Cytoplasmic dynein is a protein complex that transports molecular cargo along microtubules (MTs), playing a key role in the intracellular trafficking network. Vertebrate dynein’s movement becomes strikingly enhanced upon interacting with dynactin and a cargo adaptor such as BicaudalD2. However, the mechanisms responsible for increased transport activity are not well understood, largely owing to limited structural information. We used cryo-electron tomography (cryo-ET) to visualize the 3D structure of the MT-bound dynein–dynactin complex from Mus musculus and show that the dynactin–cargo adaptor complex binds two dimeric dyneins. This configuration imposes spatial and conformational constraints on both dynein dimers, positioning the four motor domains in proximity to one another and oriented toward the MT minus end. We propose that grouping multiple dyneins onto a single dynactin scaffold promotes collective force production, increased processivity, and unidirectional movement, suggesting mechanistic parallels to axonemal dynein. These findings provide structural insights into a previously unknown mechanism for dynein regulation.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
We thank J. C. Ducom at The Scripps Research Institute High Performance Computing for computational support and B. Anderson at The Scripps Research Institute EM facility for microscope support. We also thank E. Mattson for input on the manuscript. D.A.G. is supported by a National Sciences Foundation predoctoral fellowship. G.C.L. is supported by The Searle Scholars Program, The Pew Scholars Program, and by the National Institutes of Health (NIH) DP2EB020402. R.J.M. is supported by the National Institutes of Health (NIH) R00 grant R00NS089428 and R35GM124889. T.A.S. is supported by the Johns Hopkins Krieger School of Arts and Sciences. Computational analyses of EM data were performed using shared instrumentation funded by NIH S10OD021634 to G.C.L. R.J.M. provided purified SNAPf-Hook3.
Integrated supplementary information
Supplementary Figure 1 3D subtomogram average of the microtubule-bound dynein–dynactin–BICD2N complex
(a-c) Three views of the subtomogram average (gray density) of the MT-DDB complex are shown in orientations corresponding to the fitted model shown in Fig. 1a-c. All panels are on the same scale, scale bar in (a) corresponds to 10 nm.
(a) Representative X-Y slices of a reconstructed tomogram progressing through the z-axis, displaying MTs with associated DDB complexes, and three representative complexes highlighted by cyan dashed squares. (b) Representative X-Y slices progressing through the z-axis of extracted subvolumes demarcated by the colored dashed squares in (a). These slices display several distinct features of DDB complexes, such as dynein dimer-1 motor domains (red arrow head), dynein dimer-2 motor domains (orange arrow head), dynein motor stalk (black arrow), dynein linker arm extension (red arrow), dynactin barbed end (black bracket), dynactin pointed end (red bracket) and dynactin shoulder domain (red asterisk). Scale bar in (a) represents 100 nm. (c) Subtomogram average of the DDB-MT complex with fitted atomic models (as shown in Fig. 1) oriented to match the DDB complexes shown in the corresponding extracted subvolumes, with a dotted box depicting the region shown above.
(a) Cryo-EM reconstruction of the TBD complex (EMDB 28607) was docked into the DDB-MT and DDH-MT sub-volumes using UCSF Chimera10. Selected subvolumes (gray density) are shown with docked TBD complex shown in blue. (b) The docked densities provided the initial Euler parameters for RELION 3D autorefinement using a limited angular search, yielding subtomogram averages of the DDB and DDH complexes. (c) Focused 3D refinements were performed using local 3D ellipsoidal binary masks (transparent yellow) corresponding to dynein motor domains (left box) and dynein tail-adaptor-dynactin region (right box). In order to further resolve dimeric dynein heads belonging to each dynein (Dyn-A and Dyn-B), local masking and refinement was performed. Masking of dynactin did not lead to substantial improvement of corresponding density (right box). (d) Focused subvolume averages of different subregions of DDB-MT complex were aligned and stitched together using UCSF Chimera10 to generate a composite map of the full complex. Orthogonal views of the composite map are shown.
(a) 3D refinement focusing on the dynactin-dynein tail region of DDB-MT subvolumes results in this region becoming better resolved, while the density corresponding to the dynein motor domains worsens due to misalignment (left image). Re-refinement of the poorly-resolved dynein heads with ellipsoid binary mask (transparent yellow) results in better resolved density for the dynein motor domains, and poorly-resolved density dynactin-dynein tail (middle image). Continuing refinement with a binary mask around the dynactin-dynein tail density restores this density to a well-resolved state (right image), similar to that of the original 3D refinement of this region (left image), suggesting that the assisted subtomogram averaging procedure followed by focused 3D refinement does not impose any form of model bias to the final subtomogram averages. (b) Assigning starting Eulers from the manually-docked TDB complex density (EMDB 28607) into the DDB-MT sub-volumes results in convergence to a density that corresponds to the dynactin-dynein tail-BICD2N region of the DDB-MT complex (top panel). Random assignment of the starting Eulers for the TDB complexes using three different seed models (Seed #1, #2, #3) results in 3D reconstructions that do not resemble any distinguishable feature of the DDB-MT complex, further suggesting that the assisted subtomogram averaging procedure does not introduce model bias to the final subtomogram averages of the dynein-dynactin-cargo adaptor complexes.
Supplementary Figure 5 Focused subtomogram averages of the DDB–MT complex were resolved to ~38-Å resolution
(a-c) Fourier shell correlation plots of individual focused reconstructions are shown with resolution reported at 0.5 FSC and the reconstructed maps shown to the right of the FSC curves. The dynactin-dynein tail region was resolved to 38Å (a), and Dyn-A (b) and Dyn-B (c) were both resolved to 38Å.
Supplementary Figure 6 Two dimeric dyneins bind to a single dynactin in the presence of cargo adaptor protein BICD2N
(a) A difference map (gold), calculated by subtracting TDB density (EMD-28627) from focused subtomogram average of the dynactin-dynein tail-BICD2N region (gray), displays a density corresponding to a second dynein tail (left panel). A difference map (gold), calculated by subtracting dynactin density (EMD-62908) (gray) from focused subtomogram average of the dynactin-dynein tail-BICD2N region, displays a density corresponding to two dynein tails associated with dynactin (right panel). (b) Left panel shows two views of the initial refinement of the DDB-MT complex after assisted subtomogram averaging, highlighting the presence of four poorly resolved motor densities (circled in red). Right panel shows same views as left panel of the combined subtomogram average after focused refinement of all components (dynactin, dynein motor domains) of DDB complex for comparison. (c) Focused 3D classification of the dynein motor domains of the DDB-MT complex using an ellipsoid binary mask (transparent pink) results in one well-resolved 3D class (Class 10), containing majority of the subvolumes (74.3%), and shows the presence of four dynein motor domains corresponding to two dimeric dyneins. (d) Comparison of 2D averages of DDB-MT complex8 (left image) with DDB-MT’s dynactin-dynein tail-BICD2N subvolume average (light yellow density), by overlaying the 3D density map onto the 2D average, shows a high correlation between the two (middle image). Segmentation of the 3D subvolume average shows the presence of two dynein tails (light yellow and gold) associated with a single dynactin (blue) in DDB-MT complex (right image).
(a) Focused subvolume average (map shown in gray) of the dynactin-dynein tails-Hook3 region of the DDH-MT complex was resolved to 38Å as shown in the FSC plot. (b) SDS-Polyacrylamide gel electrophoresis (SDS-PAGE) of microtubule-bound dynein-dynactin-Hook3 complex purified from mouse brain shows that all components of the complex are present, including dynein and dynactin subunits, as well as Hook3 and tubulin. (c) A Difference map (gold), calculated by subtracting dynactin density (EMD-62908) (gray map) from focused subtomogram average of the dynactin-dynein tail-Hook3 region, displays a density corresponding to two dynein tails associated with dynactin.
(a) Histogram showing distribution of distances between motor domains of dimeric dyneins in DY-MT complexes indicates that the majority are spaced 10-15nm apart, however a wide range of distances is also observed (5-40nm), suggesting a high degree of flexibility of dimeric dynein in the absence of dynactin and cargo adaptors. Large distances (>25nm) may result from dyneins observed to be bridging neighboring MTs. (b) Representative X-Y slices from extracted subvolumes of DY-MT complexes with the center of two motor domains marked with green dots, which were used to measure the 3D motor-motor distances shown in (a). Only motors that were visibly connected as a dimer were used for distance measurements. Scale bar in (b) represents 20nm distance.
Supplementary Figure 9 Similarities between the organization of cytoplasmic and axonemal dynein motor domains
(a) Density corresponding to dynein motor domains with linker arm extensions from each of the dynactin-associated dynein dimers (Dyn-A, left column; Dyn-B, right column) (gold density) correlates well with the motor domains (gray density) present in the axonemal dynein map (EMD-575721).
Subtomogram average of the dynein–dynactin–BICD2N complex.