Franz-Ulrich Hartl

Mitochondrial protein import

The transport of nuclear-encoded proteins from the cytosol into mitochondria is mediated by targeting (signal) sequences present on precursor forms. Most precursors of the mitochondrial matrix possess amino-terminal signals which characteristically contain hydroxylated and basic amino acids and lack acidic residues. With a minority of precursor proteins, internal sequence motifs can direct proteins to the mitochondria (Pfanner, N., Hoeben, P., Tropschug, M. and Neupert, W. (1987) J. Biol. Chem. 262, 14851–14854). The presence of a mitochondrial targeting sequence alone, however, is not sufficient for specific targeting to the organelle and further to the various subcompartments. There is the need for components which recognise the targeting sequences and others which keep the precursor protein in a translocation-competent form. Beyond the recognition step, components are required which mediate translocation across the mitochondrial membranes. Mitochondria possess two translocation machineries, one in the outer membrane and one in the inner membrane. The matrix space harbors a number of factors which participate in the import of proteins, in their unfolding and folding. Energy is required at several steps of these processes.

DnaK, DnaJ and GrpE form a cellular chaperone machinery capable of repairing heat-induced protein damage.

Members of the conserved Hsp70 chaperone family are assumed to constitute a main cellular system for the prevention and the amelioration of stress‐induced protein damage, though little direct evidence exists for this function. We investigated the roles of the DnaK (Hsp70), DnaJ and GrpE chaperones of Escherichia coli in prevention and repair of thermally induced protein damage using firefly luciferase as a test substrate. In vivo, luciferase was rapidly inactivated at 42 degrees C, but was efficiently reactivated to 50% of its initial activity during subsequent incubation at 30 degrees C. DnaK, DnaJ and GrpE did not prevent luciferase inactivation, but were essential for its reactivation. In vitro, reactivation of heat‐inactivated luciferase to 80% of its initial activity required the combined activity of DnaK, DnaJ and GrpE as well as ATP, but not GroEL and GroES. DnaJ associated with denatured luciferase, targeted DnaK to the substrate and co‐operated with DnaK to prevent luciferase aggregation at 42 degrees C, an activity that was required for subsequent reactivation. The protein repair function of DnaK, GrpE and, in particular, DnaJ is likely to be part of the role of these proteins in regulation of the heat shock response.

Chaperonin-mediated protein folding: GroES binds to one end of the GroEL cylinder, which accommodates the protein substrate within its central cavity.

The mechanism of GroEL (chaperonin)‐mediated protein folding is only partially understood. We have analysed structural and functional properties of the interaction between GroEL and the co‐chaperonin GroES. The stoichiometry of the GroEL 14mer and the GroES 7mer in the functional holo‐chaperonin is 1:1. GroES protects half of the GroEL subunits from proteolytic truncation of the approximately 50 C‐terminal residues. Removal of this region results in an inhibition of the GroEL ATPase, mimicking the effect of GroES on full‐length GroEL. Image analysis of electron micrographs revealed that GroES binding triggers conspicuous conformational changes both in the GroES adjacent end and at the opposite end of the GroEL cylinder. This apparently prohibits the association of a second GroES oligomer. Addition of denatured polypeptide leads to the appearance of irregularly shaped, stain‐excluding masses within the GroEL double‐ring, which are larger with bound alcohol oxidase (75 kDa) than with rhodanese (35 kDa). We conclude that the functional complex of GroEL and GroES is characterized by asymmetrical binding of GroES to one end of the GroEL cylinder and suggest that binding of the substrate protein occurs within the central cavity of GroEL.

A zinc finger-like domain of the molecular chaperone DnaJ is involved in binding to denatured protein substrates.

The Escherichia coli heat‐shock protein DnaJ cooperates with the Hsp70 homolog DnaK in protein folding in vitro and in vivo. Little is known about the structural features of DnaJ that mediate its interaction with DnaK and unfolded polypeptide. DnaJ contains at least four blocks of sequence representing potential functional domains which have been conserved throughout evolution. In order to understand the role of each of these regions, we have analyzed DnaJ fragments in reactions corresponding to known functions of the intact protein. Both the N‐terminal 70 amino acid ‘J‐domain’ and a 35 amino acid glycine‐phenylalanine region following it are required for interactions with DnaK. However, only complete DnaJ can cooperate with DnaK and a third protein, GrpE, in refolding denatured firefly luciferase. As demonstrated by atomic absorption and extended X‐ray absorption fine structure spectroscopy (EXAFS), the 90 amino acid cysteine‐rich region of DnaJ contains two Zn atoms tetrahedrally coordinated to four cysteine residues, resembling their arrangement in the C4 Zn binding domains of certain DNA binding proteins. Interestingly, binding experiments and cross‐linking studies indicate that this Zn finger‐like domain is required for the DnaJ molecular chaperone to specifically recognize and bind to proteins in their denatured state.

HOW DO POLYPEPTIDES CROSS THE MITOCHONDRIAL MEMBRANES

Summary of the transfert of proteins across the two mitochondrial membranes (from the cytosol into the mitochondrial matrix) and proposition a hypothesis about the physical pathway and energetics by which polypeptide chains cross the membranes.

Antifolding activity of hsp60 couples protein import into the mitochondrial matrix with export to the intermembrane space

Cytochrome b2 reaches the intermembrane space of mitochondria by transport into the matrix followed by export across the inner membrane. While in the matrix, the protein interacts with hsp60, which arrests its folding prior to export. The bacterial-type export sequence in pre-cytochrome b2 functions by inhibiting the ATP-dependent release of the protein from hsp60. Release for export apparently requires, in addition to ATP, the interaction of the signal sequence with a component of the export machinery in the inner membrane. Export can occur before import is complete provided that a critical length of the polypeptide chain has been translocated into the matrix. Thus, hsp60 combines two activities: catalysis of folding of proteins destined for the matrix, and maintaining proteins in an unfolded state to facilitate their channeling between the machineries for import and export across the inner membrane. Anti-folding signals such as the hydrophobic export sequence in cytochrome b2 may act as switches between these two activities.

The processing peptidase of yeast mitochondria: the two co-operating components MPP and PEP are structurally related

Two proteins co‐operate in the proteolytic cleavage of mitochondrial precursor proteins: the mitochondrial processing peptidase (MPP) and the processing enhancing protein (PEP). In order to understand the structure and function of this novel peptidase, we have isolated mutants of Saccharomyces cerevisiae which were temperature sensitive in the processing of mitochondrial precursor proteins. Here we report on the mif2 mutation which is deficient in MPP. Mitochondria from the mif2 mutant were able to import precursor proteins, but not to cleave the presequences. The MPP gene was isolated. MPP is a hydrophilic protein consisting of 482 amino acids. Notably, MPP exhibits remarkable sequence similarity to PEP. We speculate that PEP and MPP have a common origin and have evolved into two components with different but mutually complementing functions in processing of precursor proteins.

A comment on: ‘The aromatic amino acid content of the bacterial chaperone protein groEL (cpn60): Evidence for the presence of a single tryptophan’, by N.C. Price, S.M. Kelly, S. Wood and A. auf der Mauer (1991) FEBS Lett. 292, 9–12

The groEL protein of Escherichia coli, a tetradecamer of -60 kDa subunits, functions as an ATP-dependent molecular chaperone in protein folding. Price et al. recently published a spectroscopic analysis of purified groEL in which they reported the presence of a single tryptophan per groEL subunit. The presumed absence of tryptophan from groEL, indicated by the DNA-derived sequence [l], had formed the basis for the conformational analysis of groEL-bound substrate proteins via their intrinsic tryptophan fluorescence [2]. We have re-analyzed the spectroscopic properties of our groEL preparations and conclude that groEL protein does not contain tryptophan. This is based on the following observations.

Affinity purification of molecular chaperones of the yeast Hansenula polymorpha using immobilized denatured alcohol oxidase

We used peroxisomal alcohol oxidase (AO) for the affinity purification of molecular chaperones from yeasts. Methodical studies showed that up to 0.8 mg of purified bacterial GroEL was able to bind per ml of immobilized denatured AO column material. Using crude extracts of Hansenula polymorpha or Saccharomyces cerevisiae, several proteins were specifically eluted with Mg‐ATP which were recognized by antibodies against hsp60 or hsp70. One H. polymorpha 70 kDa protein was strongly induced during growth at elevated temperatures, whereas a second 70 kDa protein as well as a 60 kDa protein showed strong protein sequence homology to mitochondrial SSCI and hsp60, respectively, from S. cerevisiae.

Import of proteins into the various submitochondrial compartments.

Summary Import of proteins into mitochondria can be subdivided into several distinct steps. (1) Mitochondrial proteins are synthesized on free ribosomes and are released into cytosolic pools. Nucleoside triphosphates are required to keep precursors in a conformation competent for import. (2) Precursors are directed to mitochondria by specific targeting signals (in most cases contained in N-terminal presequences) and by binding to receptors on the surface of the outer membrane. (3) Precursors interact with a component in the outer membrane which is believed to facilitate membrane insertion (‘general insertion protein’). (4) Outer membrane proteins are then directly routed to their final location. Proteins of all other submitochondrial compartments are directed into translocation contact sites between outer and inner membranes. Transfer into contact sites is dependent on the membrane potential (ΔΨ) across the inner membrane. (5) Presequences of precursors are cleaved in the matrix by the mitochondrial processing peptidase in cooperation with the processing enhancing protein. (6) Precursors of the intermembrane space or the outer surface of the inner membrane have to be re-translocated back across the inner membrane (‘conservative sorting’). Cytochrome c is an exception to this general import pathway. The precursor, apocytochrome c, is directly translocated across the outer membrane into the intermembrane space in a process independent of ΔΨ.