Targeting signals for protein import into peroxisomes

Natural strategies for the spatial optimization of metabolism in synthetic biology

Metabolism is a highly interconnected web of chemical reactions that power life. Though the stoichiometry of metabolism is well understood, the multidimensional aspects of metabolic regulation in time and space remain difficult to define, model and engineer. Complex metabolic conversions can be performed by multiple species working cooperatively and exchanging metabolites via structured networks of organisms and resources. Within cells, metabolism is spatially regulated via sequestration in subcellular compartments and through the assembly of multienzyme complexes. Metabolic engineering and synthetic biology have had success in engineering metabolism in the first and second dimensions, designing linear metabolic pathways and channeling metabolic flux. More recently, engineering of the third dimension has improved output of engineered pathways through isolation and organization of multicell and multienzyme complexes. This review highlights natural and synthetic examples of three-dimensional metabolism both inter- and intracellularly, offering tools and perspectives for biological design.

Peroxisome targeting signal 1: is it really a simple tripeptide?

Originally, the peroxisomal targeting signal 1 (PTS1) was defined as a tripeptide at the C-terminus of proteins prone to be imported into the peroxisomal matrix. The corresponding receptor PEX5 initiates the translocation of proteins by identifying potential substrates via their C-termini and trapping PTS1s through remodeling of its TPR domain. Thorough studies on the interaction between PEX5 and PTS1 as well as sequence-analytic tools revealed the influence of amino acid residues further upstream of the ultimate tripeptide. Altogether, PTS1s should be defined as dodecamer sequences at the C-terminal ends of proteins. These sequences accommodate physical contacts with both the surface and the binding cavity of PEX5 and ensure accessibility of the extreme C-terminus. Knowledge-based approaches in applied Bioinformatics provide reliable tools to accurately predict the peroxisomal location of proteins not yet determined experimentally.

Evidence for a novel pathway for the targeting of a Saccharomyces cerevisiae peroxisomal protein belonging to the isomerase/hydratase family

We, and others, have identified a novel Saccharomyces cerevisiae peroxisomal protein that belongs to the isomerase/hydratase family. The protein, named Dci1p, shares 50% identity with Eci1p, a delta(3)-cis-delta(2)-trans-enoyl-CoA isomerase that acts as an auxiliary enzyme in the beta-oxidation of unsaturated fatty acids. Both of these proteins are localized to peroxisomes, and both contain motifs at their amino- and carboxyl termini that resemble peroxisome targeting signals (PTS) 1 and 2. However, we demonstrate that the putative type 1 signaling motif is not required for the peroxisomal localization of either of these proteins. Furthermore, the correct targeting of Eci1p and Dci1p occurs in the absence of the receptors for the type 1 or type 2 peroxisome targeting pathway. Together, these data suggest a novel mechanism for the intracellular targeting of these peroxisomal proteins.

The difference in recognition of terminal tripeptides as peroxisomal targeting signal 1 between yeast and human is due to different affinities of their receptor Pex5p to the cognate signal and to residues adjacent to it

Pex5p is the receptor for the peroxisomal targeting signal 1 (PTS1) that consists of a C-terminal tripeptide (consensus (S/A/C)(K/R/H)(L/M)). Hexadecapeptides recognized by Pex5p from Homo sapiens and Saccharomyces cerevisiae were identified by screening a two-hybrid peptide library, and the targeting ability of the peptides was demonstrated using the green fluorescent protein as reporter. The PTS1 receptors recognized in a species-specific manner a broad range of C-terminal tripeptides, and these are reported herein. In addition, residues upstream of the tripeptide influenced the strength of the interaction in the two-hybrid system as well as in an in vitro competition assay. In peptides interacting with the human protein, hydrophobic residues were found with high frequency especially at positions -2 and -5, whereas peptides interacting with S. cerevisiae Pex5p were more hydrophilic and frequently contained arginine at position -2. In instances where the terminal tripeptide deviated from the consensus, upstream residues exerted a greater influence on the ability of the hexadecapeptides to bind Pex5p.

Metabolic effects of altering redundant targeting signals for yeast mitochondrial malate dehydrogenase

Eukaryotic cells contain highly homologous isozymes of malate dehydrogenase which catalyze the same reaction in different cellular compartments. To examine whether the metabolic functions of these isozymes are interchangeable, we have altered the cellular localization of mitochondrial malate dehydrogenase (MDH1) in yeast. Since a previous study showed that removal of the targeting presequence from MDH1 does not prevent mitochondrial import in vivo, we tested the role of a putative cryptic targeting sequence near the amino terminus of the mature polypeptide. Three residues in this region were changed to residues present in analogous positions in the other two yeast MDH isozymes. Alone, these replacements did not affect activity or localization of MDH1 but, in combination with deletion of the presequence, prevented mitochondrial import in vivo. Measurable levels of the resulting cytosolic form of MDH1 were low with expression from a centromere-based plasmid but were comparable to normal cellular levels with expression from a multicopy plasmid. The cytosolic form of MDH1 restored the ability of a deltaMDH1 disruption strain to grow on ethanol or acetate, suggesting that mitochondrial localization of MDH1 is not essential for its function in the TCA cycle. This TCA cycle function observed for the cytosolic form of MDH1 is unique to that isozyme since overexpression of MDH2 and of a cytosolic form of MDH3 in a deltaMDH1 strain failed to restore growth. Finally, only partial restoration of growth of a deltaMDH2 disruption mutant was attained with the cytosolic form of MDH1, suggesting that MDH2 may also have unique metabolic functions.

Application of yeasts in gene expression studies: a comparison of Saccharomyces cerevisiae, Hansenula polymorpha and Kluyveromyces lactis -- a review

From the onset of gene technology yeasts have been among the most commonly used host cells for the production of heterologous proteins. At the beginning of this new development the attention in molecular biology and biotechnology focused on the use of the best characterized species, Saccharomyces cerevisiae, leading to an increasing number of production systems for recombinant compounds. In recent years alternative yeasts became accessible for the techniques of modern molecular genetics and, thereby, for potential applications in biotechnology. In this respect Kluyveromyces lactis, and the methylotrophs Hansenula polymorpha and Pichia pastoris have been proven to offer significant advantages over the traditional baker's yeast for the production of certain proteins. In the following article, the present status of the various yeast systems is discussed.

Analysis of the carboxyl-terminal peroxisomal targeting signal 1 in a homologous context in Saccharomyces cerevisiae

Most peroxisomal matrix proteins contain a carboxyl-terminal tripeptide that directs them to peroxisomes. Within limits, these amino acids may be varied, without loss of function. The specificity of this peroxisomal targeting signal (PTS1) is remarkable considering its small size and its relaxed consensus sequence. Moreover, several peroxisomal proteins have a PTS1-like signal that does not fit the reported consensus sequence. Because many of these PTS1 variants seem to be functional in a species-dependent or protein context-dependent manner, we investigated the PTS1 requirements in a homologous context, using Saccharomyces cerevisiae and endogenous peroxisomal malate dehydrogenase (MDH3). Peroxisomal import of the MDH3-PTS1 variants was tested qualitatively by the ability to complement the Deltamdh3 mutant and quantitatively by subcellular fractionation. We observed efficient import of MDH3 into peroxisomes with a large variety of PTS1 tripeptides. Many of these variants do not fit the observed PTS1 requirements for heterologously expressed proteins, which suggests that additional domains in the protein may be of decisive importance whether or not a certain PTS1 variant is recognized by the components of the peroxisomal import machinery. Because we show that dimerization of MDH3 precedes import into the organelle, these domains are most likely conformational domains.

In situ heterogeneity of peroxisomal oxidase activities: an update

Oxidases are a widespread group of enzymes. They are present in numerous organisms and organs and in various tissues, cells, and subcellular compartments, such as mitochondria. An important source of oxidases, which is investigated and discussed in this study, are the (micro)peroxisomes. Oxidases share the ability to reduce molecular oxygen during oxidation of their substrate, yielding an oxidized product and hydrogen peroxide. Besides the hydrogen peroxide-catabolizing enzyme catalase, peroxisomes contain one or more hydrogen peroxide-generating oxidases, which participate in different metabolic pathways. During the last four decades, various methods have been developed and elaborated for the histochemical localization of the activities of these oxidases. These methods are based either on the reduction of soluble electron acceptors by oxidase activity or on the capture of hydrogen peroxide. Both methods yield a coloured and/or electron dense precipitate. The most reliable technique in peroxisomal oxidase histochemistry is the cerium salt capture method. This method is based on the direct capture of hydrogen peroxide by cerium ions to form a fine crystalline, insoluble, electron dense reaction product, cerium perhydroxide, which can be visualized for light microscopy with diaminobenzidine. With the use of this technique, it became clear that oxidase activities not only vary between different organisms, organs, and tissues, but that heterogeneity also exists between different cells and within cells, i.e. between individual peroxisomes. A literature review, and recent studies performed in our laboratory, show that peroxisomes are highly differentiated organelles with respect to the presence of active enzymes. This study gives an overview of the in situ distribution and heterogeneity of peroxisomal enzyme activities as detected by histochemical assays of the activities of catalase, and the peroxisomal oxidases D-amino acid oxidase, L-alpha-hydroxy acid oxidase, polyamine oxidase and uric acid oxidase.

Pay32p of the yeast Yarrowia lipolytica is an intraperoxisomal component of the matrix protein translocation machinery

Pay mutants of the yeast Yarrowia lipolytica fail to assemble functional peroxisomes. One mutant strain, pay32-1, has abnormally small peroxisomes that are often found in clusters surrounded by membraneous material. The functionally complementing gene PAY32 encodes a protein, Pay32p, of 598 amino acids (66,733 D) that is a member of the tetratricopeptide repeat family. Pay32p is intraperoxisomal. In wild-type peroxisomes, Pay32p is associated primarily with the inner surface of the peroxisomal membrane, but approximately 30% of Pay32p is localized to the peroxisomal matrix. The majority of Pay32p in the matrix is complexed with two polypeptides of 62 and 64 kD recognized by antibodies to SKL (peroxisomal targeting signal-1). In contrast, in peroxisomes of the pay32-1 mutant, Pay32p is localized exclusively to the matrix and forms no complex. Biochemical characterization of the mutants pay32-1 and pay32-KO (a PAY32 gene disruption strain) showed that Pay32p is a component of the peroxisomal translocation machinery. Mutations in the PAY32 gene prevent the translocation of most peroxisome-bound proteins into the peroxisomal matrix. These proteins, including the 62-kD anti-SKL-reactive polypeptide, are trapped in the peroxisomal membrane at an intermediate stage of translocation in pay32 mutants. Our results suggest that there are at least two distinct translocation machineries involved in the import of proteins into peroxisomes.

Cloning, sequencing and heterologous expression of the monoamine oxidase gene from Aspergillus niger

The gene encoding the flavin-containing monoamine oxidase (MAO-N) of the filamentous fungus Aspergillus niger was cloned. MAO-N is the first nonvertebrate monoamine oxidase described to date. Three partial cDNA clones, isolated from an expression library, were used to identify and clone the structural gene (maoN) from an A. niger genomic DNA library. The maoN gene was sequenced, and analysis revealed an open reading frame that codes for a protein of 495 amino acids with a calculated molecular mass of 55.6 kDa. Sequencing of an internal proteolytic fragment of the purified enzyme confirmed the derived amino acid sequence. Analysis of the deduced amino acid sequence indicates that MAO-N is structurally related to the human monoamine oxidases MAO-A and MAO-B. In particular, the regions known to be involved in the binding of the FAD cofactor show a high degree of homology; however, the conserved cysteine residue to which the flavin cofactor is covalently bound in the mammalian forms is absent in the fungal enzyme. MAO-N has the C-terminal tripeptide Ala-Arg-Leu, which corresponds to the consensus targeting sequence found in many peroxisomal enzymes. The full-length cDNA for MAO-N was expressed in Escherichia coli from the T7 promoter of the expression vector pET3a, yielding a soluble and fully active enzyme form.

ACR1, a gene encoding a protein related to mitochondrial carriers, is essential for acetyl-CoA synthetase activity in Saccharomyces cerevisiae

The utilization of ethanol via acetate by the yeast Saccharomyces cerevisiae requires the presence of the enzyme acetyl-coenzyme A synthetase (acetyl-CoA synthetase), which catalyzes the activation of acetate to acetyl-coenzyme A (acetyl-CoA). We have isolated a mutant, termed acr1, defective for this activity by screening for mutants unable to utilize ethanol as a sole carbon source. Genetic and biochemical characterization show that, in this mutant, the structural gene for acetyl-CoA synthetase is not affected. Cloning and sequencing demonstrated that the ACR1 gene encodes a protein of 321 amino acids with a molecular mass of 35370 Da. Computer analysis suggested that the ACR1 gene product (ACR1) is an integral membrane protein related to the family of mitochondrial carriers. The expression of the gene is induced by growing yeast cells in media containing ethanol or acetate as sole carbon sources and is repressed by glucose. ACR1 is essential for the utilization of ethanol and acetate since a mutant carrying a disruption in this gene is unable to grow on these compounds.

Peroxisomal disorders: a review

Until recently peroxisomal disorders were considered to be extremely rare and the diagnostic procedures available for postanatal and prenatal diagnosis were not widely known. At present, 17 human disorders are linked to peroxisomal dysfunction. The clinical, biochemical and morphological peroxisome heterogeneity described in the different diseases illustrate that only combined analysis of all the different approaches will lead to a correct diagnosis and a coherent pathophysiological model to guide ongoing research. With the study of human peroxisomal disease, advances have been gained as to the function of the peroxisome in normal and pathological conditions. Genetic analysis of peroxisome biogenesis and research on peroxisomal targeting signals are now in progress. Peroxisomal disorders are usually classified according to the degree of biochemical impairment. In this paper, a tentative classification of peroxisomal disorders will be proposed, based on the degree of biochemical abnormalities combined with new data obtained on whether or not defective peroxisome assembly is involved: (1) disorders with peroxisome assembly deficiencies; (2) disorders with single enzyme deficiencies. The clinical onset and the major symptoms of the various disorders, and the recently discovered findings are discussed.

The pas8 mutant of Pichia pastoris exhibits the peroxisomal protein import deficiencies of Zellweger syndrome cells--the PAS8 protein binds to the COOH-terminal tripeptide peroxisomal targeting signal, and is a member of the TPR protein family

We previously described the isolation of mutants of the yeast Pichia pastoris that are deficient in peroxisome assembly (pas mutants). We describe the characterization of one of these mutants, pas8, and the cloning of the PAS8 gene. The pas8 mutant is deficient for growth, but not for division or segregation of peroxisomes, or for induction of peroxisomal proteins. Two distinct peroxisomal targeting signals, PTS1 and PTS2, have been identified that are sufficient to direct proteins to the peroxisomal matrix. We show that the pas8 mutant is deficient in the import of proteins with the PTS1, but not the PTS2, targeting signal. This is the same import deficiency as that found in cells from patients with the lethal human peroxisomal disorder Zellweger syndrome. Cloning and sequencing of the PAS8 gene reveals that it is a novel member of the tetratricopeptide repeat gene family. Antibodies raised against bacterially expressed PAS8 are used to show that PAS8 is a peroxisomal, membrane-associated protein. Also, we have found that in vitro translated PAS8 protein is capable of binding the PTS1 targeting signal specifically, raising the possibility that PAS8 is a PTS1 receptor.

Two complementary approaches to study peroxisome biogenesis in Saccharomyces cerevisiae: forward and reversed genetics

In order to investigate the mechanisms of peroxisome biogenesis and to identify components of the peroxisomal import machinery we studied these processes in the yeast Saccharomyces cerevisiae. The forward genetic approach has led to pas-mutants (peroxisomal assembly) which fall into 12 complementation groups and allowed to identify 10 of the corresponding wild-type PAS genes (PAS 1-7, 9, 11 and 12). Recent sequence analysis data of some of these genes are beginning to provide first hints as to the possible function of their gene products. The PAS genes and their corresponding mutants are presently used to address some important questions of peroxisomal biogenesis. Reversed genetics has been started as a complementary approach to characterize especially the function of peroxisomal membrane proteins. For this purpose we describe a technique to isolate highly purified peroxisomes. This led to the identification of 21 polypeptides as constituents of this organelle. Some of them are presently sequenced.

Targeting sequences of the two major peroxisomal proteins in the methylotrophic yeast Hansenula polymorpha

Dihydroxyacetone synthase (DAS) and methanol oxidase (MOX) are the major enzyme constituents of the peroxisomal matrix in the methylotrophic yeast Hansenula polymorpha when grown on methanol as a sole carbon source. In order to characterize their topogenic signals the localization of truncated polypeptides and hybrid proteins was analysed in transformed yeast cells by subcellular fractionation and electron microscopy. The C-terminal part of DAS, when fused to the bacterial beta-lactamase or mouse dihydrofolate reductase, directed these hybrid polypeptides to the peroxisome compartment. The targeting signal was further delimited to the extreme C-terminus, comprising the sequence N-K-L-COOH, similar to the recently identified and widely distributed peroxisomal targeting signal (PTS) S-K-L-COOH in firefly luciferase. By an identical approach, the extreme C-terminus of MOX, comprising the tripeptide A-R-F-COOH, was shown to be the PTS of this protein. Furthermore, on fusion of a C-terminal sequence from firefly luciferase including the PTS, beta-lactamase was also imported into the peroxisomes of H. polymorpha. We conclude that, besides the conserved PTS (or described variants), other amino acid sequences with this function have evolved in nature.