Protein Import into Peroxisomes: New Developments

PDFlib PLOP: PDF Linearization, Optimization, Protection Page inserted by evaluation version www.pdflib.com – [email protected]

Protein Import into Peroxisomes: New Developments PETER REHLING, MARKUS ALBERTINI, AND WOLF-H. KUNAU" Institut fur Physiologische Chemie der Ruhr-Universitat Bochum Medizinische Fakultiit Abteilung fur Zellbiochemie 0-44780 Bochum, Germany

INTRODUCTION

A decade ago, in 1985 Lazarow and Fujiki summarized a large body of evidence indicating that the early concept of the biogenesis of peroxisomes was no longer tenable. According to the original notion, peroxisomes were formed by budding from the endoplasmic reticulum (for a review see reference 1). In contrast, the new model for peroxisome biogenesis proposed that new organelles arise by growth and division of preexisting ones (FIG. 1). As peroxisomes possess no DNA, all peroxisomal proteins are encoded by nuclear genes. After transcription, the mRNAs for peroxisomal matrix and membrane proteins are translated on free polyribosomes. Subsequently, the cytosolic proteins are post-translationally imported into preexisting organelles. Almost all peroxisomal proteins are synthesized at their final size. A well-characterized exception is the mammalian thiolase, which undergoes N-terminal ~ l e a v a g e . ~ , ~ The protease responsible for the degradation of the leader peptide released after cleavage of the thiolase precursor was recently identified.4 According to this generally accepted view, peroxisome formation closely resembled the concepts of mitochondria5 and chloroplast6 biogenesis and had nothing in coinmon with vesicle formation as this for example, occurs in the secretory pathways7 BIOGENESIS MUTANTS

A large body of data on individual peroxisomal proteins, their import into peroxisomes, and other aspects of peroxisome biogenesis have emerged over the last ten years and have allowed us to update the concept of peroxisome biogenesis. A major role in this development was played by the successful application of the highly developed methodology of classical and molecular genetics. This was especially successful with Saccharomyces cerevisiae8and some other yeast species:,10 as several techniques which are difficult to perform in mammals are easily carried out in some of "Address for correspondence: Wolf-H. Kunau, Institut fir Physiologische Chemie der RuhrUniversitat Bochum, Abt. Zellbiochemie, Universitatsstrape 150, D-44780 Bochum, Germany. 34

REHLING et al.: PROTEIN IMPORT

35

FIGURE 1. Model for peroxisome biogenesis modified (according to Lazarow and Fujiki'): (1) transcription; (2) translation; (3) posttranslational import into preexisting organelles; (4) di-

vision.

the lower single-cell eukaryotes. This in particular holds true for the isolation of mutants and the construction of null mutants of any given gene. Thus, the genetic approach has most successfully been used for the identification of fimgal mutants which were defective in either a single enzyme of one of the peroxisomal pathways or an essential step of peroxisome biogenesis. The latter group, termedpas mutants8 in S. cerevisiue, served at least three important purposes. First, the different phenotypes observed with peroxisomal mutants allowed one to distinguish distinctive aspects of peroxisome biogenesis, including targeting signal-specific import and proliferation of peroxisomes." Second, peroxisomal mutants made possible the identification of the affected genes and their wild type gene products. Third, given the lack of a reliable and efficient in vitro import assay in yeast, peroxisoma1 mutants have provided very useful tools for in vivo import studies. The importance of these in vivo import systems is emphasized by the fact that so far attempts in several laboratories to use semi-intact yeast cells to study protein import into peroxisomes were unsuccessful. The establishment of permeabilized cell systems of higher

ANNALS NEW YORK ACADEMY OF SCIENCES

36

TABLE1. Fungal Genes Essential for Peroxisome Biogenesis S. cerevisiae

H. pobrnorpha

l?pastoris

PASP2 PAS263 PAS3G4 PAS#" PASS" PAS6" PAS730 PAS847 PASP PAS1 034 PASll" PAS12" PAS20h PAS21h PAS22h

PER4' PER2' PERF PER@

PAS1 66

PERP

I:lipolytica

PAS2d PAS7e PAS1 0" PER368

PERF PER333

PAY32f PAS6'

OW.-H. Kunau and co-workers, unpublished results. bH. F. Tabak and co-workers, unpublished results. 'M. Veenhuis and co-workers, unpublished results. d S . Subramani and co-workers, unpublished results. 'S. J. Gould and co-workers, unpublished results. fR. A. Rachubinski and co-workers, unpublished results.

eukaryotic cell lines for peroxisomal import studies is a very recent a c h i e ~ e m e n t ' ~ J ~ and will certainly complement the in vivo studies. The genetic approach, in which one isolated yeast mutants defective in peroxisome assembly, was first started in S. cerevisiae8J4and has led to the discovery of 15 pas complementation g r o u p ~ . ' ~These J ~ data indicate that there must be at least 15 polypeptides required for peroxisome biogenesis. Similar mutants have also been isolated from at least four other yeast species: Pichia pastoris, 17a1* Hansenula polymorpha, l9 Yarrowia lipolytica,20and Candida tropicalis (T. Kamyrio, personal communication). Consequently, a large collection of mutants is available which already has and certainly will in future serve as invaluable tools by which to gain insights into the molecular mechanisms of peroxisome biogenesis. Over the last few years, many of the wild type alleles of the affected genes have been cloned and sequenced (TABLE 1). At present their structure-function analysis is a major topic of peroxisomal research, An important question relates to the number of pas complementation groups. Do the PAS genes defined by the 15 pas complementation groups represent all polypeptides essential for peroxisome biogenesis? An unexpected observation seems to indicate that saturation may be almost reached. Most of the identified genes which complement one of the various pas, pec andpay complementation groups of H. polymorpha, I! pastoris and !I lipolytica seem to be counterparts of one of the 15 PAS genes of S. cerevisiae (TABLE1). However, it cannot be excluded that there is a bias in all the genetic screens, or that an essential step can be accomplished by two gene prod-

REHLING et al.: PROTEIN IMPORT

37

ucts with overlapping function. Therefore, reversed genetic approaches will certainly be, and have been, useful to address this q u e ~ t i o n . ' ~ , ~ ~ Mutations that affected peroxisome assembly occur naturally in the human population and result in a newly recognized group of human inborn errors, the peroxisoma1 d i s ~ r d e r s It. ~is~widely ~ ~ ~ believed that a major group of these disorders are caused by defects in peroxisome b i ~ g e n e s i sThe . ~ ~ fact that these disorders are characterized by severe neurological impairment and are mostly fatal underlines the functional importance of peroxisomes. However, peroxisomes do not appear to be essential for cellular viability. This has allowed investigators to establish fibroblast cell lines of patients with peroxisomal disorders and to group them by somatic genetic analysis into at least ten complementation groups.24 CHO cell mutants defective in peroxisome biogenesis were initially cloned serendipitouslyZ5and then on purpose26by a screen to identify cells that are defective in the first enzyme for the biogenesis of plasmalogens. However, these attempts have so far led to the identification of only two different complementation groups. As mentioned above, analysis of phenotypes of peroxisomal mutants provides first insights into distinct aspects of peroxisome biogenesis. A striking example is the fact that among all complementation groups of human and yeast peroxisomal disorders, only two and two yeast g r o ~ p sexhibit ~ ~ , ~selective ~ import deficiencies. This correlates nicely with the fact that two peroxisomal targeting signals (PTSl and PTS2) have been identified for peroxisomal matrix protein^.^,^^ The PTS 1 comprises the tripeptide SKL and variants thereof found at the extreme C-terminus of many peroxisomal matrix proteins. The PTS2 is located within the amino-terminal sequence of a minor group of peroxisomal matrix proteins and its consensus sequence is somewhat more complex,. . . RLX,H/QL. . . (TABLE2). Indeed, the phenotypes of the known human and yeast mutants suggest that, for proteins containing PTSl or PTS2, different initial import pathways exist which may later converge upon a common translocation route. Fibroblast cell lines of complementation group 2 of peroxisomal d i s ~ r d e r and s ~ the ~ ~yeast ~ ~ mutants pas1 0 (S. cerevi~iae):~ pas8 (Ppast~ris),~ per3 ' (H. p ~ l y r n o r p h a )and , ~ ~pay32 (K lipolytica) (R. A. Rachubinski, personal communication) can all import PTS2 proteins but not PTSl proteins. In contrast, fibroblast cell lines of complementation group 427,2x and the pas7 mutants of S. cerevisiae30fail to import the PTS2 protein thiolase, while sorting of PTSl proteins is not affected. The fact that in all organisms which have been studied there is only one complementation group specific for a distinct PTS import pathway suggests two interesting points. First, the PTS-specific genes may encode the corresponding PTS receptors. Secondly, other genes defined by complementation groups with mutants that exhibit a general import deficiency for peroxisomal matrix proteins might encode components of an import machinery commonly used by both types of matrix proteins. TARGETING SIGNAL RECEPTORS

First results on the PTSl import factor were obtained when McCollum et al. identified the PAS8 gene of I? pastoris by complementing the PTSl import-deficient pas8 mutant.32They reported that the Pp. Pas8p binds to the PTSl targeting signal

Protein

KI V

QA

RLxXXXXHL

MDRL"L,ATQLEQNF'AKGLDAITSKNPDDV MERLRQIASQATAASA P L D P L S T MSKRVEVLLTQLPAYNRLKTPYEAELETAK MQPWYHKLGRQGRQLAEQWQTDAEPWGVATPT

MDRLNQLSGQLKPNAKQSILQKNPDDVVIV

MQPIPDVNQRIARISAHLHPPKSQMEESSALWCR'

MHRLQVVLGHLAGRSESSSALQAAF'CQ

MSQRLQSIKDHLVLSAMGLGESKRKNSLLEK MQRLQVVLGHLRGPADSGWMPQAAPC" MSESVGRTSAMHRLQVVLGHLAGRPESSSALQMC"

Sequence of N-terminus

TABLE 2. PTS2 Sequences of Peroxisomal Matrix Proteins Organism

~

Thiolase Thiolase Thiolase A Thiolase B Malate dehydrogenase Thiolase Thiolase Amine oxidase Aldolase Per 1p

~

S. cerevisiae Human Rat Rat Watermelon C. tropicalis K lipolytica H. polymorpha T. brucei H. polymorpha Putative consensus sequence

"Cleavage site.

Reference

71,72 73 74,75 74,75 76,77 75 78 79,80 81 65

REHLING et al.: PROTEIN IMPORT

39

by demonstrating that the in vitro translated protein preferentially binds SKL peptides as compared to the same peptide without SKL in vitro. Meanwhile, several homologues of the Pp. PAS8 have been described in other yeasts. S.C. PAS10,34H.p. PER3,33and Y1.PAY32, (R. A. Rachubinski, personal communication). The polypeptides that are encoded by these genes belong to the family of TPR proteins and are shown to be related, at least in terms of sequence similarity. Very recently Brocard et al. were able to show, using a two-hybrid system, that the S. cerevisiae homologue (S.C.PaslOp) interacts with a protein containing a canonical PTSl sequence.35 The sequence information of PTSl receptors enabled two groups independently to identify a human homologue (Pxr lp) by screening genome database^.^^,^^ Here again it was shown that in vitro transcribed Pxrlp binds to SKL peptides preferentially. Moreover Dodt et al. were able to map the PTS 1 binding region to the TPR containing portion of the protein.36 Independently, Fransen et al. were able to identify the same gene by screening a human cDNA library by means of a two-hybrid system using a Gal4-palmitoyl-CoA oxidase fusion protein as a bait.38They also showed the binding of the recombinant human PTSl receptor to SKL peptides. The data on the PTSl binding ability of the various yeast and the human proteins represent good evidence for proposing a receptor function. At present, however, it cannot be excluded that these putative receptors may have additional functions. Catalase, for example, possesses a PTSl sequence that is not essential for targeting. Internal targeting information in the protein was identified by Kragler et al.39Nevertheless, it is not imported into peroxisomes of a PTS 1 receptor-deficient mutant of S. cerevisiae (paslo). There are conflicting results concerning the subcellular distribution of the different PTS 1 receptors. The location of the proteins seem to vary significantly in different organisms. Even for a distinct receptor protein, Pxrlp, three laboratories published different results. Dodt et al. and Wiemer et al. demonstrated that the protein is predominantly cytosolic and only a minor portion is associated with peroxi~ o m e s . In ~ ~contrast, . ~ ~ Fransen et al. reported that the same protein is an integral membrane protein.38So far there is no unifying concept that might explain the underlying mechanisms of binding and subsequent translocation of peroxisomal proteins into the organelle. Only one complementation group of S. c e r e v i ~ i a eand ~ ~one , ~ ~human complementation g r o ~ phave ~ ~been , ~ identified ~ that show a selective defect for the import of the PTS2-containing protein thiolase. This suggests that there is a single peroxisome assembly factor essential for PTS2 protein import. The pas7 mutant of S. cerevisiae is able to import PTS 1-containing proteins into the peroxisomal matrix but fails to do so for t h i o l a ~ e .By ~ ~complementation ,~~ analysis Marzioch et al. as well as Zhang et al. identified the corresponding PAS7lPEB1 gene.30,40Sequence analysis of the PAS7 gene revealed an open reading frame encoding a protein of a calculated molecular mass of 42.3 kDa. Analysis of the protein sequence showed that the S.C.Pas7p was a new member of the WD-40 protein family. This sequence motif was first identified in the P-subunit of the heterotrimeric G-protein transducin? * Although the number of members belonging to the WD-40 family is steadily growing, the function of this motif still remains ~ n k n o w n ? ~Some 3 ~ ~ proteins in the WD-40 family are functionally related to proteins of the TPR family, which also comprises the above mentioned PTSl receptors. Present data suggest that at least

40

ANNALS NEW YORK ACADEMY OF SCIENCES

some WD-40 proteins are tightly associated or interact with members of the TPR family.44 Marzioch eta!. were the first to propose a model for the function of a peroxisomal targeting signal receptor (S.C. Pa~7p).~O It was suggested that S.C. Pas7p binds the PTS2 targeting sequence of thiolase in the cytosol and directs the protein to the peroxisome. At the organelle thiolase is released and imported. Subsequently, S.C.Pas7p shuttles back to the cytosol for the next cycle. The first step in this process of recognition, transport and release is based on the receptor function of the protein. A major feature of any import receptor is the specific recognition of the targeting signal. This distinguishes the receptor from a chaperone, which also binds to other parts of the protein different from the targeting sequence. To substantiate and expand our model we first investigated the S.C. Pas7p binding to thiolase. For this purpose we used a variety of independent approaches, including a two-hybrid system, coimmunoprecipitation, genetic analysis and in vitro binding assays. Our recent data demonstrate (Rehling et al., manuscript in preparation): 1. S.C.Pas7p binds to thiolase, and moreover, this interaction is dependent on the existence of the PTS2 (coimmunoprecipitation, two-hybrid system). 2. S.C. Pas7p binds specifically to the first 16 amino acids of thiolase containing the PTS2 (two-hybrid system, genetic analysis, in vitro binding studies). 3. S.C. Pas7p interaction with thiolase occurs between the folded proteins (twohybrid system, coaffinity purification). 4. No functional peroxisome is obviously required for the binding of S.C. Pas7p and thiolase (coimmunoprecipitation in different pas mutants). 5. Evidently, a free amino terminus is not required for S.C.Pas7p binding to the PTS2 of thiolase (two-hybrid system, in vitro binding assays).

Taken together, these results indicate that S.C.Pas7p possesses the properties expected for a peroxisomal targeting signal receptor. Conflicting results have been reported regarding the subcellular distribution of Pas7p, which might be due to the use of different tags at different locations within the protein. Zhang et al. proposed a predominant location of the protein in the peroxisoma1 matrix."O Marzioch et al. reported S.C. Pas7p to be in the cytosol and only a minor amount associated with peroxisome~.~~ It cannot be excluded that overproduction under the strong CUP1 promoter contributed to the observed predominantly cytosolic location of S.C.Pas7p. However, this fact cannot be entirely responsible for the observed cytosolic location because in a ,fix3 background no peroxisomal association has been observed at all.30The subperoxisomal location of myc-Pas7p was recently analyzed by a pH8-step and carbonate extraction. We found the amount of the protein associated with peroxisomes partly in the pH 8 supernatant (matrix) and partly in the carbonate pellet (integral membrane proteins). Two very distinct peroxisomal targeting signals, two import receptors with nonoverlapping binding specificities have been identified, and genetic evidence in man and yeast exists for the functional independence of the two import pathways. However, there may be a fhctional relationship of an as yet unknown nature between the two import receptors. Dodt et al, and Wiemer et al. reported the phenotype of a

REHLING et al.: PROTEIN IMPORT

41

patient defective in PTSl as well as PTS2 import into p e r o x i s o m e ~ .This ~ ~ ? defect ~~ was caused by a single mutation in the PXRl gene, resulting in a premature stop codon. An interesting observation in this context is that the PTSl receptor (S.C. PaslOp) and the PTS2 receptor (S.C. Pas7p) interact when assayed by means of a twohybrid system. POSSIBLE COMPONENTS OF THE TRANSLOCATION MACHINERY It seems reasonable to assume that candidates for subunits of the peroxisomal translocation apparatus should be found among those PAS (PER, PAX PEB) gene products which are either peripheral or integral membrane polypeptides. However, in this context it is important to note that even peroxisomal membrane proteins known to be essential for peroxisome biogenesis may also be involved in other aspects of organelle formation than protein import. For example, there is accumulating evidence that peroxisome fission is not simply triggered by the amount of protein imported into the peroxisomal matrix as originally anticipated, but requires the products of specific genes. S.C. Pas4p15 and its putative counterparts in other yeasts, Per8p of H. p~Zymorpha,~~ and Pas7p of I? pastoris ( S . J. Gould, personal communication) are the first polypeptides which have been proposed to fulfill such a function. In addition to these proteins, PMP27 of S. cerevisiae also seem to be associated with peroxisome proliferation.21,46 Of the ten PAS genes which have been cloned and sequenced in our laboratory, eight encode polypeptides, which are thought to be located in peroxisomal membranes and therefore are putative components of the peroxisomal translocation machinery. An additional candidate is the PAS8 gene product o f s. cerevisiae, which is currently under investigation by Tabak and coworkers.47The eight PAS proteins are summarized in TABLE3. TABLE 3. Candidates for Components of the Peroxisomal Translocation Apparatus in S. cerevesiae

Pas2p

Subperoxisomal

Specific Features

Reference

peripheral, outside

Member of the

63

integral

UBC family -

Amino acids, Mr (kDa)

Location

183 aa, 21 kDa

Pas3P PasSp Pas6p

44 1 aa, 5 1 kDa 271 aa, 31 kDa

?

589 aa, 68 kDa

peripheral, inside

Pas8p

103 aa, 116 kDa

integral?

Pas9p Pasllp Pasl2p

199 aa, 23 kDa 400 aa, 46 kDa 351 aa, 40 kDa

peripheral integral peripheral, outside (?)

(a) W.-H. Kunau and co-workers, unpublished results.

C3HC4-motif SKL-protein, hydrophobic Putative ATPase (AAA-family) -

Prenylated

64 a a

47 a

a a

42

ANNALS NEW Y O U ACADEMY OF SCIENCES

Mutants defective in these proteins fall into the type I class ofpas mutants, a subgroup of mutants defined by lack of morphologically detectable peroxisomes and the mislocalization of peroxisomal matrix enzymes to the cytosol." This phenotype resembles that of fibroblasts of Zellweger syndrome patients.23 All the proteins listed in TABLE3 are putative candidates for the peroxisomal translocation machinery. However, for none of them has this been shown experimentally. Nevertheless, some of these proteins show interesting features which may turn out to be relevant for this question. For example, the PAS6 gene product, S.C.Pasbp, has a canonical SKL at its extreme C-terminus but its deficiency causes a more severe phenotype than that of Paslop, the PTSl re~eptor.'~ This result is taken as an indication of its involvement in the translocation machinery. This molecule has redundant targeting information enabling it on the one hand to be an essential part of the translocation machinery and on the other hand an SKL protein depending on this machinery. Furthermore, Pas6p interacts with, and without SKL, with Paslop, as shown by means of a two-hybrid-system (W.-H. Kunau and H. F. Tabak, unpublished results). Another PAS gene product with interesting properties is S.C.Pas1 lp. It is an integral membrane protein of peroxisomes of S. cerevisiae and its two putative membrane spans share considerable sequence similarity with two membrane spans (IIS4 and IISS) of the a-subunits of voltage-dependent calcium ~ h a n n e l s . 4At~ present ~ ~ ~ it is premature to speculate what this means in terms of its function. Peroxisomal membrane proteins have also been characterized in mammalian cells. At present, four genes have been identified, of which the products PMP70,50,51 PMP22,52,53PAF1,54 and ALDp55,56have been shown to be integral proteins of the peroxisomal membrane. The mutation in the ALD gene causes X-linked adrenoleukodystrophy (ALD) and therefore its gene product cannot be a candidate for the protein import machinery. Fibroblasts of X-linked ALD patients do contain import-competent p e r o x i ~ o m e s . ~ ~ * ~ ~ Do counterparts of PMP70, PMP22 and PAFl exist in yeast? Only for the couple PAFl and S.C.Pas5p does the possibility exist that they may serve the same function. This assumption is based on two properties: size and the presence of a C3HC4-zinc finger motif. But this consensus sequence has been found in a steadily growing number of proteins associated with different fiu~ctions.'~ However, PAFl does not rescue the pas5 mutant, although it is properly inserted into the membrane of yeast peroxisomes in wild type cells (J. Jessen and W.-H. Kunau, unpublished results). SUMMARY AND PERSPECTIVES The results regarding peroxisome biogenesis obtained over the last ten years have in principle supported the model described by Lazarow and Fujiki'; however, very surprising new aspects did evolve: 1. Post-translational import of matrix enzymes is directed by targeting signals

(PTSI and PTS2) which are recognized by import receptors (PTS1- and PTS2 receptors). However, it is still uncertain in which tissue andor organisms per-

REHLING et al.: PROTEIN IMPORT

43

oxisome subpopulations exist (and how many) and which of them are import competent. 2. At present our knowledge about insertion mechanism(s) and targeting signals of membrane proteins is scanty. There is no evidence yet for a stop-transfer signal as well as a translocation machinery used in common by membrane and matrix proteins. 3. No experimental data have been found supporting the notion that import into peroxisomes requires unfolded polypeptides. On the contrary, there are results indicating that the PTS2-import receptor of S. cerevisiae (S.C.Pas7p) binds thiolase in a folded state. The crystal structure of thiolase reveals that both NH,termini of the homodimeric protein stick out of the surface and thus should be freely acce~sible.~~ 4. Furthermore, the “piggyback” e x p e r i m e n t ~suggest ~ ~ , ~ ~that oligomeric proteins can be translocated. That the peroxisomal import machinery can accommodate structures of that size was very recently demonstrated by Walton et al. 6o These authors reported that prefolded proteins stabilized with disulfide bonds and that chemical cross-linkers were shown to be substrates for peroxisoma1 import and, furthermore, that even gold particles (4-9 nm) conjugated to proteins bearing the peroxisomal targeting signal SKL are imported in a PTS1dependent manner. 5. An especially intriguing and certainly unexpected possibility is that the peroxisomal import receptors may bind the appropriate matrix proteins in the cytosol and shuttle the matrix proteins to and perhaps even into the peroxisome. Such a mechanism differs largely from mitochondria1 protein import and resembles much more the current view of protein import into the nucleus.61 However, there is certainly no evidence whatsoever for morphological structures in the peroxisomal membrane similar to nuclear pores. It is not difficult to predict that future research on peroxisome protein import will focus on the following topics: What is the precise function of the PTS 1 and PTS2 receptors? Are folded proteins imported without prior unfolding? Which PAS proteins are components of a “translocation” machinery and what is its molecular architecture? Which are the human counterparts of the yeast PAS genes?

REFERENCES 1. LAZAROW, P. B. &Y. FUJIKI. 1985. Ann. Rev. Cell Biol. 1: 489-530. 2. HIJIKATA, M., N. ISHII,H. KAGAMIJAMA, T. OSUMI& T. HASHIMOTO. 1987. J. Biol. Chem. 262: 8151-8158. 3. BODNAR, A. G. & R. A. RACHUBINSKI. 1990. Gene 91: 193-199. 4. AUTHIER, F., J, J. M. BERGERON, W.-J. Ou, R. A. RACHUBINSKI, B. I. POSNER& I? A. WALTON. 1995. Proc. Natl. Acad. Sci. USA 92: 3859-3863.

44

ANNALS NEW YORK ACADEMY OF SCIENCES

5. NEUPERT, W. 1994. Clin. Investig. 72: 251-261. 6. ROB~SON, C. & R. B. KL~SGEN. 1994. Plant Molec. Biol. 26: 15-24. 7. ROTHMAN, J. E. & L. ORCI.1992. Nature 355: 409415.

8. ERDMANN, R., M. VEENHUIS, D. MERTENS, W.-H. KUNAU.1989. Proc. Natl. Acad. Sci. USA 86: 5419-5423. 9. SUBRAMANI, S. 1993. Annu. Rev. Cell Biol. 9: 445-478. 10. LAZAROW, F! B. 1993. Trends Cell Biol. 3: 89-93. 11. H~HFELD, J., D. MERTENS, F. F. WIEBEL& W.-H. KUNAU.1992. In Membrane Biogenesis and Protein Targeting. W. Neupert & R. Lill, Eds.: 185-207. Elsevier. Amsterdam. 12. WENDLAND, M. & S. SUBRAMANI. 1993. J. Cell Biol. 120: 675-685. 13. RAPP,S., U. SOTO& W. W. JUST.1993. Exp. Cell Res. 205: 59-65. 14. ERDMANN, R. & W.-H. KUNAU.1993. Cell Biochem. Funct. 10: 167-174. 15. KUNAU,W.-H., A. BEYER,T. FRANKEN,K. G ~ T T EM. , MARZIOCH,J. SAIDOWSKY, A. SKALETZ-ROROWSKI & F. F. WIEBEL.1993. Biochimie 75: 209-224. 16. ELGERSMA, Y,, M. VAN DEN BERG,H. F.TABAK & B. DISTEL.1993. Genetics 135: 731-740. 17. Lru, H., X. TAN,M. VEENHUIS, D. MCCOLLUM & J. M. CREGG.1992. Bacteriol. 174: 4943-495 1. 18. GOULD,S. J., D. MCCOLLUM, A. P. SPONG,J. A. HEYMAN & S. SUBRAMANI. 1992. Yeast 8: 613-628. 19. CREGG, J. M., 1. J. VAN KLEI,G. J. SULTER, M. VEENHUIS & W. HARDER. 1990. Yeast 6: 87-97. 20. NUTTLEY, W. M., A. M. BRADE,C. GAILLARDIN, G. A. EITZEN,J. R. GLOVER, J. D. AITCHISON & R. A. RACHUBINSKI. 1993. Yeast 9: 507-517. 21. ERDMANN, R. & G. BLOBEL.1995. J. Cell Biol. 128: 509-523. 22. MOSER,H. W. 1993. Peroxisomal diseases. In Advances in Human Genetics, vol. 21. H. Harris & K. Hirschhorn, Eds.: 1-106. Plenum Press. New York. l? B. & H. W. MOSER.1995. In The Metabolic and Molecular Bases of Inherited 23. LAZAROW, Disease, vol. 11. C. R. Scriver, A. L. Beaudet, W. S. Sly & D. Valle, Eds.: 2287-2324. McGraw-Hill. New York. 24. SHIMOZAWA, N., T. TSUKAMOTO, Y. SUZUKI, T. ONI, Y. SHIRAYOSHI, T. MORI& Y. FUJIKI. 1992. Science 255: 1132-1134. 1986. Proc. Natl. Acad. Sci. USA 83: 5170-5174. 25. ZOELLER, R. A. & C. R. H. RAETZ. T., S. MIURA&Y. FU~KI. 1991. Nature 350: 77-81. 26. TSUKAMCYTO, 27. MOTLEY, A., E. HETTEMA, B. DISTEL& H. TABAK.1994. J. Cell Biol. 125: 755-767. A. B. MOSER,H. W. MOSER& S. J. GOULD.1995. 28. SLAWECKI, M. L., G. DODT,S. STEINBERG, J. Cell Sci. 108: 1817-1829. 29. VAN DER LEU,I., M. VAN DEN BERG,R. BOOT,M. M. FRANSE,B. DISTEL& H. F. TABAK. 1992. J. Cell Biol. 119: 153-162. 30. MARZIOCH, M., R. ERDMANN, M. VEENHUIS& W.-H. KUNAU.1994. EMBO J. 13: 490849 18. 31. DE HOOP,M. J. & G. AB. 1992. Biochem. J. 286: 657-669. 32. MCCOLLUM, D., E. MONOSOV & S. SUBRAMANI. 1993. J. Cell Biol. 121: 761-774. 33. VAN DER KLEI,I. J., R. E. HILBRANDS, G. J. SWAVING, H. R. WATERHAM, E. G . BRIELTNG, V. I. TITORENKO, J. M. CREGG,W. HARDER & M. VEENHUIS. 1995. J. Biol. Chem. In press. 34. VAN DER LEIJ,I., M. M. FRANSE, Y. ELGERSMA, B. DISTEL& H. F. TABAK.1993. Proc. Natl. Acad. Sci. USA 90: 11782-1 1786. 35. BROCARD, C., F. KRAGLER, M. M. SIMON, T. SCHUSTER &A. HARTIG.1994. Biochem. Biophys. Res. Commun. 204: 1016-1022. 36. DODT,G., N. BRAVERMAN, C. WONG,A. MOSER,H. W. MOSER,P. WATKINS, D. VALLE& S. J. GOULD.1995. Nature Genet. 9: 115-125. 37. WIEMER, E. A. C., W. M. NUTTLEY, B. L. BERTOLAET, X. LI, U. FRANCKE, M. J. WHEELOCK, U. K. ANNE,K. R. JOHNSON & S. SUBRAMANI. 1995. J. Cell Biol. 130: 51-65.

REHLING ef al.: PROTEIN IMPORT

45

38. FRANSEN, M., C. BREES,E. BAUMGART, J. C. T. VANHOOREN, M. BAES,G. P.MANNERTS & P. P. VANVELDHOVEN. 1995. J. Biol. Chem. 270: 7731-7736. 39. KRAGLER, F., A. LANGEDER, J. RAUPACHOVA, M. BINDER& A. HARTIG.1993. J. Cell Biol. 120: 665-673. 40. ZHANG,J. W. & P. B. LAZAROW. 1995.3. Cell Biol. 129: 65-80. 41. FONG,H. K. W., J. B. HURLEY, R. S. HOPKINS, R. MIAKE-LYE, M. S. JOHNSON, R. F. DOOLITTLE & M. I. SIMON. 1986. Proc. Natl. Acad. Sci. USA 83: 2162-2166. 42. VAN DER VOORN,L. & H. L. PLOEGH. 1992. FEBS Lett. 307: 131-134. 43. NEER,E. J., C. J. SCHMIDT, R. NAMBUDRIPAD & T.F. SMITH.1994. Nature 371: 297-300. 44. GOEBL,M. & M.YANAGIDA. 1991. TIBS 16: 173-177. 45. TAN,^., H. R. WATERHAM, M. VEENHUIS & J. M. CREGG.1995. J. Cell Biol. 128: 307-319. 46. MARSHALL, I? A., Y. I. KRIMKEVICH, R. H. LARK,J. M. DYER,M. VEENHUIS & J. M. GOODMAN. 1995. J. Cell Biol. 129: 345-355. 47. VOORN-BROUWER, T., I. VAN DER LEIJ, W. HEMBRIKA, B. DISTEL& H. F. TABAK.1993. Biochim. Biophys. Acta 1216: 325-328. 48. CATTERALL, W. A. 1988. Science 242: 50-61. 49. HUI,A., P. T. ELLINOR, 0. KRIZANOVA, J.-J. WANG,R. J. DIEBOLD & A. SCHWARTZ. 1991. Neuron 7: 3544. K., S. TAKETANI, S. YOKOTA, T. OSUMI& T. HASHIMOTO. 1990. J. Biol. Chem. 265: 50. KAMIJO, 45344540. 51. GARTNER, J., H. MOSER& D. VALLE.1992. Nature Genet. 1: 16-23. 52. KALDI,K., P. DIESTELKOTTER, G. STENBECK, S. AUERBACH, U. JAKLE,H.-J. MAGERT,F. T. WIELAND & W. W. JUST.1993. FEBS Lett. 315: 217-222. 53. GARTNER, J., W. W. CHEN,R. I. KELLEY, S. J. MIHALIK & H. W. MOSER.1991. Pediat. Res. 29: 141-146. Y. SUZUKI,T. ORII,Y. SHIRAYOSHI, T. MORI& Y. FUJIKI. 54. SHIMOZAWA, N., T. TSUKAMOTO, 1992. Science 255: 1132-1134. 55. MOSSER,J., A.-M. DOUAR,C.-0. SARDE,I? KIOSCHIS, R. FEIL,H. MOSER,A.-M. POUSTKA, J.-L. MANDEL & I? AUBOURG. 1993. Nature 361: 726-729. 56. CONTRERAS, M., J. MOSER,J. L. MANDEL, P. AUBOURG & I. SINGH.1994. FEBS Lett. 344: 2 11-2 15. 57. MATHIEU, M., J. I? ZEELEN, R. A. PAUPTIT, R. ERDMANN, W.-H. KUNAU & R.K. WIERENGA. 1994. Structure 2: 797-808. 58. GLOVER, J. R., D. W. ANDREWS & R. A. RACHUBINSKI. 1994. Proc. Natl. Acad. Sci. USA 91: 10541-10545. 59. MCNEW,J. A. & J. M. GOODMAN. 1994. J. Cell Biol. 127: 1245-1257. 60. WALTON, P. A., I? E. HILL& S. SUBRAMANI. 1995. Molec. Biol. Cell 6: 675-683. 61. POWERS, M. A. & D. J. FORBES.1994. Cell 79: 931-934. 62. ERDMANN, R., F. F. WIEBEL, A. FLESSAU, J. RYTKA, A. BEYER,K.-U. FROHLICH & W.-H. KUNAU. 1991. Cell 64: 499-510. 63. WIEBEL,F. F. & W.-H. KUNAU. 1992. Nature 359: 73-76. 64. HOHFELD, J., M. VEENHUIS & W.-H. KUNAU.1991. J. Cell Biol. 114: 1167-1 178. H. R., V: I. TIORENKO, I? HAIMA,J. M. CREGG,W. HARDER& M. VEENHUIS. 65. WATERHAM, 1994. J. Cell Biol. 127: 737-749. 66. HEYMAN, J. A,, E. MONOSOV & S. SUBRAMANI. 1994. J. Cell Biol. 127: 1259-1273. 67. CRANE,D. I., J. E. KALISH& S. J. GOULD.1994. J. Biol. Chem. 269: 21835-21844. & J. M. CREGG.1995. J. Biol. Chem. 270: 68. LIU,H., Y. TAN,K. A. RUSSELL, M. VEENHUIS 10940-10951. 69. SPONG, A. I? & S. SUBRAMANI. 1993. J. Cell Biol. 123: 535-548. 70. NUTTLEY, W. M., A. M. BRADE,G. A. EITZEN,M. V E E N H ~J.S D. , AITCHISON, R. K. SzrLARD,J. R. GLOVER& R. A. RACHUBMSKI. 1994, J. Biol. Chem. 269: 556-566.

46

ANNALS NEW YORK ACADEMY OF SCIENCES

71. GLOVER, J. R., D. W. ANDREWS, S. SUBRAMANI & R. A. RACHUBINSKI. 1994. J. Biol. Chem. 269: 7558-7563. 72. ERDMANN, R. 1994. Yeast 10: 935-944. 73. BOUT,A,, Y. TEUNISSEN, T. HASHIMOTO, R. BENNE& J. M. TAGER.1988. Nuc. Acids Res. 16: 10369. & T. HASHIMOTO. 1990. J. Biol. Chem. 265: 4600-4606. 74. HIJIKATA, M., J.-K. WEN,T. OSUMI 75. SWINKELS, B. W., S. J. GOULD,A. G. BODNAR, R. A. RACHUBINSKI & S. SUBRAMANI. 1991. EMBO J. 10: 3255-3262. 76. GIETL,C. 1990. Proc. Natl. Acad. Sci. USA 87: 5773-5777. 77. VAN DER KLEI, I. J., K. N. FABER,I. KEIZER-GUNNINK, C. GIETL,W. HARDER& M. VEENHUIS. 1993. FEBS Lett. 334: 128-132, 78. BERNINGER, G., R. SCHMIDTCHEN, G. CASEL,A. KNORR,K. RAUTENSTRAUSS, W.-H. KUNAU & E. SCHWEIZER. 1993. Eur. J. Biochem. 216: 607-613. 79. BRUINENBERG, P. G., M. EVERS,H. R. WATERHAM, J. KUIPERS, A. C. ARNBERG & G. AB. 1989. Biochim. Biophys. Acta 1008: 157-167. 80. FABER,K. N., I. KEIZER-GUNNINK, D. PLUIM,W. HARDER,G. AB & M. VEENHUIS. 1995. FEBS Lett. 357: 115-120. 81. CLAYTON. C. E. 1985. EMBO J. 4: 2997-3003.