Multivalency, autoinhibition, and protein disorder in the regulation of interactions of dynein intermediate chain with dynactin and the nuclear distribution protein

  1. Kayla A Jara
  2. Nikolaus M Loening
  3. Patrick N Reardon
  4. Zhen Yu
  5. Prajna Woonnimani
  6. Coban Brooks
  7. Cat H Vesely
  8. Elisar J Barbar

  • Curated by eLife

    eLife logo

    Evaluation Summary:

    The identification of an autoinhibitory mechanism of the dynein intermediate chain provides an important contribution to our understanding of dynein assembly and illustrates the plethora of regulatory mechanisms attainable by intrinsically disordered proteins. This paper provides insight into the autoinhibited inactive state of dynein as well as the activation mechanism. A wide range of biophysical approaches is used, providing a very nice example of how these diverse technologies can be applied in concert and in a synergistic manner to study an important question in the realm of "unstructured biology".

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. The reviewers remained anonymous to the authors.)

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

As the only major retrograde transporter along microtubules, cytoplasmic dynein plays crucial roles in the intracellular transport of organelles and other cargoes. Central to the function of this motor protein complex is dynein intermediate chain (IC), which binds the three dimeric dynein light chains at multivalent sites, and dynactin p150 Glued and nuclear distribution protein (NudE) at overlapping sites of its intrinsically disordered N-terminal domain. The disorder in IC has hindered cryo-electron microscopy and X-ray crystallography studies of its structure and interactions. Here we use a suite of biophysical methods to reveal how multivalent binding of the three light chains regulates IC interactions with p150 Glued and NudE. Using IC from Chaetomium thermophilum , a tractable species to interrogate IC interactions, we identify a significant reduction in binding affinity of IC to p150 Glued and a loss of binding to NudE for constructs containing the entire N-terminal domain as well as for full-length constructs when compared to the tight binding observed with short IC constructs. We attribute this difference to autoinhibition caused by long-range intramolecular interactions between the N-terminal single α-helix of IC, the common site for p150 Glued , and NudE binding, and residues closer to the end of the N-terminal domain. Reconstitution of IC subcomplexes demonstrates that autoinhibition is differentially regulated by light chains binding, underscoring their importance both in assembly and organization of IC, and in selection between multiple binding partners at the same site.

Article activity feed

  1. Version published to 10.7554/elife.80217 on eLife
  2. eLife

    Evaluation Summary:

    The identification of an autoinhibitory mechanism of the dynein intermediate chain provides an important contribution to our understanding of dynein assembly and illustrates the plethora of regulatory mechanisms attainable by intrinsically disordered proteins. This paper provides insight into the autoinhibited inactive state of dynein as well as the activation mechanism. A wide range of biophysical approaches is used, providing a very nice example of how these diverse technologies can be applied in concert and in a synergistic manner to study an important question in the realm of "unstructured biology".

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. The reviewers remained anonymous to the authors.)

  3. eLife

    Reviewer #1 (Public Review):

    Jara et al studied the interaction between the dynein intermediate chain (IC) and the three dimeric light chains (Tctex, LC7, LC8), dynactin p150, and the nuclear distribution protein (NudE). The authors are able to produce the entire intrinsically disordered N-terminal domain of IC from Chaetomium thermophilum (CT) allowing them to study the assembly and regulatory mechanism of IC with five different partners.

    The authors convincingly demonstrate that IC is maintained in an auto-inhibitory conformation through NMR titrations of separate constructs of IC. Using a combination of NMR, ITC, SV-AUC, SEC, and SEC-MALS, they demonstrate that release of this auto-inhibitory conformation, through binding of LC7, is required for binding to NudE and to some extent p150. Importantly, the presence of this auto-inhibited state is validated in the context of the full-length IC protein expressed and purified from insect cells.
    The work provides novel insight into how dynein assembly is regulated (Fig. 10) and illustrates the unique interaction mechanisms that can be exploited by intrinsically disordered proteins.

    The conclusions of the manuscript are supported, for most parts, by experimental data, however, some aspects require some clarification or should be further supported by experimental data:

    1. The authors propose that the two-step binding isotherm observed for p150 is due to binding to both the SAH and H2 regions of IC(1-88), while NudE shows a single binding event by ITC due to interaction with the SAH region only. The NMR experiments of IC(1-88) do not provide sufficient support for this hypothesis (Fig. 6B, bottom panel). Additional experimental data would be needed to fully support this conclusion.

    2. ITC, NMR, and AUC data are presented for the binding of NudE to IC(1-260). Some more clarification is needed in terms of the interpretation of these data, also in the context of the observations on IC(FL). The experimental observations do not seem to be explainable simply by a weak complex or a concentration-dependent effect as suggested by the authors.

    3. The difference in sedimentation coefficient of the dynein subcomplex + NudE and of the dynein subcomplex + p150 is surprisingly large suggesting significantly different shapes of the two bound complexes. Some discussion of this issue is present in the manuscript, but no clear explanation is provided. It would seem necessary to confirm these observations with other complementary techniques.

    4. The authors suggest that the binding of LC7 releases the auto-inhibitory interaction of IC, however, NMR does not directly support this conclusion (Fig. 7B). Some discussion of why this long-range interaction inhibits the binding of NudE, but not LC7 itself, should be included.

  4. eLife

    Reviewer #2 (Public Review):

    All protein components are highly pure. The authors characterize their oligomerization state by AUC and SEC-MALS and characterize interaction sites by NMR spectroscopy. They generate a convincing model, in which the light chains stabilize intermediate chain dimers in a ladder-like topology. P150Glued or NudE can further bind to this complex. The authors also provide insight into the formation of sub-complexes and which components are necessary for binding of p150Glued and NudE. Specifically, p150Glued can overcome the autoinhibited state of intermediate chain monomers (which is identified in this manuscript), whereas NudE cannot and requires binding of at least the light chain Lc7 first.

    The work in this manuscript provides detailed insight into the assembly. It does not provide enough background information for non-dynein expert readers to understand the function of this complex. Or why it has to assemble in the way it does. More information about the requirement of individual chains for function, why disorder may be beneficial, and whether any disease mutants are known to disrupt function, may help. If this is not included, the authors risk that their careful work goes unnoticed.
    The Discussion claims that the model derived from the totality of the data "illustrates the importance of autoinhibition in dynein regulation". This is not the case. The reader does not know why any of the assembly steps, nor the final structure is important.

    On the flip side, the authors can considerably streamline their manuscript by eliminating repetition. The discussion does not require a retelling of all detailed results.

  5. eLife

    Reviewer #3 (Public Review):

    The manuscript by Jera and coworkers describes an internal long-range interaction within the dynein intermediate chain, which can be relieved by light chain binding to provide access for additional protein ligands, and partly by binding of specific protein ligands. The work uses a suite of biophysical methods including AUC, SEC-MALS, and NMR spectroscopy, and a palette of protein constructs and complexes to assess complex sizes and stoichiometries, pinpointing by NMR the molecular details. The molecular auto-inhibition is supported by the data and is likely to be of general interest. The strength of the manuscript is the use of full-length proteins/longer regions and thus the investigation of higher-order complexes within context, which have been crucial to elucidate an important and likely biologically relevant autoinhibitory state in dynein as well as its modulation.

    The manuscript by Jera et al is in general very well written, the experiments have been thoroughly conducted and analyzed, and the conclusions are generally well supported by data. The work delivers important new insight into a case where disordered linkers may enable molecular functions. However, the significance of the finding for the biological function of dynein is not clear. How is it anticipated that the observed differential autoinhibition of dynein will affect the biological outcomes?

    Below are some recommendations that I find may improve the manuscript.

    As a non-dynein expert, I found the introduction into the protein system to be too superficial and the model shown in Fig. 1B, did not help much (e.g. the light chains were hard to acknowledge as they appear to be rather small compared to the IC chain and was at first overlooked at just the binding sites; where is the heavy chain of dynein, why is there no coiled coil of p150, etc?). The biological role of dynein is not explained particularly well in the introduction and the biological relevance of the findings is too briefly addressed. I suggest a much more detailed description of the system at the beginning of the introduction including the biological relevance of the different ligands, which should include an upgrade of figure 1b, with more detail on domains, etc. Also, the abstract would benefit from a more precise description of the biological question and why this study is relevant, and the title is also very broad. Finally, how autoinhibition plays a role in the biological function of dynein should be more clearly discussed in the discussion, e.g., what is the relevance of the differential binding of the two ligands and their differential effects on the autoinhibited state. Which biological outcomes are to be expected?

    One of the conclusions is that the internal contacts occur between the C-terminal of the IC and SAH/H2, which is seen from the intensity changes in the HSQC spectra upon addition of the 160-240 construct to the 1-88 construct. However, adding the linker part from 100-160 produces a much more pronounced effect (Fig. 5C, bottom), suggesting that residues in this region, which includes the Tctex and LC8 binding motifs play additional roles. Is the binding of the light chains to IC of higher affinity to the 100-160 protein than to the 1-260? In that case, this could suggest that also inhibitory access to these two sites occurs in the autoinhibited state. The additional effect of the 100-160 residues should be addressed.

    Can H3 be excluded as a player in the internal interactions, just because you see binding to the LC7 site when studied in isolation? Once the LC7 regions is bound, H3 may also participate, as also clearly indicated from the data shown in Fig. 4A. Using an H3 peptide would be relevant.

    Fig6B and associated text: there is a clear although weak loss in intensity/peak volume in the H2 region for the interaction with NudE. Why assume that there is no interaction? The affinity for NudE is lower, so the concentration of the complex will also be lower at similar conditions compared to that of p150, and this would give rise to the smaller effects in the spectra. In the lower panel, there is a clear indication of binding to H2 as well, and SAH and H2 binding may very well be cooperative as they are sequentially close. What are the relative concentrations of NudE and p150 in the cell? Would they be competitive despite the difference in affinities? Can a mechanism for p150 ability to relieve autoinhibition be proposed - from Fig6B, could it be able to bind first to the H2 region even though SAH is involved in the autoinhibitory interaction?

  6. Version published to 10.1101/2022.06.05.494868v1 on bioRxiv