Import of proteins into peroxisomes

Review

Import of proteins into peroxisomes

Ewald H. Hettema, Ben Distel, Henk F. Tabak *Department of Biochemistry, Academic Medical Centre, Meibergdreef 15, 1105 AZ Amsterdam, Netherlands

Received 15 March 1999; received in revised form 9 June 1999; accepted 11 June 1999

Abstract

Peroxisomes are organelles that confine an important set of enzymes within their single membrane boundaries. In man, awide variety of genetic disorders is caused by loss of peroxisome function. In the most severe cases, the clinical phenotypeindicates that abnormalities begin to appear during embryological development. In less severe cases, the quality of life ofadults is affected. Research on yeast model systems has contributed to a better understanding of peroxisome formation andmaintenance. This framework of knowledge has made it possible to understand the molecular basis of most of theperoxisome biogenesis disorders. Interestingly, most peroxisome biogenesis disorders are caused by a failure to targetperoxisomal proteins to the organellar matrix or membrane, which classifies them as protein targeting diseases. Here wereview recent fundamental research on peroxisomal protein targeting and discuss a few burning questions in the fieldconcerning the origin of peroxisomes. ß 1999 Elsevier Science B.V. All rights reserved.

Keywords: Peroxisome; Protein; Membrane

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

2. Experimental approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

3. Peroxisomal matrix protein import . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193.1. Peroxisomal targeting signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193.2. Protein folding and assembly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223.3. Molecular chaperones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233.4. Two pathways for peroxisomal matrix protein targeting . . . . . . . . . . . . . . . . . . . . . . . 233.5. The PTS1 receptor, Pex5p . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243.6. The PTS2 receptor, Pex7p . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243.7. Docking and translocation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253.8. The peroxisomal matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

4. Peroxisomal membrane proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274.1. Targeting of peroxisomal membrane proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274.2. Factors required for PMP insertion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

0167-4889 / 99 / $ ^ see front matter ß 1999 Elsevier Science B.V. All rights reserved.PII: S 0 1 6 7 - 4 8 8 9 ( 9 9 ) 0 0 0 8 7 - 7

* Corresponding author. Fax: +31-20-691-5519; E-mail : [email protected]

BBAMCR 14514 30-7-99

Biochimica et Biophysica Acta 1451 (1999) 17^34

www.elsevier.com/locate/bba

4.3. Peripheral peroxisomal membrane proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

5. Species-speci¢c di¡erences in peroxisomal protein import . . . . . . . . . . . . . . . . . . . . . . . . . 29

6. Future challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

1. Introduction

Peroxisomes are cellular organelles bounded by asingle membrane that characteristically house the en-zymes for fatty acid L-oxidation, hydrogen-peroxide-producing oxidases and catalase. Their importance inhuman metabolism is illustrated by the occurrence ofsevere genetic disorders caused by disturbances inperoxisome biogenesis and metabolism. Peroxisomesbelong to the morphologically designated group ofeukaryotic organelles known as microbodies whichalso includes glyoxysomes in plants [1] and fungi[2] and glycosomes in Kinetoplastida [3]. Both themechanism by which proteins are targeted and im-ported into microbodies and the components re-quired for peroxisome biogenesis (peroxins; abbrevi-ated as pex) are evolutionarily conserved [4].

The biogenesis of peroxisomal matrix proteins isunderstood increasingly well (for recent review see[5^8]). Peroxisomal proteins are nuclear-encodedand are synthesised in the cytoplasm on free polyri-bosomes [9]. Most peroxisomal matrix proteins aresynthesised at their mature molecular weight andcontain either a carboxy-terminal or an amino-termi-nal peroxisomal targeting signal (PTS1 or PTS2, re-spectively). The peroxisomal targeting signals (PTS)are recognised by speci¢c receptors (for review see[5,8,10]). These proteins are thought to act as recep-tors that cycle between the cytoplasm, where theybind PTS-containing matrix proteins, and the perox-isomal membrane, where they deliver their cargo.After binding of the targeting complex to peroxiso-mal membrane, both PTS1- and PTS2-containingproteins are translocated across the peroxisomalmembrane via a shared translocation apparatus,upon which they reach the peroxisomal matrix. Per-oxisomal membrane proteins are targeted and in-serted post-translationally into the peroxisomal

membrane via a di¡erent pathway than peroxisomalmatrix proteins. However, some peroxisomal mem-brane proteins have been suggested to reach theirlocation via the endoplasmic reticulum (for reviewsee [11,12]).

We will review recent progress concerning the im-port of peroxisomal proteins and the targeting andinsertion of membrane proteins. Furthermore, wewill discuss recent views with respect to the originsof peroxisomes. We will refer to comprehensive re-views for older work that has been incorporated, butnot explicitly cited.

2. Experimental approaches

In order to study the peroxisomal protein importprocess, two complementary approaches are beingfollowed. (1) Genetic screens in yeast, mammalsand Kinetoplastida have identi¢ed 22 factors (perox-ins, see Table 1) that are mutated in peroxisomeassembly mutants [8,13^16]. (2) In vitro import as-says are being developed in order to identify factorsinvolved in the import process and to test the func-tion of the various peroxins. These in vitro importassays have been developed successfully for plantperoxisomes [17^21]. Unfortunately, hardly anyplant peroxins have been identi¢ed that can be testedin these systems. Recently, however, Baker and co-workers developed an in vitro import system withsun£ower glyoxysomes and puri¢ed recombinantprotein. They could distinguish intermediates onthe import pathway, which opens the way to detailedanalysis of the import process and identi¢cation ofproteins which may associate with the import inter-mediate [21].

In yeast and mammalian cells, in vitro import sys-tems have been di¤cult to reproduce. The major

BBAMCR 14514 30-7-99

E.H. Hettema et al. / Biochimica et Biophysica Acta 1451 (1999) 17^3418

problems encountered include fragility of peroxi-somes and the lack of a solid criterium for import.In contrast to the import of precursor proteins intothe endoplasmic reticulum, chloroplasts and mito-chondria, import of peroxisomal proteins is notcoupled to processing (see also below) and thereforeimport has been addressed only by protease-protec-tion (for review see [22]). This stimulated the devel-opment of permeabilised cell systems or microinjec-tion-based import assays [23^26]. These studies plusthe in vitro import studies revealed that the importprocess can be divided into two steps: (a) an ATP-independent binding step to the peroxisomal mem-brane which can occur at low temperature (0³C) andwhich is signal-dependent and (b) a translocationstep which is energised; by ATP hydrolysis andwhich is temperature-dependent. Furthermore, thesestudies showed that matrix protein import is a satu-rable process which depends on the presence of mem-brane-associated factors and which can occur inde-pendent of a membrane potential. Although in moststudies a requirement for cytosolic factors was found,in some studies addition of cytosol had no stimula-tory e¡ect (for review see [17,22]). The lack of acytosolic requirement could be caused by the pres-ence of cytosolic factors in the translation mixture orby copuri¢cation of cytosolic factors with peroxi-somes, or by a bypass of cytosolic factors when usinga high concentration of import substrate [21].

3. Peroxisomal matrix protein import

3.1. Peroxisomal targeting signals

Peroxisomal matrix proteins are nuclear encoded,are synthesised on free polyribosomes and are re-leased into the cytoplasm before they are imported[9]. Speci¢c targeting of most peroxisomal matrixproteins is achieved by an evolutionarily conservedC-terminal tripeptide of the sequence serine^lysine^leucine or a conserved derivative thereof (consensussequence; S/C/A,K/R/H,L). This amino acid se-quence, called the peroxisomal targeting signal typeI (PTS1) has been found to be both necessary andsu¤cient to direct reporter proteins to peroxisomes[27,28]. However, various peroxisomal matrix pro-teins contain a C-terminal sequence with a two-out-

of-three ¢t with the consensus sequence and some ofthese proteins are imported in one species but not inanother. For instance, the PTS1 of human alanine:glyoxylate aminotransferase comprises -KKL, a sig-nal that is not able to target a reporter protein tohuman peroxisomes [29]. The PTS1 could be changedinto several other non-consensus PTS1 sequenceswithout interference of targeting capacity [29]. Exten-sive variation of the three C-terminal amino acids ofthe Saccharomyces cerevisiae protein Mdh3p revealedthat a much greater variety of PTS1 variants allowedimport of Mdh3p into yeast peroxisomes than theconsensus PTS1 sequence would predict [30]. Theseresults imply that besides the C-terminal tripeptideadditional information within peroxisomal proteinsis required for import. It has now become clearthat the amino acids directly preceding the PTS1provide the additional information [31^33]. La-metschwandtner et al. [33] showed, by making useof the two-hybrid system, that the PTS1 receptor isable to bind a wide variety of C-terminal tripeptidesand that the strength of the interaction is modulatedby the preceding eight amino acid residues. There-fore, the PTS1 is not con¢ned to the three C-terminalamino acids of peroxisomal matrix proteins but in-cludes also the amino acids just upstream. Mostlikely, this region binds in a single binding site onthe PTS1 receptor and multiple contacts are requiredfor e¤cient binding. In the case of the consensustripeptide sequence, the three C-terminal amino acidsbind the receptor with such a high a¤nity that noadditional interactions with the PTS1 receptor arerequired. On the other hand, non-consensus tripep-tides depend on the interaction of upstream aminoacid residues with the PTS1 receptor (see Fig. 1).Indeed, all non-consensus PTS1-containing humanproteins contain a lysine residue immediately preced-ing the C-terminal tripeptide (see Fig. 1). The humanPTS1 receptor and S. cerevisiae PTS1 receptor rec-ognise a partially overlapping set of tripeptides, andthe positions of the upstream amino acids that con-tribute to PTS1 recognition vary between man andyeast. Therefore, the observation that some peroxi-somal proteins are imported in a species-speci¢cmanner can be attributed to di¡erences in the PTS1receptors to recognise the C-terminal tripeptide andits accessory sequences [33].

The PTS2 is found in only a few peroxisomal ma-

BBAMCR 14514 30-7-99

E.H. Hettema et al. / Biochimica et Biophysica Acta 1451 (1999) 17^34 19

Table 1Description of peroxins

Peroxin Features Remarks

Pex1p AAA ATPase Interacts with Pex6p

Pex2p C3HC4 motif ( = RING domain), PMP Glycosylated in Y. lipolytica required for human PTS1receptor recycling

Pex3p PMP Interacts with Pex19p, required for PMP insertion

Pex4p Ubiquitin conjugating enzyme, peroxisome-associated PTS1 receptor accumulates in peroxisomes of H. polymorphavpex4 cells

Pex5p PTS1 receptor, 7^8 TPR motifs, cytosol and peroxisome PTS1-speci¢c cycling receptor, binds PTS1 and the dockingcomplex

Pex6p AAA ATPase Required for stability of PTS1 receptor in human ¢broblasts,interacts with Pex1p

Pex7p PTS2 receptor, 6 WD-40 motifs, cytosol and peroxisome PTS2-speci¢c import factor that binds PTS2, Pex18p, Pex21pand the docking complex

Pex8p Peroxisomal membrane-associated, contains PTS1

Pex9p PMP

Pex10p RING, PMP Required for PTS1 receptor export?

Pex11p Peroxisomal membrane-associated/PMP? Peroxisome division

Pex12p RING, PMP Required for PTS1 receptor recycling

Pex13p SH3 domain, PMP Part of docking complex

Pex14p Peroxisomal membrane-associated Part of docking complex

Pex15p PMP, phosphorylated

Pex16p Peroxisomal membrane-associated in Y. lipolytica Glycosylated in Y. lipolyticaPMP in man Required for PMP insertion in man

Pex17p Peroxisomal membrane-associated Part of docking complex

Pex18p Related to Pex21p PTS2-speci¢c import factor, interacts with Pex7p

Pex19p Farnesylated, cytosol and peroxisome Interacts with Pex3p, required for PMP insertion

Pex20p Cytosol PTS2-speci¢c import factor in Y. lipolytica

Pex21p Related to Pex18p PTS2-speci¢c import factor, interacts with Pex7p

Djp1p J domain, cytosol Stimulates matrix protein import

For uni¢ed nomenclature see [13]. For references and a more detailed description of most the peroxins see text or [8].

BBAMCR 14514 30-7-99


trix proteins in yeast, plants, mammals and Kineto-plastida (for review see [34,35]) The PTS2 is a bipar-tite signal of the following consensus: (R/K)(L/V/I)X5(H/Q)(L/A). Although this sequence is foundmainly at the N-terminus of proteins, it can functioninternally [36]. In mammals and plants, PTS2 pro-teins are synthesised as precursors, with the PTS2residing in the cleavable presequence [37^43]. In con-trast to biogenesis of mitochondrial and secretoryprecursor proteins, import of the precursor andcleavage of the presequence are not coupled [44].

Various examples of internal targeting signals havebeen reported, but they remain ill de¢ned [45^48].

Are these internal peroxisomal targeting signals re-£ecting the existence of a novel import pathway withits own unique import component(s) or are they rec-ognised by the PTS1 or PTS2 receptors? Interest-ingly, S. cerevisiae catalase A and carnitine acetyl-transferase are both mislocated to the cytosol incells lacking the PTS1 receptor. Furthermore, carni-tine acetyltransferase lacking its PTS1 could still in-teract with the PTS1 receptor in a yeast two-hybridassay. These results show that import mediated viathe internal PTS of these peroxisomal matrix pro-teins depends on the PTS1 receptor [48]. For S. cere-visiae acyl-CoA oxidase, however, it remains un-

Fig. 1. Model for PTS1 recognition by the PTS1 receptor. (A) Consensus PTS1 tripeptides (S,A,C/K,R,H/L) interact strongly in thePTS1-binding site, without the need for additional interactions between the receptor and the PTS1-containing protein. Non-consensusPTS1 tripeptides function as targeting signals only in combination with additional interactions with the PTS1 receptor. (B) In all hu-man peroxisomal proteins that contain a non-consensus PTS1, the C-terminal tripeptide is preceded by a lysine residue (K). KANL ofhuman catalase is su¤cient to direct a reporter protein to peroxisomes, whereas ANL is not [31].

BBAMCR 14514 30-7-99


clear how it reaches the peroxisomes. The group ofLazarow isolated S. cerevisiae mutants that were dis-turbed in either PTS1 import or PTS2 import. Theyreported that import of acyl-CoA oxidase is neitherdisturbed in PTS1-import-de¢cient cells nor in PTS2-import-de¢cient cells. This observation lead thempostulate the existence of a third peroxisomal proteinimport pathway [49].

Alternatively, subunits from homodimeric or ho-motrimeric peroxisomal matrix proteins lacking aPTS can be imported into peroxisomes in complexwith a PTS-containing subunit [30,50^52]. It hastherefore been postulated that some proteins (for in-stance acyl-CoA oxidase) might enter peroxisomes byassociation with PTS-containing proteins forming aheteromeric complex (for review see [5,6,35]). If thisoccurs, a multimerisation sequence would be inter-preted as a PTS.

3.2. Protein folding and assembly

It is now generally accepted that at least someproteins fold and oligomerise in the cytoplasm priorto import into peroxisomes and that complete un-folding is not a prerequisite for translocation (for

review see [6]). This is based on the following obser-vations: (1) peroxisomal matrix proteins can foldand assemble into their active conformation in thecytosol in peroxisome assembly mutants; (2) sub-units lacking a PTS can be piggy-back-importedwith PTS-containing subunits after oligomerisationin the cytoplasm [30,50^53]; (3) the subunits do notseem to mix during the transport process, suggestingthat oligomer dissociation is not required for trans-location; (4) import of a DHFR fusion protein is notinhibited in vivo by the conformation-stabilisingligand aminopterin [54]; (5) chemically cross-linkedhuman serum albumin (HSA) conjugated to multiplePTS1-containing peptides was imported into ¢bro-blast peroxisomes after microinjection into the cyto-plasm [55]; and (6) even 9-nm gold particles coatedwith HSA-PTS1 peptides were incorporated into per-oxisomes [55].

Although these studies imply that folding and as-sembly might occur prior to import and that theimport machinery can handle folded proteins as sub-strate, it does not necessarily imply that this is theway peroxisomal matrix proteins are normally im-ported. Indeed, it is now clear that proteins do notneed to be in their native conformation to be im-

Fig. 2. Model showing the two pathways for peroxisomal protein targeting. The PTS-speci¢c soluble receptors Pex5p and Pex7p rec-ognise PTS-containing proteins in the cytoplasm. The loaded receptors subsequently bind the peroxisomal membrane docking complexthat consists of Pex13p, Pex14p and Pex17p. The docking complex is required for both PTS1 and PTS2 protein import.

BBAMCR 14514 30-7-99


ported. Human alanine:glyoxylate aminotransferaseI (AGT) can be imported both as a homodimer andas a monomer [53]. Furthermore, in vitro importstudies into pumpkin glyoxysomes revealed thatmonomeric isocitrate lyase was imported more e¤-ciently than the homotetramer [20]. Alcohol oxidasesof the yeast Hansenula polymorpha, Candida boidiniiand Pichia pastoris are targeted to peroxisomes asunassembled monomers [56^58]. Assembly of alcoholoxidase into homo-octamers has been proposed tooccur in the peroxisomal matrix [56,59,60].

3.3. Molecular chaperones

Molecular chaperones act by binding to non-na-tive substrate proteins in stoichiometric amounts,keeping them in a soluble state and preventingnon-productive interactions (for review see [61]). Mo-lecular chaperones of the 70-kDa heat shock proteinfamily are involved in a variety of di¡erent processesincluding protein folding, disassembly of oligomericprotein complexes, delivery of proteins to proteasesand translocation of polypeptides across intracellularmembranes. The diverse actions of Hsp70s are regu-lated by Hsp40s, J-domain-containing proteins thatrecruit Hsp70 to their site of action. The J-domainbinds a speci¢c Hsp70 family member, whereas otherparts of the J-domain-containing protein are thoughtto interact with the HSP70 substrate (for review see[62,63]).

A requirement for a Hsp70/Hsp40 chaperone sys-tem in peroxisomal protein import is now well estab-lished. Import of a folded peroxisomal matrix pro-tein was inhibited by an antiserum against thecytoplasmic Hsp73 in a permeabilised-cell assayand in a microinjection-based import assay into hu-man ¢broblasts [64]. They reported also that Hsp73was found associated with the peroxisomal surface inrat hepatocytes, especially under conditions of per-oxisome proliferation.

Import of isocitrate lyase requires the presence ofHsp70 in an in vitro glyoxysomal import assay [20].Furthermore, a prenylated Hsp40 and an Hsp70have been shown associated with plant glyoxysomemembranes [20,65,66]. Recently, we have reportedthe identi¢cation of a cytosolic J-domain-containingprotein (Djp1p) that is speci¢cally required for per-oxisomal protein import in S. cerevisiae [67]. Cells

lacking the DJP1 gene partially mislocalise peroxiso-mal matrix proteins to the cytoplasm. The cytosoliclocation of Djp1p and its speci¢c role in peroxisomalprotein import are intriguing and suggest an earlyrole in the targeting process. Whereas the depend-ence of Hsp70 for import of glyoxysomal isocitratelyase into glyoxysomes was during protein transla-tion [20], Hsp73 and Djp1p are required at a post-translational step [64,67].

What is the post-translational role of the Hsp40/Hsp70 machinery in peroxisomal protein import?Although Hsp40s and Hsp70s have been implicatedin the folding and assembly of proteins, we considerit unlikely that Djp1p is required for folding of per-oxisomal matrix proteins into their native conforma-tion in the cytoplasm since mislocalised enzymes areactive in vdjp1 cells. Furthermore, as mentionedabove, acquiring a native conformation does not nec-essarily imply that the protein is more e¤ciently im-ported [20]. Therefore, Djp1p could be involved inputting or keeping newly synthesised peroxisomalproteins in a translocation-competent (monomeric?)conformation. However, the mainly cytoplasmic lo-calisation of Djp1p is also compatible with a role atthe level of the targeting process. It could, for in-stance, stimulate the recognition of peroxisomal pro-teins by their cognate receptors by either acting onthe PTS-containing protein or by acting on the PTSreceptor. Alternatively, Hsp40/Hsp70 may mediatethe various conformational changes that either thePTS-containing protein or the various import factorshave to undergo during the translocation process.Despite serious e¡orts to identify molecular chaper-ones in the peroxisomal matrix that could assist ei-ther in protein assembly or in the import processitself by pulling in the matrix proteins, none havebeen found. A possible exception is an Hsp70 whichis localised mainly to plastids and to a minor extentto glyoxysomes. This protein contains a weak PTS2signal that is able to direct a reporter protein intoperoxisomes in yeast [68].

3.4. Two pathways for peroxisomal matrix proteintargeting

The isolation of peroxisome assembly mutants hasresulted in the identi¢cation of a large number ofgenes encoding proteins (peroxins) involved in per-

BBAMCR 14514 30-7-99


oxisome biogenesis (for review see [8,13]). Phenotypicanalysis of S. cerevisiae peroxisome assembly mu-tants revealed that import of PTS1 proteins andPTS2 proteins could be separated genetically [69]:in vpex5 cells, PTS1 proteins were mislocalised tothe cytosol and PTS2 proteins were imported intoperoxisomal structures: in vpex7 cells, import ofPTS2 proteins was disturbed whereas import ofPTS1 proteins was una¡ected (Fig. 2). In the otherpex mutants, both PTS1 and PTS2 proteins weremislocated to the cytosol, which suggested thatsome components of the import machinery for ma-trix proteins are used by both PTS1 and PTS2 pro-teins [69]. Similar phenotypes were observed amongthe ¢broblasts derived from patients su¡ering fromperoxisome biogenesis disorders [44,70]. These di¡er-ential import mutants turned out to be mutated intheir soluble PTS receptors. On the other hand, mostpex mutants with a general import defect are mu-tated in either a peroxisomal integral-membrane pro-tein or a peroxisomal membrane-associated protein.

3.5. The PTS1 receptor, Pex5p

The PTS1 receptor has been shown to be mutatedin PTS1-speci¢c import-de¢cient mutants in yeast(pex5) and man. Pex5p has been shown to be thePTS1 receptor by its speci¢c interaction with thePTS1 in the two-hybrid assay, and its direct interac-tion with PTS1-containing peptides (for review see[35,71,72]).

The subcellular localisation of Pex5p has strongimplications as to how it functions. However, thishas been a matter of debate between various groupsthat work on di¡erent model organisms. Pex5p hasbeen found: (1) both peroxisome-associated and inthe cytosol; (2) exclusively in the peroxisomal ma-trix; and (3) mainly associated with the peroxisomalmembrane. The various localisations led to fourspeculative models of PTS1 receptor functioning(for review see [10,35,71]). Careful quantitation ofthe steady-state distribution of Pex5p in S. cerevisiae,P. pastoris and human ¢broblasts has revealed thatPex5p is located mainly in the cytoplasm with only afew percent being peroxisome-associated [73^75]. Thedual localisation of Pex5p in these experiments sug-gests that it functions as a cycling receptor whichpicks up a PTS1-containing protein in the cytosol

and delivers it at the peroxisomal membrane. Therate-limiting step in Pex5p cycling might vary de-pending on organisms and culture conditions, there-by leading to a variable steady-state distribution be-tween cytoplasm and association with peroxisomes.The observed intraperoxisomal pool of the PTS1 re-ceptor might either be formed by dead-end moleculesor might re£ect an import pathway by which thePTS1 receptor is co-imported with the PTS1-contain-ing protein. Since the PTS1 receptor is thought tofunction catalytically, it should subsequently be ex-ported out of peroxisomes back into the cytoplasmfor another round of targeting.

Experimental evidence supporting cycling of thePTS1 receptor between cytoplasm and the peroxiso-mal surface was obtained in human ¢broblasts. Con-ditions that inhibit peroxisomal protein import invitro (low temperature and ATP depletion) couldpartially trap Pex5p at the surface of peroxisomesin vivo [75]. Pex5p returned to the cytoplasm afterrestoration of import conditions. Pex5p accumula-tion at the peroxisomal membrane could be repeatedby another incubation under protein import inhibi-tory conditions.

Although early studies in yeast showed that thePTS1 receptor is required for PTS1 import only,studies in mammalian cells show a requirement forthe PTS1 receptor in PTS2 import [76^79]. Two iso-forms of mammalian Pex5p have been identi¢ed thatdi¡er in the presence of an additional exon. Thisadditional exon is required for PTS2 import [79,80].

3.6. The PTS2 receptor, Pex7p

The PTS2 receptor is encoded by the PEX7 gene[81^86]. ScPex7p has been shown to interact with thePTS2 in a two-hybrid assay and by co-immunopre-cipitation with a PTS2-containing protein. Further-more, a temperature-sensitive mutation in the thio-lase PTS2 could be suppressed by overexpression ofPex7p in S. cerevisiae [36,87]. The localisation ofyeast Pex7p has been highly debated. Based on car-boxy-terminally epitope-tagged versions of ScPex7p,Pex7p has been suggested to be located mainly insideperoxisomes [82,87]. After overexpression of amino-terminally epitope-tagged Pex7p, only a minoramount of Pex7p was found associated with perox-isomes, with most residing in the cytosol [36,81].

BBAMCR 14514 30-7-99


Elgersma et al. [86] obtained similar results by over-expression and epitope tagging of P. pastoris Pex7p.However, when they studied the subcellular distribu-tion of unmodi¢ed endogenous Pex7p (using a poly-clonal antiserum directed against PpPex7p), theyfound Pex7p located mainly in the cytoplasm withonly a minor fraction present inside peroxisomes.Two models have been postulated for Pex7p func-tion. In the cycling receptor model, Pex7p picks upa PTS2-containing protein in the cytoplasm and de-livers it to the peroxisomal membrane. Pex7p andligand might either dissociate at the peroxisomalmembrane with only the ligand being translocatedand Pex7p returning back into the cytoplasm, orligand and Pex7p may be co-imported into peroxi-somes where they dissociate, with Pex7p being ex-ported back into the cytoplasm. In an alternativemodel Pex7p functions as an intraperoxisomalPTS2 receptor that pulls PTS2-containing proteinsinto the peroxisomal matrix in a manner similar tohow mitochondrial Hsp70 and BiP are thought tocontribute to import of proteins into mitochondriaand the endoplasmic reticulum, respectively (re-viewed by [88]. Interestingly, ScPex7p is able to asso-ciate with two peroxins (Pex18p and Pex21p) that arefunctionally redundant and that display a low aminoacid sequence homology [16]. Cells lacking bothPex18p and Pex21p fail to import PTS2-containingproteins and Pex7p is located exclusively in the cy-tosol. These observations suggest that Pex18p/Pex21p function in directing Pex7p to peroxisomes.The localisation of Pex18p (mainly cytosolic, with aminor fraction associated with the cytoplasmic faceof the peroxisomal membrane) is in agreement withsuch a role. For our understanding of Pex7p func-tioning, it is crucial to ¢nd out whether Pex18p/Pex21p target the loaded PTS2 receptor or the un-loaded receptor to peroxisomes. It is interesting tonote that by using the two hybrid assay, Purdue et al.[16] were able to show that Pex18p could interactwith a PTS2-containing protein only in the presenceof the PEX7 gene.

3.7. Docking and translocation

At the peroxisomal membrane we can distinguishthree events that contribute to the translocation ofperoxisomal matrix proteins across the membrane.

First, the PTS-containing protein complexed to itscognate receptor binds to a component of the mem-brane-residing docking machinery. Second, the PTS-containing protein is delivered to the translocationmachinery and is translocated across the membrane.Third, the receptor shuttles back to the cytoplasmand the docking and translocation machinery canreceive new substrates.

Candidate proteins that could facilitate dockingand translocation were ¢rst identi¢ed in S. cerevisiae(for review see [7]). Both the PTS1 receptor andPTS2 receptor have been shown to interact with aperoxisomal membrane complex that consists of atleasts three di¡erent proteins (Fig. 2); these are ei-ther integral peroxisomal membrane proteins(Pex13p) or are associated with the cytoplasmicface of the peroxisomal membrane (Pex14p andPex17p) [14,73,74,89^93]. Their are multiple bindingsites for the PTS receptors on this complex whichmay be redundant, although the peroxins themselvesare not redundant for the import process (see below).Cells lacking either of these peroxins (Pex17p orPex13p) fail to import both PTS1 and PTS2 proteins.Therefore, loss of one of its components could a¡ectthe whole complex, and since this heteromeric com-plex contains the putative docking sites for both thePTS1 receptor and the PTS2 receptor, this wouldlead to a block in both PTS1 and PTS2 protein tar-geting and import. Alternatively, the function ofPex13p, Pex14p and Pex17p may not be limited tothe formation of docking sites for PTS receptors, butthey may also part of the translocation machinery(translocon?). The multiple PTS-receptor-bindingperoxins that are located at the peroxisomal mem-brane (Pex13p, Pex14p and Pex17p) might re£ect theexistence of an import cascade where successive in-teractions of a PTS receptor with peroxins at theperoxisomal membrane might induce conformationalchanges in the PTS receptor, which are required for:(1) docking; (2) dissociation of the PTS-containingprotein; (3) passage of the PTS-containing protein toother peroxins; and (4) recycling of the receptor tothe cytosol [7,90]. For the PTS2 receptor this cascademight also involve Pex18p/Pex21p in addition to thePex13p/Pex14p/Pex17p complex.

Additional peroxins that are likely to be involvedin peroxisomal matrix protein import are Pex2p andPex12p. These integral peroxisomal membrane pro-

BBAMCR 14514 30-7-99


teins contain a RING ¢nger domain that is thoughtto mediate protein^protein interactions [94]. In pa-tient cells, in which either their PEX2 or PEX12 geneis mutated, peroxisomal matrix import is impairedand the PTS1 receptor has been shown to accumu-late at the cytoplasmic face of the peroxisomal mem-brane [75]. It is possible that Pex2p and Pex12p arerequired for translocation of PTS-containing proteinsand that a block at this step results in accumulationof the PTS1 receptor (possibly loaded with cargo) atthe peroxisomal membrane, analogous to a block intranslocation caused by ATP depletion [75]. Alterna-tively, Pex2p and Pex12p may be required either forpassage of the PTS1 receptor through an import cas-cade or for recycling of the PTS1 receptor back intothe cytoplasm. Since PTS2 import is also impaired inthese patients' ¢broblasts, it would be interesting tostudy the localisation of the PTS2 receptor in thesepatients. The PTS1 receptor has been shown to ac-cumulate inside peroxisomes in H. polymorpha vpex4mutant cells [95] and in a patient with a mutation inits PEX10 gene [75]. It is for future research to es-tablish whether these peroxins are required for ex-port of the PTS1 receptor back into the cytoplasmor whether they normally prevent the PTS1 receptorfrom entering peroxisomes.

The mechanism by which the import machinerycan accommodate the translocation of folded andoligomerised peroxisomal proteins across the perox-isomal membrane remains enigmatic. One way toaccommodate the translocation of folded proteinsacross membranes is via large pores analogous tonuclear pores. A nuclear pore can mediate the pas-sage of large multisubunit complexes and particles upto 25 nm. In contrast to the nuclear pore, whichprovides an 9-nm di¡usion channel through whichmolecules smaller than 60 kDa can di¡use, the puta-tive peroxisomal pore should be able to open tran-siently to allow translocations to occur but shouldthen close completely to maintain the permeabilitybarrier for small molecules. Freeze-fracture electronmicroscopy studies have failed to identify large pro-tein complexes in the peroxisomal membrane with astructure similar to the nuclear pore [96]. It has beenpostulated that this putative pore assembles onlytransiently and is therefore hard to detect [6]. Analternative way to import peroxisomal proteinswould be via an endocytic event occurring at the

peroxisomal membrane [6]. Newly synthesised perox-isomal proteins are targeted to a speci¢c region onthe surface of peroxisomes, which invaginates andbuds o¡ into the peroxisomal matrix. Subsequentdegradation of the membrane or redistribution ofmembrane lipids will then lead to release of the cargoin the matrix. However, small internal vesicularstructures are found only rarely, and whether thesestructures are import vesicles is not known. Thesetwo completely opposing models illustrate the gapin our knowledge concerning the actual proteintranslocation/incorporation event. Recently, an evo-lutionarily conserved pathway has been identi¢ed fortransport of stably folded proteins across the chloro-plast thylakoid membrane and bacterial inner mem-brane. Also for this pathway the translocation mech-anism remains unclear (for review see [97]).

Jamming of the import site by proteins that haveto be transported across other membranes (mito-chondria, ER, chloroplast and bacterial plasmamembrane) has been very informative for the com-position and architecture of the translocation appa-ratus of these organelles. Unfortunately, no jammingof PTS-containing proteins in vivo has ever beenshown. The recent observation that a peroxisomalfusion protein get stuck in the import site in vitro[21] might open the way to investigate the compo-nents and architecture of the peroxisomal transloca-tion apparatus in more detail.

3.8. The peroxisomal matrix

Upon translocation of a PTS-containing proteinacross the peroxisomal membrane, it might eitherremain in the lumen of the peroxisome or associatewith the inner lea£et of the peroxisomal membrane.Whereas most PTS-containing proteins are freely dis-tributed through the peroxisomal matrix, some PTS-containing proteins, including the acyl-CoA synthe-tase Faa2p, are concentrated at the inner lea£et ofthe peroxisomal membrane [98]. Whether Faa2p as-sociates with lipids or with a peroxisomal membraneprotein is not clear.

For some proteins folding and assembly might oc-cur in the cytoplasm whereas other proteins are im-ported as monomers and have to assemble insideperoxisomes. The high concentration of monomersand cofactors (for instance FAD) inside peroxisomes

BBAMCR 14514 30-7-99


may drive oligomerisation, so that no additionalfolding enzymes/factors are required for protein as-sembly. Alternatively, folding enzymes/factors maybe required but these have not yet been identi¢ed.Various peroxins have been proposed to function in-side peroxisomes to stimulate peroxisomal proteinimport. As has been discussed above, a varying frac-tion of the PTS receptors has been found inside per-oxisome. Another peroxin that might be involved inprotein translocation is Pex8p which is an intraper-oxisomal protein that is associated with the innerlea£et of the peroxisomal membrane [99^101]. Itsrole in the import process remains to be established.

4. Peroxisomal membrane proteins

The peroxisomal membrane harbours a large set ofproteins that are required for metabolite transport,matrix protein import, membrane protein insertion,segregation to progeny cells and proteins that mod-ulate organelle shape and spacial distribution in thecell. As has been shown for peroxisomal matrix pro-teins, integral peroxisomal membrane proteins(PMPs) are synthesised on free polyribosomes [102^104] with one possible exception [104]. Pulse-chaseexperiments revealed that the rat peroxisomal ABCtransporter Pmp70p is released at its mature weightinto the cytosol (de¢ned as a 100 000Ug superna-tant) before it binds peroxisomes. Peroxisome bind-ing is rapidly followed by insertion into the mem-brane as determined by resistance to alkalineextraction [105]. These in vivo studies clearly demon-strated that Pmp70p is post-translationally imported.Furthermore, in vitro studies con¢rm and extendthese studies. Pmp70p and the multi-membrane-spanning peroxisomal membrane protein Pmp22pbind and insert speci¢cally into puri¢ed rat liver per-oxisomal membranes [105,106]. Binding and inser-tion depend on the presence of both cytosolic andperoxisome-associated factors. ATP- or GTP-hydrol-ysis are not required in this binding and insertionassay (see below).

4.1. Targeting of peroxisomal membrane proteins

Genetic studies have revealed that PMPs are tar-geted/inserted (at least partly) independently from

the peroxisomal matrix protein import pathways.Firstly, most pex mutants are disturbed in matrixprotein import but assemble their membrane proteinsproperly with only a few exceptions (see below).These observations are compatible with a model inwhich translocation of matrix proteins and insertionof PMPs require speci¢c components as well as com-mon components. However, a more likely explana-tion is that, due to a defect in PMP insertion, matrixprotein import is secondarily a¡ected since the latterprocess depends on the presence PMPs (for instancePex13p).

Secondly, the targeting signals for PMPs are dis-tinct from the PTS1 and PTS2. The PTS for C. boi-dinii Pmp47p resides in a 20 amino acid hydrophilicloop between two membrane-spanning regions [107].Similar amino acid sequences have been identi¢ed inregions of Pex3p and Pex15p that are required formembrane targeting [108^110] (see below for moredetail).

4.2. Factors required for PMP insertion

Post-translational import of hydrophobic proteinsrequires the presence of accessory proteins that keepthe protein soluble and direct it to the acceptor mem-brane. Some of the cytosolic and membrane factorsrequired for binding and insertion of PMPs into per-oxisomal membranes have been identi¢ed. Indeed,rat Pmp22p synthesised in a reticulocyte lysate as-sembles into two complexes that keep this hydropho-bic protein soluble in the absence of peroxisomalmembranes [111]. In the ¢rst complex, Pmp22p wasfound to be associated with the TCP1 ring complex,the cytosolic chaperonin. In the second complex,Pmp22p was bound to a 40-kDa polypeptide.Pmp22p present in this second complex insertedwith high e¤ciency into peroxisomal membranesand therefore the 40-kDa protein has been postu-lated to be the cytoplasmic Pmp22p receptor [111].The identity of this 40-kDa protein remains to beestablished.

As mentioned above, analysis of yeast and mam-malian pex mutants have resulted in the identi¢ca-tion of three peroxins required for insertion of per-oxisomal membrane proteins [108,109,112,113].These peroxins (Pex3p, Pex16p and Pex19p) mightbe involved in either the PMP insertion process itself

BBAMCR 14514 30-7-99


or alternatively, in the formation of peroxisomalmembranes.

There are now various observations that mightindicate that not all peroxisomal PMPs are post-translationally imported into the peroxisomal mem-branes directly but might reach peroxisomes via theendoplasmic reticulum (for review see [11,12]). Anal-yses of the topogenic signals of two yeast PMPs sug-gest a relationship between peroxisomes and ER. Theamino-terminal 40 amino acids from P. pastorisPex3p contain all the information for targeting andinsertion into peroxisomal membranes. The ¢rst 16amino acid residues, which precede the putativetransmembrane domain, can direct a reporter proteinto the ER. Whether this fusion protein resides insidethe ER or whether it associates with an ER surfacecomponent has not been investigated [109]. Overex-pression of wild-type and modi¢ed versions ofPex15p (truncation or fusion proteins) localise tothe ER or plasma membrane [110]. These resultssuggest (but do not prove) that Pex3p and Pex15pare normally targeted to the ER membrane and aresubsequently sorted to the peroxisomal membrane.This process might be disturbed by overexpressionor truncation of these proteins. In line with this,the Pex3p membrane-spanning region would containthe ER-to-peroxisome sorting information. A moredetailed analysis of this amino acid sequence willreveal whether such speci¢c sorting signals are indeedpresent.

However, we should be careful about studying thetargeting/sorting signals of peroxisomal membraneproteins after overexpression or truncation. For ex-ample, overexpression of ScPat1p, a yeast homo-logue of rat Pmp70p, results in mislocalisation tothe nuclear envelope and the lateral ER. This sug-gests that cryptic ER targeting signals are present insome peroxisomal PMPs (E.H.H. unpublished re-sults), or worse, that the ER constitutes a sink thatattracts overproduced hydrophobic proteins.

Experimental evidence has been provided by Tito-renko and Rachubinski [114] who showed that atleast two peroxins of the yeast Yarrowia lipolytica,Pex2p and Pex16p, are targeted ¢rst to the ER,where they obtain ER-speci¢c modi¢cations, andare subsequently sorted to peroxisomes. Signalsthat mediate ER targeting and subsequent sortingto peroxisomes have not yet been identi¢ed. Despite

many attempts have been made, no PMPs have beenobserved to reach peroxisomes via the endoplasmicreticulum in other model systems (see, for instance,[110,112,115]).

4.3. Peripheral peroxisomal membrane proteins

Several peroxisomal membrane proteins have beenidenti¢ed that associate with the peroxisomal mem-brane. Are these latter proteins speci¢cally binding toprotein determinants of the peroxisomal membraneor are these proteins recognising speci¢c lipid deter-minants? Alternatively, speci¢city might be achievedby a combination of both, as has been described forthe endosomal RING FYVE domain-containing pro-tein EEA1 [116]. For several peripheral membrane-associated peroxins it has now been shown that theyindeed interact with PMPs directly or indirectly. Forinstance, the farnesylated peroxin Pex19p associateswith the PMP Pex3p [15] and Pex14p interacts withthe PMP Pex13p [90].

A major obstacle in research of peroxisomal mem-brane protein studies is the lack of solid experimentaltechniques to verify whether proteins are integralmembrane proteins or whether they are associatedwith the peroxisomal membrane. This has led tomany contradictory results. A well-established wayof addressing this question has been extraction ofmembranes with 0.1 M sodium carbonate, pH 11.5,followed by pelleting of membranes at 100 000Ug.However, it seems that the outcome of this experi-ment is in£uenced by the source of the membranes[117^119]. Another major disadvantage is the lack ofmodi¢cations of peroxisomal proteins on the luminalside of the organelle. Therefore, topology studieshave to rely on either protease protection experi-ments or analysis of fusion proteins (which mightin£uence topology), or on experiments in which anti-sera against (certain regions of) membrane proteinsare tested for their ability to bind antigen after se-lective permeabilisation of the plasma membrane.This limited set of techniques combined with theuse of various model systems have lead to con£ictingdata concerning the topology of a number of perox-ins including for instance Pex2p, Pex3p, Pex11p andPex12p. Therefore, new tools have to be developedand inter-laboratory di¡erences have to be sorted outsystematically.

BBAMCR 14514 30-7-99


5. Species-speci¢c di¡erences in peroxisomal proteinimport

Some details of the import process vary betweenspecies. As discussed above, the mammalian PTS1receptor is required for PTS2 protein import whereasin yeast, the PTS1 receptor functions only in PTS1protein import. H. polymorpha Pex4p is required spe-ci¢cally for PTS1 import whereas in other yeast spe-cies, Pex4p is required for import of both PTS1- andPTS2-containing proteins [95,120,121]. In particular,the yeast Y. lipolytica seems to be an exception withrespect to both matrix and membrane protein im-port. For instance, Pex2p and Pex16p reach peroxi-somes via the endoplasmic reticulum [114]. This hasnot been observed for other Pex2p and Pex16p or-thologues. Furthermore, the peroxins Pex1p andPex6p are required for both peroxisome assemblyand secretion of some proteins [11]. This is in con-trast to all other organisms in which pex1 and pex6mutants have been isolated.

6. Future challenges

Although genetic and biochemical approacheshave resulted in the identi¢cation of most of thecomponents of the matrix protein import machinery,the mechanism of peroxisomal protein translocationremains unsolved. For instance, are proteins trans-located through a channel or are they engulfed byperoxisomal membranes which dissolve after enteringthe peroxisomal lumen? Where do PTS receptors andPTS-containing proteins dissociate, in other words,are PTS receptors co-imported and subsequently ex-ported or does dissociation take place at the cyto-plasmic surface of the peroxisomal membrane? Isthere a permanent or a transient docking or importcomplex? Furthermore, it would be rather ine¤cientif PTS receptors could dock at the peroxisomal mem-brane without being loaded with a PTS-containingprotein. Therefore, it is likely that binding of ligandinduces a conformational change in the PTS receptorwhich might lead to exposure of a domain that caninteract with the docking complex. So what is thatdocking domain and how are conformationalchanges mediated?

These kind of questions need experimental ap-

proaches whereby we study the dynamics of the per-oxisomal protein import process. One approachwould be to use conditional alleles for the variousperoxins in combination with radioactive pulse-chaseexperiments to study the e¡ects on other peroxinsand the import process. Additionally, we have tostudy the peroxins at the structural level to ¢nd outwhat the protein^protein interaction domains areand how binding to these domains results in con-formational changes of the various peroxins. Fur-thermore, the promising in vitro import systemsdeveloped for plant glyoxysomes will certainly con-tribute to insight in the role of peroxins in the importprocess and the sequence of events during transloca-tion.

The targeting and insertion of membrane proteinsis less well characterised compared to import of per-oxisomal matrix proteins. Novel genetic screenscould identify more components for this process.The role of the endoplasmic reticulum in peroxiso-mal membrane protein sorting is unclear for mostorganisms and needs to be veri¢ed. Since most genet-ic screens did not select for conditional mutants, itremains possible that we have missed mutants in es-sential genes that might function in both peroxisomeassembly and the secretory pathway or other essen-tial processes.

Acknowledgements

We thank Ineke Braakman, Alison Motley and themembers of the peroxisome biogenesis group forstimulating discussions. This work was supportedby a grant from the Netherlands Organisation ofScienti¢c Research (NWO) to H.F.T.

References

[1] R.W. Breidenbach, H. Beevers, Association of glyoxylatecycle enzymes in a novel subcellular particle from castorbean endosperm, Biochem. Biophys. Res. Commun. 27(1967) 462^469.

[2] S. Fukui, A. Tanaka, Peroxisomes of alkane- and methanol-grown yeasts, J. Appl. Biochem. 1 (1979) 171^202.

[3] F. Opperdoes, P. Borst, Localisation of nine glycolytic en-zymes in a microbody-like organelle in T. brucei : the glyco-some, FEBS Lett. 80 (1977) 360^364.

BBAMCR 14514 30-7-99


[4] S. Subramani, Convergence of model systems for peroxisomebiogenesis, Curr. Opin. Cell Biol. 8 (1996) 513^518.

[5] Y. Elgersma, H.F. Tabak, Proteins involved in peroxisomebiogenesis and functioning, Biochim. Biophys. Acta 1286(1996) 269^283.

[6] J.A. McNew, J.M. Goodman, The targetting and assemblyof peroxisomal proteins: some old rules do not apply,Trends Biochem. Sci. 21 (1996) 54^58.

[7] R. Erdmann, M. Veenhuis, W.-H. Kunau, Peroxisomes: or-ganelles at the cross roads, Trends Cell Biol. 7 (1997) 400^407.

[8] S. Subramani, Components involved in peroxisome import,biogenesis, proliferation, turnover and movement, Physiol.Rev. 78 (1998) 171^188.

[9] P.B. Lazarow, Y. Fujiki, Biogenesis of peroxisomes, Annu.Rev. Cell Biol. 1 (1985) 489^530.

[10] H.R. Waterham, J.M. Cregg, Peroxisome biogenesis, BioEs-says 19 (1997) 57^66.

[11] V.I. Titorenko, R.A. Rachubinski, The endoplasmic reticu-lum plays an essential role in peroxisome biogenesis, TrendsBiochem. Sci. 23 (1998) 231^233.

[12] W.-H. Kunau, R. Erdman, Peroxisome biogenesis : Back tothe endoplasmic reticulum?, Curr. Biol. 8 (1998) 299^302.

[13] B. Distel, R. Erdmann, S.J. Gould, G. Blobel et al., A uni-¢ed nomenclature for peroxisome biogenesis factors, J. CellBiol. 135 (1996) 1^3.

[14] B. Huhse, P. Rehling, M. Albertini, L. Blank, K. Meller,W.-H. Kunau, Pex17p of Saccharomyces cerevisiae is a novelperoxin and component of the peroxisomal protein translo-cation machinery, J. Cell Biol. 140 (1998) 49^60.

[15] K. Go«tte, W. Girzalsky, M. Linkert, E. Baumgart, S. Kam-merer, W.-H. Kunau, R. Erdmann, Pex19p, a farnesylatedprotein essential for peroxisome biogenesis, Mol. Cell. Biol.18 (1998) 616^628.

[16] P.E. Purdue, X. Yang, P.B. Lazarow, Pex18p and Pex21p, anovel pair of related peroxins essential for peroxisomal tar-getting by the PTS2 pathway, J. Cell Biol. 143 (1998) 1859^1869.

[17] A. Baker, In vitro systems in the study of peroxisomal pro-tein import, Experientia 52 (1996) 1055^1062.

[18] D.G. Brickner, J.J. Harada, L.J. Olsen, Protein transportinto higher plant peroxisomes: in vitro import assay pro-vides evidence for receptor involvement, Plant Physiol. 113(1997) 1213^1221.

[19] D.G. Brickner, L.J. Olsen, Nucleotide triphosphates are re-quired for the transport of glycolate oxidase into peroxi-somes, Plant Physiol. 116 (1998) 309-317.

[20] W. Crookes, L.J. Olsen, The e¡ects of chaperones and thein£uence of protein assembly on peroxisomal protein import,J. Biol. Chem. 273 (1998) 17236^17242.

[21] M.R. Pool, E. Lopez-Huertas, A. Baker, Characterisation ofintermediates in the process of plant peroxisomal proteinimport, EMBO J. 17 (1998) 6854^6862.

[22] S. Subramani, Protein import into peroxisomes and biogen-esis of the organelle, Annu. Rev. Cell Biol. 9 (1993) 445^478.

[23] S. Rapp, U. Soto, W.W. Just, Import of ¢re£y luciferase intoperoxisomes of permeabilized Chinese hamster ovary cells : a

model system to study peroxisomal protein import in vitro,Exp. Cell Res. 205 (1993) 59^65.

[24] M. Wendland, S. Subramani, Cytosol-dependent peroxiso-mal protein import in a permeabilized cell system, J. CellBiol. 120 (1993) 675^685.

[25] P.A. Walton, S.J. Gould, J.R. Feramisco, S. Subramani,Transport of microinjected proteins into peroxisomes ofmammalian cells : inability of Zellweger cell lines to importproteins with the SKL tripeptide peroxisomal targeting sig-nal, Mol. Cell. Biol. 12 (1992) 531^541.

[26] P.A. Walton, S.J. Gould, R.A. Rachubinski, S. Subramani,J.R. Feramisco, Transport of microinjected alcohol oxidasefrom Pichia pastoris into vesicles in mammalian cells : in-volvement of the peroxisomal targeting signal, J. Cell Biol.118 (1992) 499^508.

[27] S.J. Gould, G.-A. Keller, S. Subramani, Identi¢cation ofperoxisomal targeting signals located at the carboxy termi-nus of four peroxisomal proteins, J. Cell Biol. 107 (1988)897^905.

[28] S.J. Gould, G.A. Keller, M. Schneider, S.M. Howell, L.Garrard, J.M. Goodman, B. Distel, H.F. Tabak, S. Subra-mani, Peroxisomal protein import is conserved betweenyeast, insects and mammals, EMBO J. 9 (1990) 85^90.

[29] A. Motley, M.J. Lumb, P.B. Patel, P.R. Jennings, P.A. DeZoysa, R.J.A. Wanders, H.F. Tabak, C.J. Danpure, Identi-¢cation of the peroxisomal targeting sequence of mammalianalanine:glyoxylate aminotransferase 1. Increased degeneracyand context speci¢city of the mammalian PTS1 motif andimplications for the peroxisome-to-mitochondrion mistarget-ing of AGT in primary hyperoxaluria type 1, J. Cell Biol.131 (1995) 95^109.

[30] Y. Elgersma, A. Vos, M. van den Berg, C.W.T. van Roer-mund, P. van der Sluijs, B. Distel, H.F. Tabak, Analysis ofthe carboxy-terminal peroxisomal targetting signal 1 in ahomologous context in Saccharomyces cerevisiae, J. Biol.Chem. 271 (1996) 26375^26382.

[31] P.E. Purdue, P.B. Lazarow, Targetting of human catalase toperoxisomes is dependent upon a novel COOH-terminal per-oxisomal targetting sequence, J. Cell Biol. 134 (1996) 849^862.

[32] R.T. Mullen, M.S. Lee, C.R. Flynn, R.N. Trelease, Diverseamino acid residues function within the type 1 peroxisomaltargetting signal, Plant Physiol. 115 (1997) 881^889.

[33] G. Lametschwandtner, C. Brocard, M. Fransen, P. vanVeldhoven, J. Berger, A. Hartig, The di¡erence in recogni-tion of terminal tripeptides as peroxisomal targeting signal 1between yeast and human is due to di¡erent a¤nities of theirreceptor Pex5p to the cognate signal and to residues adjacentto it, J. Biol. Chem. 273 (1998) 33635^33643.

[34] P.E. Purdue, P.B. Lazarow, Peroxisomal biogenesis : multi-ple pathways of protein import, J. Biol. Chem. 269 (1994)30065^30068.

[35] R.A. Rachubinski, S. Subramani, How proteins penetrateperoxisomes, Cell 83 (1995) 525^528.

[36] P. Rehling, M. Marzioch, F. Niesen, E. Wittke, M. Veen-huis, W.-H. Kunau, The import receptor for the peroxisomal

BBAMCR 14514 30-7-99


targeting signal2 (PTS2) in Saccharomyces cerevisiae is en-coded by the PAS7 gene, EMBO J. 15 (1996) 2901^2913.

[37] B.W. Swinkels, S.J. Gould, A.G. Bodnar, R.A. Rachubinski,S. Subramani, A novel, cleavable peroxisomal targeting sig-nal at the amino-terminus of the rat 3-ketoacyl-CoA thio-lase, EMBO J. 10 (1991) 3255^3262.

[38] T. Osumi, T. Tsukamoto, S. Hata, S. Yokota, S. Miura, Y.Fujiki, M. Hijikata, S. Miyazawa, T. Hashimoto, Amino-terminal presequence of the precursor of peroxisomal 3-ke-toacyl-CoA thiolase is a cleavable signal peptide for perox-isomal targeting, Biochem. Biophys. Res. Commun. 181(1991) 947^954.

[39] R. Preisig-Muller, H. Kindl, Thiolase mRNA translated invitro yields a peptide with a putative N-terminal prese-quence, Plant Mol. Biol. 22 (1993) 59^66.

[40] C.F. Gietl, K.N. Faber, I.J. van der Klei, M. Veenhuis,Mutational analysis of the N-terminal topogenic signal ofwatermelon glyoxysomal malate dehydrogenase using theheterologous host Hansenula polymorpha, Proc. Natl.Acad. Sci. USA 91 (1994) 3151^3155.

[41] A. Kato, M. Hayashi, M. Kondo, M. Nishimura, Targetingand processing of a chimeric protein with the N-terminalpresequence of the precursor to glyoxysomal citrate syn-thase, Plant Cell 8 (1996) 1601^1611.

[42] E.C.J.M. de Vet, A.W.M. Zomer, G.J.H.T.J. Lahaut, H. vanden Bosch, Polymerase chain reaction-based cloning of al-kyl-dihydroxyacetonephosphate synthase complementaryDNA from guinea pig liver, J. Biol. Chem. 272 (1997)798^803.

[43] G.A. Jansen, R. Ofman, S. Ferdinandusse, L. Ijlst, T.O.Muijsers, O.H. Skjeldal, O. Stokke, C. Jakobs, G.T.N. Bes-ley, E. Wraith, R.J.A. Wanders, Refsum disease is caused bymutations in the phytanoyl-CoA hydroxylase gene, Nat.Genet. 17 (1997) 190^193.

[44] A. Motley, E. Hettema, B. Distel, H.F. Tabak, Di¡erentialprotein import de¢ciencies in human peroxisome assemblydisorders, J. Cell Biol. 125 (1994) 755^767.

[45] G.M. Small, L.J. Szabo, P.B. Lazarow, Acyl-CoA oxidasecontains two targetting sequences each of which can mediateprotein import into peroxisomes, EMBO J. 7 (1988) 1167^1173.

[46] T. Kamiryo, Y. Sakasegawa, H. Tan, Expression and trans-port of Candida tropicalis peroxisomal acyl-coenzyme A ox-idase in the yeast Candida maltosa, Agric. Biol. Chem. 53(1989) 179^186.

[47] F. Kragler, A. Langeder, J. Raupachova, M. Binder, A.Hartig, Two independent peroxisomal targeting signals incatalase-A of Saccharomyces cerevisiae, J. Cell Biol. 120(1993) 665^673.

[48] Y. Elgersma, C.W.T. Van Roermund, R.J.A. Wanders, H.F.Tabak, Peroxisomal and mitochondrial carnitine acetyltrans-ferases of Saccharomyces cerevisiae are encoded by a singlegene, EMBO J. 14 (1995) 3472^3479.

[49] J.W. Zhang, Y. Han, P.B. Lazarow, Novel peroxisome clus-tering mutants and peroxisome biogenesis mutants of Sac-charomyces cerevisiae, J. Cell Biol. 123 (1993) 1133^1147.

[50] J.R. Glover, D.W. Andrews, R.A. Rachubinski, Saccharo-myces cerevisiae peroxisomal thiolase is imported as a dimer,Proc. Natl. Acad. Sci. USA 91 (1994) 10541^10545.

[51] J.A. McNew, J.M. Goodman, An oligomeric protein is im-ported into peroxisomes in vivo, J. Cell Biol. 127 (1994)1245^1257.

[52] M.S. Lee, R.T. Mullen, R.N. Trelease, Oilseed isocitratelyases lacking their essential type I peroxisomal targettingsignal are piggybacked to glyoxysomes, Plant Cell 9 (1997)185^197.

[53] J.M. Leiper, P.B. Oatey, C.J. Danpure, Inhibition of alani-ne:glyoxylate aminotransferase I dimerisation is a prerequi-site for its peroxisome-to-mitochondrion mistargetting in pri-mary hyperoxaluria type I, J. Cell Biol. 135 (1996) 939^951.

[54] T. Ha«usler, Y.D. Stierhof, E. Wirtz, C. Clayton, Import of aDHFR hybrid protein into glycosomes in vivo is not inhib-ited by the folate-analogue aminopterin, J. Cell Biol. 132(1996) 311^324.

[55] P.A. Walton, P.E. Hill, S. Subramani, Import of stablyfolded proteins into peroxisomes, Mol. Biol. Cell 6 (1995)675^683.

[56] E. Bellion, J.M. Goodman, Proton ionophores prevent as-sembly of a peroxisomal protein, Cell 48 (1987) 165^173.

[57] B. Distel, M. Veenhuis, H.F. Tabak, Import of alcohol ox-idase into peroxisomes of Saccharomyces cerevisiae, EMBOJ. 6 (1987) 3111^3116.

[58] H.R. Waterham, K.A. Russell, Y. de Vries, J.M. Cregg,Peroxisomal targetting, import, and assembly of alcohol ox-idase in Pichia pastoris, J. Cell Biol. 139 (1997) 1419^1431.

[59] M.E. Evers, V. Titorenko, I.J. van der Klei, W. Harder, M.Veenhuis, Assembly of alcohol oxidasein peroxisomes of theyeast Hansenula polymorpha requires the cofactor £avin ad-enine dinucleotide, Mol. Biol. Cell 5 (1994) 829^837.

[60] M.E. Evers, V. Titorenko, W. Harder, I.J. van der Klei, M.Veenhuis, Flavin dinucleotide binding is a crucial step inalcohol oxidase assembly in the yeast Hansenula polymor-pha, Yeast 12 (1996) 917^923.

[61] F.U. Hartl, Molecular chaperones in cellular protein folding,Nature 381 (1996) 571^580.

[62] J. Rassow, O. von Ahsen, U. Boemer, N. Pfanner, Molec-ular chaperones: towards a characterization of the heat-shock protein 70 family, Trends Cell Biol. 7 (1997) 129^133.

[63] W.L. Kelley, The J-domain family and the recruitment ofchaperone power, Trends Biochem. Sci. 23 (1998) 222^227.

[64] P.A. Walton, M. Wendland, S. Subramani, R.A. Rachubin-ski, W.J. Welch, Involvement of 70 kD heat-shock pro-teins in peroxisomal import, J. Cell Biol. 125 (1994) 1037^1046.

[65] R. Preisig-Mueller, G. Munster, H. Kindl, Heat shock en-hances the amount of prenylated Dnaj protein at membranesof glyoxysomes, Eur. J. Biochem. 219 (1994) 57^63.

[66] F.J. Corpas, R.N. Trelease, The plant 73 kDa peroxisomalmembrane protein (PMP73) is immuno-related to molecularchaperones, Eur. J. Cell Biol. 65 (1997) 280^290.

[67] E.H. Hettema, C.C.M. Ruigrok, M. Groot Koerkamp, M.

BBAMCR 14514 30-7-99


van den Berg, H.F. Tabak, B. Distel, I. Braakman, Thecytosolic DnaJ-like protein Djp1p is involved speci¢cally inperoxisomal protein import, J. Cell Biol. 27 (1998) 421^434.

[68] B. Wimmer, F. Lottspeich, I. van der Klei, M. Veenhuis, C.Gietl, The glyoxysomal and plastid molecular chaperones(70-kDa heat shock protein) of watermelon cotyledons areencoded by a single gene, Proc. Natl. Acad. Sci. USA 94(1997) 13624^13629.

[69] I. Van der Leij, M. Van der Berg, R. Boot, M.M. Franse, B.Distel, H.F. Tabak, Isolation of peroxisome assembly mu-tants from Saccharomyces cerevisiae with di¡erent morphol-ogies using a novel positive selection procedure, J. Cell Biol.119 (1992) 153^162.

[70] M.L. Swalecki, G. Dodt, S. Steinberg, A.B. Moser, H.W.Moser, S.J. Gould, Identi¢cation of three distinct peroxiso-mal protein import defects in patients with peroxisome bio-genesis disorders, J. Cell Sci. 108 (1995) 1817^1829.

[71] S. Subramani, Protein translocation into peroxisomes,J. Biol. Chem. 271 (1996) 32483^32486.

[72] H.F. Tabak, Y. Elgersma, E.H. Hettema, M.M. Franse,T. Voorn-Brouwer, B. Distel, Transport of proteins andmetabolites across the impermeable membrane of peroxi-somes, Cold Spring Harbor Symp. Quant. Biol. 60 (1995)649^655.

[73] Y. Elgersma, L. Kwast, A. Klein, T. Vorrn-Brouwer, M. vanden Berg, B. Metzig, T. America, H.F. Tabak, B. Distel, TheSH3 domain of the Saccharomyces cerevisiae peroxisomalmembrane protein Pex13p functions as a docking site forPex5p, a mobile receptor for the import of PTS1-containingproteins, J. Cell Biol. 135 (1996) 97^109.

[74] S.J. Gould, J.E. Kalish, J.C. Morrell, J. Bjorkman, A.J. Ur-quhart, D.I. Crane, An SH3 protein in the peroxisome mem-brane is a docking factor for the PTS1 receptor, J. Cell Biol.135 (1996) 85^95.

[75] G. Dodt, S.J. Gould, Multiple PEX genes are required forproper subcellular distribution and stability of Pex5p, thePTS1 receptor: evidence that PTS1 protein import is medi-ated by a cycling receptor, J. Cell Biol. 135 (1996) 1763^1774.

[76] G. Dodt, N. Braverman, C. Wong, A. Moser, H.W. Moser,P. Watkins, D. Valle, S.J. Gould, Mutations in the PTS1receptor gene, PXR1, de¢ne complementation group 2 ofthe peroxisome biogenesis disorders, Nat. Genet. 9 (1995)115^125.

[77] E.A.C. Wiemer, W.M. Nuttley, B.L. Bertolaet, X. Li,U. Franke, M.J. Wheelock, W.K. Anne, K.R. Johnson, S.Subramani, Human peroxisomal targeting signal-1 receptorrestores peroxisomal protein import in cells from patientswith fatal peroxisomal disorders, J. Cell Biol. 130 (1995)51^65.

[78] M. Baes, P. Gressens, E. Baumgart, P. Carmeliet, M. Cas-teels, M. Fransen, P. Evrard, D. Fahimi, P.E. Declercq,D. Collen, P.P. van Veldhoven, G.P. Mannaerts, A mousemodel for Zellweger syndrome, Nat. Genet. 17 (1997) 49^57.

[79] H. Otera, K. Okumoto, K. Tateishi, Y. Ikoma, E. Matsuda,M. Nishimura, T. Tsukamoto, T. Osumi, K. Ohashi, O.Higuchi, Y. Fujiki, Peroxisome targeting signal type 1(PTS1) receptor is involved in import of both PTS1 andPTS2: studies with PEX5-defective CHO cell mutants,Mol. Cell. Biol. 18 (1998) 388^399.

[80] N. Braverman, G. Dodt, S.J. Gould, D. Valle, An isoform ofPex5p, the human PTS1 receptor, is required for the importof PTS2 proteins into peroxisomes, Hum. Mol. Genet. 7(1998) 1195^1205.

[81] M. Marzioch, R. Erdmann, M. Veenhuis, W.-H. Kunau,PAS7 encodes a novel yeast member of the WD-40 proteinfamily essential for import of 3-oxoacyl-CoA thiolase, aPTS2-containing protein, into peroxisomes, EMBO J. 13(1994) 4908^4917.

[82] J.W. Zhang, P.B. Lazarow, PEB1 (PAS7) in Saccharomycescerevisiae encodes a hydrophilic, intra-peroxisomal proteinthat is a member of the WD repeat family and is essentialfor the import of thiolase into peroxisomes, J. Cell Biol. 129(1995) 65^80.

[83] A.M. Motley, E.H. Hettema, E.M. Hogenhout, P. Brites,A.L.M.A. ten Asbroek, F.A. Wijburg, F. Baas, H.S. Heij-mans, H.F. Tabak, R.J.A. Wanders, B. Distel, Rhizomelicchondrodysplasia punctata is a peroxisomal protein target-ting disease caused by a non-functional PTS-2 receptor, Nat.Genet. 15 (1997) 377^380.

[84] N. Braverman, G. Steel, C. Obie, A. Moser, H. Moser, S.J.Gould, D. Valle, Human PEX7 encodes the peroxisomalPTS2 receptor and is responsible for rhizomelic chondrodys-plasia punctata, Nat. Genet. 15 (1997) 369^376.

[85] P.E. Purdue, J.W. Zhang, M. Skoneczny, P.B. Lazarow,Rhizomelic chondrodysplasia punctata is caused by de¢-ciency of human PEX7, a homologue of the yeast PTS2receptor, Nat. Genet. 15 (1997) 381^384.

[86] Y. Elgersma, M. Elgersma-Hooisma, T. Wenzel, J.M.McCa¡ery, M.G. Farquhar, S. Subramani, A mobile PTS2receptor for peroxisomal protein import in Pichia pastoris,J. Cell Biol. 140 (1998) 807^820.

[87] J.W. Zhang, P.B. Lazarow, Peb1p (Pas7p) is an intraperox-isomal receptor for the NH2-terminal type 2, peroxisomaltargetting sequence of thiolase: Peb1p itself is targetted toperoxisomes by an NH2-terminal peptide, J. Cell Biol. 132(1996) 325^334.

[88] G. Schatz, B. Dobberstein, Common principles of proteintranslocation across membranes, Science 271 (1996) 1519^1525.

[89] G. Erdmann, B. Blobel, Identi¢cation of Pex13p, a peroxi-somal membrane receptor for the PTS1 recognition factor,J. Cell Biol. 135 (1996) 111^121.

[90] M. Albertini, P. Rehling, R. Erdmann, W. Girzalsky,J.A.K.W. Kiel, M. Veenhuis, W.-H. Kunau, Pex14p, a per-oxisomal membrane protein binding both receptors of thePTS-dependent import pathways, Cell 89 (1997) 83^92.

[91] C. Brocard, G. Lametschwandtner, R. Koudelka, A. Hartig,Pex14p is a member of the protein linkage map of Pex5p,EMBO J. 16 (1997) 5491^5500.

BBAMCR 14514 30-7-99


[92] W. Girzalsky, P. Rehling, K. Stein, J. Kipper, L. Blank,W.-H. Kunau, R. Erdmann, Involvement of Pex13p inPex14p localisation and peroxisomal targeting signal2-de-pendent protein import into peroxisomes, J. Cell Biol. 144(1999) 1151^1162.

[93] W. Schliebs, J. Saidowsky, B. Agianian, G. Dodt, F.W.Herberg, W.-H. Kunau, Recombinant human peroxisomaltargetting signal receptor PEX5: structural basis for inter-action of PEX5 with PEX14, J. Biol. Chem. 274 (1999)5666^5673.

[94] A.J. Saurin, K.L.B. Borden, M.N. Boddy, P.S. Freemont,Does this have a familiar RING?, Trends Biochem. Sci. 21(1996) 208^214.

[95] I.J. van der Klei, R.E. Hilbrands, J.A.K.W. Kiel, S.W.Rasmussen, J.M. Cregg, M. Veenhuis, The ubiquitin-con-jugating enzyme Pex4p of Hansenula polymorpha is re-quired for e¤cient functioning of the PTS1 import machi-nery, EMBO J. 17 (1998) 3608^3618.

[96] H. Kryvi, J. Kvannes, T. Flatmark, Freeze-fracture studyof rat liver peroxisomes: evidence for an induction of intra-membrane particles by agents stimulating peroxisomal pro-liferation, Eur. J. Cell Biol. 53 (1990) 227^233.

[97] A.M. Settles, R. Martienssen, Old and new pathways ofprotein export in chloroplasts and bacteria, Trends CellBiol. 8 (1998) 494^501.

[98] E.H. Hettema, C.W.T. Van Roermund, B. Distel, M. Vanden Berg, C. Vilela, C. Rodrigues-Pousada, R.J.A. Wan-ders, H.F. Tabak, The ABC transporter proteins Pat1 andPat2 are required for import of long-chain fatty acids intoperoxisomes of Saccharomyces cerevisiae, EMBO J. 15(1996) 3813^3822.

[99] H.R. Waterham, V.I. Titorenko, P. Haima, J.M. Cregg, W.Harder, M. Veenhuis, The Hansenula polymorpha PER1gene is essential for peroxisome biogenesis and encodes aperoxisomal matrix protein with both carboxy- and amino-terminal targeting signals, J. Cell Biol. 127 (1994) 737^749.

[100] H. Liu, X. Tan, M. Veenhuis, J.M. Cregg, PER3, a generequired for peroxisome biogenesis in Pichia pastoris, en-codes a peroxisomal membrane protein involved in proteinimport, J. Biol. Chem. 270 (1995) 10940^10951.

[101] J.J. Smith, R.K. Szilard, M. Marell, R.A. Rachubinski, Theperoxin Pex17p of the yeast Yarrowia lipolytica is associ-ated peripherally with the peroxisomal membrane and isrequired for the import of a subset of matrix proteins,Mol. Cell. Biol. 17 (1997) 2511^2520.

[102] Y. Fujiki, R.A. Rachubinski, P.B. Lazarow, Synthesis of amajor membrane polypeptide of rat liver peroxisomes onfree polyribosomes, Proc. Natl. Acad. Sci. USA 81 (1984)7127^7131.

[103] Y. Suzuki, T. Orii, M. Takiguchi, M. Mori, M. Hijikata, T.Hashimoto, Biosynthesis of membrane polypeptides of ratliver peroxisomes, J. Biochem. 101 (1987) 491^496.

[104] A.G. Bodnar, R.A. Rachubinski, Characterization of theintegral membrane polypeptides of rat liver peroxisomesisolated from untreated and clo¢brate-treated rats, Bio-chem. Cell. Biol. 69 (1991) 499^508.

[105] T. Imanaka, Y. Shiina, T. Takano, T. Hashimoto, T. Osu-mi, Insertion of the 70 kDa peroxisomal membrane proteininto peroxisomal membranes in vivo and in vitro, J. Biol.Chem. 271 (1996) 3706^3713.

[106] P. Dieselkotter, W.W. Just, In vitro insertion of the 22-kDperoxisomal membrane protein into isolated rat liver per-oxisomes, J. Cell Biol. 123 (1993) 1717^1725.

[107] J.M. Dyer, J.A. McNew, J.M. Goodman, The sorting se-quence of the peroxisomal integral membrane proteinPMP47 is contained within a short hydrophilic loop,J. Cell Biol. 133 (1996) 269^280.

[108] R.J.S. Baerends, S.W. Rasmussen, R.E. Hilbrands, M. vander Heide, K.N. Faber, P.T.W. Reuvekamp, J.A.K.W.Kiel, J.M. Cregg, I.J. van der Klei, M. Veenhuis, The Han-senula polymorpha PER9 gene encodes a peroxisomalmembrane protein essential for peroxisome assembly andintegrity, J. Biol. Chem. 271 (1996) 8887^8894.

[109] E.A.C. Wiemer, G.H. Luers, K.N. Faber, T. Wenzel, M.Veenhuis, S. Subramani, Isolation and characterisation ofPas2p, a peroxisomal membrane protein essential for per-oxisome biogenesis in the methylotrophic yeast Pichia pas-toris, J. Biol. Chem. 271 (1996) 18973^18980.

[110] Y. Elgersma, L. Kwast, M. van den Berg, W.B. Snyder, B.Distel, S. Subramani, H.F. Tabak, Overexpression ofPex15p, a phosphorylated peroxisomal integral membraneprotein required for peroxisome assembly in S. cerevisiae,causes proliferation of the endoplasmtic reticulum mem-brane, EMBO J. 16 (1997) 7326^7341.

[111] B. Pause, P. Dieselkotter, H. Heid, W.W. Just, Cytosolicfactors mediate protein insertion into the peroxisomalmembrane, FEBS Lett. 414 (1997) 95^98.

[112] A.T. South, S.J. Gould, Peroxisome synthesis in the ab-sence of preexisting peroxisomes, J. Cell Biol. 144 (1999)255^266.

[113] M. Honsho, S. Tamura, N. Shimozawa, Y. Suzuki, N.Kondo, Y. Fujiki, Mutation in PEX16 is causal in theperoxisome-de¢cient Zellweger syndrome of complemen-tation group D, Am. J. Hum. Genet. 63 (1998) 1622^1630.

[114] V.I. Titorenko, R.A. Rachubinski, Mutants of the yeastYarrowia lipolytica defective in protein exit from the endo-plasmic reticulum are also defective in peroxisome biogen-esis, Mol. Cell. Biol. 18 (1998) 2789^2803.

[115] M. Passreiter, M. Anton, D. Lay, R. Frank, C. Harter, F.Wieland, K. Gorgas, W.W. Just, Peroxisome biogenesis :involvement of ARF and coatomer, J. Cell Biol. 141(1998) 373^383.

[116] A. Simonsen, R. Lippe, S. Christoforidis, J.-M. Gaullier, A.Brech, J. Callaghan, B.-H. Toh, C. Murphy, M. Zerial, H.Stenmark, EEA1 links PI(3)K function to Rab5 regulationof endosome fusion, Nature 394 (1998) 494^498.

[117] R. Erdmann, G. Blobel, Giant peroxisomes in oleic acid-induced Saccharomyces cerevisiae lacking the peroxisomalmembrane protein Pmp27p, J. Cell Biol. 128 (1995) 509^523.

[118] P.A. Marshall, Y. Krimkevich, R. Lark, J.M. Dyer, M.

BBAMCR 14514 30-7-99


Veenhuis, J.M. Goodman, Pmp27 promotes peroxisomeproliferation, J. Cell Biol. 129 (1995) 345^355.

[119] P.A. Marshal, J.M. Dyer, M.E. Quick, J.M. Goodman,Redox-sensitive homodimerisation of Pex11p: a proposedmechanism to regulate peroxisomal division, J. Cell Biol.135 (1996) 123^137.

[120] F. Wiebel, W.H. Kunau, The Pas2 protein essential for

peroxisome biogenesis is related to ubiquitin-conjugatingenzymes, Nature 359 (1992) 73^76.

[121] D.I. Crane, J.E. Kalish, S.J. Gould, The Pichia pastorisPAS4 gene encodes a ubiquitin-conjugating enzyme re-quired for peroxisome assembly, J. Biol. Chem. 269(1994) 21835^21844.

BBAMCR 14514 30-7-99


Import of proteins into peroxisomes

Documents

Transcript of Import of proteins into peroxisomes