Introduction
Mitochondria are defining eukaryotic organelles that are essential for cell viability. It is well established that they contain respiratory complexes and the F1Fo-ATPsynthase, enabling bulk ATP production through oxidative phosphorylation. Moreover, mitochondria are also the site of primary energy-yielding pathways such as the tricarboxylic acid cycle (1). Owing through their crucial role in energy metabolism, mitochondria are historically known as the cellular powerhouse. However, accumulating evidence demonstrates that mitochondria take center stage in numerous metabolic processes and signaling pathways (figure 1 adapted from 1). For example, they are involved in the metabolism of amino acids, lipids and nucleotides as well as the production of iron-sulfur clusters and cofactors. Moreover, mitochondria are key players in activating apoptosis, cell signaling and redox homeostasis (1). Based on their central biochemical role it is not surprising that mitochondrial dysfunction is tightly associated with different pathologies that mainly affect energy-demanding tissues, such as brain and muscles, and thereby often result in neurodegenerative diseases e.g. Alzheimer’s, Parkinson’s and Huntington’s disease (2). Furthermore, mitochondria are important players in cancer and diabetes. It is widely accepted that mitochondria are descendants of a free living α-proteobacterium that was internalized by a primordial eukaryotic cell about 2 billion years ago. During their subsequent endosymbiotic evolution, mitochondria retained their two concentric membranes: the outer membrane (OM) and inner membrane (IM) with the inter membrane space (IMS) located in between, while the IM separates the matrix from the IMS (figure 1). The genome of the endosymbiont was, however, vastly reduced as part of it was lost and some genes were transferred to the nucleus. Consequently, the current genomes of mitochondria are very small, encoding 13 proteins in humans and 8 in yeast. These are typically hydrophobic subunits that make up the core of respiratory complexes (1).
The mitochondrial proteome comprises about 1000 to 1500 proteins in yeast and humans, respectively. About 99% of these are nuclear-encoded and are imported as precursors from the cytosol into the organelle following their translation (1). Mitochondria therefore evolved dedicated translocases located in the OM and IM that recognize these precursor proteins and facilitate their import and sorting (3,4). Proper delivery at the mitochondrial surface is ensured by the presence of targeting signals within precursors, which either exist as a N-terminal presequence that is proteolytically removed after import, or an internal targeting signal that is not removed. The vast majority of precursor proteins is imported into mitochondria through the translocase of the outer membrane (TOM complex), serving as the main mitochondrial entry gate (3,4). The TOM complex is made up of 6 to 7 subunits forming core and peripheral components. The core comprises Tom40, a membrane-embedded β-barrel protein that functions as protein-conducting channel, as well as Tom22, Tom5, Tom6 and Tom7 that comprise single pass α-helical membrane proteins and serve as receptor for precursors or assist in the assembly of the core complex. The peripheral subunits include Tom20 and Tom70, single membrane spanning α-helical proteins, that function as primary import receptors (3,4). The scant structural information available on the TOM complex primarily stems from crosslinking experiments and a medium-resolution cryo-EM structure (5,6). To fully understand the architecture and functional mechanism of the TOM complex, detailed structural information is required. With regards to this it should be noted that recently a high-resolution cryo-EM reconstruction of the TOM complex was reported (7). Here, I will discuss this structure as well as the current functional understanding of mitochondrial protein import.
Mitochondrial protein import
It has been estimated that about 99% of the proteins that make up the mitochondrial proteome, which typically ranges between 1000 and 1500 proteins, are nuclear-encoded and imported into the organelle following their translation on cytosolic ribosomes. To ensure efficient targeting to mitochondria as well as their respective sub compartment, these proteins are commonly produced as precursors with either an N-terminal presequence that is clipped off after import or an internal targeting signal that is not removed after import. Approximately 60% of all proteins imported by mitochondria are equipped with an N-terminal presequence, forming positively charged amphipathic α-helices (1). Presequences are typically present on precursor proteins destined for the matrix and IM. In contrast, carrier proteins, a subset of IM proteins with six transmembrane domains (TMDs), lack a presequence but contain poorly defined internal targeting signals with hydrophobic elements (1). Presequences and other mitochondrial targeting signals are recognized by dedicated receptor proteins of the TOM complex, which serves as main entry gate for most mitochondrial proteins.
Five major import pathways of precursor proteins into mitochondria have been identified that are essential for cell viability (1,8). In four of these, precursor proteins are transported through the TOM complex in an unfolded conformation (figure 2 adapted from 8). Presequence-bearing precursors are typically recognized by Tom20 and Tom22 (pathway 1) and are subsequently passed on to Tom5 and in to the translocation channel of Tom40. Following transport of the precursor across the OM, its presequence is bound by Tim50, the receptor of the TIM23 complex on the IMS side of the OM. TIM23 serves as translocase of the inner membrane and facilitates the passage of precursors across the IM into the matrix. The presequence translocase-associated motor (PAM or MMC) with mtHsp70, an ATP-driven chaperone, as central component is located in the matrix. After crossing the IM, the presequence is bound by mtHsp70 that functions as import motor and completes the translocation of the precursor, while its presequence is removed by the mitochondrial processing peptidase (1,8). Newly imported proteins are folded subsequently into their active conformation with the help of matrix chaperones. Polytopic IM proteins of the carrier class lack a cleavable presequence and are bound by cytosolic chaperones (Hsp70 and Hsp90) probably during translation to prevent their aggregation. The chaperone-membrane precursor protein complex docks at the Tom70 receptor of the TOM complex, which recognizes internal targeting sequences (pathway 2). After chaperone release, the precursor moves through the Tom40 channel in a loop conformation across the IM into the IMS. Here, it is bound by small chaperones (Tim9-Tim10 or Tim8-Tim13) to prevent its aggregation. The small TIM chaperones also deliver their substrate to the carrier translocase of the IM, the so-called TIM22 complex, which facilitates their insertion into the IM (1,8). Similar to the OM of Gram-negative bacteria, the mitochondrial OM contains β-barrel membrane proteins. These are initially synthesized in the cytosol without a cleavable presequence and are translocated to the IMS via the TOM complex (pathway 3). Within the IMS, precursors of β-barrel membrane proteins are also bound by small TIM chaperones, that deliver their cargo at the TOB/SAM complex. This represents a translocator for the assembly of β-barrel membrane proteins into the lipid phase of the OM (1,8). A substantial number of IMS proteins contain functionally essential disulfide bonds, which are formed via oxidative folding of the imported precursors through the Tom40 channel in the IMS (pathway 4). The mitochondrial intermembrane space import and assembly (MIA) system catalyzes the oxidative folding of newly imported precursors equipped with an IMS-specific targeting signal. These precursors are recognized by Mia40, that functions as an oxidoreductase, and cooperates with Erv1 in a cyclic fashion, thereby accepting transient disulfide bonds from Erv1 and introducing these into imported precursors (1,8).
General architecture of the TOM complex
The vast amount of proteins that make up the mitochondrial proteome are post-translationally imported from the cytosol into the organelle through one of the five major import pathways. In four of these pathways, precursor proteins are translocated through the TOM complex, implying that this translocase is able to handle a variety of structurally and biophysically different substrate proteins, ranging from cysteine-rich precursors to hydrophobic ones destined for the IM with multiple TMDs. The yeast Saccharomyces cerevisiae has been a powerful model for several decades to study mitochondrial protein import biochemically and genetically, providing a thorough molecular understanding of the import machineries and their mechanism. For example, the molecular composition of the yeast TOM complex was elucidated by blue native gel electrophoresis, revealing that it made up of six to seven subunits and can be separated in a stable core, historically known as the general import pore (GIP), and several distinct subcomplexes (figure 2). The holo-complex has a molecular weight of about 450 kDa, while the core is approximately 400 kDa and comprises Tom40, Tom22, Tom5, Tom6 and Tom7. The peripheral complexes, that are weakly associated with the core, are formed by Tom20 and Tom70 (9). Tom40, the central component of the outer membrane import machinery, is an integral β-barrel membrane protein of about 40 kDa embedded into the OM with 19 predicted β-sheets. Tom40 homologs are ubiquitously present in eukaryotic cells and are essential for cell viability. Similar to other β-barrel proteins, Tom40 adopts a cylindrical conformation with a central pore that functions as protein-conducting channel. Yeast Tom22 is an integral membrane protein of 18 kDa with a single α-helical TMD. Tom22 is stably associated with Tom40 in the core complex mainly through its TMD, that is also is required for proper assembly of the TOM core. Like Tom40, Tom22 is evolutionarily conserved throughout all eukaryotes unlike the other core subunits. Within the core complex, Tom22 interacts with the peripheral subunits Tom20 and Tom70, the main receptors of the TOM complex. In addition to its TMD, Tom20 contains two soluble domains that are exposed to either the cytosol or IMS. Its cytosolic domain binds presequence-bearing precursors, thereby functioning as secondary receptor, while its IMS exposed domain accepts precursors that emerge from the Tom40 translocation channel following passage across the OM. These are subsequently passed on to the TIM23 translocase. The other subunits of the core complex, Tom5, Tom6 and Tom7, are all single pass α-helical membrane proteins that assist in the assembly of the core complex. Yeast Tom20 and Tom70 are also single spanning α-helical membrane proteins of about 20 and 70 kDa, respectively, that only weakly associate with the core complex. Both proteins are equipped with cytosolically exposed domains that contain numerous so-called tetratricopeptide repeat (TPR) motifs that mediate protein-protein interactions and ensure the capture of precursor proteins. It has been established that Tom20 specifically binds presequences after which the precursors are handed over to Tom22, while Tom70 recognizes internal targeting signals of carrier proteins. However, Tom70 is also able to bind precursor proteins with presequences and, therefore, the substrate specificity of Tom22 and Tom70 partially overlaps. Thus, the TOM complex represents a unique α/β translocator made up of a β-barrel channel and α-helical subunits that coordinate its activity. Further insight into the molecular architecture of the yeast TOM complex was provided by in vivo and in organello crosslinking experiments (7), showing that the translocator contains three β-barrel channels that are flanked by different Tom22 receptors as well as the small Tom subunits. Interestingly, an N-terminal segment of Tom40 extends through the translocation channel into the IMS and probably recruits chaperones from the IMS to the exit of the Tom40 pore, enabling the transfer of precursor proteins. These studies, moreover, confirmed that precursor proteins move through the Tom40 β-barrel pore during passage of the OM, thereby following different transport paths inside the protein-conducting channel. Substrates with a presequence move along an acidic path on the interior of the Tom40 channel, while carrier proteins follow a hydrophobic path.
Structure of the TOM complex
To fully understand how the TOM complex is able to process the wide variety of precursors, detailed structural information is required. However, cryo-EM studies only produced moderate resolution structures of the fungal and yeast TOM core complex (6,9). The left panel of figure 3 shows the structure of the core complex from Saccharomyces cerevisiae at 18 Å resolution, revealing a complex with an overall triangular shape and three globular domains that are localized to the OM. Within the structure, three pores are visible with each of these corresponding to the β-channel of one Tom40 protomer (9). Hence, the TOM core comprises most likely three Tom40 barrels, consistent with the crosslinking results discussed above. A more detailed view of the TOM complex was provided recently by assessing the cryo-EM structure of the TOM core complex from Neurosporra crassa at 6.8 Å (6).
This structure is presented in the right panel of figure 3 and shows that it is a symmetrical dimer made up of ten membrane-embedded subunits with two Tom40 barrels forming identical protein-conducting channels across the OM. These are joined together by two α-helical TMDs of two Tom22 molecules, while the small Tom subunits surround each Tom40 protomer. Furthermore, the structure confirms that the N-terminal soluble domain of Tom22 is exposed to the cytosol, ideally positioned for the capture of precursor proteins, whereas its C-terminal soluble domain is localized to the IMS and is thought to accept precursors that emerge from the other side of the Tom40 channel that are subsequently transferred to the TIM23 complex. Thus, the structural evidence presented above indicates that the core complex exists either as a dimer or trimer. Conceivably, the dimer represents an intermediate that serves as an assembly platform for the mature trimeric form through incorporating an additional Tom40 subunit (5).
In addition to the moderate resolution structures discussed above, a detailed structure of the yeast TOM core complex was solved at 3.8 Å recently (7). To this end, the complex was isolated from S. cerevisiae as a mixture of dimers and trimers and subjected to cryo-EM analysis. This structure is presented in figure 4 in surface (left panel) and ribbon orientation (right panel), revealing a dimeric complex made up two molecules of Tom40, Tom22, Tom5, Tom6 and Tom7. A top view of the structure is provided in the left bottom panel, showing that each Tom40 protomer is surrounded by the other Tom subunits and comprises a single protein-conducting channel. A surface representation of the core complex with coloring based on hydrophobicity (red) is presented in the right bottom panel, demonstrating that the channel interior is mainly hydrophilic, whereas each Tom40 protomer is surrounded by hydrophobic subunits as expected. Tom40 is folded into a cylindrical barrel with 19 transmembrane β-strands as predicted, while Tom22, Tom5, Tom6 and Tom7 are α-helical membrane-embedded subunits that are packed tightly against the β-barrel. Intriguingly, Tom7 and Tom22 exhibit a kinked transmembrane α-helix owing to the presence of helix-breaking proline residues. Two Tom40 barrels are joined together through association with two Tom22 α-helices. The α/β interactions between Tom40 and Tom22 are mainly mediated by hydrophobic residues. The complex is further stabilized by the transmembrane α-helix of Tom6, thereby explaining why this subunit is essential for the proper assembly of the core complex. Both Tom40 subunits directly contact each other at the membrane cytosol interface. The central channel of the pore is closed by the first and last β-strands that associate through hydrogen bonds.
Protein import through the TOM complex
In organello and in vitro crosslinking experiments showed that precursors with a presequence and membrane protein precursors contact different residues of the Tom40 channel. Specifically, presequences follow an acidic path along the inner wall of the Tom40 channel, whereas membrane protein precursors interact mainly with hydrophobic residues of the translocator pore. The detailed molecular information of the TOM core complex discussed above allows structural rationalization of these crosslinking data (figure 5 adopted from 7). Each Tom40 has two exit sites for precursor proteins, namely: one located centrally in the dimer that is used for the presequence-containing precursors and a peripheral site for precursors with internal targeting information. Moreover, Tom40 contains an N-terminal extension that adopts a helix structure which located inside the translocation channel. This helix plays a crucial role in the import of various proteins destined for the IMS. Precursors with a presequence are directed to the IMS side via the IMS facing regions of Tom40, Tom22 and Tom7 and leave the β-pore in the middle of the dimer. Subsequently, the IMS exposed domains of Tom40 and Tom22 transfer the precursor to Tim50 of the TIM23 complex. In contrast, Tom5 and the N-terminal extension guide presequence-less precursors, including carrier proteins and MIA substrates, to the peripheral exit site. For MIA substrates, this N-terminal extension is involved in both the early and final import steps as well as oxidative folding.
Conclusion
Mitochondria are key players in a variety of cellular processes, ranging from energy metabolism to apoptosis, cell signaling and redox homeostasis. Owing to this central biochemical role, proper mitochondrial activity is essential for cell viability and this depends on the correct functioning of its proteome. A typical mitochondrial proteome comprises 1000 – 1500 different proteins of which 99% is nuclear-encoded and imported into the organelle following translation on cytosolic ribosomes. To enable protein import, mitochondria are equipped with oligomeric protein complexes localized to the OM and IM that function as specialized protein translocation nanomachines. For example, the translocase of the outer membrane (TOM complex) serves as main mitochondrial entry gate and is responsible for the import of most mitochondrial proteins. These proteins are typically produced as precursors containing specific information responsible for efficient mitochondrial targeting, including an N-terminal presequence that is clipped off after import or an internal targeting signal that is not removed after import. This targeting information is recognized by dedicated receptor proteins of the TOM complex. The TOM complex comprises six to seven subunits and can be separated in a stable core and several distinct subcomplexes (figure 2). The core contains Tom40, Tom22, Tom5, Tom6 and Tom7. The peripheral complexes, that are weakly associated with the core, are formed by the receptor proteins Tom20 and Tom70. Tom40, the central component of the outer membrane import machinery, is an integral β-barrel membrane protein anchored into the OM and comprises the protein-conducting channel. It is surrounded the other Tom proteins, which are α-helical membrane-embedded subunits. Recent detailed structural information showed that the yeast core complex is dimeric and comprises two molecules of Tom40, Tom22, Tom5, Tom6 and Tom7. Importantly, the core subunits are conserved in all eukaryotes and the structure of the yeast complex is therefore representative for all eukaryotic TOM complexes. The high-resolution of the TOM core revealed that each Tom40 subunit possesses two exit sites for precursor proteins of which one is located centrally in the dimer that is used for the presequence-containing precursors and a peripheral site for precursors with internal targeting information. These precursors are therefore guided via different pathways within the Tom40 translocation channel to their respective exit site (figure 5). Tom22, Tom40 and Tom7 direct precursors with a presequence to the central exit site, while Tom5 and Tom40 guide precursors with internal targeting signals to the peripheral exit site. Thus, the detailed molecular information of the TOM complex reveals how this unique α/β translocator is able the facilitate the import of variety of structurally and biophysically different substrate proteins, ranging from cysteine-rich precursors to hydrophobic ones destined for the IM with multiple TMDs.
References
1. Pfanner N, Warscheid B, and Wiedemann N. 2019. Mitochondrial proteins: from biogenesis to functional networks. Nat Rev Mol Cell Biol. 20: 267-284.
2. Nunnari J, and Suomalainen A. (2012). Mitochondria: in sickness and in health. Cell. 148: 1145-1159.
3. Mani J, Meisinger C, and Schneider A. (2015). Peeping at TOMs—Diverse Entry Gates to Mitochondria Provide Insights into the Evolution of Eukaryotes. Mol Biol Evol. 33: 337-351.
4. Perry AJ, Rimmer KA, and Mertens HDT. et al. (2008). Structure, topology and function of the translocase of the outer membrane of mitochondria. Plant Physiol Biochem. 46: 265-274.
5. Shiota T, Imai K, and Qiu J. et al. (2015). Molecular architecture of the active mitochondrial protein gate. Science. 349: 1544-1548.
6. Bausewein T, Mills DJ, and Langer JD. et al. (2017). Cryo-EM Structure of the TOM Core Complex from Neurospora crassa. Cell. 170: 693-700.
7. Araiso Y, Tsutsumi A, and Qiu J. et al. (2019). Structure of the mitochondrial import gate reveals distinct preprotein paths. Nature. doi: 10.1038/s41586-019-1680-7.
8. Endo T, Yamano K. 2009. Multiple pathways for mitochondrial protein traffic. Biol Chem. 390: 723-730.
9. Dekker PJ, Ryan MT, Brix J. et al. (1998). Preprotein translocase of the outer mitochondrial membrane: molecular dissection and assembly of the general import pore complex. Mol Cell Biol. 18: 6515-24.
10. Model K, Meisinger C, Kühlbrandt W. 2008. Cryo-electron microscopy structure of a yeast mitochondrial preprotein translocase. J Mol Biol. 383: 1049-1057.
Reactie schrijven