Structure of the AAA protein Msp1 reveals mechanism of mislocalized membrane protein extraction
- Howard Hughes Medical Institute, United States
- Department of Biochemistry and Biophysics, University of California, San Francisco, United States
- Centre for Integrative Biology, Department of Integrated Structural Biology, IGBMC, CNRS, Inserm, Université de Strasbourg, France
- UCSF/UCB Graduate Program in Bioengineering, University of California, San Francisco, United States
Figures
Figure 1 with 6 supplements
Architecture of the Msp1-substrate complexes.
(A to C) Final reconstructions of Δ30-Msp1 (open), Δ30-Msp1 (closed) and Δ30-Msp1E214Q complexes shown in top and side views. Each subunit (M1 to M6) is assigned a distinct color, and the substrate …
Figure 1—figure supplement 1
Sequence comparison of AAAMC ATPases to the mitochondrial AAA proteases suggests structural similarity within the meiotic clade.
The sequences of five major members of the meiotic clade AAA (AAAMC) proteins are aligned by the Clustal Omega software (Sievers et al., 2011). From top to bottom are fidgetin (H. sapiens), Msp1 (C. …
Figure 1—figure supplement 2
Primary sequence alignment of Msp1 homologs showing conserved structural elements.
The primary sequences of Msp1 from D. melanogaster; ATAD1 from M. musculus, H. sapiens, and Msp1 from S. cerevisiae and C. thermophilum are aligned using Clustal Omega (Sievers et al., 2011). The …
Figure 1—figure supplement 3
The SEC traces of the Δ30-Msp1 and the Δ30-Msp1E214Q proteins.
The SEC traces of both proteins show that the wild-type Msp1 (red line) forms a homogeneous oligomer on the gel filtration column, whereas Msp1 with a Walker B mutation (blue line) forms higher …
Figure 1—figure supplement 4
Data analysis scheme of the Δ30-Msp1 structures.
(A) Representative micrograph showing the quality of data used for the final reconstruction of the Δ30-Msp1 (open) and Δ30-Msp1 (closed) structures. (B) Data processing scheme showing the 2D and 3D …
Figure 1—figure supplement 5
Data analysis flow for the Δ30-Msp1E214Q structure.
(A) Representative micrograph showing the quality of the data used to generate the 3D reconstruction. (B) Data processing scheme showing that the RELION software was used for 2D classification, and …
Figure 1—figure supplement 6
Structure of the larger oligomer shows potential steric clash between the additional subunit and the TMD of existing subunits.
Figure 2 with 1 supplement
Structural details of the LD.
(A) Cryo-EM map of Δ30-Msp1E214Q showing the arrangement of the fishhook motifs in the spiral. M1-M5 shows significant density for the entire fishhook motif (α0 and the L1), whereas M6 shows density …
Figure 2—figure supplement 1
Peptide array and molecular modeling suggest Msp1’s substrate specificity.
(A) A peptide array showing the binding of Δ30-Msp1E214Q to the selected peptides. (B) Amino acid fold-enrichment is plotted against their hydrophobicity scale (Wimley and White, 1996). (C) As one …
Figure 3 with 2 supplements
Msp1 interacts with the substrate via unique pore-loops.
(A) Cut-away view of the Δ30-Msp1E214Q map showing the substrate density (highlighted in white dashed lines) in the central pore. (B) Cartoon representation of the three pore-loops. Pore-loop one is …
Figure 3—figure supplement 1
Structural details of the pore-loops’ interactions with the substrate.
(A) Pore-loops 1 (W187 and Y188) form a staircase around the peptide. (B) Pore-loops two from M2-M5 are well ordered, and H227 from M2-M4 directly contact the peptide backbone. The cryo-EM density …
Figure 3—figure supplement 2
Pore-loop two forms a web of interactions with pore-loops 1 and 3 from subunits on both sides.
View of the pore-loops in the Δ30-Msp1E214Q structure showing that pore-loop 2 interacts with pore-loop 1 through π - π stacking and pore-loop 3 via electrostatic and polar interactions. The π - π …
Figure 4
Yeast growth assays.
Yeast growth assay showing mutations in the pore-loops, the WD motif and the ISS disrupt Msp1’s activity in vivo. All mutations are introduced to the S. cerevisiae Msp1 (S.c. Msp1) in the get3Δ msp1Δ…
Figure 5 with 1 supplement
The NCL communicates the nucleotide-bound state between adjacent subunits.
(A) The cryo-EM map of Δ30-Msp1 showing that the NCLs interacting with the ATP-bound subunits (M2–M4) are well ordered, whereas those interacting with the ADP (M5) or the Apo (M1) subunits are …
Figure 5—figure supplement 1
Structural details of nucleotide binding pockets in the Δ30-Msp1 (closed) complex.
The cryo-EM density of the entire nucleotide binding pockets with all the subunits in all three structures (except for M1 in Δ30-Msp1, which has poor density due to its movement) is shown. In the …
Figure 6 with 3 supplements
Mechanistic model for Msp1-mediated peptide extraction.
(A) Model for Msp1’s mechanism illustrated in three major steps. The Msp1 models on the left and the middle are of Δ30-Msp1E214; the right one is generated by rotating the Δ30-Msp1E214Q model …
Figure 6—figure supplement 1
Comparison of ISS and NCL.
(A), (B), and (C) show the schematic views of the Yme1 (PDB ID: 6azo), Vps4 (PDB ID: 6ap1) and Msp1 hexamers, showing the nucleotide-bound states of each subunit. The mobile subunit (M1) is shown in …
Figure 6—figure supplement 2
Cryo-EM map of the ISS motif in M1.
(A) Cryo-EM map of the Δ30-Msp1 shown at σ = 6.0. The M1 subunit is colored in red. (B) Zoom-in of the ISS region of M1 in panel (A). (C) Cryo-EM map of the Δ30-Msp1 shown at σ = 4.0. (D) Zoom-in of …
Figure 6—figure supplement 3
Overlay of the Δ30-Msp1 (closed) to the Vps4-substrate complex structures.
(A) M2 of the Δ30-Msp1 (closed) structure is superimposed to the corresponding subunit of the Vps4-substrate complex structure. As indicated by the arrow, Msp1 forms a tighter ring around the …
Videos
Video 1
The linker domain (LD) of Msp1.
Video 2
The substrate interactions in the central pore.
Tables
Table 1
Data collection, reconstruction, and model refinement statistics.
| Data collection | |||
|---|---|---|---|
| Δ30-Msp1E214Q | Δ30-Msp1 | ||
| Microscope | Titan Krios | Titan Krios | |
| Voltage (keV) | 300 | 300 | |
| Nominal magnification | 22500x | 22500x | |
| Exposure navigation | Stage shift | Stage shift | |
| Electron exposure (e-Å−2) | 70 | 70 | |
| Exposure rate (e-/pixel/sec) | 7.85 | 7.85 | |
| Detector | K2 summit | K2 summit | |
| Pixel size (Å) | 1.059 | 1.059 | |
| Defocus range (μm) | 0.6–2.0 | 0.6–2.0 | |
| Micrographs | 1443 | 2502 | |
| Total extracted particles (no.) | 502534 | 902573 | |
| Reconstruction | |||
| Δ30-Msp1E214Q | Δ30-Msp1 (closed) | Δ30-Msp1 (open) | |
| EMDB ID | 20320 | 20318 | 20319 |
| Final particles (no.) | 45687 | 48861 | 29723 |
| Symmetry imposed | C1 | C1 | C1 |
| FSC average resolution at 0.143/0.5, unmasked (Å) | 4.6/8.2 | 4.1/7.8 | 6.8/9.6 |
| FSC average resolution at 0.143/0.5, masked (Å) | 3.5/4.0 | 3.1/3.6 | 3.7/4.1 |
| Applied B-factor (Å) | 89.9 | 83.7 | 70.8 |
| Final reconstruction package | cryoSPARC v0.55 private beta | ||
| Local resolution range | 2.8–6.0 | 2.5–5.5 | 2.5–6.0 |
| Refinement | |||
| PDB ID | 6PE0 | 6PDW | 6PDY |
| Protein residues | 1672 | 1469 | 1660 |
| Ligands | 10 | 11 | 13 |
| RMSD Bond lengths (Å) | 0.003 | 0.003 | 0.002 |
| RMSD Bond angles (o) | 0.685 | 0.671 | 0.639 |
| Ramachandran outliers (%) | 0.06 | 0.07 | 0.06 |
| Ramachandran allowed (%) | 12.25 | 10.63 | 10.90 |
| Ramachandran favored (%) | 88.69 | 89.30 | 89.04 |
| Poor rotamers (%) | 0.14 | 0.25 | 0.00 |
| CaBLAM outliers (%) | 6.09 | 6.74 | 6.86 |
| Molprobity score | 1.99 | 2.06 | 2.14 |
| Clash score (all atoms) | 7.40 | 9.27 | 11.29 |
| B-factors (protein) | 73.26 | 69.33 | 107.50 |
| B-factors (ligands) | 54.73 | 46.51 | 78.24 |
| EMRinger Score | 2.00 | 2.92 | 1.62 |
| Model refinement package | phenix.real_space_refine (1.13-2998-000) | ||
Key resources table
| Reagent type (species) or resource | Designation | Source or reference | Identifiers | Additional information |
|---|---|---|---|---|
| Gene (Chaetomium thermophilum) | Msp1 | Uniprot | G0S654 | |
| Genetic reagents (S. cerevisiae) | MATα leu2-3,112 TRP1 can1-100 ura3-1 ADE2 his3-11,15 (wild-type) | PMID: 24821790 | PWY1944 in the lab stock | |
| Genetic reagents (S. cerevisiae) | msp1Δ::HpHR | PMID: 24821790 | PWY1947 in the lab stock | |
| Genetic reagents (S. cerevisiae) | get3Δ::NATR | PMID: 24821790 | PWY1950 in the lab stock | |
| Genetic reagents (S. cerevisiae) | msp1Δ::HpHR get3Δ::NATR | PMID: 24821790 | PWY1953 in the lab stock | |
| Recombinant DNA reagent | GST-thrombin-C.thermo Msp1 (plasmid) | This paper | Materials and method section: cloning of Msp1 | |
| Recombinant DNA reagent | GST-thrombin-C. thermo Msp1 (E214) (plasmid) | This paper | Materials and method section: cloning of Msp1 | |
| Software, algorithm | MotionCor2 | PMID: 28250466 | RRID: SCR_016499 | |
| Software, algorithm | Relion | PMID: 23000701 | RRID: SCR_016274 | |
| Software, algorithm | Cryosparc | PMID: 28165473 | RRID: SCR_016501 | |
| Software, algorithm | UCSF Chimera | PMID: 15264254 | RRID: SCR_004097 | |
| Software, algorithm | GCTF | PMID: 26592709 | RRID: SCR_016500 | |
| Software, algorithm | Phenix | PMID: 20124702 | RRID: SCR_014224 | |
| Software, algorithm | Coot | PMID: 20383002 | RRID: SCR_014222 | |
| Software, algorithm | Pymol | Schrödinger, LLC | RRID: SCR_000305 |
