Henk F. Tabak

Contribution of the Endoplasmic Reticulum to Peroxisome Formation

How peroxisomes are formed in eukaryotic cells is unknown but important for insight into a variety of diseases. Both human and yeast cells lacking peroxisomes due to mutations in PEX3 or PEX19 genes regenerate the organelles upon reintroduction of the corresponding wild-type version. To evaluate how and from where new peroxisomes are formed, we followed the trafficking route of newly made YFP-tagged Pex3 and Pex19 proteins by real-time fluorescence microscopy in Saccharomyces cerevisiae. Remarkably, Pex3 (an integral membrane protein) could first be observed in the endoplasmic reticulum (ER), where it concentrates in foci that then bud off in a Pex19-dependent manner and mature into fully functional peroxisomes. Pex19 (a farnesylated, mostly cytosolic protein) enriches first at the Pex3 foci on the ER and then on the maturing peroxisomes. This trafficking route of Pex3-YFP is the same in wild-type cells. These results demonstrate that peroxisomes are generated from domains in the ER.

A role for Vps1p, actin, and the Myo2p motor in peroxisome abundance and inheritance in Saccharomyces cerevisiae

In vivo time-lapse microscopy reveals that the number of peroxisomes in Saccharomyces cerevisiae cells is fairly constant and that a subset of the organelles are targeted and segregated to the bud in a highly ordered, vectorial process. The dynamin-like protein Vps1p controls the number of peroxisomes, since in a vps1Δ mutant only one or two giant peroxisomes remain. Analogous to the function of other dynamin-related proteins, Vps1p may be involved in a membrane fission event that is required for the regulation of peroxisome abundance. We found that efficient segregation of peroxisomes from mother to bud is dependent on the actin cytoskeleton, and active movement of peroxisomes along actin filaments is driven by the class V myosin motor protein, Myo2p: (a) peroxisomal dynamics always paralleled the polarity of the actin cytoskeleton, (b) double labeling of peroxisomes and actin cables revealed a close association between both, (c) depolymerization of the actin cytoskeleton abolished all peroxisomal movements, and (d) in cells containing thermosensitive alleles of MYO2, all peroxisome movement immediately stopped at the nonpermissive temperature. In addition, time-lapse videos showing peroxisome movement in wild-type and vps1Δ cells suggest the existence of various levels of control involved in the partitioning of peroxisomes.

The ABC transporter proteins Pat1 and Pat2 are required for import of long-chain fatty acids into peroxisomes of Saccharomyces cerevisiae.

Peroxisomes of Saccharomyces cerevisiae are the exclusive site of fatty acid beta‐oxidation. We have found that fatty acids reach the peroxisomal matrix via two independent pathways. The subcellular site of fatty acid activation varies with chain length of the substrate and dictates the pathway of substrate entry into peroxisomes. Medium‐chain fatty acids are activated inside peroxisomes hby the acyl‐CoA synthetase Faa2p. On the other hand, long‐chain fatty acids are imported from the cytosolic pool of activated long‐chain fatty acids via Pat1p and Pat2p, peroxisomal membrane proteins belonging to the ATP binding cassette transporter superfamily. Pat1p and Pat2p are the first examples of membrane proteins involved in metabolite transport across the peroxisomal membrane.

The membrane of peroxisomes in Saccharomyces cerevisiae is impermeable to NAD(H) and acetyl-CoA under in vivo conditions

We investigated how NADH generated during peroxisomal beta‐oxidation is reoxidized to NAD+ and how the end product of beta‐oxidation, acetyl‐CoA, is transported from peroxisomes to mitochondria in Saccharomyces cerevisiae. Disruption of the peroxisomal malate dehydrogenase 3 gene (MDH3) resulted in impaired beta‐oxidation capacity as measured in intact cells, whereas beta‐oxidation was perfectly normal in cell lysates. In addition, mdh3‐disrupted cells were unable to grow on oleate whereas growth on other non‐fermentable carbon sources was normal, suggesting that MDH3 is involved in the reoxidation of NADH generated during fatty acid beta‐oxidation rather than functioning as part of the glyoxylate cycle. To study the transport of acetyl units from peroxisomes, we disrupted the peroxisomal citrate synthase gene (CIT2). The lack of phenotype of the cit2 mutant indicated the presence of an alternative pathway for transport of acetyl units, formed by the carnitine acetyltransferase protein (YCAT). Disruption of both the CIT2 and YCAT gene blocked the beta‐oxidation in intact cells, but not in lysates. Our data strongly suggest that the peroxisomal membrane is impermeable to NAD(H) and acetyl‐CoA in vivo, and predict the existence of metabolite carriers in the peroxisomal membrane to shuttle metabolites from peroxisomes to cytoplasm and vice versa.

Excised group II introns in yeast mitochondria are lariats and can be formed by self-splicing in vitro

Excised group II introns in yeast mitochondria appear as covalently closed circles under the electron microscope. We show that these circular molecules are branched and resemble the lariats arising through splicing of nuclear pre-mRNAs in yeast and higher eukaryotes. One member of this intron class (aI5c in the gene for cytochrome c oxidase subunit I) is capable of self-splicing in vitro, giving correct exon-exon ligation and resulting in the appearance of both linear and lariat forms of the excised intron. Nuclease digestion of the latter molecules reveals the presence of a complex oligonucleotide with the probable structure AGU, which thus resembles the branch point formed in the spliceosome-dependent reactions undergone by nuclear pre-mRNAs. Unlike group I introns, this group II intron is not demonstrably dependent on GTP for self-splicing and circularization of the isolated, linear intron is not observed. A model accounting for these observations is presented.

Peroxisomal protein import is conserved between yeast, plants, insects and mammals.

We have previously demonstrated that firefly luciferase can be imported into peroxisomes of both insect and mammalian cells. To determine whether the process of protein transport into the peroxisome is functionally similar in more widely divergent eukaryotes, the cDNA encoding firefly luciferase was expressed in both yeast and plant cells. Luciferase was translocated into peroxisomes in each type of organism. Experiments were also performed to determine whether a yeast peroxisomal protein could be transported to peroxisomes in mammalian cells. We observed that a C‐terminal segment of the yeast (Candida boidinii) peroxisomal protein PMP20 could act as a peroxisomal targeting signal in mammalian cells. These results suggest that at least one mechanism of protein translocation into peroxisomes has been conserved throughout eukaryotic evolution.

Overexpression of Pex15p, a phosphorylated peroxisomal integral membrane protein required for peroxisome assembly in S.cerevisiae, causes proliferation of the endoplasmic reticulum membrane

We have cloned PEX15 which is required for peroxisome biogenesis in Saccharomyces cerevisiae. pex15Δ cells are characterized by the cytosolic accumulation of peroxisomal matrix proteins containing a PTS1 or PTS2 import signal, whereas peroxisomal membrane proteins are present in peroxisomal remnants. PEX15 encodes a phosphorylated, integral peroxisomal membrane protein (Pex15p). Using multiple in vivo methods to determine the topology, Pex15p was found to be a tail‐anchored type II (Ncyt–Clumen) peroxisomal membrane protein with a single transmembrane domain near its carboxy‐terminus. Overexpression of Pex15p resulted in impaired peroxisome assembly, and caused profound proliferation of the endoplasmic reticulum (ER) membrane. The lumenal carboxy‐terminal tail of Pex15p protrudes into the lumen of these ER membranes, as demonstrated by its O‐glycosylation. Accumulation in the ER was also observed at an endogenous expression level when Pex15p was fused to the N‐terminus of mature invertase. This resulted in core N‐glycosylation of the hybrid protein. The lumenal C‐terminal tail of Pex15p is essential for targeting to the peroxisomal membrane. Furthermore, the peroxisomal membrane targeting signal of Pex15p overlaps with an ER targeting signal on this protein. These results indicate that Pex15p may be targeted to peroxisomes via the ER, or to both organelles.

Peroxisomal and mitochondrial carnitine acetyltransferases of Saccharomyces cerevisiae are encoded by a single gene.

Carnitine acetyltransferase (CAT) is present in mitochondria and peroxisomes of oleate‐grown Saccharomyces cerevisiae. Both proteins are encoded by the same gene, YCAT, which encodes a protein with a mitochondrial targeting signal (MTS) at the N‐terminus, and a peroxisomal targeting signal type 1 (PTS‐1) at the C‐terminus. Deletion of both motifs revealed the presence of an additional internal targeting sequence. Import of CAT via this internal signal was shown to be dependent on PAS10, a protein which is required for the import of PTS‐1 containing proteins. An interaction of PAS10 with this internal targeting signal was demonstrated using the yeast two‐hybrid technique. Expression of the YCAT gene behind a heterologous promoter resulted in loss of peroxisomal targeting, indicating that differential targeting is controlled at transcriptional or translational level. Determination of the 5′‐ends of YCAT mRNAs revealed that YCAT transcripts initiating after the first AUG were present in oleate‐grown cells. These transcripts were virtually absent in acetate‐ or glycerol‐grown cells. We propose that in response to oleate, shorter transcripts are produced from which the peroxisomal form of CAT is translated, resulting in a CAT protein without a MTS, which can be targeted to peroxisomes.

Molecular characterization of carnitine-dependent transport of acetyl-CoA from peroxisomes to mitochondria in Saccharomyces cerevisiae and identification of a plasma membrane carnitine transporter, Agp2p.

In Saccharomyces cerevisiae, β‐oxidation of fatty acids is confined to peroxisomes. The acetyl‐CoA produced has to be transported from the peroxisomes via the cytoplasm to the mitochondrial matrix in order to be degraded to CO2 and H2O. Two pathways for the transport of acetyl‐CoA to the mitochondria have been proposed. The first involves peroxisomal conversion of acetyl‐CoA into glyoxylate cycle intermediates followed by transport of these intermediates to the mitochondria. The second pathway involves peroxisomal conversion of acetyl‐CoA into acetylcarnitine, which is subsequently transported to the mitochondria. Using a selective screen, we have isolated several mutants that are specifically affected in the second pathway, the carnitine‐dependent acetyl‐CoA transport from the peroxisomes to the mitochondria, and assigned these CDAT mutants to three different complementation groups. The corresponding genes were identified using functional complementation of the mutants with a genomic DNA library. In addition to the previously reported carnitine acetyl‐CoA transferase (CAT2), we identified the genes for the yeast orthologue of the human mitochondrial carnitine acylcarnitine translocase (YOR100C or CAC) and for a transport protein (AGP2) required for carnitine transport across the plasma membrane.

Import of proteins into peroxisomes

Peroxisomes are organelles that confine an important set of enzymes within their single membrane boundaries. In man, a wide variety of genetic disorders is caused by loss of peroxisome function. In the most severe cases, the clinical phenotype indicates that abnormalities begin to appear during embryological development. In less severe cases, the quality of life of adults is affected. Research on yeast model systems has contributed to a better understanding of peroxisome formation and maintenance. This framework of knowledge has made it possible to understand the molecular basis of most of the peroxisome biogenesis disorders. Interestingly, most peroxisome biogenesis disorders are caused by a failure to target peroxisomal proteins to the organellar matrix or membrane, which classifies them as protein targeting diseases. Here we review recent fundamental research on peroxisomal protein targeting and discuss a few burning questions in the field concerning the origin of peroxisomes.