Dear Editor,
Mitochondria have more than 1000 proteins, most of which are encoded by nuclear DNA and imported from the cytosol.1 A large group of multi-spanning membrane proteins, the solute carrier proteins, are imported from the cytosol by the translocase of the outer membrane (TOM) and translocated into the inner membrane (IM) by the TIM22 complex.2 The TIM22 complex in Saccharomyces cerevisiae (S. cerevisiae) has seven subunits: Tim22, Tim18, Tim54, Sdh3, Tim9, Tim10, and Tim12. Tim22 is the core translocase subunit. The three small Tim proteins, Tim9, Tim10, and Tim12, are homologous to each other. Tim9 and Tim10 form hexameric chaperones in the intermembrane space (IMS) to deliver the polypeptide substrates from the TOM to the TIM22.3 Tim9 and Tim10 become part of the TIM22 complex with the help of Tim12.4 Tim54 likely interacts with Tim9-Tim10-Tim12 in the complex.5 Sdh3 is a membrane component of respiratory complex II (succinate dehydrogenase, SDH). Tim18 is homologous to Sdh4, the partner of Sdh3 in respiratory complex II. The functions of Sdh3 and Tim18 in the TIM22 complex are less clear. Previous studies have suggested that driven by the membrane potential, the TIM22 complex might function as a twin-pore translocase.6 However, it is poorly understood how the subunits of TIM22 assemble and accomplish protein translocation due to lack of high-resolution structural information. To address some of the fundamental questions in the field, we determined the electron cryo-microscopy (cryo-EM) structure of the endogenous TIM22 complex from S. cerevisiae.
To facilitate purification of the endogenous TIM22 complex, Tim18 in the yeast genome was fused with a C-terminal twin-strep tag. The structure of the purified TIM22 complex was determined at a resolution of 3.8 Å (Supplementary information, Figs. S1–S3). All the subunits could be identified in the structure.
The small Tim subunits in TIM22 form a hexameric ring and sit on the membrane with a ~45° tilt (Fig. 1a, b). The ring is similar to the crystal structure of the Tim9-Tim10 complex,3 except that one of the Tim10 subunits is replaced by Tim124 (Fig. 1b). Tim12 is homologous to Tim10 and has an additional C-terminal helix that intercalates between the N- and C-terminal helices of one Tim9 subunit (Supplementary information, Fig. S4). Interestingly, strong density blobs are also observed in the same cleft of the other two Tim9 subunits (Supplementary information, Fig. S4b, c). Although the exact sequences could not be identified, one of the density blobs is extended from Tim54 and the other one is likely from the N-terminal segment of Tim22. Therefore, the hydrophobic clefts of all three Tim9 subunits are occupied by the peptides, whereas the hydrophobic clefts of Tim10 and Tim12 are open for substrate binding.7
The core structure of Tim54 is well resolved in the density map, although the model is incomplete due to large disordered regions. Tim54 has a typical α/β motif in the IMS. A four-stranded β sheet is sandwiched by four α-helices, with α1 and α4 on one side, and α2 and α3 on the other side (Supplementary information, Fig. S5a). Similar α/β motifs are found in the lipid kinases, such as diacylglycerol kinases (DGK) and sphingosine kinases8 (Supplementary information, Fig. S5b, c). Interestingly, the human TIM22 complex contains mitochondrial acylglycerol kinase (AGK), but not Tim54. Tim54 and AGK may have similar folds and functions in the TIM22 complex, although the two subunits have no sequence homology. Tim54 is anchored on the membrane by an N-terminal transmembrane segment (TM) (residues 37–71, α0) and an amphipathic helix from α4 (Supplementary information, Fig. S5a). Tim54 may help to hold the ring of the small Tim subunits in the 45°-tilted conformation as it is the only subunit that has interactions with the ring in the IMS (Fig. 1a).
Tim18, Sdh3, and Tim22 are close to each other in the membrane. Together, the three membrane subunits provide a docking platform for the small Tim subunits (Fig. 1a). Both Tim18 and Sdh3 have three TMs and an amphipathic helix on the IMS side (Fig. 1c). Tim18 and Sdh3 form a heterodimer with pseudo two-fold symmetry, similar to the organization of Sdh3-Sdh4 in mitochondrial respiratory complex II.9 Tim22 has four TMs that form a curved surface, with the concave side opening to the lipids and the convex side facing Tim18 (Fig. 1c). Consistent with the biochemical data,10 the conserved cysteine residues, 42 and 141, form a disulfide bond between TMs 1 and 2 (Fig. 1c).
The TIM22 complex has some unusual electrostatic features in the membrane (Fig. 1d). While the other subunits have the typical charge and hydrophobicity distributions of membrane proteins, the concave surface of Tim22 has a large negatively charged patch exposed to the membrane on the IMS side. The transmembrane distance is relatively short due to the TM2-TM3 split on the matrix side (Fig. 1e). Indeed, on the IMS side, Tim22 has a completely conserved glutamic acid (E140) of TM2 next to the disulfide bond, as well as a conserved aspartic acid (D190) in TM411 (Fig. 1e). The TM2-TM3 split is bordered by two lysine residues pointing to each other, a conserved lysine (K127) of TM2 and a lysine in the loop between TMs 3 and 4 (K169) (Fig. 1e). Mutagenesis studies show that the single mutations, E140A and K127A, greatly impaired yeast cell growth (Fig. 1f), while the assembly of the TIM22 complex in the mutants was not affected (Supplementary information, Fig. S6). The double mutants, E140A/D190A and K127A/K169A, and the quadruple mutant are essentially lethal to yeast growth (Fig. 1f). The results highlight the critical roles of the charged residues for the proper functioning of Tim22.The structure of the TIM22 complex does not reveal an obvious translocation channel for substrate import. It is rather surprising as previous studies have suggested that the TIM22 complex might act as a twin-pore translocase.6 One possible scenario could be that the current structure may represent an idle state or an intermediate state. Driven by the membrane potential, the TIM22 complex, particularly the Tim22 subunit, may undergo conformational changes to oligomerize and form pores. Alternatively, the single Tim22 subunit may be sufficient to act as a TM insertase, similar to the bacterial insertase, YidC.12 The completely conserved E140 and K127 are only 12 residues away in TM2, with a linear distance of ~20 Å, about half of the thickness of the regular membranes (Fig. 1e). The distribution of the charged residues in the TMs of Tim22 may decrease the thickness of the local lipid membrane. The charged environment in the thinner membrane could greatly reduce the energy barrier for TM insertion.
The Tim9-Tim10 complex chaperones the polypeptide substrates from the TOM complex to TIM22. Tim9-Tim10 may deliver the substrates by making direct contact with the Tim22 complex. Indeed, during 2D classification, a top view of the complex with two rings was occasionally observed (Supplementary information, Fig. S2b). The “two-ring” view was similar to the results from negative stain EM,6 both of which had a diameter of ~16 Å. The additional ring might be the Tim9-Tim10 complex. The polypeptide substrates carried by the Tim9-Tim10 ring might be passed directly to the Tim9-Tim10-Tim12 ring and then to Tim22, whose concave surface is well aligned with a substrate-binding cleft of Tim9-Tim10-Tim12 (Supplementary information, Fig. S7).
In summary, the structure of the TIM22 complex reveals a sophisticated organization of the seven subunits. Tim18-Sdh3 and Tim22 provide a docking platform for anchoring Tim9-Tim10-Tim12. Tim9-Tim10-Tim12 is at the center of the assembly, likely to receive the polypeptide substrates from Tim9-Tim10 and pass them to Tim22. Tim54 helps to hold Tim9-Tim10-Tim12 in a tilted conformation. Tim22 may create a hydrophilic environment either inside a channel or in the membrane to facilitate TM insertion.
References
Sickmann, A. et al. Proc. Natl. Acad. Sci. USA 100, 13207–13212 (2003).
Wiedemann, N. & Pfanner, N. Annu. Rev. Biochem. 86, 685–714 (2017).
Baker, M. J. et al. Mol. Biol. Cell 20, 769–779 (2009).
Gebert, N. et al. EMBO Rep. 9, 548–554 (2008).
Wagner, K. et al. Mol. Biol. Cell 28, 4251–4260 (2008).
Rehling, P. et al. Science 299, 1747 (2003).
Weinhäupl, K. et al. Cell 175, 1365–1379 (2018).
Ma, Q., Gabelli, S. B. & Raben, D. M. Adv. Biol. Regul. 71, 104–110 (2019).
Sun, F. et al. Cell 121, 1043–1057 (2005).
Wrobel, L., Sokol, A. M., Chojnacka, M. & Chacinska, A. Sci. Rep. 6, 27484 (2016).
Žárský, V. & Doležal, P. Biol. Direct 11, 54–54 (2016).
Hennon, S. W., Soman, R., Zhu, L. & Dalbey, R. E. J. Biol. Chem. 290, 14866–14874 (2015).
Acknowledgements
We thank Dr. Ningning Li, Dr. Guopeng Wang, and Dr. Ye Xiang for assistance with EM data analysis, Dr. Ning Gao for critical reading of the manuscript, the National Center for Protein Sciences at Peking University in Beijing, China, for assistance with protein purification and mass spectrometry, the Core Facilities at School of Life Sciences Peking University for assistance with negative staining EM, and the Electron Microscopy Laboratory of Peking University and the cryo-EM platform of Peking University for help with data collection. Computation was supported by the High-performance Computing Platform of Peking University. Q.L. is supported by Beijing Outstanding Young Scientist Program (BJJWZYJH01201910001005).
Author information
Authors and Affiliations
Contributions
Y.Z., X.O., D.S., and X.Z. prepared the samples. Y.Z., X.O., and X.Wu performed cryo-EM sample preparation and data collection. X.Wang performed yeast genetics, Y.Z. and L.L. determined the structures. L.L. and Q.L. supervised the project. All authored are involved in preparing the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Supplementary information
Rights and permissions
About this article
Cite this article
Zhang, Y., Ou, X., Wang, X. et al. Structure of the mitochondrial TIM22 complex from yeast. Cell Res 31, 366–368 (2021). https://doi.org/10.1038/s41422-020-00399-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41422-020-00399-0
This article is cited by
-
Central role of Tim17 in mitochondrial presequence protein translocation
Nature (2023)
-
Molecular pathway of mitochondrial preprotein import through the TOM–TIM23 supercomplex
Nature Structural & Molecular Biology (2023)
-
Structural basis of mitochondrial protein import by the TIM23 complex
Nature (2023)
-
Mitochondrial complexome reveals quality-control pathways of protein import
Nature (2023)
-
A unified evolutionary origin for the ubiquitous protein transporters SecY and YidC
BMC Biology (2021)