Peroxisome biogenesis.

Fifteen years ago, we had a model of peroxisome biogenesis that involved growth and division of preexisting peroxisomes. Today, thanks to genetically tractable model organisms and Chinese hamster ovary cells, 23 PEX genes have been cloned that encode the machinery ("peroxins") required to assemble the organelle. Membrane assembly and maintenance requires three of these (peroxins 3, 16, and 19) and may occur without the import of the matrix (lumen) enzymes. Matrix protein import follows a branched pathway of soluble recycling receptors, with one branch for each class of peroxisome targeting sequence (two are well characterized), and a common trunk for all. At least one of these receptors, Pex5p, enters and exits peroxisomes as it functions. Proliferation of the organelle is regulated by Pex11p. Peroxisome biogenesis is remarkably conserved among eukaryotes. A group of fatal, inherited neuropathologies are recognized as peroxisome biogenesis diseases; the responsible genes are orthologs of yeast or Chinese hamster ovary peroxins. Future studies must address the mechanism by which folded, oligomeric enzymes enter the organelle, how the peroxisome divides, and how it segregates at cell division. Most pex mutants contain largely empty membrane "ghosts" of peroxisomes; a few mutants apparently lacking peroxisomes entirely have led some to propose the de novo formation of the organelle. However, there is evidence for residual peroxisome membrane vesicles ("protoperoxisomes") in some of these, and the preponderance of data supports the continuity of the peroxisome compartment in space and time and between generations of cells.

Pex18p is constitutively degraded during peroxisome biogenesis.

Pex18p and Pex21p are structurally related yeast peroxins (proteins required for peroxisome biogenesis) that are partially redundant in function. One or the other is essential for the import into peroxisomes of proteins with type 2 peroxisomal targeting sequences (PTS2). These sequences bind to the soluble PTS2 receptor, Pex7p, which in turn binds to Pex18p (or Pex21p or possibly both). Here we show that Pex18p is constitutively degraded with a half-time of less than 10 min in wild-type Saccharomyces cerevisiae. This degradation probably occurs in proteasomes, because it requires the related ubiquitin-conjugating enzymes Ubc4p and Ubc5p and occurs normally in a mutant lacking the Pep4p vacuolar protease. The turnover of Pex18p stops, and Pex18p accumulates to a much higher than normal abundance in pex mutants in which the import of all peroxisomal matrix proteins is blocked. This includes mutants that lack peroxins involved in receptor docking at the membrane (Deltapex13 or Deltapex14), a mutant that lacks the peroxisomal member of the E2 family of ubiquitin-conjugating enzymes (Deltapex4), and others (Deltapex1). This stabilization in a variety of pex mutants indicates that Pex18p turnover is associated with its normal function. A Pex18p-Pex7p complex is detected by immunoprecipitation in wild type cells, and its abundance increases considerably in the Deltapex14 peroxisome biogenesis mutant. Cells that lack Pex7p fail to stabilize and accumulate Pex18p, indicating an important role for complex formation in the stabilization. Mono- and diubiquitinated forms of Pex18p are detected in wild-type cells, and there is no Pex18p turnover in a yeast doa4 mutant in which ubiquitin homeostasis is defective. These data represent, to the best of our knowledge, the first instance of an organelle biogenesis factor that is degraded constitutively and rapidly.

Eci1p uses a PTS1 to enter peroxisomes: either its own or that of a partner, Dci1p.

Saccharomyces cerevisiae delta3,delta2-enoyl-CoA isomerase (Eci1p), encoded by ECI1, is an essential enzyme for the betaoxidation of unsaturated fatty acids. It has been reported, as well as confirmed in this study, to be a peroxisomal protein. Unlike many other peroxisomal proteins, Ecilp possesses both a peroxisome targeting signal type 1 (PTS1)-like signal at its carboxy-terminus (-HRL) and a PTS2-like signal at its amino-terminus (RIEGPFFIIHL). We have found that peroxisomal targeting of a fusion protein consisting of Eci1p in front of green fluorescent protein (GFP) is not dependent on Pex7p (the PTS2 receptor), ruling out a PTS2 mechanism, but is dependent on Pex5p (the PTS1 receptor). This Pex5p-dependence was unexpected, since the putative PTS1 of Ecilp is not at the C-terminus of the fusion protein; indeed, deletion of this signal (-HRL-) from the fusion did not affect the Pex5p-dependent targeting. Consistent with this, Pex5p interacted in two-hybrid assays with both Eci1p and Eci1PdeltaHRL. Ecilp-GFP targeting and Eci1pdeltaHRL interaction were abolished by replacement of Pex5p with Pex5p(N495K), a point-mutated Pex5p that specifically abolishes the PTS1 protein import pathway. Thus, Eci1p peroxisomal targeting does require the Pex5p-dependent PTS1 pathway, but does not require a PTS1 of its own. By disruption of ECI1 and DCI1, we found that Dci1p, a peroxisomal PTS1 protein that shares 50% identity with Eci1p, is necessary for Eci1p-GFP targeting. This suggests that the Pex5p-dependent import of Eci1p-GFP is due to interaction and co-import with Dci1p. Despite the dispensability of the C-terminal HRL for import in wild-type cells, we have also shown that this tripeptide can function as a PTS1, albeit rather weakly, and is essential for targeting in the absence of Dci1p. Thus, Eci1p can be targeted to peroxisomes by its own PTS1 or as a hetero-oligomer with Dcilp. These data demonstrate a novel, redundant targeting pathway for Eci1p.

Peroxisomal ghosts are intracellular structures distinct from lysosomal compartments in Zellweger syndrome: a confocal laser scanning microscopy study.

Peroxisome ghosts are aberrant peroxisomal structures found in cultured skin fibroblasts from patients affected by Zellweger Syndrome (ZS), a genetic disorder of peroxisomal assembly. They contain peroxisomal integral membrane proteins (PxIMPs) and they lack most of the matrix enzymes that should be inside the organelle (Santos et al., Science 239 (1988) 1536-1538). Considerable evidence indicates that these ghosts result from genetic defects in the cellular machinery for importing newly-synthesized peroxisomal proteins into the organelle. In contrast to these observations, (Heikoop et al., Eur. J. Cell Biol. 57 (1992) 165-171) report that in Zellweger Syndrome, peroxisomal membranes are located within lysosomes and/or contain lysosomal enzymes. We have undertaken a more detailed and systematic investigation of this matter, employing confocal laser scanning microscopy (CLSM). In fibroblasts derived from ZS patients belonging to different complementation groups, peroxisomes were labeled with antibodies against PxIMPs and lysosomes were labeled with an antibody against a lysosome associated membrane protein (LAMP-2) or with LysoTracker. The results unambiguously demonstrated no appreciable colocalization of PxIMPs and LAMPs (or LysoTracker), indicating that peroxisomal ghosts are distinct subcellular structures, occupying separate subcellular locations.

Rhizomelic chondrodysplasia punctata, a peroxisomal biogenesis disorder caused by defects in Pex7p, a peroxisomal protein import receptor: a minireview.

Rhizomelic chondrodysplasia punctata (RCDP) is a lethal autosomal recessive disease corresponding to complementation group 11 (CG11), the second most common of the thirteen CGs of peroxisomal biogenesis disorders (PBDs). RCDP is characterized by proximal limb shortening, severely disturbed endochondrial bone formation, and mental retardation, but there is an absence of the neuronal migration defect found in the other PBDs. Plasmalogen biosynthesis and phytanic acid oxidation are deficient, but very long chain fatty acid (VLCFA) oxidation is normal. At the cellular level, RCDP is unique in that the biogenesis of most peroxisomal proteins is normal, but a specific subset of at least four, and maybe more, peroxisomal matrix proteins fail to be imported from the cytosol. In this review, we discuss recent advances in understanding RCDP, most prominently the cloning of the affected gene, PEX7, and identification of PEX7 mutations in RCDP patients. Human PEX7 was identified by virtue of its sequence similarity to its Saccharomyces cerevisiae ortholog, which had previously been shown to encode Pex7p, an import receptor for type 2 peroxisomal targeting sequences (PTS2). Normal human PEX7 expression rescues the cellular defects in cultured RCDP cells, and cDNA sequence analysis has identified a variety of PEX7 mutations in RCDP patients, including a deletion of 100 nucleotides, probably due to a splice site mutation, and a prevalent nonsense mutation which results in loss of the carboxyterminal 32 amino acids. Identification of RCDP as a PTS2 import disorder explains the observation that several, but not all, peroxisomal matrix proteins are mistargeted in this disease; three of the four proteins deficient in RCDP have now been shown to be PTS2-targeted.

Pex18p and Pex21p, a novel pair of related peroxins essential for peroxisomal targeting by the PTS2 pathway.

We have identified ScPex18p and ScPex21p, two novel S. cerevisiae peroxins required for protein targeting via the PTS2 branch of peroxisomal biogenesis. Targeting by this pathway is known to involve the interaction of oligopeptide PTS2 signals with Pex7p, the PTS2 receptor. Pex7p function is conserved between yeasts and humans, with defects in the human protein causing rhizomelic chondrodysplasia punctata (RCDP), a severe, lethal peroxisome biogenesis disorder characterized by aberrant targeting of several PTS2 peroxisomal proteins, but uncertainty remains about the subcellular localization of this receptor. Previously, we have reported that ScPex7p resides predominantly in the peroxisomal matrix, suggesting that it may function as a highly unusual intraorganellar import receptor, and the data presented in this paper identify Pex18p and Pex21p as key components in the targeting of Pex7p to peroxisomes. They each interact specifically with Pex7p both in two-hybrid analyses and in vitro. In cells lacking both Pex18p and Pex21p, Pex7p remains cytosolic and PTS2 targeting is completely abolished. Pex18p and Pex21p are weakly homologous to each other and display partial functional redundancy, indicating that they constitute a two-member peroxin family specifically required for Pex7p and PTS2 targeting.

Isolation of a Chinese hamster fibroblast variant defective in dihydroxyacetonephosphate acyltransferase activity and plasmalogen biosynthesis: use of a novel two-step selection protocol.

We have developed a two-step selection protocol to generate a population of Chinese hamster ovary (CHO) cell variants that are plasmalogen-deficient, but contain intact, functional peroxisomes (plasmalogen-/peroxisome+). This involved sequential exposures of a mutagenized cell population to photodynamic damage by using two different pyrene-labelled sensors, 9-(1'-pyrene)nonanol and 12-(1'-pyrene)dodecanoic acid. By this procedure we generated several isolates, all except one of which displayed a severe decrease in plasmalogen biosynthesis. Further characterization of one of the plasmalogen-deficient isolates, NRel-4, showed that it contained intact, functional peroxisomes. Whole-cell homogenates from NRel-4 displayed severely decreased dihydroxyacetone phosphate acyltransferase, which catalyses the first step in plasmalogen biosynthesis. NRel-4 and another, recently described, plasmalogen-deficient cell line, NZel-1 [Nagan, Hajra, Das, Moser, Moser, Lazarow, Purdue and Zoeller (1997) Proc. Natl. Acad. Sci. U.S. A. 94, 4475-4480] were hypersensitive to singlet oxygen, supporting the notion of plasmalogens as radical oxygen scavengers. Wild-type-like resistance could be conferred on NRel-4 upon restoration of plasmalogen content by supplementation with a bypass compound, sn-1-hexadecylglycerol. NRel-4 and other plasmalogen-/peroxisome+ strains will allow us to examine further the role of ether lipids in cellular functions without complications associated with peroxisome deficiency, and might serve as an animal cell model for certain forms of the human genetic disorder rhizomelic chondrodysplasia punctata.

A fibroblast cell line defective in alkyl-dihydroxyacetone phosphate synthase: a novel defect in plasmalogen biosynthesis.

Using fluorescence-activated cytotoxicity selection, followed by colony autoradiographic screening of the surviving population, we have isolated a unique plasmalogen-deficient Chinese hamster ovary (CHO) cell line. The mutant, NZel-1, showed a dramatic (90%) reduction in the rate of biosynthesis and levels of plasmalogens, as determined using short- and long-term labeling with 32Pi. Enzymatic assays and lipid supplementation studies showed that NZel-1 was defective in a single step in the biosynthetic pathway for plasmalogens. This step, catalyzed by the peroxisomal enzyme, alkyl-dihydroxyacetone phosphate (DHAP) synthase, is responsible for the introduction of the ether bond found in plasmalogens. The activity of alkyl-DHAP synthase was reduced in whole-cell homogenates from NZel-1 to 18% of wild-type values. Unlike previously described plasmalogen-deficient mutants, NZel-1 contained peroxisomes, as confirmed by immunofluorescence microscopy and catalase release by digitonin. Peroxisomal functions, including the breakdown of very long-chain (>20 carbons) fatty acids, phytanic acid oxidation, and the acylation of DHAP, were normal. Cell fusion studies revealed that the mutation is recessive and belongs to a new complementation group. To our knowledge this is the first report describing the isolation and characterization of a mutant CHO cell line defective in plasmalogen biosynthesis which contains intact, functional peroxisomes. These cells will allow us to examine the role of ether lipids in cellular functions without complications associated with peroxisome deficiency.

Rhizomelic chondrodysplasia punctata is caused by deficiency of human PEX7, a homologue of the yeast PTS2 receptor.

The rhizomelic form of chondrodysplasia punctata (RCDP) is an autosomal recessive disease of peroxisome biogenesis characterized by deficiencies in several peroxisomal proteins, including the peroxisomal enzymes of plasmalogen biosynthesis and peroxisomal 3-ketoacyl thiolase. In cultured fibroblasts from patients with this disorder, both the peroxisomal targeting and proteolytic removal of the amino-terminal type 2 peroxisomal targeting sequence (PTS2) of thiolase are defective, whereas the biogenesis of proteins targeted by carboxyterminal type 1 peroxisomal targeting sequences (PTS1) is unimpaired. We have previously isolated a Saccharomyces cerevisiae peroxisomal biogenesis mutant, pex7 (formerly peb1/pas7), which demonstrates a striking similarity to the cellular phenotype of RCDP fibroblasts in that PTS1 targeting is functional, but the peroxisomal packaging of PTS2 targeted thiolase is lacking. Complementation of this mutant has led to the identification of the protein ScPex7p, a PTS2 receptor. In this paper we report cloning of the human orthologue of ScPEX7, and demonstrate that this is the defective gene in RCDP. We show that expression of human PEX7 in RCDP cells rescues PTS2 targeting and restores some activity of dihydroxyacetone phosphate acyltransferase (DHAP-AT), a peroxisomal enzyme of plasmalogen biosynthesis, and we identify the mutations responsible for loss of function of PEX7 in a compound heterozygote RCDP patient. These results imply that several peroxisomal proteins are targeted by PTS2 signals and that the various biochemical and clinical defects in RCDP result from a defect in the receptor for this class of PTS.

Targeting of human catalase to peroxisomes is dependent upon a novel COOH-terminal peroxisomal targeting sequence.

We have identified a novel peroxisomal targeting sequence (PTS) at the extreme COOH terminus of human catalase. The last four amino acids of this protein (-KANL) are necessary and sufficient to effect targeting to peroxisomes in both human fibroblasts and Saccharomyces cerevisiae, when appended to the COOH terminus of the reporter protein, chloramphenicol acetyl transferase. However, this PTS differs from the extensive family of COOH-terminal PTS tripeptides collectively termed PTS1 in two major aspects. First, the presence of the uncharged amino acid, asparagine, at the penultimate residue of the human catalase PTS is highly unusual, in that a basic residue at this position has been previously found to be a common and critical feature of PTS1 signals. Nonetheless, this asparagine residue appears to constitute an important component of the catalase PTS, in that replacement with aspartate abolished peroxisomal targeting (as did deletion of the COOH-terminal four residues). Second, the human catalase PTS comprises more than the COOH-terminal three amino acids, in that COOH-terminal-ANL cannot functionally replace the PTS1 signal-SKL in targeting a chloramphenicol acetyl transferase fusion protein to peroxisomes. The critical nature of the fourth residue from the COOH terminus of the catalase PTS (lysine) is emphasized by the fact that substitution of this residue with a variety of other amino acids abolished or reduced peroxisomal targeting. Targeting was not reduced when this lysine was replaced with arginine, suggesting that a basic amino acid at this position is required for maximal functional activity of this PTS. In spite of these unusual features, human catalase is sorted by the PTS1 pathway, both in yeast and human cells. Disruption of the PAS10 gene encoding the S. cerevisiae PTS1 receptor resulted in a cytosolic location of chloramphenicol acetyl transferase appended with the human catalase PTS, as did expression of this protein in cells from a neonatal adrenoleukodystrophy patient specifically defective in PTS1 import. Furthermore, through the use of the two-hybrid system, it was demonstrated that both the PAS10 gene product (Pas10p) and the human PTS1 receptor can interact with the COOH-terminal region of human catalase, but that this interaction is abolished by substitutions at the penultimate residue (asparagine-to- aspartate) and at the fourth residue from the COOH terminus (lysine-to-glycine) which abolish PTS functionality. We have found no evidence of additional targeting information elsewhere in the human catalase protein. An internal tripeptide (-SHL-, which conforms to the mammalian PTS1 consensus) located nine to eleven residues from the COOH terminus has been excluded as a functional PTS. Additionally, in contrast to the situation for S. cerevisiae catalase A, which contains an internal PTS in addition to a COOH-terminal PTS1, human catalase lacks such a redundant PTS, as evidenced by the exclusive cytosolic location of human catalase mutated in the COOH-terminal PTS. Consistent with this species difference, fusions between catalase A and human catalase which include the catalase A internal PTS are targeted, at least in part, to peroxisomes regardless of whether the COOH-terminal human catalase PTS is intact.

Peb1p (Pas7p) is an intraperoxisomal receptor for the NH2-terminal, type 2, peroxisomal targeting sequence of thiolase: Peb1p itself is targeted to peroxisomes by an NH2-terminal peptide.

Peb1 is a peroxisome biogenesis mutant isolated in Saccharomyces cerevisiae that is selectively defective in the import of thiolase into peroxisomes but has a normal ability to package catalase, luciferase and acyl-CoA oxidase (Zhang, J. W., C. Luckey, and P. B. Lazarow. 1993. Mol. Biol. Cell. 4:1351-1359). Thiolase differs from these other peroxisomal proteins in that it is targeted by an NH2-terminal, 16-amino acid peroxisomal targeting sequence type 2 (PTS 2). This phenotype suggests that the PEB1 protein might function as a receptor for the PTS2. The PEB1 gene has been cloned by functional complementation. It encodes a 42,320-D, hydrophilic protein with no predicted transmembrane segment. It contains six WD repeats that comprise the entire protein except for the first 55 amino acids. Peb1p was tagged with hemagglutinin epitopes and determined to be exclusively within peroxisomes by digitonin permeabilization, immunofluorescence, protease protection and immuno-electron microscopy (Zhang, J. W., and P. B. Lazarow. 1995. J. Cell Biol. 129:65-80). Peb1p is identical to Pas7p (Marzioch, M., R. Erdmann, M. Veenhuis, and W.-H. Kunau. 1994. EMBO J. 13: 4908-4917). We have now tested whether Peb1p interacts with the PTS2 of thiolase. With the two-hybrid assay, we observed a strong interaction between Peb1p and thiolase that was abolished by deleting the first 16 amino acids of thiolase. An oligopeptide consisting of the first 16 amino acids of thiolase was sufficient for the affinity binding of Peb1p. Binding was reduced by the replacement of leucine with arginine at residue five, a change that is known to reduce thiolase targeting in vivo. Finally, a thiolase-Peb1p complex was isolated by immunoprecipitation. To investigate the topogenesis of Peb1p, its first 56-amino acid residues were fused in front of truncated thiolase lacking the NH2-terminal 16-amino acid PTS2. The fusion protein was expressed in a thiolase knockout strain. Equilibrium density centrifugation and immunofluorescence indicated that the fusion protein was located in peroxisomes. Deletion of residues 6-55 from native Peb1p resulted in a cytosolic location and the loss of function. Thus the NH2-terminal 56-amino acid residues of Peb1p are necessary and sufficient for peroxisomal targeting. Peb1p is found in peroxisomes whether thiolase is expressed or not. These results suggest that Peb1p (Pas7p) is an intraperoxisomal receptor for the type 2 peroxisomal targeting signal.

Identification of peroxisomal membrane ghosts with an epitope-tagged integral membrane protein in yeast mutants lacking peroxisomes.

Many yeast peroxisome biogenesis mutants have been isolated in which peroxisomes appear to be completely absent. Introduction of a wild-type copy of the defective gene causes the reappearance of peroxisomes, despite the fact that new peroxisomes are thought to form only from pre-existing peroxisomes. This apparent paradox has been explained for similar human mutant cell lines (from patients with Zellweger syndrome) by the discovery of peroxisomal membrane ghosts in the mutant cells (Santos, M. J., T. Imanaka, H. Shio, G. M. Small and P. B. Lazarow. 1988. Science 239, 1536-1538). Introduction of a wild-type gene is thought to restore to the ghosts the ability to import matrix proteins, and thus lead to the refilling of the peroxisomes. It is vitally important to our understanding of peroxisome biogenesis to determine whether the yeast mutants contain ghosts. We have solved this problem by introducing an epitope-tagged version of Pas3p, a peroxisome integral membrane protein (that is essential for peroxisome biogenesis). Nucleotides encoding a nine amino acid HA epitope were added to the PAS3 gene immediately before the stop codon. The tagged gene (PAS3HA) was inserted in the genome, replacing the wild-type gene at its normal locus. It was fully functional (the cells assembled peroxisomes normally and grew on oleic acid) but the expression level was too low to detect the protein with monoclonal antibody 12CA5. PAS3HA was expressed in greater quantity from an episomal plasmid with the CUP1 promoter. The gene product, Pas3pHA, was detected by immunogold labelling on the membranes of individual and clustered peroxisomes; the clusters appeared as large spots in immunofluorescence. PAS3HA was similarly expressed in peroxisome biogenesis mutants peb2 and peb4, which lack morphologically recognizable peroxisomes. Gold-labelled membranes were clearly visible in both mutants: in peb2 the labelled membrane vesicles were generally much smaller than those in peb4, which resembled normal peroxisomes in size.

Peroxisome structure, function, and biogenesis--human patients and yeast mutants show strikingly similar defects in peroxisome biogenesis.

Peroxisomes are found in almost all eukaryotic cells. Two major functions of the organelle are in lipid metabolism: peroxisomes catalyze the initial steps in the biosynthesis of plasmalogens, which are phospholipids that are present in large amounts in myelin. Peroxisomes also catalyze the beta-oxidation of fatty acids; this pathway is essential for the catabolism of a variety of substrates that are not oxidized by mitochondria. A third important function is in cellular respiration, involving the metabolism of H2O2, for which the peroxisome is named. Peroxisomes increase in size by the post-translational import of newly synthesized proteins from the cytosol; these pre-existing peroxisomes divide to form new peroxisomes. Proteins are targeted to peroxisomes by three different types of topogenic sequences, and it is hypothesized that a receptor exists for each type. The newly made proteins are translocated through the peroxisomal membrane into the interior by a machinery that is energized by ATP hydrolysis. Human patients and yeast mutants have remarkably similar defects in peroxisome biogenesis. Some such mutants are defective in the import of a subset of peroxisomal proteins that share a topogenic sequence type; other mutants fail to import all newly made proteins into peroxisomes, regardless of the type of targeting sequence they possess. These mutants might be defective in receptors and in translocation machinery components, respectively. Cloned genes that are essential for peroxisome biogenesis encode diverse proteins: some likely receptors, some transmembrane proteins possibly involved in translocation, and others hydrophilic proteins that may play other roles in peroxisome assembly.

PEB1 (PAS7) in Saccharomyces cerevisiae encodes a hydrophilic, intra-peroxisomal protein that is a member of the WD repeat family and is essential for the import of thiolase into peroxisomes.

We have previously described mutant S. cerevisiae that are defective in peroxisome biogenesis (peb mutants) (Zhang, J. W., Y. Han, and P. B. Lazarow. 1993. J. Cell Biol. 123:1133-1147.). In some mutants, peroxisomes are undetectable. Other mutants contain normal-looking peroxisomes but fail to package subsets of peroxisomal proteins into the organelle (Zhang, J. W., C. Luckey, and P. B. Lazarow. 1993. Mol. Biol. Cell. 4:1351-1359.). In peb1 (pas7) cells, for example, the peroxisomes contain proteins that are targeted by COOH-terminal tripeptides and contain acyl-CoA oxidase (which is probably targeted by internal oligopeptides), but fail to import thiolase (which is targeted by an NH(2)-terminal 16-amino acid sequence). These and other data suggest that there are three branches in the pathway for the import of proteins into peroxisomes, each of which contains a receptor for one type of peroxisomal topogenic information. Here, we report the cloning and characterization of the PEB1 gene, that encodes a 42,320-Da hydrophilic protein with no predicted transmembrane segment. The protein contains six WD repeats, a motif which has been found in 27 proteins involved in diverse cellular functions. The PEB1 gene product was tagged with the hemagglutinin epitope and found to rescue thiolase import in the peb1 null mutant. The epitope-tagged protein was shown to be inside of peroxisomes by immunofluorescence, digitonin permeabilization, equilibrium density centrifugation, immunoelectron microscopy, and proteinase K protection studies. The PEB1 gene product does not cleave the thiolase-targeting sequence. It may function to draw thiolase into peroxisomes.