extracted_captions/fasler2020novel.json

{"CAPTION FIG S1.png": "'\\n\\n**Fig. S1.: Cryo-ET of the actin network in a NIH3T3 fibroblast lamellipodium. (a) Computational slice through a binned, non-CTF corrected tomogram (17.096A/px) of a NIH3T3 fibroblast lamellipodium. Protein density is black. The dense actin network is visible and the helical appearance of individual actin filaments can be clearly appreciated. Several branch junctions are highlighted in red circles. The dashed line indicates the cell edge. Scale bar is 100nm. (b) Gallery of selected branch junctions. The density corresponding to the Arp2/3 complex at branch junctions is visible in the individual examples.**'", "CAPTION FIG S2.png": "'\\n\\n**Fig. S2.: Graphical workflow of image processing.** Flow chart indicating the data processing steps involved in generating the structure of the active Arp2/3 complex within the branch junction. Colored boxes indicate usage of specific software packages. For simplicity, the procedure to produce a reference for template matching via manual picking and averaging branch junctions is not depicted here, but is described within the material and methods section.\\n\\n'", "CAPTION FIG S5.1.png": "'\\n\\n**Fig. S5.: Comparison of the active Arp2/3 complex conformation in branch junctions in cells to the _in vitro_ Dip1-activated Arp2/3 complex and to a previously published MD-derived _in vitro_ branch model. (a) Molecular models of the Arp2/3 complex in the active conformation (shown as density maps filtered to 9.5A resolution) as observed in cells (left, this study), the _in vitro_ Dip1-activated Arp2/3 complex (middle) [29], and a MD-derived active Arp2/3 complex [12], which is based on a low-resolution negative stain ET reconstruction (right) [9]. The models were aligned on the ArpC2 subunits to visualize the different conformations of the Arp2/3 complex and in case for the branch junction also the varying position on the mother filament between the _in situ_ and _in vitro_'", "CAPTION FIG S5.2.png": "'model. **(b-c)** RMSD values (in A) calculated between the subunits of the three models shown in **(a)**. Rows indicate which subunit was used for aligning full models to each other, prior to measurements between individual subunits (indicated in the columns) of the different models of the active Arp2/3 complex. **(b)** RMSD values of differences between the Arp2/3 complex in branch junctions in cells and the _in vitro_ Dip1-activated Arp2/3 complex. In order to calculate C-alpha RMSD values between our model and pdb 6W17 only primary structure areas are considered, in which the _B. Taurus_ (as used in our model) and _S. pombe_ protein sequences are in the same register, hence omitting inserts present in only one species. This comparison reveals differences between the Dip1-activated Arp2/3 complex in vitro and the activated state of the Arp2/3 complex in cells, in particular with respect to ArpC3. **(c)** RMSD values of differences between the Arp2/3 complex in branch junctions in cells and the MD-derived _in vitro_ branch junction model.\\n\\n'", "CAPTION FIG S6.png": "'\\n\\n**Fig. S6.: Mother actin filament conformation in the actin-Arp2/3 complex branch junction. (a) Fit of the final model of the mother actin filament after MD-refinement into the EM density of the branch junction using ISOLDE. The empty density close to M4, corresponding to the helix of ArpC1 is annotated with a green ellipsoid. (b) Superimposition between the starting model used for fitting (pdb 6T20, pink) and the final model after MD-refinement. Small deviations between the filament assemblies can be observed. (c) Superimposition between one monomer of pdb 6T20 (pink) and monomer M4 of the mother filament in our branch junction model. No large-scale deviations in the monomer conformation are observed. (d) RMSD calculations between the C-alpha atoms of one monomer of pdb 6T20 and the monomers of the mother filament. The average RMSD value is 1.76A.**'", "CAPTION FIG.S4.png": "'\\n\\n**Fig. S4.: Comparison of the short pitch actin dimer conformation to Arp2 and Arp3 in their inactive and active conformation.** Density maps of molecular models of an actin dimer, or of Arp2 and Arp3 in their active and inactive conformation, respectively, filtered to a resolution of 15A. The models of the active Arp2/3 conformation were derived from our model of the actin filament Arp2/3 complex structure in cells described in this manuscript and the _in vitro_ Dip1-activated _S. pombe_ Arp2/3 complex (pdb 6W17) [29]. The shown models of filamentous actin were derived from pdb 6T20 [30], and for the inactive ATP-bound Arp2/3 complex from pdb 1TYQ [8]. Maps were oriented by fitting Arp3 subunits and the actin monomer M1 to each other. While Arp2 and Arp3 of the active complex adopt a similar short pitch conformation as the monomers of the actin dimer, this is not the case for the subunits of the inactive complex. Subunit identity is indicated by the color scheme.\\n\\n'", "CAPTION FIG1.png": "'\\nFigure 1: **Subnanometer Structure of the actin filament Arp2/3 complex branch junction in cells.****(a)** Isosurface representation of the actin filament Arp2/3 complex branch junction in cells at 9\u00c5 resolution. The structure is shown from three orientations. A guide for orientation is given in (b). **(b)** The electron microscopy density map (shown transparent) with the flexibly fitted models of the Arp2/3 complex subunits and actin filaments. Schematic guides indicate the positions of the individual subunits of the complex and the monomers of both the mother and daughter filaments. A colour legend provides the color code for the fitted models. The color code and legend are used throughout the manuscript to aid the reader.\\n\\n'", "CAPTION FIG2.png": "'\\n\\n**Fig. 2: Comparison of Arp2/3 complex in its inactive conformation and in the branch junction in cells. (a) Molecular models of the Arp2/3 complex in the inactive (derived from pdb 1TYQ) and the active conformation shown as density maps filtered to 9.5A resolution. The models are shown from three orientations, corresponding to the views in figure 1. (b) RMSD values (in A) calculated between the inactive and active conformations of the individual Arp2/3 subunits (based on the models used in (a)). Rows indicate which subunit was used for aligning the full models against each other, prior to measurements between individual subunits of the inactive and active Arp2/3 complexes (indicated in the columns). The RMSD analysis reveals that the structural transition upon complex activation is accommodated by two subcomplexes consisting of Arp2, ArpC1, ArpC4 and ArpC3, ArpC2 and ArpC3, respectively, that rotate against each other along an axis formed by the large helices of ArpC2 and ArpC4 (Movie S4). RMSD variations between the same subunits in the inactive and active conformations derive from changes upon MD-based modelling of the x-ray crystal structure-derived model into the electron microscopy density map of the branch junction.**'", "CAPTION FIG3.png": "'\\n\\n**(a)** Interaction surfaces between the Arp2/3 complex and the actin mother and daughter filament. The surfaces for the Arp2/3 complex, mother and daughter filament are shown as density maps at 9.5A resolution generated from their respective models. Grey coloring on the surface of Arp2/3 subunits indicates contact sites with actin and coloring of actin monomers in a specific color indicates a contact site with the associated Arp2/3 subunit (color code is given within the figure). Coloring was applied via the color zone command in ChimeraX in a 5.5A radius around each C-alpha atom of the underlying model, which was positioned in a 10A radius to a C-alpha atom of its putative interactor. **(b)** Surface representation of the interaction of ArpC2 and ArpC4 with the mother filament monomers M6 and M7 (the interaction of ArpC2 with M5 is omitted for clarity). **(c)** The protrusion helix of ArpC1 fitted into its density close to subdomains 1 and 3 of M4. **(d)** Surface representation of the interactions of Arp3 and ArpC3 with the mother filament. Note the cavity below Arp2, where no contacts between the Arp2 subunit and the mother filament are observed. ArpC3 acts as a linker between Arp2 and the mother filament. **(e)** Surface representation of the interaction between Arp2, Arp3 and the first two monomers of the daughter filament.\\n\\nFig. 3: **Actin-Arp2/3 complex interaction surfaces within the branch junction**\\n\\n**(a)** Interaction surfaces between the Arp2/3 complex and the actin mother and daughter filament. The surfaces for the Arp2/3 complex, mother and daughter filament are shown as density maps at 9.5\u00c5 resolution generated from their respective models. Grey coloring on the surface of Arp2/3 subunits indicates contact sites with actin and coloring of actin monomers in a specific color indicates a contact site with the associated Arp2/3 subunit (color code is given within the figure). Coloring was applied via the color zone command in ChimeraX in a 5.5\u00c5 radius around each C-alpha atom of the underlying model, which was positioned in a 10\u00c5 radius to a C-alpha atom of its putative interactor. **(b)** Surface representation of the interaction of ArpC2 and ArpC4 with the mother filament monomers M6 and M7 (the interaction of ArpC2 with M5 is omitted for clarity). **(c)** The protrusion helix of ArpC1 fitted into its density close to subdomains 1 and 3 of M4. **(d)** Surface representation of the interactions of Arp3 and ArpC3 with the mother filament. Note the cavity below Arp2, where no contacts between the Arp2 subunit and the mother filament are observed. ArpC3 acts as a linker between Arp2 and the mother filament. **(e)** Surface representation of the interaction between Arp2, Arp3 and the first two monomers of the daughter filament.\\n\\n'", "CAPTION FIG4.png": "'\\n\\n**Fig. 4: Structural changes in Arp3 and ArpC5 upon branch junction formation**\\n\\n**(a)** Comparison of the conformation of the Arp3 C-terminal tail in the inactive and active conformation. No density is observed for the C-terminal tail of Arp3 in its inactive conformation (top). Instead, the C-terminal tail can flip towards Arp2 and ArpC4, where it is accommodated by an empty density present in the branch junction structure (bottom). **(b)** No density is observed for the ArpC5 N-terminus at its binding side in the inactive complex (top). Instead, the ArpC5 N-terminus can be fitted into a density between ArpC1 and Arp2. **(c)** Positively charged residues in ArpC1 and Arp2 could coordinate the negatively charged N-terminus of ArpC5. **(d)** Electrostatic potential map of the area shown in **(c)**.\\n\\n'", "CAPTION FIGS1.png": "'\\n\\n**Table S1.**\\n\\n**Data acquisition and image processing parameters**'", "CAPTION FIGS3.png": "'\\n\\n**Fig. S3.: Structural details of the actin filament Arp2/3 complex branch junction.** Densities for the individual subunits of the Arp2/3 complex plus their fitted models are shown. Subunit colors are annotated and identical to the schematic guide given in Figure 1. Secondary structure detail (i.e. alpha helices) are clearly visualized, for example for Arp3, ArpC2 and ArpC4. The beta-propellers of ArpC1 allow unambiguous fitting of the subunit into the density of the branch junction. The increased apparent flexibility at the N-terminus of the ArpC5 helical core (annotated by an arrow) is visible, resulting in reduced density for the N-terminal helices of this subunit. The subnanometer resolution of our structure also allows to clearly visualize secondary structure details in the actin filament, further highlighted by the visibility of an additional density accommodating Phalloidin (annotated by an arrow).\\n\\n'", "CAPTION TABS2.png": "'\\n\\n**Table S2.**\\n\\n**Summary of model content**\\n\\nThe pdb files used to generate the final model of the active Arp2/3 complex are listed. Residues that had to be added or removed from the original models are indicated. Residue stubs in the original models were completed to contain their entire side chains for MD-modelling in ISOLDE.\\n\\n'", "CAPTION TABS3.png": "'\\n\\n**Table 53.**\\n\\n**Modelling parameters and statistics**'", "CAPTION TABS4.png": "'\\n\\n**Table S4. Arp2/3 complex residues contacting the actin mother and daughter filaments**\\n\\nSummary of residues forming interactions between the Arp2/3 complex and the actin filaments, defined by a 10A C-alpha to C-alpha distance cutoff. The UniProt identifiers for the individual proteins are given.\\n\\n'"}