where an effective fraction cutoff was used from 0.1 to 1.0, as opposed to the number of conformational states. What are the numbers of conformational states for these oligomers? This should be quantifiable, e.g., defining the conformational differences by chi_2.

    The reviewer is correct that the entropy-based term for preventing overfitting is a key aspect of the method. In contrast to some of the other methods to combine experiments with simulations, our approach does, however, not require us to define individual conformational states. Instead, the weights in the entropy term refer to individual configurations rather than states, and we can thus integrate the SAXS experiments and simulations without, for example, clustering the conformations. Indeed, for most of the collective variables that we have calculated from the ensembles, such as the radii of gyration, end-to-end distances, and MATH-MATH distances, we observe continuous monomodal probability distributions, which suggests that it might be difficult to define a few distinct conformational states. For the MATH-BTB/BACK distance, we observe a trimodal distribution, and these distinct conformational states are shown as overlaid structures in Fig. 4i. Thus, while these “states” change populations during reweighting, this is the result from changing weights of the individual configurations.

    Reviewer 3 (Public Review):

    Molecular-level interpretations of SAXS data are challenging, especially for oligomeric systems of variable length with intrinsic flexibility and the possibility of multiple association interfaces. In order to make this challenge tractable, a number of assumptions are made here: 1) There is a single pathway by which individual domains associate first into homodimers and then into longer oligomers; 2) the association kinetics is isodesmic, which allows the direct calculation of oligomer distributions based on the given value of a single dissociation constant; 3) the internal dynamics within dimers is restricted essentially to relative domain-domain motions, that are sampled comprehensively via MD simulations. As a result, excellent fits to the SAXS data are obtained and the underlying conformational ensembles are highly plausible. The resulting models are useful to further understand SPOP function, especially in the context of liquidliquid phase separation.

    We thank the reviewer for taking time to read our work and for their various suggestions.

  3. eLife

    eLife assessment

    In this important paper, the authors have developed an approach for simultaneously optimizing the conformational ensemble and degrees of oligomerization, and this has been tested by applying it to a specific protein (SPOP). Comparison of the quality of fits with different models also provides valuable insights into structural features important to the assembly of oligomers. The approach, presented with compelling experimental support, is potentially applicable to other systems as well.

  4. eLife

    Reviewer #1 (Public Review):

    Protein oligomerization is essential to their in vivo function, and it is generally challenging to determine the distribution of oligomeric states and the corresponding conformational ensembles. By combining coarse-grained molecular dynamics simulations and experimental small-angle X-ray scattering profiles at different protein concentrations, the authors have established a robust approach to self-consistently determine the oligomeric state(s) and the conformational ensemble. The approach has been applied specifically to the speckle-type POZ protein (SPOP) and generated new insights into the conformational ensemble and structural features that determine the ensemble. The model was further tested by the analysis of several relevant mutants as well as models with different types of structural restraints. The results also support the isodesmic self-association model, with KD values comparable to those measured from independent experiments in the literature. The approach is potentially applicable to a broad set of systems.

  5. eLife

    Reviewer #2 (Public Review):

    This manuscript applied the SAXS data analysis of protein self-assembly by implementing the simultaneous fitting of intra- and inter-molecular motions/conformations against SAXS data at a series of oligomerization states/concentrations. Despite several major assumptions hinted, a diverse pool of conformational and oligomeric candidates was generated from CG simulations, and more importantly, these candidates were fitted into these SAXS data to reach a reasonable agreement, suggesting a somewhat convergence (even if the ensemble-fitting could well be at a local minimal). This is considered a technical advance, given the fairly large numbers of both the oligomer fraction phi_i (i=1, ..., N) and the conformational weight w_k (k=1, ..., n), where N is the number of oligomers and n is the number of internal conformational states.

    Central is optimizing phi_i and w_k, simultaneously. The former has been illustrated in Fig. 4 and SI-Fig. 7 for the total number of 60-mers. The latter relies on an overfitting-preventing strategy, as shown in SI_Fig. 1, where an effective fraction cutoff was used from 0.1 to 1.0, as opposed to the number of conformational states. What are the numbers of conformational states for these oligomers? This should be quantifiable, e.g., defining the conformational differences by chi_2.

  6. eLife

    Reviewer #3 (Public Review):

    Molecular-level interpretations of SAXS data are challenging, especially for oligomeric systems of variable length with intrinsic flexibility and the possibility of multiple association interfaces. In order to make this challenge tractable, a number of assumptions are made here: 1) There is a single pathway by which individual domains associate first into homodimers and then into longer oligomers; 2) the association kinetics is isodesmic, which allows the direct calculation of oligomer distributions based on the given value of a single dissociation constant; 3) the internal dynamics within dimers is restricted essentially to relative domain-domain motions, that are sampled comprehensively via MD simulations. As a result, excellent fits to the SAXS data are obtained and the underlying conformational ensembles are highly plausible. The resulting models are useful to further understand SPOP function, especially in the context of liquid-liquid phase separation.

  7. Version published to 10.1101/2022.10.08.511432v1 on bioRxiv