Scholarly article on topic 'Biogenesis of peroxisomes and glycosomes: trypanosomatid glycosome assembly is a promising new drug target'

Biogenesis of peroxisomes and glycosomes: trypanosomatid glycosome assembly is a promising new drug target Academic research paper on "Biological sciences"

0
0
Share paper
Academic journal
FEMS Microbiology Reviews
OECD Field of science
Keywords
{""}

Academic research paper on topic "Biogenesis of peroxisomes and glycosomes: trypanosomatid glycosome assembly is a promising new drug target"

ELSEVIER

FEMS Microbiology Reviews 28 (2004) 603-643

www.fems-microbiology.org

Biogenesis of peroxisomes and glycosomes: trypanosomatid glycosome assembly is a promising new drug target

Juliette Moyersoen a'!, Jungwoo Choe b'2, Erkang Fan b, Wim G.J. Hol b'c,

Paul A.M. Michels a'*

a Research Unit for Tropical Diseases, Christian de Duve Institute of Cellular Pathology and Laboratory of Biochemistry, Université Catholique de Louvain, ICP-TROP 74.39, Avenue Hippocrate 74, B-1200 Brussels, Belgium b Department of Biochemistry and Department of Biological Structure, Biomolecular Structure Center, University of Washington,

P.O. Box 357742, 98195-7742 Seattle, USA c Howard Hughes Medical Institute, University of Washington, Seattle, USA

Received 9 March 2004; received in revised form 14 June 2004; accepted 15 June 2004 First published online 21 July 2004

Abstract

In trypanosomatids (Trypanosoma and Leishmania), protozoa responsible for serious diseases of mankind in tropical and subtropical countries, core carbohydrate metabolism including glycolysis is compartmentalized in peculiar peroxisomes called glycosomes. Proper biogenesis of these organelles and the correct sequestering of glycolytic enzymes are essential to these parasites. Biogenesis of glycosomes in trypanosomatids and that of peroxisomes in other eukaryotes, including the human host, occur via homologous processes involving proteins called peroxins, which exert their function through multiple, transient interactions with each other. Decreased expression of peroxins leads to death of trypanosomes. Peroxins show only a low level of sequence conservation. Therefore, it seems feasible to design compounds that will prevent interactions of proteins involved in biogenesis of trypanosomatid glycosomes without interfering with peroxisome formation in the human host cells. Such compounds would be suitable as lead drugs against trypanosomatid-borne diseases.

© 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. Keywords: Trypanosomatids; Glycosome and peroxisome biogenesis; Peroxin; Protein-protein interaction; Drug design

Contents

1. Introduction........................................................................................................................................................604

2. Glycolysis in trypanosomatids................................................................................................................................605

3. Glycosomes, structure and function........................................................................................................................606

4. Proper compartmentation of glycosomal enzymes and intact glycosomes are vital to trypanosomes..............................607

5. Overview of peroxisome biogenesis........................................................................................................................610

* Corresponding author. Tel.: +32 2 7647473; fax: +32 2 7626853. E-mail address: michels@bchm.ucl.ac.be (P.A.M. Michels).

1 Present address: Division of Molecular Biology, The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands.

2 Present address: Scripps Research Institute, San Diego, USA.

0168-6445/$22.00 © 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.femsre.2004.06.004

5.1. Import of matrix proteins; receptor recognition..............................................................................................611

5.2. The docking complex for import of matrix proteins........................................................................................615

5.3. Import of matrix proteins; cargo-receptor complex translocation and receptor recycling....................................616

5.4. Receptor recycling and the 'extended shuttle' model for the import of matrix proteins........................................618

5.5. Three models for the translocation mechanism of matrix proteins....................................................................619

5.6. Preimplex hypothesis for the import of matrix proteins....................................................................................620

5.7. Insertion of peroxisomal membrane proteins..................................................................................................620

5.8. Proliferation of peroxisomes and the formation of their membrane..................................................................622

5.9. Growth and division model versus de novo formation model..........................................................................622

5.10. A role for the ER in peroxisomal membrane formation and insertion of peroxisomal membrane proteins?..........624

6. Aspects of glycosome biogenesis............................................................................................................................625

6.1. Targeting and import of glycosomal matrix proteins......................................................................................625

6.2. Peroxins involved in glycosome biogenesis......................................................................................................626

6.3. Protein interactions in glycosome biogenesis and possibilities for drug design ....................................................630

7. Feasibility to target transient protein-protein interactions........................................................................................633

8. Conclusions..........................................................................................................................................................633

Acknowledgements ..............................................................................................................................................634

References............................................................................................................................................................634

1. Introduction

Trypanosomatids are parasitic organisms belonging to a higher taxonomic group of protists - usually assigned the rank of an order - called Kinetoplastida that probably diverged early in evolution from the other eukaryotic lineages. It is one of the several major lineages that separated from each other in the early, as yet unresolved, diversification of eukaryotes [1,2]. The Kinetoplastida are characterized by a number of peculiarities such as the presence of a special part of the single mitochondrion, where DNA with a highly unusual structure is found, called the kineto-plast, hence the name of this group of organisms; an unusual, often highly complex form of editing of the mitochondrial RNA; the organization of nuclear genes in very long multicistronic transcription units; precursor RNA molecules are processed to mRNAs by trans-splicing, probably while transcription continues downstream; the apparent absence of specific RNA polymerase II promoters; and a unique way of metabolic compartmentation [3,4]. The most peculiar aspect of the metabolic compartmentation in Kinetoplastida is the sequestering of part of the glycolytic pathway and some other enzyme systems within peroxisome-like organelles called glycosomes [5-9].

Several species within the kinetoplastid family of the parasitic Trypanosomatidae are responsible for serious, but largely neglected diseases of humans and domestic animals, particularly in tropical and subtropical areas of the world. In sub-Saharan Africa, two subspecies of Trypanosoma brucei (Trypanosoma brucei gambiense and Trypanosoma brucei rhodesiense) cause sleeping sickness or human African trypanosomiasis. These

extracellularly multiplying parasites are transmitted from animal reservoirs to human beings by the tsetse fly. When introduced into the host by the bite of the insect taking its blood meal, the trypanosomes proliferate initially in the blood and lymphatic systems, before invading the central nervous system. T. b. gambiense causes a chronic disease that takes several years to progress to the meningoencephalitic stage. In contrast, infection by T. b. rhodesiense leads to an acute form that takes only a few weeks to reach the second stage. Without appropriate treatment, both forms of the disease are fatal. The World Health Organization estimates that nearly 60 million people in 36 of Africa's 52 countries are at risk. An estimated half a million people are probably infected. Unfortunately, current treatments are largely unsatisfactory and the situation is even worsening [10,11]. Chemotherapy is the mainstay for control because effective vaccines are not available and antigenic variation makes the prospects for vaccination gloomy. Nevertheless, most currently used drugs are inefficient and toxic, and the development and spreading of drug-resistant parasites are becoming a major problem [12]. Furthermore, several Trypanosoma species (Trypanosoma brucei brucei, Trypanosoma vivax and Trypanosoma congolense) are responsible for cattle trypanosomiasis also called 'nagana'. This disease has an important impact on the development of rural Africa by reducing cattle production and the availability of animal traction for farming.

In Central and South America, Trypanosoma cruzi causes Chagas' disease. About 100 million people are at risk, an estimated 11-18 million people are infected and annually 13,000 deaths are reported. A large range of wild and domestic mammals act as a reservoir for the parasites, which are transmitted by a blood-sucking

reduviid bug, which deposits its infective faeces on the skin. After entry in human beings, T. cruzi invade host cells. After an initial, acute stage, usually a balance between host and parasite is established in which patients are asymptotic. A chronic form of the disease often develops after 10-25 years, causing irreversible damage to the heart, digestive or nervous systems. Patients become progressively more ill and ultimately die, often by sudden heart failure. Also the available drugs for Chagas' disease may induce serious side effects and lack high efficacy. Although success has been achieved in reducing transmission due to vector control programs, the large number of people still infected, and the limitations of the available drugs make the development of new medicines urgent [10,13].

At least 17 different species of the trypanosomatid genus Leishmania are pathogenic for humans. These parasites cause a spectrum of diseases, called leishmani-ases, ranging from cutaneous and mucocutaneous lesions to life-threatening visceral diseases. About 350 million people in 88 countries worldwide, but mainly in the tropics and subtropics are at risk, with 12 million people affected and approximately two million new cases each year. In recent years, co-infection of HIV patients with Leishmania has also become a serious problem. The parasites are transmitted between human beings and animal reservoirs by the bite of a phlebotomine sandfly. Inside the mammalian host, Leishmania cells live within phagolysosomes of macrophages. Although new drugs against leishmaniasis have become available in recent years, the chemotherapy currently available is far from satisfactory [14]. Older drugs are toxic and less efficient, and resistance is spreading. Recently introduced drugs overcome these problems but are highly expensive or difficult to administer. A continued effort to develop efficient, safe and affordable drugs remains essential.

The deplorable situation with regard to treatment of these diseases should be attributed to several causes. First, vaccine development is difficult because of the intricate mechanisms used by the parasites to evade the host's immune defense mechanisms; African trypan-osomes employ a highly elaborate system of antigenic variation (reviewed in [15]), American trypanosomes and Leishmania species hide intracellularly (reviewed in [16]). But also very importantly, these diseases affect mainly the poorest people in the world. There is little financial incentive for pharmaceutical companies to invest in research and trials for new drugs against these tropical diseases. Fortunately, the many peculiarities in the cell biology and metabolism of kinetoplastids have been subject of intense research. This has led to the identification of potential targets - enzymes or subcellular structures - for new trypanocidal drugs. The potential targets identified are unique to the parasites, or at least sufficiently different from corresponding host molecules

or structures, rendering feasible the development of compounds that may selectively interfere with their function in trypanosomatids without affecting a corresponding host function (reviewed in [17-19]). Moreover, several of these peculiar traits have been validated as drug targets; they proved to be essential for the survival of the parasites.

In this review, we will discuss the structure, function and biogenesis of glycosomes in the trypanosomatids, and compare the biogenesis of glycosomes with that of the related peroxisomes in other organisms. We provide arguments that interference with biogenesis offers good prospects for the development of selective trypanocidal drugs.

2. Glycolysis in trypanosomatids

Glycolysis has been perceived as a potentially excellent target for the design of new anti-trypanosomatid drugs [6,19-23]. This is particularly the case for African trypanosomes which are entirely dependent on glycoly-sis for their ATP supply when living in the mammalian bloodstream. The mitochondrion is poorly developed during this stage of the life cycle; the organelle has no enzymes of the tricarboxylic acid cycle and no cyto-chrome-containing respiratory chain. Glucose, abundantly present in the blood, is taken up by the parasites and metabolized into pyruvate that is excreted. Indeed, bloodstream-form trypanosomes starved at 37 °C without glucose, or incubated with an inhibitor of the plasma-membrane glucose transporter such as phloretin, die within several minutes [24]. The trypano-somes are also killed when incubated with pentalenolac-tone, a specific inhibitor of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) [25], or a-cyanocinnamic acid derivatives, which specifically inhibit transporters of monocarboxylic acids such as pyruvate [26]. However, these inhibitors are not selective for the enzymes of the parasite. Therefore, structure-based drug design is being employed to develop compounds that only inhibit try-panosomatid glycolytic enzymes without having any effect on corresponding human enzymes. Indeed, adeno-sine derivatives were designed that compete with NAD+ binding by the trypanosomatid GAPDH without affecting the human enzyme, even at 1000-fold higher concentration. Within minutes after adding these compounds to cultured bloodstream-form T. brucei, production of glucose-derived pyruvate ceased, parasite motility was lost, and lysis of parasites was observed [27,28]. Similarly, the designed compounds stunted the growth of intracellular T. cruzi without having any effect on growth of cultured human fibroblasts. Structure- and catalytic-mechanism-based approaches are also applied to design selective inhibitors of other glycolytic enzymes

of trypanosomes. The progress in this research has been summarized in a recent review [19].

Whereas glycolysis is the only ATP-generating process in the bloodstream-form of African trypanosomes, this is not the case for the mammalian-intracellular stage (amastigote) of Leishmania spp. Leishmania amastigotes have a functional mitochondrion with enzymes of the tricarboxylic acid cycle and a respiratory chain [29-32]. Fatty acids and amino acids may play a more important role in the free energy supply of these cells [33]. However, a complete set of enzymes for carbohydrate metabolism is present [30-32,34] and indeed glucose catabolism has been measured albeit at a lower rate than in bloodstream-form T. brucei [35] or cultured insect-stage (pro-mastigote) Leishmania cells. Moreover, both the presence of glucose and the possession of a functional glucose transporter have been shown to be essential for cultured Leishmania mexicana amastigotes [36].

Not much information is as yet available about the energy and carbohydrate metabolism of T. cruzi amasti-gotes. However, it has been shown that cultured amasti-gote-like forms readily ferment glucose to succinate, have a high activity of some glycolytic enzymes, a low activity of tricarboxylic acid cycle enzymes and a low cytochrome content [37]. The insect stages of all human-pathogenic trypanosomatids (T. brucei, T. cruzi and Leishmania spp.) seem to be capable of metabolizing both amino acids and sugars for their free energy supply; they have a large repertoire of enzymes for carbohydrate metabolism, including the glycolytic pathway, and a well-developed mitochondrion with a respiratory-chain linked system for oxidative phos-phorylation (reviewed in [9,38,39]).

3. Glycosomes, structure and function

Whereas in almost all organisms studied glycolysis is a process that occurs in the cytosol, in Kinetoplastida the major part of the pathway is localized in organelles called glycosomes. The glycosomes of bloodstream-form T. brucei contain the enzymes converting glucose into 3-phosphoglycerate; only the last three enzymes of the gly-colytic pathway are present in the cytosol (Fig. 1) [5-9]. A consequence of this organization is that, inside the organelle, the consumption of ATP by hexokinase (HXK) and phosphofructokinase (PFK) equals the ATP production by phosphoglycerate kinase (PGK). Net ATP production occurs only in the cytosol, in the reaction catalyzed by pyruvate kinase (PYK). Similarly, the NADH formed in the reaction catalyzed by GAPDH is re-oxidized inside the organelle by a NAD+-dependent glycerol-3-phosphate dehydrogenase (GPDH). Subsequently, the electrons are transferred to a mitochondrial glycerol-3-phosphate oxidase via a redox shuttle comprising a putative transporter in the

glycosomal membrane which exchanges glycerol 3-phos-phate for dihydroxyacetone phosphate (Fig. 1). The transporter remains to be identified, but its existence is inferred from the low permeability of the glycosomal membrane (see below) and the requirement for strict coupling of the fluxes by which the two triosephosphates are exchanged [5,40]. The glycerol-3-phosphate oxidase complex contains a FAD-linked GPDH, ubiquinone and a terminal oxidase, known as the alternative oxidase, that is insensitive to cyanide, but can be inhibited by salicylhydroxamic acid. The redox process catalyzed by the glycerol-3-phosphate oxidase is nonprotonmotive and thus not coupled to ATP formation.

As described above, amastigotes of Leishmania spp. and T. cruzi, and insect-stage form parasites (called 'pro-cyclic' form in T. brucei, 'epimastigote' form in T. cruzi, and 'promastigote' form in Leishmania) have a more elaborate energy- and carbohydrate-metabolic network. In these cells, other ATP-dependent kinases (e.g., phos-phoeno/pyruvate carboxykinase and pyruvate phosphate dikinase) may be found in the glycosome, whereas PGK is relocated to the cytosol. These alterations have as apparent consequence that the ATP/ ADP balance is maintained. Similarly, the presence of other NAD+-dependent oxidoreductases (malate dehy-drogenase and fumarate reductase) in the glycosomal matrix of these cells is accompanied by a significant drop in the activity of GPDH, suggesting that also the intraglycosomal production and consumption of NADH remain balanced. Although the enzyme distribution and stoichiometric relationship of metabolites have not yet been studied in much detail in other kinetoplast-ids than T. brucei, the available evidence suggests that the intraglycosomal balance in formation and consumption of both ATP and NADH is a general feature of these organisms and thus possibly important for proper functioning of the organelle.

In bloodstream-form T. brucei, glycolysis is virtually the only process occurring in the glycosomes. Glyco-lytic enzymes may comprise up to 90% of the matrix proteins of the organelles in these cells. In contrast, in other trypanosomatids, or other T. brucei life-cycle stages, the concentration of glycolytic enzymes in glyco-somes is less important. Other metabolic systems that -at least in part - may be found in the glycosomes are the pentose-phosphate and gluconeogenic pathways, enzymes of purine salvage and pyrimidine biosynthesis, p-oxidation of fatty acids, ether-lipid biosynthesis, isoprenoid biosynthesis and oxidant defense mechanisms [6,8,9]. Despite the presence of this unique variety of metabolic systems in glycosomes, these organelles should be considered as authentic peroxi-somes on the basis of the following criteria: (1) glyco-somes, like other members of the peroxisome family (peroxisomes, yeast microbodies and plant glyoxy-somes), are bounded by a single phospholipid bilayer

Fig. 1. Scheme of glycolysis in bloodstream-form T. brucei. Enzymes: 1, hexokinase; 2, glucose-6-phosphate isomerase; 3, phosphofructokinase; 4, aldolase; 5, triosephosphate isomerase; 6, glyceraldehyde-3-phosphate dehydrogenase; 7, phosphoglycerate kinase; 8, glycerol-3-phosphate dehydrogenase; 9, glycerol kinase; 10, phosphoglycerate mutase; 11, enolase; 12, pyruvate kinase; 13, glycerol-3-phosphate oxidase. The nutrient (glucose) and end-product of the metabolism (pyruvate) are boxed.

membrane, have an electron-dense proteinaceous matrix and do not contain any detectable DNA; (2) although glycolytic enzymes are the most prominent proteins in glycosomes, some other enzymes or enzyme systems are shared with peroxisomes, notably enzymes of peroxide metabolism, fatty-acid oxidation and ether-lipid biosynthesis; (3) glycosomes and other members of the peroxisome family have a similar process of biogenesis involving several cascades of proteins which are homologous in the different organelles [7,41,42]. The process of biogenesis is discussed below. It may be concluded, therefore, that the common ancestor of Kinetoplastida and other eukaryotes already contained peroxisomes and that an ancestral organism in the kin-etoplastid lineage additionally acquired the capability of compartmentalizing the major part of the glycolytic pathway in these organelles. How this may have happened has been discussed elsewhere [9,43,44].

Although the presence of glycolysis in peroxisomes is unique, it should be appreciated that the enzymic content of the peroxisomal family may differ considerably

between different organisms. Indeed, the astounding metabolic diversity of the members of the peroxisome family has been already identified many years ago and reviewed by de Duve [45] and Borst [46].

4. Proper compartmentation of glycosomal enzymes and intact glycosomes are vital to trypanosomes

As described in Section 2, glycolysis is essential for trypanosomes and is considered as a validated drug target. One could wonder if proper compartmentation of glycolytic enzymes inside glycosomes and possession of intact organelles are equally essential. If so, compounds that will interfere with glycosome biogenesis or integrity may offer an additional possibility for therapy against trypanosomatid-borne diseases.

Indeed, several experiments have provided evidence that mislocalization of glycolytic enzymes is deleterious for the trypanosome. Trypanosomes have two distinct,

but highly similar isoenzymes of PGK, a glycosomal one (gPGK) only expressed in the bloodstream form and a cytosolic one (cPGK) only found in the procyclic trypanosomes growing in the gut of tsetse flies. The signal for glycosome-targeting of gPGK has been located within the C-terminal six residues [47,48]. Bloodstream-form trypanosomes are rapidly killed when a transgene of cPGK or a C-terminally truncated form of gPGK is expressed [49]. The toxicity depends both on enzyme activity - expression of a mutant PGK with a single residue substitution, rendering the enzyme inactive, has no effect - and cytosolic location. Expression of a transgene of gPGK in procyclic trypanosomes had a less dramatic effect; nonetheless, the growth rate was slowed down and the morphology of the cells was changed. Also, expression of cytosolic triosephosphate isomerase (TIM), after transfection of bloodstream-form T. brucei with the gene of the Saccharomyces cerevisiae enzyme, inhibited cell growth [50].

To understand why the expression of an additional copy of a glycolytic enzyme in the wrong compartment is detrimental for trypanosomes, one has to know the function or consequences of this form of metabolic compartmentation. For years, it was thought that the confinement of glycolytic enzymes inside an organelle was an evolutionary adaptation of trypanosomatids to sustain a high glycolytic flux [6]. All glycosomes together constitute about 4% of the total cellular volume of T. brucei, and the enzymes are present at a relatively high concentration. It was therefore assumed that the sequestering of enzymes overcomes a diffusion limitation of glycolytic metabolites and thus enables a high flux. However, this notion is now considered unlikely. First, a relatively high glycolytic flux is only characteristic of the bloodstream-form T. brucei, not of most other trypanosom-atids living in environments where sugars are less abundant. There is no reason to assume that the capacity for a high glycolytic flux must have been a selective advantage for the ancestral kinetoplastids [51,52]. Second, it was shown that S. cerevisiae can have a 2-fold higher flux than T. brucei, with a less than 2-fold higher glycolytic protein content that is not compartmentalized [53]. Third, calculations have shown that, even if the enzymes were dispersed over the entire cytosol, the glycolytic flux of bloodstream-form T. brucei would not be limited by diffusion of metabolites from one enzyme to the next, but rather by the catalytic activity of the enzymes [52]. Computer simulation of metabolism of bloodstream-form T. bru-cei has, however, provided a new clue about the possible function of the glycosome and a possible explanation why proper compartmentation of enzymes may be essential. The explanation provided by this computer analysis is dependent on two premises: first that the glycosomal membrane has a poor permeabil-

ity for most glycolytic intermediates, adenine nucleo-tides and nicotinamide adenine nucleotides, and second that the glycolytic enzymes inside the glyco-somes lack the usual mechanisms of activity regulation.

Several studies have provided strong evidence that the glycosomal and cytosolic compartments are indeed separated by a permeability barrier. First, the kinetics of incorporation of radioactivity in glycolytic intermediates, upon a pulse of radioactively labelled glucose added to trypanosomes, indicated the existence of two pools of intermediates: a rapidly labelled pool, representing 20-30% of the total cellular metabolites, and a slowly labelled pool representing 70-80% of the total. The rapidly labelled pool is directly involved in glyco-lysis and was interpreted as being present in the glyco-somes, whereas the second pool, in the cytosolic compartment, is not directly involved in glycolysis. Exchange between the two pools occurs only very slowly [54]. Second, under anaerobic conditions, bloodstream-form trypanosomes produce one molecule of pyruvate and one molecule of glycerol per molecule of glucose, instead of two molecules of pyruvate [55]. This is due to the presence of a glycerol kinase present in the glycosome. It catalyses the thermodynamically unfavourable 'reverse' reaction with ADP and glycerol 3-phosphate as substrates (Fig. 1). However, this would only be feasible at high concentrations of these substrates. Such conditions would be difficult to attain in the entire cell, but can occur in a small compartment, such as the glycosome, with a membrane impermeable for the substrates. The consequence of this reverse reaction is that glycolysis can proceed under anaerobic reactions, while maintaining the ATP/ADP and NADH/NAD+ balances within the glycosome, but with a net production of only one molecule of ATP per molecule of glucose instead of two ATP under aerobic conditions.

The notion that the glycosomal membrane is poorly permeable is in line with current ideas about the membrane of mammalian and yeast peroxisomes. Although peroxisomes were, for many years, considered leaky for low-molecular weight compounds, a considerable amount of evidence is now available supporting the view that peroxisomes form closed compartments under in vivo conditions (reviewed in [56]). This evidence includes genetic data showing the need for shuttle systems for redox equivalents [57]. The notion that peroxi-somes are surrounded by a membrane that is poorly permeable for most metabolites implies that specific transporter molecules should exist for the transmembrane movement of metabolites. Indeed, two classes of transporters in the peroxisomal membrane have been identified: ABC transporters (reviewed in [58]), some of which have been shown to be involved in transport of fatty acids in S. cerevisiae [59] and have

possibly similar functions in other organisms, and transporters homologous to the mitochondrial solute carriers or AAC transporters. Recently, homologues of the ABC transporters involved in fatty-acid transport have been identified in the glycosomal membrane of procyclic T. brucei (C. Yernaux and PM, unpublished).

It should be noted that some researchers working on plant peroxisomes claim that compartmentalization of plant peroxisomal metabolism is in major parts not caused by a boundary function of the membrane but is primarily due to the specific structure of the protein matrix. Transfer of metabolites across the membrane would proceed through relatively unspecific porin-like channels (reviewed in [60]). However, this claim is not supported by available data for peroxisomes from other organisms [56-59].

With regard to the second premise made for the computer analysis of glycosome metabolism, indeed, the regulation of glycolysis in trypanosomes differs largely from that in other organisms. There is an apparent lack of activity regulation of the trypanosomatid HXK and PFK [61-64]. In most organisms, the activities of these two glycolytic enzymes are highly regulated by their products, by metabolites further downstream in the metabolism or by other (allosteric) effectors. This regulation serves two purposes. First, it prevents the loss of ATP by futile cycling when glycolysis and glu-coneogenesis occur simultaneously. Second, it has been argued, and experimentally confirmed by using S. cere-visiae mutants, that a tight regulation of the first steps of glycolysis is vital [65]. Indeed absence of regulation may lead to unrestricted accumulation of glycolytic intermediates, a situation that will be highly toxic for the cell.

Using available kinetic data of the different glyco-lytic enzymes, a mathematical model of glycolysis in bloodstream-form T. brucei was developed [66] and used to assess the metabolic consequences of compart-mentation [53]. The question addressed was how the functional behaviour of trypanosome glycolysis would change if the pathway were not compartmentalized. According to the computer model this would not significantly affect the steady-state glycolytic flux, in agreement with the conclusions described above. But strikingly, it would lead to toxic accumulation of hex-ose phosphates upon addition of glucose. Such toxic accumulation was prevented by the existence of distinct glycosomal and cytosolic adenine nucleotide pools. The kinases in the beginning of the pathway respond to the glycosomal ATP/ADP ratio, not the cytosolic one. The glycosomal ATP/ADP ratio, which - according to the computer simulation - is usually low, controls the activity of HXK and PFK and maintains the concentration of the hexose phosphates constrained within a narrow range. If the enzymes

would sense the higher cytosolic ATP/ADP ratio, their activity would not be restrained and the products would accumulate. Therefore, the computer analysis supports the notion that the compartmentation of gly-colysis serves a regulatory function that compensates for the lack of activity regulation of its enzymes, and that compartmentation of the process within a membrane that is poorly permeable for solutes is essential for the parasite. Would such a function also provide an explanation why the additional presence of a glycosomal glycolytic enzyme in the cytosol is highly detrimental to the cell?

It is essential that a stoichiometric relationship of all metabolites - phosphorylated glycolytic intermediates, adenine nucleotides and nicotinamide adenine nucleo-tides - participating in the reactions inside the glyco-some is maintained, to allow glycolysis to proceed [66]. If such a metabolite would leave the glycosome, it would affect the stoichiometric relationship and cause arrest of glycolysis by accumulation of another metabolite and/or disruption of the ATP/ADP and NADH/NAD+ balance. For some metabolites transport mechanisms should exist across the membrane, for example for glyc-erol 3-phosphate and dihydroxyacetone phosphate, which are involved in the redox shuttle to the mitochondrion. Furthermore, the pulse-labelling experiments mentioned above [54], in conjunction with measurements of intracellular metabolite concentrations [67] indicated, under steady-state conditions, similar concentrations of glycolytic intermediates in the glycosomal and cytosolic pools, despite the slow equilibration across the membrane. It is conceivable that the partial mislo-calization of a glycosomal enzyme to the cytosol would catalyze a reaction between the cytosolic intermediates, causing a drain on glycosomal metabolites that are capable of crossing the membrane and thus disturb the stoichiometry of the intraglycosomal reactions. This effect would even be reinforced if an extraglycosomal shunt of part of the glycolytic pathway could be formed [49,50]. This is feasible, because some enzymes have a constitutively dual subcellular localization: part is in the glycosomes, another part in the cytosol (reviewed in [68]). Trypanosomatids have two different isoenzymes of GAPDH. The metabolic role of the cytosolic enzyme is not known yet, but it may be involved in redox buffering of the cytosol. In addition, the parasites have only one isoenzyme of glucose-6-phosphate isomerase, but part of it is permanently found in the cytosol. Possibly, this cytosolic activity is involved in the pentose-phos-phate pathway that also has a dual localization: it may isomerize fructose 6-phosphate, one of the products of the pathway's activity, into the substrate glucose 6-phos-phate. Thus, the incorrect compartmentation of even a low amount of a glycosomal enzyme may thus have a highly detrimental effect on trypanosomes, as most clearly shown for PGK [49].

5. Overview of peroxisome biogenesis

Peroxisomes, like mitochondria and chloroplasts, multiply by growth and division, and segregate at cell division. However, contrary to these other organelles, peroxisomes do not contain any DNA. All their proteins are encoded by nuclear genes, are synthesized in the cytosol on free ribosomes and post-translationally imported into pre-existing organelles. Moreover, the mechanism of peroxisome biogenesis is fundamentally different from that of mitochondria and chloroplasts, as revealed by studies performed during the last two decades, mainly on various yeasts and mammalian cells (reviewed in [69,70]). Conceptually, the biogenesis of peroxisomes can be considered as a number of different processes: recruitment of lipids for membrane formation, synthesis and insertion of membrane proteins, synthesis and import of matrix proteins, and organelle fission and inheritance. To date, 32 proteins, called

peroxins (or with the acronym PEX) [71], have been identified as required for the different aspects of perox-isome biogenesis (Table 1). The first peroxins were discovered in yeasts by screens of mutant cells presenting defects in the assembly of peroxisomes [72]. These cells were unable to grow on oleate as unique source of carbon. This substrate is known to induce the proliferation of peroxisomes [73]. Similarly, Chinese Hamster Ovary (CHO) mutant cells affected in plasmalogen synthesis turned out to be also impaired in peroxisome biogenesis [74]. Complementation studies of these two types of mutant cells with genomic and cDNA libraries, respectively, led to the discovery of proteins involved in the biogenesis of the organelles. Subsequently, other perox-ins were identified by additional genetic studies, and upon the detection of interactions between known peroxins and other ones, as revealed by yeast or bacterial two-hybrid studies and coprecipitation analyses and afterwards confirmed by in vitro binding studies.

Table 1

Identified peroxins

Peroxin Organisms Peroxin function Localization

Yeasts/fungi Mammals Plants Other

Sc Yl Other3

PEX1 X X X X X D.m./C.e. Matrix protein import - receptor recycling or vesicle fusion cyt/MA

PEX2 X X X X X C.e. Matrix protein import - translocation IMP

PEX3 X X X X PMP membrane insertion IMP

PEX4 X X X ? cyt/MA

PEX5 X X X X X D.m./C.e. Matrix protein import - PTS1 receptor cyt/MA/matrix

PEX6 X X X X X C.e. Matrix protein import - receptor recycling or vesicle fusion cyt/MA

PEX7 X X X X Matrix protein import - PTS2 receptor cyt/MA/matrix

PEX8 X X X Matrix protein import matrix/MA

PEX9 X ? IMP

PEX10 X X X X X Matrix protein import - translocation IMP

PEX11 X X X X C.e. regulation peroxisome proliferation IMP

PEX12 X X X X C.e. Matrix protein import - translocation IMP

PEX13 X X X X C.e. Matrix protein import - docking IMP

PEX14 X X X X X Matrix protein import - docking IMP

PEX15 X Recruitment PEX6 (and PEX1) to peroxisomal membrane IMP

PEX16 X X X X PMP membrane insertion IMP

PEX17 X X X Matrix protein import - docking IMP

PEX18 X Matrix PTS2 protein import cyt/MA

PEX19 X X X X X C.e. PMP - membrane insertion cyt/MA

PEX20 X X Matrix PTS2 protein import cyt/MA

PEX21 X Matrix PTS2 protein import cyt/MA

PEX22 X X Recruiting of PEX4 to peroxisomal membrane IMP

PEX23 X X ? IMP

PEX24 X X ? IMP

PEX25 X Regulation peroxisome proliferation IMP

PEX26 X Recruitment PEX6 (and PEX1) to peroxisomal membrane IMP

PEX27 X Regulation peroxisome proliferation IMP

PEX28 X Regulation peroxisome proliferation IMP

PEX29 X Regulation peroxisome proliferation IMP

PEX30 X Regulation peroxisome proliferation IMP

PEX31 X Regulation peroxisome proliferation IMP

PEX32 X Regulation peroxisome proliferation IMP

Abbreviations: S.c., Saccharomyces cerevisiae; Y.l., Yarrowia lipolytica; D.m., Drosophila melanogaster; C.e., Caenorhabditis elegans; cyt, cytosol; MA, membrane associated; IMP, integral membrane protein.

a Other yeasts/fungi: Pichia pastoris, Hansenula polymorpha and Neurospora crassa.

Results from a large number of genetic, molecular and cell-biological studies show convincingly that the general principles of peroxisome biogenesis have been maintained in all organisms analyzed and are similar for the different members of the organelle family (per-oxisomes, microbodies, glyoxysomes, glycosomes). Models for various steps of peroxisome assembly emerged from these studies and will be discussed in the following sections. However, it will also be shown that differences have been found in details of the process and that the molecules (peroxins) involved may differ considerably between species. In some cases, different species seem to use non-homologous proteins to perform corresponding steps in the organelle-assembly process, or homologous proteins could not (yet) be identified. In Section 6, available data about the biogenesis of trypanosomatid glycosomes will be discussed in much detail. The presently known differences between glyco-some biogenesis in the parasites and peroxisome biogenesis in humans will be highlighted and the possible exploitation of these differences for the design of trypan-ocidal drugs will be discussed.

5.1. Import of matrix proteins; receptor recognition

The majority of the peroxins discovered until now are involved in the routing of newly synthesized proteins to the matrix of peroxisomes (Fig. 2). These matrix pro-

teins are synthesized on free ribosomes in the cytosol where they interact with a cytosolic receptor, either PEX5 or PEX7, dependent on their peroxisome-target-ing signal (PTS).

The principal peroxisomal matrix protein-import pathway involves the proteins harbouring a perox-isome-targeting signal of type 1 (PTS1) which is recognized by the receptor PEX5. Early studies on PTS1 revealed that the signal is a C-terminal tripeptide with the consensus sequence (S/A/C)-(K/R/H)-L [75]. Later, it was shown that this motif may vary considerably more regarding to its length and sequence, as also the specificity of the signal may vary in different organisms [7678]. Recently, Neuberger et al. [79] reanalyzed all published data on PTS1-PEX5 interaction and arrived at the conclusion that at least the 12 C-terminal residues of a given protein are implicated in PTS1 recognition. This 12 amino-acid motif can be structurally and functionally divided into three regions: (1) the C-terminal tripeptide, (2) a region interacting with the surface of PEX5 (about four residues further upstream) and (3) a polar-solvent accessible and unstructured region with linker function (the five remaining residues). Gatto et al. [80] studied in great detail the affinities of variant PTS1-containing peptides to investigate the sequence requirement for receptor binding. Their results suggested that there is a threshold affinity of the peptide for the PTS1-binding domain of PEX5. If its affinity

Fig. 2. General model of peroxisomal matrix protein import. Matrix proteins harbouring a PTS1 or a PTS2 are recognized in the cytosol by the receptors, PEX5 and PEX7, respectively. A cascade of protein-protein interactions occurs in the consecutive steps of peroxisomal protein import which are: docking, translocation, cargo release and receptor recycling. For details, see text.

for the receptor is tighter than this threshold, a given sequence will function as a peroxisome-targeting signal.

The PTS1 receptor, PEX5, is one of the most intensively studied and better understood peroxins. First identified in the yeast Pichia pastoris [81], it has then been cloned in numerous organisms: mammals, other yeast species, Caenorhabditis elegans, plants [82-90] and the trypanosomatids T. brucei and Leishmania donovani (see Section 6). Subcellular localization experiments indicated that PEX5 is mostly cytosolic with a smaller proportion associated with peroxisomes [81,83,87,90-93]. But Szilard et al. [86] observed that PEX5 was entirely localized inside the peroxisome of the yeast Yarrowia lipolytica and Terlecky et al. found PEX5 tightly associated with the matrix face of the peroxisomal membrane of P. pastoris. The different localizations of PEX5 are in agreement with results of

recent studies in which PEX5 (and PEX7, another PTS receptor - see below) were found to undergo multiple rounds of entry into the peroxisome lumen. This gave rise to the 'extended shuttle' hypothesis in which both receptors actually enter the peroxisome together with their bound cargo molecules and release them in the lumen before shuttling back to the cytosol (see Section 5.4).

The C-terminal half of PEX5 contains a region of 6-7 tetratricopeptide repeats (TPRs), a 34 amino-acid degenerate motif (Fig. 3). These TPRs have been shown to be essential for the interaction between PEX5 and the PTS1 [83,84,94,95]. The three-dimensional structure of human PEX5, complexed with a PTS1-containing pen-tapeptide, has been resolved by X-ray crystallography unravelling the nature of the interaction between PEX5 and the PTS1 [96] (Fig. 4). In addition, random

(a) mammals

PXXP-region

PTS2 protein

(b) yeast

PXXP-region

PTS2 protein

Fig. 3. Scheme of known interactions between proteins to be imported into the peroxisomal matrix, their respective receptors and the docking-complex peroxins. Yeast and mammalian PEX5 possess, in their C-terminal part, seven TPR motifs (indicated as white boxes) responsible for the recognition of the PTS1. PTS2 proteins are recognised by PEX7. (a) In mammals, a proline-rich motif (PXXP) in PEX14 seems to be responsible for interaction of this peroxin with the SH3 domain of PEX13. The N-terminal part of PEX13 binds to the WXXXF/Y motifs (indicated as small black bars) no. 2 to 4 of PEX5, whereas other WXXXF/Y motifs are involved in binding the N-terminal part of PEX14. The longer version of PEX5, PEX5L, possesses a putative PEX7 binding box. (b) Also in the yeast S. cerevisiae, the proline-rich PXXP sequence of PEX14 allows the binding of this peroxin to the SH3 domain of PEX13. Moreover, this PEX13 SH3 domain interacts also with the first of only two WXXXF/Y motifs present in S. cerevisiae PEX5. PEX18 and PEX21 are yeast specific peroxins which bind to PEX7, and are necessary for PTS2 import.

Fig. 4. The structure of the C-terminal TPR domain of the human PEX5 in complex with a PTS1 containing pentapeptide (YQSKL), based on the crystal structure reported by Gatto et al. [96] (Protein Data Bank accession code 1FCH). The two TPR clusters surrounding the peptide, TPR1-3 and TPR5-7, are coloured blue and yellow, respectively. TPR4, forming a hinge region, is in green and other regions are in grey. The dashed lines indicate the disordered regions of the protein. The peptide is drawn as a ball-and-stick model with carbon coloured in grey, oxygen in red and nitrogen in blue.

and site-directed mutagenesis of S. cerevisiae PEX5 (ScPEX5) confirmed the importance of the TPR domain for PTS1 binding [97]. These studies showed that the seven TPR segments of PEX5 form two clusters (TPR 1-3 and TPR 5-7) with TPR 4 acting as a hinge. The PTS1 sequence binds to the groove between these two TPR clusters.

In mammals, two isoforms of PEX5 are produced by alternative splicing [98,99]. The isoforms are called short-form and long-form of PEX5, PEX5S and PEX5L, respectively. This longer form has a 37 amino-acid insertion located upstream of the TPR domain. It has been shown that PEX5L, but not PEX5S, interacts with PEX7, the cytosolic receptor of PTS2 proteins. In these organisms, PEX7 does not bind directly to PEX14, a protein of the docking complex at the peroxisomal membrane (see also Section 5.2). Therefore, it has been proposed that PEX5L interacts with PEX7 in order to enable the docking of the PTS2 receptor-ligand complex to the membrane of the peroxisome. Thus PEX5 seems in these organisms not only responsible for the PTS1 import pathway but also involved in the PTS2 pathway.

Protein-protein interaction studies demonstrated that PEX5 binds to proteins forming the docking complex at the membrane: PEX13 [92,93,100-105] and PEX14 [101,104-111] (Figs. 2 and 3). The interactions with

these two peroxins, PEX13 and PEX14, are mediated via pentapeptides WXXXF/Y localized in the N-termi-nal domain of PEX5. The WXXXF/Y motifs are present in two or more copies depending on the species: two in S. cerevisiae, seven in human or 11 in tobacco. PEX5 has also been shown to interact with PEX10 and PEX12 [112-114], peroxins involved in the translocation step, and to PEX8 [115] leading to an extensive network of interactions. A scheme of the different mapped regions of PEX5 is shown in Fig. 3. These interactions will be described in more detail in Sections 5.2 and 5.3.

The second import pathway involves the other known cytosolic receptor, PEX7, and proteins harbouring the peroxisomal-targeting signal of type 2 (PTS2). This PTS2 is a relatively rare topogenic signal and the PTS2—PEX7 interaction as well as the PTS2-protein import mechanism are far less understood. The first PTS2 to be identified was the one of rat thiolase [116]. Despite the fact that only a few proteins employ the PTS2 pathway, their import is absolutely essential. In human, an inability to import PTS2 proteins leads to the disease rhizomelic chondrodysplasia punctata that causes death of the patient within two years of life [117,118]. The PTS2 consists of a nonapeptide (R/ K)(L/V/I)X5(Q/H)(L/A) close to the N-terminus of the protein [119,120]. The five amino-acid linker between the conserved residues is crucial for recognition by the PTS2 receptor. However, it seems that the sequence required for import is somewhat different between species and therefore the above consensus may be an oversimplification [121-123]. In the cytosol, PTS2 proteins are recognized by PEX7. This receptor has been found in mammals and yeasts [119,124-129], but is apparently absent from C. elegans. Indeed C. elegans, and perhaps even more species, does not utilize this targeting pathway. Every known peroxisomal matrix protein from this organism possesses a PTS1 sequence and is imported via PEX5 [130].

The protein PEX7 consists of six or seven WD repeats, which are approximately 40 amino acids long and contain a central tryptophan (W)-aspartate (D) motif. These repeats, constituting almost the entire protein, are preceded by a distinct N-terminal region of approximately 60 amino acids. The subcellular localization of PEX7 has been described as being in the cytosol but the protein has also been found in the lumen of the per-oxisome [119,125,126,129-133]. This intraperoxisomal localization found for PEX7 and PEX5 (see above) gave evidence for the 'extended shuttle' model that invokes entry of the receptor-cargo complex into the peroxi-somal matrix (see Section 5.4). Little is known about the critical regions of PEX7 involved in protein-protein interactions. However, PEX7 has been shown to interact with several peroxins. These peroxins are PEX14 [106,107,109], PEX13 [134,135] and, only observed in S. cerevisiae, both PEX18 and PEX21 [136].

As mentioned above, it has been shown that in human cells PEX7 interacts with the longer isoform, of PEX5, PEX5L, which has a 37 amino-acid insertion located upstream of the TPR domain, but does not bind to PEX14 [98,99]. Initially, this additional sequence was thought to be responsible for the interaction of PEX5L with PEX7. Later, studies to unravel this interaction demonstrated that a region upstream of the 37 residues is also crucial for the interaction with PEX7. Moreover, the C-terminal 18-21 amino acids of the 37-residues insert can be removed without significantly reducing the ability of PEX5L to mediate PTS2 protein import. However, a sequence of at least eight residues is needed in order to bind PEX7 [99,131]. Recently, two-hybrid analysis of Arabidopsis thaliana peroxins showed that PEX14 interacts with PEX5, but not with PEX7. Instead, PEX7 binds PEX5 indicating that, in Arabidopsis, the PTS2 pathway depends also on the PTS1 pathway. The interaction analysis showed that the binding domain of AtPEX7 was restricted to the very C-terminal region which in Arabidopsis does not include the WD40 repeats. Indeed, the WD repeats were identified as the binding domain for PTS2 proteins. And an N-terminal region of AtPEX5, which does not overlap with the binding domain for AtPEX14, is responsible for the interaction with AtPEX7 [137]. In that perspective, the PEX7 binding partner in Arabid-opsis is similar to that of long-form PEX5 in human, even though AtPEX5 is not found as differently spliced isoforms.

PEX18 and PEX21 are two structurally related peroxins involved in PTS2 protein import in S. cerevis-iae. Comparison of both protein sequences revealed a weak but significant level of similarity distributed throughout the protein sequences (23% identity). The level of identity is most prominent at the C-terminus (35% for the last 60 amino acids). Absence of both peroxins abolishes the PTS2 branch of peroxisome biogenesis without affecting the PTS1 branch [136]. They are found mostly in the cytosol where they bind PEX7 through their C-terminal regions. It seems that either PEX18 or PEX21 is required prior to the docking event for the formation of a PTS2 protein-containing import-competent complex. So in the cytosol, PEX18 and PEX21 would promote the formation of an oligo-meric complex formed by PEX7-PTS2 ligand and PEX18/PEX21 [132,135]. No orthologues of PEX18 or PEX21 have ever been found in higher eukaryotes. In Y. lipolytica, PEX20 is an orthologue of S. cerevisiae PEX18 and PEX21. YlPEX20 shows no sequence homology with ScPEX18 or ScPEX21 and databases searches in S. cerevisiae with the YlPEX20 sequence failed to identify the presence of a corresponding per-oxin [136]. Indeed, a S. cerevisiae Apex18/pex21 double mutant could be partially complemented by expression of YlPEX20 [138,139]. Similarly, PEX20 of Neurospora

crassa can also substitute for ScPEX18/PEX21 [140]. Therefore, it is likely that YlPEX20 and NcPEX20 replace PEX18 and PEX21 in forming an import-competent complex and function together with PEX7 in PTS2 import. Recently, a 21 amino-acids long peptide motif of mammalian PEX5L, shared by S. cerevisiae PEX18 and PEX21 and Y. lipolytica PEX20, was identified that overlaps the regions shown to be essential for PTS2 protein import [131,139]. Thus HsPEX5L, ScPEX18/ ScPEX21 and YlPEX20 all contain a conserved sequence which most likely represents a common PEX7 binding site.

A few proteins lack both PTS1 and PTS2 but are nevertheless targeted to the peroxisome. The first example of such a peroxisomal enzyme studied was Candida tropicalis acyl-CoA oxidase, which was shown to be targeted to the peroxisome by a polypeptide-internal sequence (I-PTS) [141]. Recently, Klein et al. [142] demonstrated the existence of a novel, PEX5-dependent non-PTS1 pathway for S. cerevisiae acyl-CoA oxidase, POX1, another I-PTS protein. POX1 and PEX5 can interact directly with each other without support of other proteins, but the binding involves novel contact sites in both proteins. The binding region in PEX5 is located in a defined area of the N-terminal part of the protein, thus outside the C-terminal TPRs constituting the PTS1-recognition domain, and the interaction site of the POX1 polypeptide is indeed located internally. Internal peroxisome-targeting signals might thus be recognized by the cytosolic receptors, PEX5 or PEX7, or by another yet unknown receptor.

However, a direct interaction between a cytosolic receptor and the I-PTS proteins is not the only import mechanism considered nowadays. In contrast to mitochondria, peroxisomes can import folded and even oligomeric proteins (see Section 5.5). Various studies showed that, for both PTS1 and PTS2 proteins, subunits containing the import signal could assemble with sub-units from which the motif had been removed and be transported as multimeric molecules into the peroxi-somes of yeast, plant and mammalian cells [121,143,144]. This is suggestive for import by 'piggy backing' as a natural mechanism, a process that would involve the formation of a heteromeric complex of the protein harbouring no consensus targeting motif with a PTS1- or PTS2-containing protein to enable the import of the former [145,146]. Strong support for 'piggy-backing' targeting has been obtained for a yeast enzyme involved in p-oxidation of unsaturated fatty acids (A3-cisA2-trans-enoyl-CoA isomerase) that has only a weakly functioning PTS1 but is imported due to interaction with a related protein having an efficient PTS1 [147]. It is therefore conceivable that an I-PTS could serve as an anchor enabling a protein harbouring such a sequence to form a complex with a PTS1 or PTS2 protein.

5.2. The docking complex for import of matrix proteins

Following the recognition in the cytosol, the receptor with its bound cargo docks on the peroxisomal membrane at a complex formed by PEX14, PEX13 and PEX17; this latter peroxin is only found in yeasts. PEX14 has been cloned from different species: mammals [108-110], various yeast species (S. cerevisiae, P. pastoris and Hansenula polymorpha) [104,106,107,148,149], plants [150,151], T. brucei and L. donovani (see Section 6).

In S. cerevisiae and P. pastoris, PEX14 has been shown to be a peripheral peroxisomal membrane protein tightly bound to the cytosolic face [106,148,149]. Despite the lack of a membrane-spanning domain, PEX14 presents short hydrophobic regions that could possibly interact with membrane proteins and lipids and may therefore be responsible for the protein's association with the membrane. Contrary to yeast, mammalian PEX14 behaves like an integral membrane protein, with its C-terminal two-third largely exposed to the cytosol [109,110,152]. By a very complete protease protection assay, it was shown that a small domain of 16 kDa is embedded in the peroxisomal membrane. So this domain alone is responsible for the fact that this PEX14 displays the characteristics of an intrinsic membrane protein [152]. As in mammals, plant PEX14 behaves as an integral membrane protein having at least two hydro-phobic segments and part of the polypeptide located in the cytosol [150,151].

Again, extensive studies have been performed on the interacting partners of PEX14. In addition to its interaction with PEX5 and PEX7, it was shown to bind PEX13 [106,107] and PEX17 [153,154]. Some of these studies also showed that PEX14 interacts with itself [106,107], and would exist predominantly as a dimer in vivo [155]. A predicted coiled-coil region was found in the mammalian and yeast peroxin and may be responsible for dimerization of the protein [108-110,148,149,152].

The interaction of PEX14 with both the PTS1 and PTS2 receptor, PEX5 and PEX7, in yeasts, and with PEX5 in mammals and plants, was shown using two-hybrid and co-immunoprecipitation studies [104-108, 110,151]. Mammalian PEX5 possesses multiple high-affinity binding sites for PEX14, which appear to be distributed throughout its N-terminal half. Pentapeptide motifs WXXXF/Y (present seven times in the protein) were proposed to be responsible for the interaction with PEX14 [101]. Later it was demonstrated that each of these PEX5 motifs are capable of forming a single, high-affinity binding site for the N-terminal PEX5-bind-ing domain of PEX14. Indeed, both aromatic amino acids tryptophan and phenylalanine/tyrosine at positions 1 and 5, respectively, are indispensable for the high affinity interaction with PEX14. The authors hypothesized that each WXXXF/Y motif, supposed to be part

of an amphipathic a-helix, binds directly to a hydrophobic region of PEX14. Anyhow, it was concluded from these studies that the pentapeptide motifs form multiple PEX14-binding sites, in a mutually independent manner [111,156]. In S. cerevisiae and P. pastoris, PEX5 possesses, respectively, two and three WXXXF/Y motifs. However, in both yeasts, no interaction between PEX5 fragments containing one or more WXXXF/Y motifs and PEX14 could be detected using overlay and pulldown assays [102-104]. In contrast, some of these motifs have been shown to be essential for the association of PEX5 with PEX13 (see below). In A. thaliana, two distinct regions in the N-terminal half of PEX14, i.e., the segments I58-L65 and R78-R97, interact with PEX5. Therefore, it was proposed that these regions might form a binding pocket for PEX5 [137].

PEX14 has been postulated to be the initial docking site for cargo-loaded PEX5 and PEX7 and would be therefore the point of convergence of the PTS1 and PTS2 pathways. In mammalian cells, overexpression of PEX14 led to the accumulation of PEX5 at the peroxi-somal membrane. In contrast, overexpression of PEX13, PEX10 or PEX12 did not cause apparent PEX5 accumulation at the peroxisomal membrane, and import of PTS1 and PTS2 proteins remained unaffected. Moreover, in mutant cells lacking PEX14, PEX5 is found in the cytosol whereas it is found at the membrane in cells lacking PEX13, PEX10 or PEX12 [156]. Later, it was shown that PEX5 loaded with PTS1-containing protein interacted only with PEX14, while unloaded PEX5 bound to PEX13. This implies that PEX13 functions at a step following the detachment of the cargo-protein from the receptor. The authors proposed a model for peroxisomal protein import mediated by the receptor PEX5 and the import machinery comprising the initial PEX5-anchoring site, PEX14, and PEX13 [156].

As mentioned above, PEX14 interacts also with another component of the docking complex, PEX13. This interaction has been well studied in both yeasts (S. cere-visiae and P. pastoris) and mammalian cells. Contrary to PEX14, it has been well established that PEX13 is an integral peroxisomal membrane protein and its termini are both oriented toward the cytosol [134]. PEX13 possesses, in its C-terminal part, a SH3 domain which is involved in protein-protein interactions. In S. cerevisiae, both PEX5 and PEX14 are able to bind PEX13 via its SH3 domain [92,93,100,106]. Competition assays demonstrated that the two ligands can bind simultaneously to this domain [157]. In this organism, PEX14 is the classical SH3 ligand as it presents a PXXP motif in the N-terminal part of the protein [134]. Indeed, the PXXP motif has been shown to interact directly with SH3 domains (reviewed in [158]). Surprisingly, PEX5 does not possess such a PXXP motif in its amino-acid sequence but still interacts directly with

PEX13. Instead, PEX5 presents an a-helical element that binds to a novel interaction site on the SH3 domain which is distinct from the classical PXXP binding cleft [102]. In P. pastoris, one identifiable feature of the SH3 binding site in PEX5 is the presence of three pen-tapeptide repeat motifs [104]. Bottger et al. [103] reported that the first WXXXF/Y motif of S. cerevisiae PEX5 is important for the interaction with PEX13 whereas the second motif does not present any affinity for either PEX13 or PEX14. The residues Trp and Phe of the first pentapeptide, as well as a negatively charged residue outside this motif, were shown to be indispensable for the binding of PEX5 to the SH3 domain of PEX13 [102,103,157]. Later, it was proven that the two ligands, ScPEX14 and ScPEX5, bind in unrelated conformation to patches located at opposite surfaces of this SH3 domain [159]. In human, several PXXP motifs in PEX14 are likely to be responsible for the interaction with PEX13, as reported for S. cerevisiae. However, human PEX5 does not bind to the SH3 domain of PEX13 but rather to the N-terminal part of the protein. This interaction occurs via the WXXXF/ Y motifs 2-4 of PEX5 [156]. Therefore, it is most likely that PEX5 interacts with PEX13 differently in mammalian and yeast cells.

The third component attributed to the docking complex is PEX17. It is only found in yeast species, Y. lipoly-tica, S. cerevisiae and P. pastoris, where it is peripherally but tightly associated with the peroxisomal membrane [153,154,160]. In S. cerevisiae, PEX17 was shown to interact with PEX14 and therefore proposed to be part of the docking complex [153]. Nevertheless, the presence of PEX17 is not a prerequisite for the docking of PEX5 and PEX7 to the peroxisomal membrane [134]. Based on studies with Apex17 mutants of P. pastoris and S. cerevisiae it was suggested that PEX17 may also be involved in the insertion of PMPs into the peroxisomal membrane [153,154,161,162]. However, such mutant cells often displayed only slight effects on PMP abundance and localization, similar to the changes sometimes observed in Apex5 mutants. Therefore, it is unlikely that PEX17 plays an important role in the insertion of PMPs into the membrane [162].

5.3. Import of matrix proteins; cargo-receptor complex translocation and receptor recycling

Following the docking step, the cargo-receptor complex will enter the peroxisome where the receptor will release its cargo protein before being recycled back to the cytosol (Fig. 2). These last processes are far less well understood and questions still arise about many aspects. PEX2, PEX10 and PEX12 are thought to be involved in the translocation step that follows the docking of PEX5 and PEX7 together with their respective ligand on the membrane. The import thus

results from a cascade of events ending eventually in the internalization of the matrix proteins. The hypothesis is that PEX2, PEX10 and PEX12 form another protein complex in the membrane distinct from the docking machinery. These three peroxins are integral peroxisomal membrane proteins, possessing a 'RING finger' domain exposed to the cytosol [112,113,163165]. The RING domains of PEX10 and PEX12 enable the interaction between the two proteins as well as with PEX5 [112,113]. In vitro, PEX10 also interacts with PEX2, a peroxin that has been suggested to act downstream of PEX10 and PEX12 [113]. Furthermore, PEX2 is also essential for PTS2 protein import. However, a single substitution in the zinc-binding RING motif results in loss of the PTS1 but not PTS2 import implying that the protein possesses two distinct domains necessary for the PTS1 and PTS2 pathways, respectively [166]. The main reason to group PEX2, PEX10 and PEX12 in the translocation machinery, and not in the docking complex, came from the observation that human cells defective in PEX2, PEX10 and PEX12 are still able to recruit PEX5 to the peroxisomal membrane, presumably through PEX14 [105,112]. Yet, Reguenga et al. [114] demonstrated that PEX2, PEX12, PEX5 and PEX14 are constituents of the same protein assembly. This again shows the complexity of the import mechanism in which each step interconnects with the other in an intimate manner. On the other hand, Dodt and Gould [91] isolated cells from a patient suffering from a disorder caused by a mutation in PEX12 and found PEX5 accumulating inside the peroxisome. From this observation and later studies, it was concluded that PEX12 actually appears to have a role in recycling PEX5 across the membrane [112]. Furthermore, Eckert and Erdmann [70] state that, based on the available evidence, it is not impossible that PEX14, PEX13 and PEX17 could also act in the translocation event in addition to their role in the docking step.

The identification and characterization of a T. brucei and L. donovani PEX2 has also been reported, and the importance of this peroxin glycosome biogenesis has been confirmed (see further details in Section 6.2).

PEX4 is a member of the family of ubiquitin-conju-gating enzymes and can bind to ubiquitin in vitro [167,168]. In S. cerevisiae and P. pastoris, deletion of the PEX4 gene disrupts targeting via both the PTS1 and PTS2 pathways [167,169]. The protein is anchored to the cytosolic face of the peroxisome by the integral membrane protein PEX22. The model predicts that PEX4 and PEX22 act together in the import of peroxi-somal matrix proteins [170]. Recent studies strongly suggested that the complex of PEX4 and PEX22 is linked to the protein import apparatus via a unique binding to PEX10 [171]. The exact role of the two peroxins PEX4 and PEX22 in the import process remains unclear. In

P. pastoris Dpex4 and Dpex22 mutants, the abundance of PEX5 is drastically decreased. Although the phenomenon is not understood [170], it has been used in an epis-tasis analysis which revealed that PEX13, PEX14, PEX10, PEX12, PEX2 and PEX8 function at the initial stage of PTS1 protein import, followed by PEX1 and PEX6 and subsequently PEX22 and PEX4 [172]. It has also been suggested that H. polymorpha PEX4 may be involved in PEX5 recycling to the cytosol [173]. In addition, an entirely different role has been proposed for PEX4. Several studies support the idea that degradation of peroxins occurs in proteasomes and that this process would be dependent on PEX4 and its ubiq-uitination. Purdue and Lazarow [174] showed that in wild-type S. cerevisiae PEX18 undergoes a rapid degradation with a half-time of less than 10 min. In these cells, mono- and di-ubiquitinated forms of PEX18 could be detected during peroxisome biogenesis. However, a role for PEX4 in this PEX18 ubiquitination remains hypothetical and more experiments will be required to investigate whether this might be the case.

Another peroxin required for both the PTS1 and PTS2 import pathways is PEX8 whose role remains to be elucidated. It is localized at the matrix side of the per-oxisomal membrane but nevertheless interacts with PEX5 in S. cerevisiae [115] and with PEX20 in Y. lipoly-tica [175]. In S. cerevisiae, Agne et al. [176] have defined two core complexes of the peroxisomal import machinery: the docking complex comprising PEX14 and PEX17, with the loosely associated PEX13 on one side and the RING finger complex containing PEX2, PEX10 and PEX12 on the other side. They conclude that PEX8 organizes the formation of a larger import complex from the matrix side of the peroxisome membrane and thus might enable functional communication between both sides of the membrane. But how PEX8 promotes the association of the docking and the RING finger complex from the inner face of the membrane will be a challenging question to answer.

PEX1 and PEX6 are membrane-bound peroxins belonging to the family of AAA (ATPase-associated activity) proteins [177,178] and have been shown to interact with each other in an ATP-dependent manner [179-181]. Their two characteristic conserved AAA domains (denoted D1 and D2), containing Walker ATP-binding sequences, are responsible for the ATP-ase activity. Studies on the localisation of PEX1 and PEX6 showed striking differences among species. It runs from peroxisomal [181], to peripherally associated with peroxisomes [182], to cytosolic when overex-pressed [183,184], and even to be present in vesicular structures distinct from mature peroxisomes [179,185,186]. The role of these peroxins is still a matter of debate. In P. pastoris, PEX1 and PEX6 have been proposed to be involved in matrix protein import. They were shown to be part of the translocation

machinery where they act downstream of the docking and translocation steps but upstream of PEX4 and PEX22 [172]. The results of this study suggest, therefore, that these peroxins may play a role in the recycling of PEX5 [187] (Fig. 2). On the other hand, it is also tempting to speculate that these two ATPase peroxins could provide the energy needed for matrix protein import from ATP hydrolysis. These two functions are not necessarily contradictory. It has been reported in many publications that peroxisomal matrix protein import is known to be dependent on hydrolysis of ATP (e.g., see [188,189]). However, Oliveira et al. [190], performing in vitro import assays with rat liver peroxisomes, showed that the ATP hydrolysis is not required for the insertion of PEX5 into the per-oxisomal membrane. They argued that protein translocation across the peroxisomal membrane may be driven by both energy released during the interactions of peroxins (e.g., PEX5 and PEX14) and ATP hydrolysis downstream of the translocation step to reset the PEX5-mediated transport system (e.g., recycling of the receptor by PEX1 and PEX6). Also other functions than energization of receptor recycling have been proposed for PEX1 and PEX6, such as mediating membrane fusion in peroxisomal membrane formation in Y. lipolytica and P. pastoris [179,185,186,191]. Clearly, further studies will be essential to elucidate the precise role of PEX1 and PEX6 in peroxisome biogenesis in different organisms.

Out of the 32 known peroxins, some remain with an as yet undefined or unclear role. PEX9 and PEX23 are integral membrane proteins in Y. lipolytica, with unknown function in peroxisome biogenesis [192,193]. Recently, a Blast search of the genome in N. crassa revealed the presence of a putative PEX23 [140]. PEX24 is a novel integral membrane protein of Y. lipolytica [194] which is also present in N. crassa [140]. Studies of a Y. lipolytica pex24 mutant revealed that these cells are defective in matrix and membrane protein import. Furthermore, PEX26 was identified using a mammalian CHO mutant cell line [195,196]. It is a per-oxisomal membrane protein. In immunoprecipitation, PEX1 and PEX6 co-precipitate with PEX26. Epitope-tagged PEX1 and PEX6 could not be detected as puncta in PEX26 defective cells, but PEX26 expression re-established colocalization, in a PEX6-dependent manner. PEX26 appeared to recruit PEX6-PEX1 complexes to the peroxisomes; PEX26 anchors PEX6 and PEX1 through PEX26-PEX6 and PEX6-PEX1 interactions, respectively.

Interestingly, these observations made on PEX26 are similar to those recently made on PEX15 in S. cerevisiae [197]. This latter membrane-bound peroxin had previously been identified in this yeast, but its function remained unknown. Now it has been shown that it is an integral membrane protein with most of its sequence,

including the N-terminus, facing the cytosol. It binds, both in vivo and in vitro, to the AAA peroxin PEX6. This suggests that PEX15 in S. cerevisiae exerts a similar function as PEX26 in mammalian cells, i.e., the recruitment of PEX6 to the peroxisomal membrane. The N-terminal part of PEX6 contains the binding site for PEX15. Mutagenesis of the Walker motifs of the two AAA cassettes D1 and D2 of PEX6 show that D1 and D2 have opposite effects on the PEX6-PEX15 interaction. The cycle of recruitment and release seems dependent on ATP hydrolysis by PEX6.

5.4. Receptor recycling and the 'extended shuttle' model for the import of matrix proteins

Initially, it was proposed by Marzioch et al. [124] that the receptors first recognize their cargo proteins in the cytosol. Subsequently, the receptor-cargo complexes would dock at the peroxisomal membrane where the receptors deliver their cargo without entering the perox-isome. After release of the cargo proteins, the free receptors would then return to the cytosol. This model

became known as the 'simple shuttle' mechanism. Later, experimental evidence that PEX5 actually enters the per-oxisome, subsequently releases its PTS1 protein and recycles back to the cytosol, was provided by Dodt and Gould [91]. Reversible accumulation of PEX5 in the peroxisomal compartment could be demonstrated in vivo by manipulating the ATP level and temperature in cultured human fibroblasts and it seemed that PEX5 exits the peroxisomal compartment by a process that requires ATP. More recently, Dammai and Subramani [198] elegantly showed that PEX5 enters human peroxi-somes during the course of its normal function and ree-merges in the cytosol to carry out further rounds of import. This led to the 'extended shuttle' model in which both receptors enter the peroxisome together with their cargo and release them in the lumen before shuttling back to the cytosol (Fig. 2). However, nothing is known yet about a protein-export machinery or PEX5 and PEX7 sorting machinery [199]. In this context, it is tempting to speculate that some peroxins proposed to play a role in the import of proteins might in fact act in receptor export.

Fig. 5. Three models for import of folded and oligomeric proteins across the peroxisomal membrane (b, c and d) compared with the general model for transmembrane transport of unfolded proteins (a). (a) Chaperonins, indicated by green-coloured dots, keep the newly synthesized protein unfolded at the cis side of the membrane, where it is delivered to a receptor. Subsequently, it is translocated through a channel and, at the trans side, properly folded with the assistance of chaperonins. (b) Proteins are folded and form oligomers in the cytosol, prior to their import through a regulated pore in the peroxisomal membrane. (c) Cytosolic oligomeric proteins are wrapped by a newly formed double membrane; the outer membrane of the protein-containing vesicle fuses with the peroxisomal membrane and subsequently the inner membrane is degraded, releasing the protein in the peroxisomal matrix. (d) The oligomeric protein in the cytosol is surrounded by an invaginating peroxisomal membrane and internalized by a process reminiscent of endocytosis. At the matrix side, the membrane is degraded to release the translocated protein. (Figure adapted from [200]).

5.5. Three models for the translocation mechanism of matrix proteins

What happens after the docking event still remains largely unknown. Whereas several peroxins are known to be required for the translocation step as discussed above, the mechanism of internalisation of matrix proteins itself has not yet been unravelled. Several data suggest that protein unfolding is not a requirement for the import of matrix proteins as it is the case for protein translocation across the membranes of the endoplasmic reticulum (ER), mitochondria and chloroplasts (reviewed in [145,200]). Indeed, it has been demonstrated that peroxisomes can import folded and oligomeric proteins [121,143,201-203]. However, import of oligomeric proteins is not a general feature of peroxisomal matrix proteins [204]. A recent study demonstrated that proteins are imported into the peroxisome either folded or in unfolded state [205]. This study supports the idea that ternary structure has little or no effect on protein import into the peroxisomal matrix. Misfolded reporter pro-

teins bearing a PTS1 appeared to be imported in association with chaperones, molecules which have never been described within the matrix of the mammalian peroxi-somes. However this process might not be functionally relevant under normal conditions.

Three possible mechanisms for transport of oligo-meric proteins across the peroxisomal membrane have been considered over the years (Fig. 5). Because large complexes of oligomeric proteins and even 4-9 nm gold beads coated with PTS1-containing protein can be imported into the peroxisome [206], the existence of a pore has been postulated as is the case in the nuclear membrane [200,207]. But freeze-fracture studies have not given any indications of the presence of such a structure in the peroxisomal membrane. Along the same range of ideas, the existence of a channel assembled transiently in response to cargo delivery to the membrane has been postulated. Again, no clear indications for the presence of such a channel have been provided. Another mode of internalization would be through invagination of a portion of the peroxisomal

Cytosol

PTS1 protein

EB > I

I-PTS protein

PTS2 protein

O PEX7 O PEX18 O PEX21

Peroxisome

Fig. 6. Model of formation of a pre-import complex by multiple protein-protein interactions involving cytosolic peroxins and peroxisomal matrix proteins to be imported ('preimplex hypothesis'). (Figure adapted from [208]). For a detailed description, see text.

membrane where the cargo proteins are docked [143]. After the internalization of the vesicle by a process reminiscent to endocytosis, the membrane surrounding the cargo proteins must be degraded in order to deliver its content in the peroxisomal matrix. Finally, the last mode of transport hypothesized is the wrapping of the cargo proteins by a double membrane [70]. The first step of this process is the formation of a vesicle around the matrix proteins. The outer membrane of the vesicle will then fuse to the peroxisomal membrane. In this process, the inner membrane has also to be degraded in order to release the vesicle content inside the peroxi-somal lumen. There is little evidence for any of these three pathways in peroxisome biogenesis and speculations on this subject are still numerous.

5.6. Preimplex hypothesis for the import of matrix proteins

As mentioned before, folded and oligomeric proteins can be imported into the peroxisome. Gould and Collins [208] have proposed the 'preimplex' hypothesis that serves to explain this unique feature of peroxisomal matrix protein import. It describes a mechanistic model for the early steps of the import. The authors hypothesize that newly synthesized matrix proteins, together with the PTS receptors, form large protein complexes shortly after their synthesis but before their import (Fig. 6). These pre-import complexes are referred to as preim-plexes. It has been shown that one PEX5 monomer binds one PTS1 protein. Because PEX5 seems to exist in vivo as a tetramer [101], it is potentially able to bind several PTS1 proteins at the same time and so assemble the preimplex. Moreover PEX14, the primary PEX5-docking site at the peroxisomal membrane, is a dimeric protein with at least one PEX5 binding site per monomer [101,111]. Preimplex expansion may be further promoted by the fact that each PEX5 monomer has several binding sites for PEX14. Eckert and Erdmann [70] proposed that PEX7 together with PEX18/PEX21 would contribute to forming preimplexes in a comparable manner as PEX5.

What are the benefits of assembling large complexes at the membrane before the translocation step? It allows the rapid and specific delivery of matrix proteins using only diffusion. It also generates an extremely high concentration of matrix proteins near the docking machinery, thus creating a situation that is likely to promote the subsequent translocation of the proteins. Bellion and Goodman [209] indeed observed that several peroxisomal enzymes enter into an extremely large protein complex immediately after their synthesis but before their translocation into the peroxisomal lumen. The preimplex model may nicely accommodate the 'piggy-backing' hypothesis and thus offers an explanation to the mode of I-PTS protein import. Finally, the inter-

nalization mechanism itself needs to be able to accommodate these large preimplexes. Some models for this mechanism, all compatible with the existence of preim-plexes, have already been discussed in Section 5.5. In conclusion, the preimplex hypothesis offers explanations for different aspects of matrix protein import and proposes a possible mechanism of how the matrix proteins are translocated.

5.7. Insertion of peroxisomal membrane proteins

Matrix protein import and peroxisomal membrane proteins (PMPs) insertion occur via two distinct pathways using different sets of peroxins. Evidence for separate processes came from the observation that in cells of Zellweger syndrome patients, empty peroxisomal membranes called ghosts were found; such ghosts still contain peroxisomal membrane proteins but lack matrix proteins [210-212]. Later, it was demonstrated that depletion of essential peroxins for the import of matrix proteins did not affect the topogenesis of peroxisomal membrane proteins [93,106,213,214]. It indicates that indeed two different pathways are involved in the insertion of PMPs into the membrane and the import of matrix proteins.

It has been accepted for a long time that peroxisomal membrane proteins are synthesized on free ribosomes in the cytosol and post-translationally inserted into the membrane of peroxisomes [215-218]. Some peroxins, including integral membrane proteins, have been identified that are responsible for the correct insertion of PMPs in the peroxisomal membrane, but their individual contribution is still not fully understood. Moreover, a recent publication is challenging the old view on the origin and the formation of peroxisomes, therefore reevaluating the mechanism of the PMPs insertion process [219]. The hypothesis presented in this paper will be discussed in Section 5.10.

Peroxisomal membrane proteins contain a peroxi-somal membrane-targeting signal (mPTS), which is presumably recognized by a mPTS-receptor that targets the proteins to the peroxisomal membrane, where they are inserted [220,221]. But so far, no general consensus for a mPTS has been identified. The first mPTS, described for Candida boidinii, is present in a hydrophobic loop within PMP47 [220]. It was shown to be necessary and sufficient for targeting the protein to the peroxisomal membrane. Later, peptides containing a mPTS were also defined for PMP34 (the human homologue of CbPMP47), PEX3, PMP22, PEX13 and PEX16. These PMPs do not possess mPTS with a consensus sequence apart from a common short stretch of basic amino acids associated with transmembrane domains. Table 2 summarizes the information presently available about these mPTS. The conclusion that can be drawn from the data in this table is that a mPTS consists of a hydrophilic

Table 2

Peroxisomal membrane targeting signals

Protein Organism

References

PMP47 Cb, Sc Cytosolic basic cluster of amino acids adjacent to a membrane anchoring transmembrane domain

Existence of two independent mPTS PMP34 Hs Basic loop + three transmembrane domains. Existence of two independent mPTS

PEX13 Hs Existence of two independent mPTS

PMP22 Rn Two transmembrane domains adjacent to a loop containing a basic cluster of amino acids

Existence of two independent mPTS Basic cluster of amino acids + a transmembrane segment

PEX16 Hs

Sc, Pp, Hs, Rn N-terminal block of positively charged amino acids

[303,304]

[305,306]

[307,308]

[225,224,236,309]

Proteins PMP47, PMP34, PEX13, PMP22, PEX16 and PEX3 from different species; Cb, C. boidinii; Hs, Homo sapiens; Rn, Rattus norvegicus; Sc, S. cerevisiae; Pp, P. pastoris.

peptide containing a group of basic amino acids adjacent to at least one hydrophobic patch or transmembrane domain. It is not surprising to find the presence of a transmembrane fragment included in the mPTS as the protein needs to be anchored into the membrane.

From the 32 peroxins identified to date, only three have been shown to be involved in the synthesis of peroxisomal membrane proteins and/or their insertion into the membrane of the peroxisome: PEX3, PEX19 and PEX16. However, the exact function of these peroxins has not yet been identified. First cloned in S. cerevisiae [222], PEX3 is an integral membrane protein [223,224] with its C-terminus facing the cytosol [225-227]. The location of its N-terminus is more controversial, reported to be either the cytosol [227] or the matrix [222,226]. PEX3 was shown to interact with PEX19 via its C-terminal domain [226-228]. It is widely accepted that PEX3-PEX19 interaction is required for the targeting of other PMPs to the peroxisomal membrane. Possibly, PEX3 interacts also with several other peroxins. This is suggested by immunoprecipitation experiments with P. pastoris which showed that PEX3 can be present in a complex with both the proteins of the docking subcomplex, PEX13, PEX14 and PEX17, and the RING-finger peroxins constituting the putative translocation subcomplex, PEX2, PEX12 and PEX10 [229]. PEX3 seems a shared component of both subcomplexes. Several interactions detected in wild-type cells between components of both subcomplexes are unde-tectable in Dpex3 cells. Together, these data strongly support the notion that PEX3 is involved in the assembly of proteins into the peroxisomal membrane and/or the assembly of peroxisomal membrane complexes necessary for matrix protein import. It should be noted that the role of PEX3 as proposed by Hazra et al. [229] on the basis of their studies with P. pastoris, i.e., linking the docking and translocation subcomplexes, is similar to that proposed by Agne et al. [176] for PEX8 in S. cerevisiae (see Section 5.3).

PEX19 is localized in the cytosol and partially in the membrane of peroxisomes [230-233]. As mentioned above, PEX19 interacts with the C-terminal part of

PEX3 via its N-terminal 42 amino acids [231]. It also interacts with a range of other PMPs [233,234] including PEX10 which requires its C-terminus for this interaction [231]. In several organisms (e.g., human, S. cerevisiae) PEX19 appears farnesylated at its C-terminus, but such modification was not observed in P. pastoris PEX19 [231]. The role of farnesylation is still under debate, but it has been shown for human PEX19 that it is an important determinant in its affinity for other peroxins [235]. Because of its extensive network of interactions, PEX19 has been proposed to act as the cytosolic receptor of PMPs, subsequently targeting them to PEX3 [233]. However this is still a matter of debate. Indeed, other authors suggested that PEX19 could act as a chap-erone. This idea was based on the observation that PEX19, in some cases, interacts with regions of PMPs which are different from the sequences required for their targeting to the peroxisomal membranes, defined as the mPTS [234,235]. It is however clear that PEX3 and PEX19 act at the earliest stage of peroxisome biogenesis which is the insertion of PMPs into the membrane [231,227,236,237].

In human, PEX16 is an integral membrane protein with both termini exposed to the cytosol [238-240]. The complementation of Dpex16 mutant fibroblasts by PEX16 expression, restored the biogenesis of peroxi-somal membranes, subsequently followed by peroxi-somal matrix protein import restoration [238,239]. The participation of PEX16 in the biogenesis of perox-isome membranes is therefore obvious but its precise role in the import of PMPs or formation of the lipid bi-layer is unclear. Recent studies on Dpex3 mutant cells showed that upon expression of PEX3, the co-expression of the cytosolic C-terminal part of PEX16 in these cells led to an abolition of peroxisome assembly restoration. These results imply that most likely PEX16 functions upstream of PEX3 in peroxisome membrane assembly [240]. In contrast, Y. lipolytica PEX16 is a peripheral membrane protein that does not seem to participate in the biogenesis of the peroxisomal membrane

[241] and it appears to be absent from S. cerevisiae

[242].

5.8. Proliferation of peroxisomes and the formation of their membrane

PEX11 has been postulated to play a role in the regulation of peroxisome proliferation since cells lacking PEX11 contain few large peroxisomes [213,243,244] and overexpression of the peroxin led to hyperprolifera-tion of small peroxisomes [245,246]. However, in S. cere-visiae, PEX11 has been proposed to be indirectly involved in peroxisome proliferation by mediating the production of a signaling molecule that is responsible for modulating peroxisome abundance [257]. In S. cere-visiae PEX11 was shown to form homodimers [244]. When present in monomeric conformation, S. cerevisiae PEX11 is active in stimulating proliferation. In contrast, PEX11 dimerization leads to inhibition of organelle division [244].

Mammalian cells contain up to three different PEX11 isoforms, a, p and y [248,249]. The PEX11P gene is con-stitutively expressed. Mice lacking this PEX11p have no apparent lack in peroxisomal import and only mild effects in peroxisomal metabolic functions. However, they show neuropathological features characteristic of the known peroxisomal disorder 'Zellweger syndrome'. Expression of the PEX11a gene is inducible by peroxisomal proliferators agents. PEX11a overexpression promotes peroxisome division. However, pex11 a-deficient mice have normal peroxisome abundance. The role of PEX11y remains to be studied.

Recently, several additional peroxins have been identified which appear to be involved in peroxisome proliferation. First, S. cerevisiae cells lacking a protein subsequently denoted PEX25 have fewer but larger per-oxisomes [250]. Subsequently, a PEX27 was identified in this yeast species [251,252]. PEX25 and PEX27 are a homologous pair of proteins whose C-termini are similar to the entire PEX11. All three proteins localize to the peroxisomal membrane. Deletion of any of these PEX genes resulted in enlarged peroxisomes. Cells deleted for PEX27 show few enlarged peroxisomes, cells harbouring double deletions of PEX27 and PEX25, PEX27 and PEX11, and PEX25 and PEX11 show greatly enlarged peroxisomes. A severe peroxisomal import defect was observed in a triple deletion mutant strain. Overexpression of PEX25 in the triple mutant strain resulted in a massive proliferation of laminar membranes and peroxisomes. Overexpression of PEX11, PEX25 or PEX27 led to peroxisome proliferation and the formation of small peroxisomes. Two-hybrid analysis showed that PEX27 interacts with PEX25 and itself, PEX25 interacts with PEX27 and itself and PEX11 interacts only with itself. Together, the data indicate that these three peroxins build a family of proteins whose members are required for peroxisome biogenesis and play a role in the regulation of perox-isome size and number.

The existence of a family of PEX11-like proteins is not limited to mammalian cells (PEX11a, p and y) and S. cerevisiae (PEX11, PEX25 and PEX27), but extends to trypanosomatids, where recently also several PEX11 homologues have been identified: PEX11, GIM5A and GIM5B, as discussed in detail in Section 6.2.

In recent years, PEX24 has been reported for Y. lipolytica and N. crassa. It is an integral peroxisomal membrane protein with unknown function. Mutant cells are defective in matrix and membrane protein import. Recently, two PEX24 homologues have been identified in the S. cerevisiae genome [253]. The encoded proteins are called PEX28 and PEX29. They are constitutively expressed integral membrane proteins. Peroxisomes of cells deleted for either or both the PEX28 and PEX29 genes are increased in number, exhibit extensive clustering and are smaller in area than peroxisomes of wildtype cells. They also often show membrane thickening between adjacent peroxisomes in a cluster. The buoyant density of peroxisomes from double deletion cells is decreased. Overexpression of PEX25 (not PEX11) can restore the wild-type phenotype in both single and double mutant cells. These data suggest that PEX28 and PEX29, together with PEX25, regulate peroxisome number, size and distribution in S. cerevisiae.

Furthermore, in the S. cerevisiae genome, three genes were found coding for proteins with high similarity to PEX23, also with unknown function so far, of Y. lipolytica [254]. These three proteins, called PEX30, PEX31 and PEX32, are integral to the peroxisomal membrane. Cells deleted for PEX30 exhibit increased numbers of peroxisomes, whereas cells deleted for PEX31 and PEX32 exhibit enlarged peroxisomes. PEX30 and PEX32 cannot functionally substitute for one another or for PEX31, whereas PEX31 shows partial functional redundancy with PEX30 and PEX32. PEX30, PEX31 and PEX32 interact with each other and with PEX28 and PEX29. Systematic deletion studies demonstrated that PEX28 and PEX29 function upstream of PEX30, PEX31 and PEX32 in the regulation of peroxisome proliferation. These data suggest a role for PEX30, PEX31 and PEX32, together with PEX28 and PEX29, in controlling peroxisome size and proliferation in S. cerevisiae.

5.9. Growth and division model versus de novo formation model

Over the years, the origin of peroxisomes was argued and two points of view arose. The earliest one proposed that peroxisomes proliferate and multiply by growth and division [255]. In this model, preexisting peroxi-somes posttranslationally import new matrix and membrane proteins and subsequently divide into new peroxisomes. These preexisting peroxisomes are small and of low density and will mature into large peroxi-

Fig. 7. Model proposed for the proliferation of peroxisomes. The model is a compilation of two earlier hypotheses: 'growth and division' and 'de novo formation' of peroxisomes. In the left part, peroxisomes arise de novo from the endoplasmic reticulum (ER), upon post-translational insertion of peroxisomal membrane proteins (PMPs) into its membrane, either in the membrane of the rough ER and/or the membrane of a pre-peroxisomal part of the ER. Subsequently, pre-peroxisomes which are competent for matrix protein import, emerge from the ER by budding and divide into mature peroxisomes. The right part of the model refers to the 'growth and division' model in which new peroxisomes arise from pre-existing mature peroxisomes by fission. (Figure adapted from [302]).

somes of higher density upon import of these new proteins [256]. Recently, in vivo studies demonstrated that peroxisomes have indeed the ability to multiply by division [257].

In the second half of the 1990s, various lines of evidence for de novo synthesis of peroxisomes, or formation of new peroxisomes in the absence of preexisting ones, were brought to light. Early analysis of Dpex3 and Dpex19 mutants of various yeasts and human cells, showed that neither of these cells contained detectable peroxisomes nor peroxisomal remnants (e.g., ghosts) and mislocalize PMPs to the cytosol [162,223,224, 228,230,232,233,258,259]. When mutant cells were complemented with their respective proteins, PEX3 or PEX19, the peroxisomal activity was restored. This was considered to be strong evidence for de novo formation of peroxisomes. But doubt has been shed on this hypothesis by other studies. The detection of these particular membrane structures was performed using electron microscopic observation, immunofluorescence or immunoelectron microscopy with antibodies against, e.g., PEX10, PEX14 or also PEX11 [162,222,224]. These methods could have easily missed very small peroxi-somal remnants [229,231]. Indeed, in recent studies on Dpex3 and Dpex19 cells of P. pastoris and Y. lipolytica, small vesicles and tubules, resembling early peroxi-

somes, were observed containing PMPs or peroxins [227,229,231,260]. Scientists in favour of de novo formation, argued that depletion of PEX19 and PEX3 could lead to a mistargeting of PMPs to other structures, incorrectly interpreted as preperoxisomes, in the absence of normal target membranes. So there is still an intense debate today whether de novo formation of peroxisomes indeed exists.

For Y. lipolytica, evidence has been presented for the existence of a dynamic population of organelles that differ in various aspects such as protein import competence [185,261,262]. Titorenko and Rachubinski [191,263] presented a model for a multistep peroxisome assembly pathway. Initially pre-peroxisomal structures, thought to arise from a subdomain of the ER (see also Section 5.10), transform into nascent peroxisomes as a result of the uptake of the membrane peroxins PEX3 and PEX16. This second population of peroxisomes is competent to import new matrix and membrane proteins and finally grows into mature peroxisomes which are fully active.

Some authors tried to reconcile both models (Fig. 7). This led to the creation of a two-way model for the biogenesis of peroxisomes [70,218]. The first pathway includes the 'growth and division' model in which early peroxisomes grow into mature peroxisomes by import

of PMPs and matrix proteins. Subsequently, these mature peroxisomes will divide to create new early peroxi-somes competent for newly synthesized protein import. The second half of the pathway is issued from the de novo formation hypothesis. Some pre-peroxisomes originate from an undefined membrane source, e.g., the ER. The newly formed peroxisomes would first acquire PEX3 and PEX16, peroxins involved in the earliest step of peroxisome biogenesis. Subsequently, with the help of PEX19, PEX3 and PEX16 they could import other PMPs involved in matrix protein import, thereby starting the process leading to mature peroxisomes.

5.10. A role for the ER in peroxisomal membrane formation and insertion of peroxisomal membrane proteins?

How the peroxisomal membranes acquire their lipid components is still largely unknown. Phosphatidyl cho-line and phosphatidyl ethanolamine are the main phosp-holipids of the lipid bilayer [264]. The site of synthesis of these phospholipids has been shown to be the ER [255]. The transport from the ER to the peroxisome may occur via specific carrier proteins, transport vesicles or through intermittent connections between the ER and the peroxisomal membrane [69]. This subject is still very controversial because there is no clear evidence of any phospholipid transport mechanism from the ER to the peroxisomes.

In de novo formation, the obvious source of new membranes is the ER. Newly synthesized PMPs could

transit via the ER in order to be incorporated into membranes and subsequently transported to the peroxisomal membranes. Elgersma et al. [221] showed that upon overexpression of PEX15, this peroxin was indeed targeted to ER membranes. But later, it was shown that the insertion of PEX15 into the ER membranes should be interpreted as mistargeting resulting from the overexpression of the protein [162]. Such artifact could be due to the protein's hydrophobic segments which would insert the protein into the membranes of the ER or other compartments if peroxisomal membranes are not available.

A very recent study shed new light on this controversial question of the ER's involvement in peroxisome biogenesis [219,265]. The authors presented evidence that peroxisomal membranes are derived from the ER in non-manipulated mouse dendritic cells. Using immu-no-electronmicroscopy and electron tomography, they observed special lamellar structures that were directly connected to the ER and that contained PEX13 and the peroxisomal transporter protein PMP70. These lamellar structures, named 'lamellae', present a special density and striping and were observed to encircle per-oxisomal clusters. Moreover, by manually tracing the membranes in tomographic slices, they reconstructed the three-dimensional volume of a portion of a cell. In these slices, they showed the membrane continuity of the ER with the lamellae and the connection between a peroxisome with the lamellar reticulum. Based on these observations, the authors proposed the following model (Fig. 8). Initially, a few integral membrane pro-

plasma membrane

nucleus

Fig. 8. ER-peroxisome connection and protein trafficking between the different compartments. Some integral membrane proteins [PEX(X)] are inserted into the ER membranes which become the lamellae. Insertion of additional proteins [PEX(Y), PEX(Z)] which will be components of the protein import machinery or involved in peroxisomal metabolism, occurs in a second step. Later, matrix proteins are imported and the lamellae give rise to mature peroxisomes. (Figure adapted from [219]).

teins enter the ER giving rise to the formation of the specialized ER, the lamellae. Subsequently, additional integral and peripheral proteins are recruited to assemble a competent protein import machinery. At that moment, matrix proteins start to be imported and globular, mature peroxisomes can bud off the lamellar structures. This model explains nicely how peroxisomal membranes are formed and also accommodates previous observations in favour of the division model. However, it remains to be determined how the very first peroxisomal membrane peroxins, necessary for the subsequent insertion of other PMPs in the pre-peroxisomes, are introduced into the ER membrane. Moreover, the results of this study are only based on morphological observations and therefore, it will be crucial to confirm these data by genetic and biochemical analysis.

6. Aspects of glycosome biogenesis

Like peroxisomes in other organisms, no DNA is detectable in trypanosomatid glycosomes. Glycosomal enzymes are synthesized on free polysomes in the cyto-sol, and then transferred into glycosomes [266]. The polypeptides are synthesized at their mature size and do not undergo any detectable proteolytic cleavage or

secondary modification, such as phosphorylation, glyco-sylation or covalent linkage of lipids, upon transfer into the glycosomes. The half-life in the cytosol before being imported into the glycosomes has been measured for various glycolytic enzymes of procyclic T. brucei and was found to be short, between 1.0 and 3.4 min [266,267], indicating that import into glycosomes can be a rapid process. These values are comparable to the fastest translocation rates reported for rat-liver and yeast peroxisomal proteins. The glycosomal enzymes of both bloodstream-form and procyclic T. brucei are very stable; no significant turnover could be measured during a 3-h period [267,268].

6.1. Targeting and import of glycosomal matrix proteins

Most of the glycosomal enzymes analyzed to date have a PTS1, some of them a PTS2, and in a small subset of glycosomal enzymes no consensus motif that might be responsible for protein targeting to the glyco-some could be detected. Table 3 presents the PTS motifs of glycosomal enzymes involved in carbohydrate metabolism in the various life-cycle stages of T. brucei and some Leishmania species.

Analysis of the trypanosomatid genome projects databases revealed the presence of about 250 predicted

Table 3

Glycosome-targeting signals of Trypanosoma brucei and Leishmania spp. enzymes involved in carbohydrate metabolism

Type of signal Enzyme T. brucei Leishmania spp.

PTS1 Extended glycolysis PGI -SHL -AHL

PFK -AKL -SKV

GAPDH -AKL -SKM

PGK-C -SSL -

GPDH -SKM -SKL

GK -AKL -AKL

PEPCK -SRL -AHL

MDH -SKL -SHL

FR -SKI -AKI

PPDK -AKL -SKL

PPP and gluconeogenesis G6PDH - -NKL

6PGL -AKF -

FBPase -SKL -SKL

SBPase -SKL enzyme absent

PTS2 Extended glycolysis HXK -RL(X)5HI- -RL(X)5HI-

ALD -RV(X)5QL- -RV(X)5QL-

PPP and gluconeogenesis 6PGL - -KV(X)5KV-/-KI(X)5HI-

FBPase -RV(X)5QF- -

I-PTS Extended glycolysis TIM amino acids 140 to 161 ?

PGK-A amino acids 24 to 91 enzyme absent

Abbreviations: PPP, pentose-phosphate pathway. Enzymes: ALD, aldolase; FBPase, fructose-1,6-bisphosphatase; FR, fumarate reductase; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GK, glycerol kinase; GPDH, glycerol-3-phosphate dehydrogenase; G6PDH, glucose-6-phoshate dehydrogenase; HXK, hexokinase; TIM, triosephosphate isomerase; PEPCK, phosphoenolpyruvate carboxykinase; 6PGL, 6-phosphogluconol-actonase; PGI, glucose-6-phosphate isomerase; PFK, phosphofructokinase; PGK, phosphoglycerate kinase (isoenzyme A or C); PPDK, pyruvate phosphate dikinase; SBPase, sedoheptulose-1,7-bisphosphatase. -, The consensus motif of the PTS is not present in the available amino-acid sequence; ?, the targeting signal has not yet been identified in the protein.

proteins with a candidate PTS motif (F. Opperdoes, unpublished). Approximately two-third of these proteins have a putative PTS1, one-third a PTS2. For about 50% of these predicted PTS-containing proteins no function is known as yet, but the remainder comprises enzymes of known glycosomal processes (see Section 3): glycolysis, pentose-phosphate pathway, purine salvage, pyrimidine biosynthesis, p-oxidation of fatty acids, fatty-acid elongation, ether-lipid biosynthesis, several reactions of the isoprenoid biosynthetic pathway and proteins involved in protection against oxidant stress.

Detailed studies have been done with stably trans-fected bloodstream-form T. brucei expressing reporter proteins to determine which variations of the C-termi-nal tripeptide (PTS1) can route a reporter protein to glycosomes [47,48,269,270]. These analyses, as well as inspection of the sequences of glycosomal enzymes, show a spectrum of variants of the canonical PTS1, -SKL, that are functional, albeit at different efficacies. The variety of functional PTS1 permutations is markedly larger in T. brucei than observed for mammalian cells [76,79].

Two proteins of glycolysis of trypanosomatids have a PTS2 motif, hexokinase and aldolase. In addition, a few other proteins involved in core carbohydrate metabolism have such a signal (Table 3). Indeed, the PTS2 motif-containing N-terminal sequence of aldolase was capable of directing a reporter protein to glycosomes [271]. The presence of an N-terminal PTS2 in aldolase, instead of a C-terminal PTS1, can be rationalized because a free C-terminal tyrosine residue is essential for the catalytic activity of this enzyme.

For two glycolytic enzymes of T. brucei, the involvement of an I-PTS in glycosome targeting was shown. A minor glycosomal PGK isoenzyme (PGK-A) is targeted to the glycosomes by an internal sequence located between amino acids 24 and 91 [272]. A second glycosomal enzyme from T. brucei, TIM, was found to be imported via a 22 amino-acid internal sequence (de Walque, S., Chevalier, N. and PM, unpublished) (Table 3). A pep-tide comprising residues 140-161 of the 250 amino acids long TIM polypeptide was shown to be sufficient and necessary to direct the reporter protein chloramphenicol acetyl transferase to the glycosomes when fused to its N-terminus. The corresponding peptide of S. cerevisiae TIM could not do so. Indeed, the trypanosomatid (T. brucei, T. cruzi, L. mexicana) TIM peptide has some specific residues (residues only conserved in trypano-somatids and a two-residue insertion) that may be responsible for the targeting. We hypothesize that this peptide, which in the three-dimensional TIM structure contributes to a surface-exposed loop [273], is responsible for interaction with a different protein. This other protein may be a peroxin or a different glycosomal matrix protein enabling TIM import by piggy-backing (see also Section 5.1).

Interestingly, a considerable number of enzyme activities of kinetoplastids have a dual localization; part of the enzyme activity is found in the glycosome, another part in the cytosol (reviewed in [68]). The extent of com-partmentation may differ between the various kineto-plastid organisms and between cells from different life cycle stages. Some activities with a dual localization are represented in each compartment by different iso-forms, encoded by distinct genes (e.g., PGK and GAPDH in T. brucei [274,275]). In most cases studied, however, no differences could be detected between the enzyme present in the glycosomes and cytosol, and the differently located activities had to be attributed to the product encoded by a single gene. Examples are glucose-6-phosphate isomerase, glucose-6-phosphate dehydrogenase and 6-phosphogluconolactonase [276,277] (see also Section 4). How such enzymes are distributed in the kinetoplastid cell in a manner that is physiologically appropriate for each species and its life cycle stage is not clear yet. Several possible mechanisms, separately or acting together, could be considered, such as different efficiencies of the targeting signal or the structural context in which the targeting signal is present in the protein, not yet detected posttranslational mechanisms for selective retention of part of a protein in a compartment, or a kinetic control determined by different rates of biosynthesis, import and turnover of different proteins. These and other possible mechanisms have been discussed in detail by Hannaert et al. [68].

Also for glycosomes, protein unfolding is not a prerequisite for import. One of the first, and most convincing examples that proteins in folded form can enter peroxisomes was reported for T. brucei glycosomes. Hausler et al. [201] employed a methodology previously applied to show that a polypeptide during its transit into mitochondria must be in an unfolded state. Methotrex-ate, a folate analogue, binds with high affinity to dihydrofolate reductase (DHFR), stabilizing its folded state and thereby inhibiting unfolding [278]. DHFR coupled to a mitochondrial signal peptide can be imported into the organelle, but upon addition of methotrexate or aminopterin, a membrane-permeable analogue, the stably folded complex gets stuck in the import channel. When a similar experiment was done in trypanosomes, targeting of the DHFR fusion protein to the mitochondrion was also inhibited. However, import of a fusion protein of DHFR and the C-terminus of gPGK containing a PTS1, into glycosomes was not inhibited upon addition of the same DHFR inhibitors.

6.2. Peroxins involved in glycosome biogenesis

Various homologues of mammalian and yeast perox-in genes have been identified in the genomes of T. brucei, T. cruzi and Leishmania spp (Table 4A). These genes were identified by different methods. Flaspohler et al.

Table 4

Peroxin homologues in Trypanosomatidae

Peroxin Trypanosoma brucei Leishmania spp. Trypanosoma cruzi

Homologue Peroxin identity Essentiality Homologue Peroxin identity Essentiality Homologue

identified confirmed confirmed identified confirmed confirmed identified

A. Identification, confirmation of identity and essentiality of the peroxins for the parasites

PEX1 X X X

PEX2 X X X X X X X

PEX5 X X X X X X X

PEX6 X X X X

PEX7 X X X

PEX10 X X X X

PEX11 X X X X X

PEX11/GIM5A X X X X

PEX11/GIM5B X X X X

PEX12 X X X X

PEX14 X X X X X X

PEX19 X X X

B. Percentages of identity between peroxin amino-acid sequences

Peroxin T.b.-T.c. T.b.-L. spp T.c.-L. spp. T.b. -human T.c. -human L. spp.-human T.b.-S.c. T.c.-S.c. L. spp.-S.c.

PEX1 60 49 49 27 28 27 23 23 23

PEX2 65 51 56 24 26 26 15 18 16

PEX5 64 52 52 26 27 26 22 22 21

PEX6 76 65 68 32 34 33 34 36 34

PEX7 76 65 68 32 34 33 34 36 34

PEX10 65 53 54 27 26 28 25 25 26

PEX11 61 59 62

PEX11/GIM5A 66a 56a 56a 12-17b 12-16b 12-19b 11-13b 12-15b 12-13b

PEX11/GIM5B

PEX12 57 40 43 20 20 18 17 18 13

PEX14 45 36 34 16 15 14 13 17 14

PEX19 36 27 26 19 13 17 17 17 17

References for the trypanosomatid peroxins experimentally analyzed are given in the text. Confirmation of essentiality of peroxins is based on published experiments described in the text and unpublished RNAi experiments (for T. brucei bloodstream-form, PEX6, PEX10 and PEX12; H. Krazy and PM, unpublished). The identification of putative peroxins was done by searches in the GeneDB database of the trypanosomatid genome projects (http://www.geneDB.org/). Leishmania sequences used for the data in Table 4 B are either of L. donovani (PEX2, PEX5, PEX14), if characterized and described in the literature, and otherwise of L. major, as identified in the genome database. Human and S. cerevisiae sequences were taken from the SwissProt database (http:// www.expasy.org/sprot/) or GenBank (http://www.ncbi.nlm.nih.gov/). A multiple alignment of the amino-acid sequences of each of the peroxins of human, S. cerevisiae and the three trypanosomatid species was prepared with the program ClustalX [310] and used to calculate the percentages of identity given in Table 4 B. Abbreviations: GIM, glycosomal integral membrane protein; PEX, peroxin; L. spp, Leishmania species; T.b., T.brucei; T.c., T.cruzi; S.c., S.cerevisiae.

a average of identities of GIM5A and GIM5B homologues in the trypanosomatid species indicated.

b range of the identities between the different trypanosomatid PEX11 isoforms (PEX11, GIM5A and GIM5B) and the three isoforms of human PEX11: PEX11a, b and y.

[279,280] identified a L. donovani PEX2 homologue by a functional screen. Promastigotes were first transfected with a gene coding for a PTS1-containing protein rendering cells resistant to the drug bleomycin. The rationale of the screen is that the cells remain susceptible to bleomycin or phleomycin when the protein is sequestered inside the glycosomes, but become resistant when drug uptake into the organelles is abolished. Upon random mutagenesis, resistant cells were obtained which showed a defect in the import of glycosomal protein. A clone thus obtained (called gim1-1, for 'glycosome import mutant') showed indeed a cytosolic localization of the drug resistance protein and a partial mislocalization of a glycosomal protein involved in purine salvage, hypoxanthine-guanine phosphoribosyltransferase (HG-PRT), a PTS1-protein (-SKV). Other glycosomal enzymes, GAPDH and PGK-C (having a PTS1 and an unknown targeting signal, respectively), remained correctly compartmentalized and glycosome morphology and number were not significantly affected [279]. Strikingly, the mutant cells lacked lipid bodies, which have been postulated to serve as an energy reserve. The gim1-1 mutant gene was identified as a PEX2 homologue. The phenotype was due to the dominant-negative nonsense mutation in one of the two PEX2 alleles of the diploid organism [280], resulting in a protein lacking one transmembrane domain and the RING finger. The same methodology has subsequently yielded two additional mutant clones (called gim2 and gim4) that have partial defects in glycosomal import of HGPRT and TIM, but not GAPDH and PGK-C, but no obvious glyco-somal or other structural changes compared to wildtype cells [281]. The gim2 and gim4 genes remain to be characterized.

The T. brucei PEX2 was identified by homology search of the sequence database [282]. It has 51% amino-acid sequence identity with L. donovani PEX2, and about 15-25% with all other known PEX2 sequences (Table 4B). Northern and Western blotting experiments suggest that PEX2 is expressed at low abundance in try-panosomes, and overexpression seemed not possible. Expression of an additional, truncated PEX2, similar to the Leishmania mutant protein, appeared lethal in bloodstream-form trypanosomes. When PEX2 expression was silenced by RNA interference (RNAi), bloodstream-form trypanosomes died rapidly upon PEX2 depletion. PEX2 knockdown in procyclic forms led to relocation of part of the tested glycosomal matrix proteins (PFK and GAPDH, both PTS1 proteins, and aldo-lase, a PTS2 protein) to the cytosol and eventually also these cells died.

Purified glycosomal membrane fractions of T. brucei contain two predominant protein species of 24 and 26 kDa [283]. Both proteins were purified, subjected to peptide sequencing, and the genes cloned [284,285]. The protein encoded by the 24 kDa protein has low se-

quence identity with PEX11 from other organisms (less than 20%, see Table 4B), but nevertheless was identified as an authentic PEX11 by complementation of a S. cerevisiae Dpex11 mutant which is defective in perox-isome proliferation [284]. Most conservation is found in the N- and C-terminal domains of this integral membrane protein; both domains are oriented towards the cytosol. The protein can form homodimers, like its yeast counterparts. The TbPEX11 gene appeared essential in trypanosomes; by inducible knockout in bloodstream-form and procyclic trypanosomes, the expression could be reduced to 10% of the wild-type level, but not lower. This reduced expression resulted in a decrease in the number of glycosomes, with corresponding increase in size. However, this reduction did initially not result in slowed growth. TbPEX11 overexpression caused growth arrest and transformed the globular glycosomes into clusters of long tubules filling a large part of the cytoplasm. Therefore, it seems that in trypanosomes TbPEX11 plays a role homologous to that of PEX11 in yeast and mammalian cells, i.e., an involvement in peroxisome proliferation (see Section 5.8) [247]. Two related genes, called GIM5A and GIM5B (GIM stands for glycosomal integral membrane protein), encode the 26 kDa glycosomal membrane protein [285]. GIM5A and GIM5B are isoforms of PEX11, albeit with a low level of sequence identity (both 15% compared to T. brucei PEX11). They have all a similar predicted topology in the glycosomal membrane, but unlike trypanosome PEX11, the GIM5 proteins have no nucleotide-binding motif, nor a potential consensus peroxisomal membrane-targeting signal. GIM5 is essential for bloodstream-form T. brucei [285,286]. Deletion of the GIM5A gene from the bloodstream-form parasite has no effect on trypanosome growth, but depletion of GIM5B to below 10% of wild-type levels in a gim5A null background leads to rapid cell death. In contrast, procyclic trypanosomes survive without GIM5A and with very low levels of GIM5B. Attempts to completely delete these PEX11-related genes were unsuccessful, suggesting that they are essential for the survival of the parasite. Dgim5AI GIM5B depleted procyclic cells have a drastically reduced number of glycosomes and are osmotically fragile. The import of aldolase and GAPDH into glycosomes was not significantly affected in these cells, but some leakage of matrix proteins into the cytosol occurred. Other changes noticed in these cells are an altered mitochondrial morphology and a drastically reduced level of ether lipid species. The bloodstream-form Dgim5A cells in which also GIM5B depletion was induced, showed, before GIM5B depletion became critical, a shift in metabolic pattern: no glycerol was produced anymore from glucose, whereas pyruvate production increased. In contrast, glycerol could still be metabolized when given as substrate, indicating that

glycerol uptake and metabolism by glycosomes were not affected.

Overexpression of GIM5B inhibits growth. It was also shown that TbPEX11, GIM5A and GIM5B have non-redundant, but overlapping functions. GIM5A and GIM5B form heterodimers, but at present no other proteins interacting with GIM5 are known. From all data obtained, it was concluded that the GIM5 proteins are required for the maintenance of normal glycosome numbers in T. brucei. The authors proposed that the function of PEX11-family proteins is to determine the overall structure and composition of the peroxisomal membrane [286]. Homologues of L. major GIM5A and GIM5B have also been detected, but not studied in any detail [285].

It should be noted that the PEX11IGIM5 situation in trypanosomatids is reminiscent of that in mammalian cells and S. cerevisiae, where also three PEX11 family members have been detected: the PEX11 isoforms a, p and y and PEX11, PEX25 and PEX27, respectively (see Section 5.8).

The genes of PEX5, the PTS1 receptor, of T. brucei and L. donovani were both cloned after successful PCR amplification of a gene fragment using primers corresponding to regions conserved in the peroxin of mammalian cells and yeasts [287,288]. The overall identity with other PEX5 sequences is low (21-27%), but the (sub)domain organization of the trypanosomid PEX5 homologue is similar to that of the PTS1 receptors of other organisms, with seven copies of the TPR motif in the C-terminal domain, and a N-terminal half containing three copies of the WXXXFIY pentapep-tide motif. The amino-acid sequences of the two try-panosomatid PEX5 sequences are 52% identical. The identity of the T. brucei gene product as a PTS1 receptor was proven by its specific interaction in vitro with glycosomal PGK-C, with an equilibrium dissociation constant (Kd) of 40 nM [287]. A C-terminally truncated PGK-C, without PTS1 did not bind to purified trypanosome PEX5, produced in E. coli. Similarly, the L. donovani protein displayed specific binding of the PTS1 containing xanthine phosphorib-osyltransferase (XPRT), with a Kd of 4.2 nM, but not of XPRT in which the PTS1 had been deleted [288]. Using biochemical, immuno-cytochemical and immuno-fluorescence techniques, a predominantly cytosolic localization was found for both TbPEX5 and LdPEX5, in agreement with findings for the PTS1 receptor in other organisms. The repeated inability to generate a Dpex5 null mutant by targeted gene replacement from Dpex5IPEX5 heterozygotes suggested that the PEX5 is essential for L. donovani promastigotes. In both the bloodstream and procyclic form of T. brucei PEX5 was shown to be essential by applying RNAi (JM, N. Galland and PM, unpublished).

Fig. 9. The overall structure of the T. brucei PEX5 fragment comprising the first three TPR motifs, based on the crystal structure determined by Kumar et al. [289] (Protein Data Bank accession code 1HXI). Note that the third TPR motif does not conform to the standard two-helix TPR motif as shown for TPR1 and 2 (and compare with Fig. 4), but exists as a continuous long helix.

A fragment of the TbPEX5 containing the first three TPR motifs has been crystallized and its structure determined at 1.6 A resolution [289]. The first two of the three TPR motifs appear to adopt the canonical antiparallel helix hairpin structure of TPRs. In contrast, the third TPR motif has a dramatically different conformation: the two helices that usually form a hairpin are folded into a single, long continuous helix never observed before for a TPR motif (Fig. 9). The potential functional importance of this different conformation has not yet been established.

Also the gene of a T. brucei homologue of PEX7, the PTS2 receptor, has been cloned and sequenced (V. Han-naert, JM and PM, unpublished). This was possible after recognition of an EST coding for a fragment with a typical PEX7 sequence in the T. cruzi sequence database. The putative T. brucei PEX7, as well as the corresponding sequence recently revealed in the L. major and T. cruzi databases, have 32-36% amino-acid sequence identity with human and yeast PEX7 (Table 4B), and 40% with A. thaliana PEX7. The T. brucei, T. cruzi and L. major sequences are 65-76% identical. The try-panosomatid sequences have a similar organization as the PEX7 sequences from other organisms: a core of six predicted WD motifs with an N-terminal region of about 60 residues. Surprisingly, they have an approximately 40 residues long extension at the C-terminus, whereas other PEX7s have only 5-10 residues. Furthermore, this C-terminal extension is very rich in proline residues. The identity of these sequences as the trypan-osomatid PTS2 receptors is very likely, but remains to be proven.

Short potential PEX14 fragments were recognized in the databases of T. brucei and L. major, by BLAST searches with yeast and mammalian PEX14 sequences. This led to the cloning and sequencing of T. brucei [290,291] and L. donovani [292] PEX14 homologues. The overall identity with PEX14 sequences of other organisms is only around 15-25%; the T. brucei and

L. donovani sequences share 34% identity. The protein was found to be associated with glycosomes by immunofluorescence microscopy. Indeed, the sequence displays a stretch of 19 consecutive hydrophobic residues extending from residue 148 to 166, with a similar hydrophobic nature as the equivalent region in other PEX14 sequences, and that is presumably responsible for its anchoring to the membrane of the organelle [291]. The N-terminal domain, upstream of this hydrophobic region, was shown to be responsible for specific binding to TbPEX5 and LdPEX5, respectively [291,292], proving the identity of these trypanosomatid proteins as PEX14, a constituent of the receptor docking complex. Moreover, detailed analysis of the interaction of the L. dono-vani protein with the glycosomal membrane indicated that LdPEX14 is tightly associated to the outer side of the membrane rather than being an integral membrane protein. The N-terminal region also has the PXXP box typically for SH3-domain binding. The equivalent region in PEX14's of other organisms has been shown to be responsible for interaction with PEX13 (see Section 5.2).

Depletion of PEX14 by RNAi showed that the protein is essential to both procyclic and bloodstream-form T. brucei [290,291]. RNAi results, in cells of both life-cycle stages, in a partial mislocalization of glycosomal matrix proteins to the cytosol. The mislocalization was observed for different classes of matrix proteins: PTS1 (GAPDH), PTS-2 (aldolase), and interestingly also the I-PTS protein TIM [291]. The partial mislocal-ization was interpreted as the retainment of newly synthesized proteins in the cytosol due to the inability to import matrix proteins by the fraction of growing and dividing glycosomes. The miscompartmentation of both PTS-1 and PTS-2 proteins strongly suggests that also in T. brucei, PEX14 is the point of convergence of both matrix proteins import pathways. To our knowledge, no evidence for the involvement of PEX14 in I-PTS protein import has been reported so far for peroxi-somes of any organism. This observation provides support for the notion that TbTIM enters glycosomes by "piggy backing'', involving the formation of a hetero-meric complex with a PTS-1 or PTS-2 protein (see Section 5.1).

The reduction of growth rate upon induction of RNAi appears in the bloodstream-form cells already after 24 hours, whereas the effect on growth of the pro-cyclic form is seen only after 4 days of treatment. Bloodstream-form trypanosomes are entirely dependent on glycolysis and glycosomes for their ATP supply. As discussed in Section 4, it was concluded from computer simulations of the metabolism that an intact glycosomal membrane and proper enzyme compartmentation are of vital importance for these cells to compensate for a lack of activity regulation of hexokinase and phosphofruc-tokinase by feed-back loops [53]. In contrast, procyclic

cells have a highly branched metabolic network by which the harmful effects of the unregulated kinases may be alleviated. The presence of an active pentosephosphate pathway in procyclics, with its enzymes present not only in the glycosomes but also in the cytosol [277,293], may provide an extra-glycosomal shunt for glucose catabolism that partially overcomes the problems caused by the miscompartmentation of glycolytic enzymes. Thus a mistargeting of glycosomal enzymes could result in a more severe phenotype in bloodstream-form trypanosomes than in procyclics. Similar differences between bloodstream-form and procyclic T. brucei have been observed when PEX2 was targeted by RNAi [282] (see above).

Interestingly, Furuya et al. [290] demonstrated that removal of simple sugars from the medium allowed the procyclic trypanosomes in which PEX14 was depleted to survive. We interpret these data as additional support for the hypothesis that proper compartmenta-tion is particularly important during growth on sugars (as occurs in the bloodstream form) to prevent the unrestrained accumulation of sugar-phosphates that would otherwise result from the lack of regulation of sugar kin-ases, as mentioned above [53]. During growth on other carbon sources such as amino acids, trypanosomes do not rely on glycosomes for catabolism and, consequently, lesser harm may occur to the cell.

Several other potential peroxins have been identified in the T. brucei, T. cruzi and L. major genome databases by Blast searches with mammalian or yeast sequences: PEX1, PEX6, PEX10, PEX12 and PEX19 (Table 4). Like the other peroxins, their amino-acid sequences display low identities with that of their human homologues, in the range of 15-35%. The putative T. brucei PEX6, PEX10 and PEX12 genes have been cloned and expressed in E. coli as recombinant proteins (H. Krazy and PM, unpublished). The predominant association of the PEX6 and PEX10 homologues with glycosomes has been confirmed by cell fractionation through differential centrifugation followed by immunoblot analysis. RNAi was used to show that the putative PEX6, PEX10 and PEX12 are essential proteins in both bloodstream-form and procyclic T. brucei.

6.3. Protein interactions in glycosome biogenesis and possibilities for drug design

As described in Section 5, the biogenesis of peroxi-somes and glycosomes occurs via a multi-step process involving a large number of proteins that form, transiently or permanently, specific interactions with each other. Most of these steps require the formation of complexes of several proteins. Because of the vital importance of proper glycosome biogenesis to the parasite (see Section 4), and the low sequence identities between the peroxins of the trypanosomatids and their human

Fig. 10. The PTS2 dimer of T. brucei aldolase. Panels a and b show, from two different angles, the aldolase surface comprising the PTS2 dimer formed by the intertwined N-termini of subunits A and B. Subunits A and B are coloured green and blue, respectively, and their N-terminal residues comprising the PTS2 (SKRVEVLLTQLPAY-; residues in bold correspond to the consensus PTS2 motif) are labeled. The numbers indicate the position of each residue in the polypepide (starting from the N-terminal Ser). The flanking subunits C and D of the tetrameric aldolase are coloured in purple. Notice the deep grooves between the PTS2 dimer and the subunits C and D, and which might be targets for drug design, by exploiting structural features of both the PTS2 dimer and flanking subunits C and D, a combination that is unique to the aldolase of the parasite. Such drugs would prevent interaction of trypanosome aldolase with PEX7, either directly by steric occlusion, or by allosteric inhibition if a change of PTS2 conformation is needed for its recognition by PEX7. The T. brucei aldolase crystal structure was determined by Chudzik et al. [296]; the coordinates are available from the Protein Data Bank via accession code 1F27.

host (Table 4B), these peroxins may be considered as highly promising anti-parasite drug targets. Targeting peroxins with compounds that should interfere with the function of these proteins in glycosome biogenesis implies targeting in a selective way the interactions between the trypanosome peroxins, without disrupting the interactions between the corresponding human peroxins. Although this research is still in a preliminary stage, important information about the interactions of some trypanosomatid proteins involved in glycosome biogenesis has already been obtained, providing clues for strategies of future design of selective inhibitors.

6.3.1. PTS1 protein-PEX5 interaction

No crystal structure is as yet available for the complete TPR domain, responsible for PTS1 binding, of a trypanosomatid PEX5. However, the structure of a fragment of human PEX5 that includes all seven TPR motifs in complex with a pentapeptide containing a PTS1 sequence has been reported (Fig. 4) [96]. This structure, in conjunction with binding studies with fluorescent peptides bearing PTS1 variants [80], revealed the molecular basis for PTS1 recognition. The low percentage of identity between the human and trypanosomatid PEX5 (36% for the TPR domain) and the markedly larger variety of

functional PTS1 permutations observed for T. brucei compared to mammalian cells (see Section 6.1) suggest that the PTS1 binding groove of the parasites' receptors may be targeted by selective inhibitors. The finding by Gatto et al. [80] that a binding energy threshold may determine the functionality of PTS1 sequences made these authors suggest that inhibitors should be designed that interfere with the binding to the trypanosomatid PEX5 but do not cross the functional threshold in humans.

6.3.2. PTS2 protein-PEX7 interaction

Very little is known about the mode of interaction of PTS2 with PEX7. This analysis is hampered by the apparent difficulty to overexpress and purify PEX7 in soluble form to perform in vitro binding studies. PEX7 belongs to the family of WD-repeat proteins, which are known to be involved in interactions with other proteins. Crystal structures of two WD-repeat proteins have been reported - the p subunit of the G-protein involved in signaling in mammalian cells [294] and an actin-interact-ing protein from S. cerevisiae [295] - and shown to have a circularized seven-bladed p-propeller structure. All WD-repeat proteins are expected to have such a p-propeller structure, but defining the binding domains remains a challenge. Although no structure is yet known of the PTS2 receptor, crystallographic analysis of T. brucei and L. mexicana aldolase has provided information about the PTS2-structure. Aldolase is a tetrameric enzyme. In both trypanosomatid enzymes, the PTS2s from two different subunits form two closely intertwined structures (Fig. 10) [296]. The finding or this specific, ordered structure offers promise for the design of trypanocidal drugs. Any compound binding tightly to this structure would most likely interfere with PTS2-PEX7 binding and thus prevent aldolase from entering the glycosome. Human aldolase is a cytosolic enzyme, not containing a PTS2. Selectivity of an inhibitor, to prevent interference with the import of any host PTS2 protein could be obtained by designing compounds that also recognize features specific of the trypanosomatid aldolase. Indeed, flanking the PTS2 dimer is a groove at the aldolase surface containing a number of unique residues that could be exploited (Fig. 10). A similar approach is possibly feasible to inhibit the import of other PTS2-containing enzymes into the glycosomes.

6.3.3. PEX5-PEX14 interaction

The interaction of trypanosomatid (T. brucei and L. donovani) PEX5 and PEX14 has been studied in considerable detail and compared with that of the corresponding human and yeast peroxins [291,292,297]. Using purified recombinant forms of PEX5 and PEX14 specific interactions were shown in vitro, using different techniques, both for the Trypanosoma and Leishmania proteins. A Kd of 2.75 iM was determined

for the L. donovani peroxins [292]. When the LdPEX5 was preloaded with a PTS1 peptide, the affinity increased to 0.40 iM. The N-terminal halves of both peroxins were shown to be responsible for the interaction. The binding sites, and the strength of the interactions between the two peroxins were characterized in detail in T. brucei by a fluorescence assay, using a panel of N-terminal regions of TbPEX14 variants and a series of different peptides derived from TbPEX5 [297]. On the PEX5 side, WXXXF/Y motifs in the N-terminal domain were shown to be responsible for PEX14 binding, similar to earlier findings in human and yeast cells. However, only two of these pentapeptide motifs showed affinity for PEX14, the second motif peptide did not. Studies using different variants of the peptides suggested that the propensity to form an a-helix, as well as the nature of residues outside the pentapeptide, notably the absence of a negative charge at the 9th residue counting from the conserved tryptophan, are important for binding. Indeed, peptides with a length of 13 or 16 residues show comparable affinity, in the low nanomolar range, whereas the affinity of shorter peptides (10- and 7-mers) for PEX14 is considerably (10- to 100-fold) reduced. The non-binding second motif has a low, predicted, tendency to form an a-helical conformation and contains a negative charge on the ninth residue (E instead of Q in motifs 1 and 3). Interestingly, WXXXF/Y motifs with a negatively charged residue at the ninth position have been shown to be responsible for PEX5 interaction with PEX13, another component of the docking complex, in P. pastoris, S. cerevisiae and Chinese hamster [102,103,156] (see Section 5.2). On the T. brucei PEX14 side, the region for binding was specified within the N-terminal 84 residues. Site-directed mutagenesis studies identified a number of specific residues, most of them hydropho-bic and conserved in all known PEX14 sequences, which seem important for binding. Similar findings have been reported for A. thaliana [137]. Particularly two phenylalanines (F35 and F52 in T. brucei) are critical. The results together suggest that hydrophobic interactions are important for PEX5-PEX14 interaction. The likely a-helical conformation of the PEX5 peptides containing the WXXXF/Y motif implies that the Trp and Phe/Tyr residues are facing the same side. They may interact with the hydrophobic residues in PEX14. Other residues in the vicinity also play an important role in determining the affinity. Interestingly, not only the overall identities of PEX5 and PEX14 between parasite and its human host are low, but there exists also considerable variation in the residues around the critical residues involved in transient PEX5:PEX14 interactions. This may have great potential for the design of selective, peptidomi-metic inhibitors.

6.3.4. Putative interactions of I-PTS proteins

As described in Sections 5.1 and 6.2, there is ample support for the notion that some enzymes enter the per-oxisomalIglycosomal matrix using a piggy-backing strategy, by forming a heteromeric complex with a PTS1 or PTS2 containing protein. Such a mechanism opens additional possibilities to design compounds that may selectively interfere with the biogenesis of the organelle of the parasite, because it is highly unlikely that the areas responsible for interaction of both proteins are conserved in host and parasite. For example, the 22 amino-acid sequence, present on the surface of T. brucei TIM and that was identified as responsible for the targeting of the protein to the glycosome (see Section 6.1), is very different from the corresponding region in the enzyme of non-trypanosomatids. This sequence has also not been found in any other glycosomal protein. Presumably, it interacts with an equally unique area of another glycosomal matrix protein. Therefore, it may be feasible to develop a compound with specificity to prevent or disrupt this interaction and thus to prevent glycosome import of TIM. Similar approaches may be feasible to prevent import of other I-PTS proteins.

7. Feasibility to target transient protein-protein interactions

Designing inhibitors that will interfere with glyco-some biogenesis involves other challenges than designing inhibitors of (glycosomal or cytosolic) enzymes. Many enzyme inhibitors are designed or selected to bind in pockets, the binding sites for natural ligands, of the proteins where they exploit the enlarged surface with its specific conformation and presence of reactive residues. This enables the development of inhibitors that interact with very high affinity or irreversibly, and with high specificity to the target. In contrast, compounds meant to disrupt glycosome biogenesis will target protein-protein interactions. For some time, the design of drugs that act through disrupting protein-protein interactions has been a controversial issue in pharmaceutics. The fact that small molecules should insert themselves between proteins in which large, complementary surfaces are responsible for interactions often with high affinities was considered as a formidable task. Usually, the interfaces are relatively flat, thus not providing a defined binding pocket for an antagonistic small molecule. However, drugs that target protein-protein interactions have already been in use for many years. For example vinca alkaloids, used in treatment of cancer, have been identified in a natural screening program about 50 years ago. But only in 1996 it was shown that vinblastin acts by interfering with the polymerization of the a— p het-erodimer of tubulin and consequently with the forma-

tion of the mitotic spindle. In the last decade, considerable progress has been made in developing antagonists of a variety of inter-protein interactions. Ribonucletide reductase antagonists have been obtained which prevent assembly of the two components of Herpes simplex virus enzyme with a Ki of 0.3 nM [298]. Another interesting example involves the trisoxazole macrolide toxins which target actins with high affinity and specificity and have promising pharmacological properties. Recently, the binding site of these compounds to actin has been revealed; the results suggested that these macrolides may be viewed as small molecule biomimetics of actin-binding proteins [299].

A number of small molecule antagonists of proteinprotein interactions have reached the market or are in various stages of trials for use against all kinds of different diseases, including cancers, vascular and coronary diseases and viral infections (cited in [300]). All these cases do not involve tightly bound proteins as found in stable homodimeric proteins, but heteromeric complexes of which the units are in dynamic equilibrium. It appears to be perfectly possible to disrupt such dynamic protein complexes with small molecules, by pepti-domimics or inhibitors (reviewed in [300,301]). The weaker such critical protein-protein interactions are, and the more convoluted the interface, the larger the chance that successful small molecule inhibitors will be developed.

The process of peroxisome and glycosome biogenesis involves transient interactions of a large number of different proteins. In view of the success achieved in developing antagonists of transient protein-protein interactions as drugs andIor lead compounds for other diseases, success with a similar approach for developing anti-trypanosome drugs seems realistic.

8. Conclusions

A major part of the core carbohydrate metabolism, most notably glycolysis, of Trypanosomatidae occurs in glycosomes, remarkable trypanosomatid organelles related to peroxisomes. It has been shown that the correct compartmentalization of glycosomal enzymes involved in glycolysis in bloodstream-form T. brucei and the integrity of the organelles are essential to the parasites. Therefore, compounds that would interfere with glycosome biogenesis would potentially be effective drugs.

In the last two decades, much research has been performed to unravel the routes of peroxisome biogenesis. At present, 32 peroxins involved in different aspects of the organelle's formation have been identified and it has been shown that many of these proteins exert their function by transient interactions with one or multiple other peroxins. These studies have provided a general

picture of peroxisome biogenesis, conserved among euk-aryotes, but at the same time also revealed significant differences between organisms such as mammals and yeasts. However, almost all details of the molecular mechanisms of processes such as matrix protein import, formation of the lipid membrane, insertion of proteins in the membrane and energetics of the organelle formation are currently still poorly understood.

Biogenesis of trypanosomatid glycosomes appears to be, in general, similar to that of peroxisomes. Several peroxins have been identified in different human-pathogenic Trypanosomatidae; they are homologous to their counterparts in mammals, yeasts and plants, however with very low percentages of amino-acid sequence identity. The presently available data on the assembly of the different members of the peroxisome family and the sequence and properties of peroxins do not permit us to conclude that glycosome biogenesis more closely resembles the corresponding process in yeasts, plants or mammals.

For various trypanosomatid peroxins it has been shown that decreasing their expression by reverse genetics methodologies leads to growth arrest and death of the parasites. The fact that the peroxins responsible for glycosome biogenesis are essential for the viability of trypanosomes, and the low degree of sequence conservation of peroxins, render these proteins promising targets for the design of anti-parasite drugs that act through preventing the formation of functional organelles and proper metabolic compartmentation. Current research is devoted to determining the transient interactions between different sets of proteins involved in glyco-some biogenesis at the atomic and molecular level. This will provide a solid basis for future design of peptidomi-metic and other compounds that will selectively interfere with these interactions without affecting protein-protein interactions in the human host.

Acknowledgements

The research on glycosome biogenesis and anti-trypanosomatid drug design in the laboratories of the authors is financially supported through the following grants: to PM from the Belgian 'Fonds de la Recherche Scientifique Medicale' (FRSM) and Interuniversity Attraction Poles - Federal Office for Scientific Technical and Cultural Affairs, to EF from the W.M. Keck Foundation Center for Microbial Pathogens at the University of Washington. JM acknowledges a PhD scholarship from the 'Fonds a la Recherche dans l'Industrie et l'Agriculture (FRIA, Belgium). We are grateful to our cowork-ers Veronique Hannaert, Hanane Krazy, Nathalie Galland, Cedric Yernaux, Nathalie Chevalier and Stephane de Walque (ICP, Brussels) for their contribution to the work cited in this review and sharing unpublished

data. We would like to thank Claudia Roach, Stephen Suresh, Abhinav Kumar and David Chudzik (UoW, Seattle) for their contributions to structural biology of trypanosomatid glycosomal proteins. We thank Dr Marc Fransen ('Katholieke Universiteit Leuven', Belgium) for stimulating discussions and comments on preliminary versions of the manuscript and Nathalie Chevalier for assistance in the preparation of figures and tables.

References

[1] Dacks, J.B. and Doolittle, W.F. (2001) Reconstructing/deconstructing the earliest eukaryotes: how comparative genomics can help. Cell 107, 419-425.

[2] Stechmann, A. and Cavalier-Smith, T. (2002) Rooting the eukaryote tree by using a derived gene fusion. Science 297, 8991.

[3] Vickerman, K. (1994) The evolutionary expansion of the trypanosomatid flagellates. Int. J. Parasitol. 24, 1317-1331.

[4] De Souza, W. (2002) Basic cell biology of Trypanosoma cruzi. Curr. Pharm. Des. 8, 269-285.

[5] Opperdoes, F.R. and Borst, P. (1977) Localization of nine glycolytic enzymes in a microbody-like organelle in Trypanosoma brucei: the glycosome. FEBS Lett. 80, 360-364.

[6] Opperdoes, F.R. (1987) Compartmentation of carbohydrate metabolism in trypanosomes. Annu. Rev. Microbiol. 41, 127151.

[7] Hannaert, V. and Michels, P.A.M. (1994) Structure, function, and biogenesis of glycosomes in Kinetoplastida. J. Bioenerg. Biomembr. 26, 205-212.

[8] Michels, P.A.M., Hannaert, V. and Bringaud, F. (2000) Metabolic aspects of glycosomes in Trypanosomatidae - new data and views. Parasitol. Today 16, 482-489.

[9] Hannaert, V., Bringaud, F., Opperdoes, F.R. and Michels, P.A.M. (2003) Evolution of energy metabolism and its compart-mentation in Kinetoplastida. Kinetoplastid Biology and Disease (BioMed Central) 2 (11), 1-30 http://www.kinetoplastids.com/ content/2/1/11 .

[10] Barrett, M.P., Burchmore, R.J.S., Stich, A., Lazzari, J.O., Frasch, A.C., Cazzulo, J.J. and Krishna, S. (2003) The trypan-osomiases. Lancet 362, 1469-1480.

[11] Fairlamb, A.H. (2003) Chemotherapy of human African tryp-anosomiasis: current and future prospects. Trends Parasitol. 19, 488-494.

[12] Stich, A., Barrett, M.P. and Krishna, S. (2003) Waking up to sleeping sickness. Trends Parasitol. 19, 195-197.

[13] Urbina, J.A. and Docampo, R. (2003) Specific chemotherapy of Chagas disease: controversies and advances. Trends Parasitol. 19, 495-501.

[14] Croft, S.L. and Coombs, G.H. (2003) Leishmaniasis - current chemotherapy and recent advances in the search for novel drugs. Trends Parasitol. 19, 502-508.

[15] Barry, J.D. and McCulloch, R. (2001) Antigenic variation in trypanosomes: enhanced phenotypic variation in a eukaryotic parasite. Adv. Parasitol. 49, 1-70.

[16] Sacks, D. and Sher, A. (2002) Evasion of innate immunity by parasitic protozoa. Nat. Immunol. 3, 1041-1047.

[17] Barrett, M.P., Mottram, J.C. and Coombs, G.H. (1999) Recent advances in identifying and validating drug targets in trypano-somes and leishmanias. Trends Microbiol. 7, 82-88.

[18] Werbovetz, K.A. (2000) Target-based drug discovery for malaria, leishmaniasis, and trypanosomiasis. Curr. Med. Chem. 7, 835-860.

[19] Verlinde, C.L.M.J., Hannaert, V., Blonski, C., Willson, M., Perie, J.J., Fothergill-Gilmore, L.A., Opperdoes, F.R., Gelb, M.H., Hol, W.G.J. and Michels, P.A.M. (2001) Glycolysis as a target for the design of new anti-trypanosome drugs. Drug Resist. Updat. 4, 50-65.

[20] Michels, P.A.M. (1988) Compartmentation of glycolysis in trypanosomes: a potential target for new trypanocidal drugs. Biol. Cell 64, 157-164.

[21] Wang, C.C. (1995) Molecular mechanisms and therapeutic approaches to the treatment of African trypanosomiasis. Annu. Rev. Pharmacol. Toxicol. 35, 93-127.

[22] Bakker, B.M., Westerhoff, H.V., Opperdoes, F.R. and Michels, P.A.M. (2000) Metabolic control analysis of glycolysis in trypanosomes as an approach to improve selectivity and effectiveness of drugs. Mol. Biochem. Parasitol. 106, 1-10.

[23] Opperdoes, F.R. and Michels, P.A.M. (2001) Enzymes of carbohydrate metabolism as potential drug targets. Int. J. Parasitol. 31, 482-490.

[24] Seyfang, A. and Duszenko, M. (1991) Specificity of glucose transport in Trypanosoma brucei. Effective inhibition by phlore-tin and cytochalasin B. Eur. J. Biochem. 202, 191-196.

[25] Duszenko, M. and Mecke, D. (1986) Inhibition of glyceralde-hyde-3-phosphate dehydrogenase by pentalenolactone in Trypanosoma brucei. Mol. Biochem. Parasitol. 19, 223-229.

[26] Wiemer, E.A.C., Michels, P.A.M. and Opperdoes, F.R. (1995) The inhibition of pyruvate transport across the plasma membrane of the bloodstream form of Trypanosoma brucei and its metabolic implications. Biochem. J. 312, 479-484.

[27] Aronov, A.M., Verlinde, C.L.M.J., Hol, W.G.J. and Gelb, M.H. (1998) Selective tight binding inhibitors of trypanosomal glyc-eraldehyde-3-phosphate dehydrogenase via structure-based drug design. J. Med. Chem. 41, 4790-4799.

[28] Aronov, A.M., Suresh, S., Buckner, F.S., Van Voorhis, W.C., Verlinde, C.L.M.J., Opperdoes, F.R., Hol, W.G.J. and Gelb, M.H. (1999) Structure-based design of submicromolar, biologically active inhibitors of trypanosomatid glyceraldehyde-3-phosphate dehydrogenase. Proc. Natl. Acad. Sci. USA 96, 4273-4278.

[29] Hart, D.T., Vickerman, K. and Coombs, G.H. (1981) Respiration of Leishmania mexicana amastigotes and promastigotes. Mol. Biochem. Parasitol. 4, 39-51.

[30] Meade, J.C., Glaser, T.A., Bonventre, P.F. and Mukkada, A.J.

(1984) Enzymes of carbohydrate metabolism in Leishmania donovani amastigotes. J. Protozool. 31, 156-161.

[31] Mottram, J.C. and Coombs, G.H. (1985) Leishmania mexicana: enzyme activities of amastigotes and promastigotes and their inhibition by antimonials and arsenicals. Exp. Parasitol. 59, 151160.

[32] Mottram, J.C. and Coombs, G.H. (1985) Leishmania mexicana: subcellular distribution of enzymes in amastigotes and promas-tigotes. Exp. Parasitol. 59, 265-274.

[33] Burchmore, R.J. and Barrett, M.P. (2001) Life in vacuoles -nutrient acquisition by Leishmania amastigotes. Int. J. Parasitol. 31, 1311-1320.

[34] Coombs, G.H., Craft, J.A. and Hart, D.T. (1982) A comparative study of Leishmania mexicana amastigotes and promastigotes. Enzyme activities and subcellular locations. Mol. Biochem. Parasitol. 5, 199-211.

[35] Mukkada, A.J., Meade, J.C., Glaser, T.A. and Bonventre, P.F.

(1985) Enhanced metabolism of Leishmania donovani amasti-gotes at acid pH: an adaptation for intracellular growth. Science 229, 1099-1101.

[36] Burchmore, R.J., Rodriguez-Contreras, D., McBride, K., Merkel, P., Barrett, M.P., Modi, G., Sacks, D. and Landfear, S.M. (2003) Genetic characterization of glucose transporter function in Leishmania mexicana. Proc. Natl. Acad. Sci. USA 100, 39013906.

[37] Engel, J.C., Franke de Cazzulo, B.M., Stoppani, A.O., Cannata, J.J. and Cazzulo, J.J. (1987) Aerobic glucose fermentation by Trypanosoma cruzi axenic culture amastigote-like forms during growth and differentiation to epimastigotes. Mol. Biochem. Parasitol. 26, 1-10.

[38] Cazzulo, J.J. (1992) Aerobic fermentation of glucose by trypan-osomatids. FASEB J. 6, 3153-3161.

[39] Tielens, A.G.M. and Van Hellemond, J.J. (1998) Differences in energy metabolism between Trypanosomatidae. Parasitol. Today 14, 265-271.

[40] Bakker, B.M., Michels, P.A.M., Opperdoes, F.R. and Westerhoff, H.V. (1999) What controls glycolysis in bloodstream form Trypanosoma brucei!. J. Biol. Chem. 274, 14551-14559.

[41] Clayton, C., Hausler, T. and Blattner, J. (1995) Protein trafficking in kinetoplastid protozoa. Microbiol. Rev. 59, 325344.

[42] Parsons, M., Furuya, T., Pal, S. and Kessler, P. (2001) Biogenesis and function of peroxisomes and glycosomes. Mol. Biochem. Parasitol. 115, 19-28.

[43] Hannaert, V., Saavedra, E., Duffieux, F., Szikora, J.P., Rigden, D.J., Michels, P.A.M. and Opperdoes, F.R. (2003) Plant-like traits associated with metabolism of Trypanosoma parasites. Proc. Natl. Acad. Sci. USA 100, 1067-1071.

[44] Waller, R.F., McConville, M.J. and McFadden, G.I. (2004) More plastids in human parasites?. Trends Parasitol. 20, 54-57.

[45] de Duve, C. (1982) Peroxisomes and related particles in historical perspective. Ann. N. Y. Acad. Sci. 386, 1-4.

[46] Borst, P. (1989) Peroxisome biogenesis revisited. Biochim. Biophys. Acta 1008, 1-13.

[47] Fung, K. and Clayton, C. (1991) Recognition of a peroxisomal tripeptide entry signal by the glycosomes of Trypanosoma brucei. Mol. Biochem. Parasitol. 45, 261-264.

[48] Sommer, J.M., Peterson, G., Keller, G.A., Parsons, M. and Wang, C.C. (1993) The C-terminal tripeptide of glycosomal phosphoglycerate kinase is both necessary and sufficient for import into the glycosomes of Trypanosoma brucei. FEBS Lett. 316, 53-58.

[49] Blattner, J., Helfert, S., Michels, P. and Clayton, C. (1998) Compartmentation of phosphoglycerate kinase in Trypanosoma brucei plays a critical role in parasite energy metabolism. Proc. Natl. Acad. Sci. USA 95, 11596-11600.

[50] Helfert, S., Estevez, A.M., Bakker, B., Michels, P. and Clayton, C. (2001) Roles of triosephosphate isomerase and aerobic metabolism in Trypanosoma brucei. Biochem. J. 357, 117-125.

[51] Michels, P.A.M. and Hannaert, V. (1994) The evolution of kinetoplastid glycosomes. J. Bioenerg. Biomembr. 26, 213-219.

[52] Bakker, B.M., Westerhoff, H.V. and Michels, P.A.M. (1995) Regulation and control of compartmentalized glycolysis in bloodstream form Trypanosoma brucei. J. Bioenerg. Biomembr. 27, 513-525.

[53] Bakker, B.M., Mensonides, F.I.C., Teusink, B., Van Hoek, P., Michels, P.A.M. and Westerhoff, H.V. (2000) Compartmenta-tion protects trypanosomes from the dangerous design of glycolysis. Proc. Natl. Acad. Sci. USA 97, 2087-2092.

[54] Visser, N., Opperdoes, F.R. and Borst, P. (1981) Subcellular compartmentation of glycolytic intermediates in Trypanosoma brucei. Eur. J. Biochem. 118, 521-526.

[55] Eisenthal, R. and Panes, A. (1985) The aerobic/anaerobic transition of glucose metabolism in Trypanosoma brucei. FEBS Lett. 181, 23-27.

[56] Wanders, R.J.A. and Tager, J.M. (1998) Lipid metabolism in relation to human disease. Chapter 7, Permeability properties of peroxisomes. Molec. Aspects Med. 19, 113-120.

[57] Van Roermund, C.W., Elgersma, Y., Singh, N., Wanders, R.J.A. and Tabak, H.F. (1995) The membrane of peroxisomes in Saccharomyces cerevisiae is impermeable to NAD(H) and acetyl-CoA under in vivo conditions. EMBO J. 14, 3480-3486.

[58] Hettema, E.H. and Tabak, H.F. (2000) Transport of fatty acids and metabolites across the peroxisomal membrane. Biochim. Biophys. Acta 1486, 18-27.

[59] Hettema, E.H., Van Roermund, C.W., Distel, B., Van den Berg, M., Vilela, C., Rodrigues-Pousada, C., Wanders, R.J.A. and Tabak, H.F. (1996) The ABC transporter proteins Patl and Pat2 are required for import of long-chain fatty acids into peroxisomes of Saccharomyces cerevisiae. EMBO J. 15, 38133822.

[60] Reumann, S. (2000) The structural properties of plant peroxisomes and their metabolic significance. Biol. Chem. 381, 639648.

[61] Nwagwu, M. and Opperdoes, F.R. (1982) Regulation of glycolysis in Trypanosoma brucei: hexokinase and phosphofruc-tokinase activity. Acta Trop. 39, 61-72.

[62] Caceres, A.J., Portillo, R., Acosta, H., Rosales, D., Quifiones, W., Avilan, L., Salazar, L., Dubourdieu, M., Michels, P.A.M. and Conception, J.L. (2003) Molecular and biochemical characterization of hexokinase from Trypanosoma cruzi. Mol. Biochem. Parasitol. 126, 251-262.

[63] Cronin, C.N. and Tipton, K.F. (1985) Purification and regulatory properties of phosphofructokinase from Trypanosoma (Trypanozoon) brucei brucei. Biochem. J. 227, 113-124.

[64] Lopez, C., Chevalier, N., Hannaert, V., Rigden, D.J., Michels, P.A.M. and Ramirez, J.L. (2002) Leishmania donovani phospho-fructokinase. Gene characterization, biochemical properties and structure-modeling studies. Eur. J. Biochem. 269, 3978-3989.

[65] Teusink, B., Walsh, M.C., Van Dam, K. and Westerhoff, H.V.

(1998) The danger of metabolic pathways with turbo design. Trends Biochem. Sci. 23, 162-169.

[66] Bakker, B.M., Michels, P.A.M., Opperdoes, F.R. and Westerhoff, H.V. (1997) Glycolysis in bloodstream form Trypanosoma brucei can be understood in terms of the kinetics of the glycolytic enzymes. J. Biol. Chem. 272, 3207-3215.

[67] Visser, N. and Opperdoes, F.R. (1980) Glycolysis in Trypanosoma brucei. Eur. J. Biochem. 103, 623-632.

[68] Hannaert, V., Adje, C.A., de Walque, S. and Michels, P.A.M.

(1999) Compartmentation of glycolytic enzymes in Kinetoplast-ida. Res. Rev. Parasitol. 59, 85-100.

[69] Purdue, P.E. and Lazarow, P.B. (2001) Peroxisome biogenesis. Annu. Rev. Cell. Dev. Biol. 17, 701-752.

[70] Eckert, J.H. and Erdmann, R. (2003) Peroxisome biogenesis. Rev. Physiol. Biochem. Pharmacol. 147, 75-121.

[71] Distel, B., Erdmann, R., Gould, S.J., Blobel, G., Crane, D.I., Cregg, J.M., Dodt, G., Fujiki, Y., Goodman, J.M., Just, W.W., Kiel, J.A., Kunau, W.H., Lazarow, P.B., Mannaerts, G.P., Moser, H.W., Osumi, T., Rachubinski, R.A., Roscher, A., Subramani, S., Tabak, H.F., Tsukamoto, T., Valle, D., Van der Klei, I., Van Veldhoven, P.P. and Veenhuis, M. (1996) A unified nomenclature for peroxisome biogenesis factors. J. Cell Biol. 135, 1-3.

[72] Erdmann, R., Veenhuis, M., Mertens, D. and Kunau, W.H. (1989) Isolation of peroxisome-deficient mutants of Saccharom-yces cerevisiae. Proc. Natl. Acad. Sci. USA 86, 5419-5423.

[73] Veenhuis, M., Mateblowski, M., Kunau, W.H. and Harder, W. (1987) Proliferation of microbodies in Saccharomyces cerevisiae. Yeast 3, 77-84.

[74] Tsukamoto, T., Yokota, S. and Fujiki, Y. (1990) Isolation and characterization of Chinese hamster ovary cell mutants defective in assembly of peroxisomes. J. Cell Biol. 110, 651-660.

[75] Gould, S.J., Keller, G.A., Hosken, N., Wilkinson, J. and Subramani, S. (1989) A conserved tripeptide sorts proteins to peroxisomes. J. Cell Biol. 108, 1657-1664.

[76] Swinkels, B.W., Gould, S.J. and Subramani, S. (1992) Targeting efficiencies of various permutations of the consensus C-terminal tripeptide peroxisomal targeting signal. FEBS Lett. 305, 133136.

[77] Elgersma, Y., Vos, A., Van den Berg, M., Van Roermund, C.W.T., Van der Sluijs, P., Distel, B. and Tabak, H.F. (1996) Analysis of the carboxyl-terminal peroxisomal targeting signal 1 in a homologous context in Saccharomyces cerevisiae. J Biol. Chem. 271, 26375-26382.

[78] Lametschwandtner, G., Brocard, C., Fransen, M., Van Veldhoven, P., Berger, J. and Hartig, A. (1998) The difference in recognition of terminal tripeptides as peroxisomal targeting signal 1 between yeast and human is due to different affinities of their receptor Pex5p to the cognate signal and to residues adjacent to it. J. Biol. Chem. 273, 33635-33643.

[79] Neuberger, G., Maurer-Stroh, S., Eisenhaber, B., Hartig, A. and Eisenhaber, F. (2003) Motif refinement of the peroxisomal targeting signal 1 and evaluation of taxon-specific differences. J. Mol. Biol. 328, 567-579.

[80] Gatto Jr., G.J., Maynard, E.L., Guerrerio, A.L., Geisbrecht, B.V., Gould, S.J. and Berg, J.M. (2003) Correlating structure and affinity for PEX5:PTS1 complexes. Biochemistry 42, 16601666.

[81] McCollum, D., Monosov, S. and Subramini, S. (1993) The pas8 mutant of Pichia pastoris exhibits the peroxisomal protein import deficiencies of Zellweger syndrome cells - the PAS8 protein binds to the COOH-terminal tripeptide peroxisomal targeting signal, and is a member of the TPR protein family. J. Cell Biol. 121, 761-774.

[82] Van der Leij, I., Franse, M.M., Elgersma, Y., Distel, B. and Tabak, H.F. (1993) PAS10 is a tetratricopeptide-repeat protein that is essential for the import of most matrix proteins into peroxisomes of Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 90, 11782-11786.

[83] Dodt, G., Braverman, N., Wong, C., Moser, A., Moser, H.W., Watkins, P., Valle, D. and Gould, S.J. (1995) Mutations in the PTS1 receptor gene, PXR1, define complementation group 2 of the peroxisome biogenesis disorders. Nat. Genet. 9, 115-125.

[84] Fransen, M., Brees, C., Baumgart, E., Vanhooren, J.C.T., Baes, M., Mannaerts, G.P. and Van Veldhoven, P.P. (1995) Identification and characterization of the putative human peroxisomal C-terminal targeting signal import receptor. J. Biol. Chem. 270, 7731-7736.

[85] Nuttley, W.M., Szilard, R.K., Smith, J.J., Veenhuis, M. and Rachubinski, R.A. (1995) The PAH2 gene is required for peroxisome sssembly in the methylotrophic yeast Hansenula polymorpha and encodes a member of the tetratricopeptide repeat family of proteins. Gene 160, 33-39.

[86] Szilard, R.K., Titorenko, V.I., Veenhuis, M. and Rachubinski, R.A. (1995) Pay32p of the yeast Yarrowia lipolytica is an intraperoxisomal component of the matrix protein translocation machinery. J. Cell Biol. 131, 1453-1469.

[87] Van der Klei, I.J., Hilbrands, R.E., Swaving, G.J., Waterham, H.R., Vrieling, E.G., Titorenko, V.I., Cregg, J.M., Harder, W. and Veenhuis, M. (1995) The Hansenula polymorpha PER3 gene is essential for the import of PTS1 proteins into the peroxisomal matrix. J. Biol. Chem. 270, 17229-17236.

[88] Wiemer, E.A.C., Nuttley, W.M., Bertolaet, B.L., Li, X., Francke, U., Wheelock, M.J., Anne, U.K., Johnson, K.R. and Subramani, S. (1995) Human peroxisomal targeting signal-1 receptor restores peroxisomal protein import in cells from patients with fatal peroxisomal disorders. J. Cell Biol. 130, 5165.

[89] Kragler, F., Lametschwandtner, G., Christmann, J., Hartig, A. and Harada, J.J. (1998) Identification and analysis of the plant peroxisomal targeting signal 1 receptor NtPEX5. Proc. Natl. Acad. Sci. USA 95, 13336-13341.

[90] Wimmer, C., Schmid, M., Veenhuis, M. and Gietl, C. (1998) The plant PTS-1 receptor: similarities and differences to its human and yeast counterparts. Plant J. 16, 453-464.

[91] Dodt, G. and Gould, S.J. (1996) Multiple PEX genes are required for proper subcellular distribution and stability of Pex5p, the PTS1 receptor: evidence that PTS1 protein import is mediated by a cycling receptor. J. Cell Biol. 135, 1763-1774.

[92] Elgersma, Y., Kwast, L., Klein, A., Voorn-Brouwer, T., Van den Berg, M., Metzig, B., America, T., Tabak, H.F. and Distel, B. (1996) The SH3 domain of the Saccharomyces cerevisiae peroxi-somal membrane protein Pex13p functions as a docking site for Pex5p, a mobile receptor for the import of PTS1-containing proteins. J. Cell Biol. 135, 97-109.

[93] Gould, S.J., Kalish, J.E., Morrell, J.C., Bjorkman, J., Urquhart, A.J. and Crane, D.I. (1996) Pex13p is an SH3 protein of the peroxisome membrane and a docking factor for the predominantly cytoplasmic PTS1 receptor. J. Cell Biol. 135, 85-95.

[94] Terlecky, S.R., Nuttley, W.M., McCollum, D., Sock, E. and Subramani, S. (1995) The Pichia pastoris peroxisomal protein Pas8p is the receptor for the C-terminal tripeptide peroxisomal targeting signal. EMBO J. 14, 3627-3634.

[95] Brocard, C., Kragler, F., Simon, M.M., Schuster, T. and Hartig,

A. (1994) The tetratricopeptide repeat-domain of the PAS10 protein of Saccharomyces cerevisiae is essential for binding the peroxisomal targeting signal -SKL. Biochem. Biophys. Res. Commun. 204, 1016-1022.

[96] Gatto Jr., G.J., Geisbrecht, B.V., Gould, S.J. and Berg, J.M. (2000) Peroxisomal targeting signal-1 recognition by the TPR domains of human PEX5. Nat. Struct. Biol. 7, 1091-1095.

[97] Klein, A.T.J., Barnett, P., Bottger, G., Konings, D., Tabak, H.F. and Distel, B. (2001) Recognition of peroxisomal targeting signal type 1 by the import receptor Pex5p. J. Biol. Chem. 276, 1503415041.

[98] Braverman, N., Dodt, G., Gould, S.J. and Valle, D. (1998) An isoform of Pex5p, the human PTS1 receptor, is required for the import of PTS2 proteins into peroxisomes. Hum. Mol. Genet. 7, 1195-1205.

[99] Otera, H., Okumoto, K., Tateishi, K., Ikoma, Y., Matsuda, E., Nishimura, M., Tsukamoto, T., Osumi, T., Ohashi, K., Higuchi, O. and Fujiki, Y. (1998) Peroxisome targeting signal type 1 (PTS1) receptor is involved in import of both PTS1 and PTS2: studies with Pex5-defective CHO cell mutants. Mol. Cell. Biol. 18, 388-399.

[100] Erdmann, R. and Blobel, G. (1996) Identification of Pex13p, a peroxisomal membrane receptor for the PTS1 recognition factor. J. Cell Biol. 135, 111-121.

[101] Schliebs, W., Saidowsky, J., Agianian, B., Dodt, G., Herberg, F.W. and Kunau, W.H. (1999) Recombinant human peroxi-somal targeting signal receptor PEX5. Structural basis for interaction of PEX5 with PEX14. J. Biol. Chem. 274, 5666-5673.

[102] Barnett, P., Bottger, G., Klein, A.T.J., Tabak, H.F. and Distel,

B. (2000) The peroxisomal membrane protein Pex13p shows a novel mode of SH3 interaction. EMBO J. 19, 6382-6391.

[103] Bottger, G., Barnett, P., Klein, A.T.J., Kragt, A., Tabak, H.F. and Distel, B. (2000) Saccharomyces cerevisiae PTS1 receptor Pex5p interacts with the SH3 domain of the peroxisomal membrane protein Pex13p in an unconventional, non-PXXP-related manner. Mol. Biol. Cell 11, 3963-3976.

[104] Urquhart, A.J., Kennedy, D., Gould, S.J. and Crane, D.I. (2000) Interaction of Pex5p, the type 1 peroxisome targeting signal receptor, with the peroxisomal membrane proteins Pex14p and Pex13p. J. Biol. Chem. 11, 4127-4136.

[105] Otera, H., Harano, T., Honsho, M., Ghaedi, K., Mukai, S., Tanaka, A., Kawai, A., Shimizu, N. and Fujiki, Y. (2000) The mammalian peroxin Pex5pL, the longer isoform of the mobile peroxisome targeting signal (PTS) type 1 transporter, translocates the Pex7p.PTS2 protein complex into peroxisomes via its initial docking site, Pex14p. J. Biol. Chem. 275, 21703-21714.

[106] Albertini, M., Rehling, P., Erdmann, R., Girzalsky, W., Kiel, J.A.K.W., Veenhuis, M. and Kunau, W.H. (1997) Pex14p, a

peroxisomal membrane protein binding both receptors of the two PTS-dependent import pathways. EMBO J. 89, 83-92.

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

[108] Fransen, M., Terlecky, S R. and Subramani, S. (1998) Identification of a human PTS1 receptor docking protein directly required for peroxisomal protein import. Proc. Natl. Acad. Sci. USA 95, 8087-8092.

[109] Shimizu, N., Itoh, R., Hirono, Y., Otera, H., Ghaedi, K., Tateishi, K., Tamura, S., Okumoto, K., Harano, T., Mukai, S. and Fujiki, Y. (1999) The peroxin Pex14p. cDNA cloning by functional complementation on a Chinese hamster ovary cell mutant, characterization, and functional analysis. J. Biol. Chem. 274, 12593-12604.

[110] Will, G.K., Soukupova, M., Hong, X., Erdmann, K.S., Kiel, J.A.K.W., Dodt, G., Kunau, W.H. and Erdmann, R. (1999) Identification and characterization of the human orthologue of yeast Pex14p. Mol. Cell. Biol. 19, 2265-2277.

[111] Saidowsky, J., Dodt, G., Kirchberg, K., Wegner, A., Nasta-inczyk, W., Kunau, W.H. and Schliebs, W. (2001) The di-aromatic pentapeptide repeats of the human peroxisome import receptor PEX5 are separate high affinity binding sites for the peroxisomal membrane protein PEX14. J. Biol. Chem. 276, 34524-34529.

[112] Chang, C.C., Warren, D.S., Sacksteder, K.A. and Gould, S.J. (1999) PEX12 interacts with PEX5 and PEX10 and acts downstream of receptor docking in peroxisomal matrix protein import. J. Cell Biol. 147, 761-773.

[113] Okumoto, K., Abe, I. and Fujiki, Y. (2000) Molecular anatomy of the peroxin Pex12p - ring finger domain is essential for Pex12p function and interacts with the peroxisome-targeting signal type 1-receptor Pex5p and a ring peroxin, Pex10p. J. Biol. Chem. 275, 25700-25710.

[114] Reguenga, C., Oliveira, M.E.M., Gouveia, A.M.M., Sa-Mir-anda, C. and Azevedo, J.E. (2001) Characterization of the mammalian peroxisomal import machinery. J. Biol. Chem. 276, 29935-29942.

[115] Rehling, P., Skaletz-Rorowski, A., Girzalsky, W., Voorn-Brou-wer, T., Franse, M.M., Distel, B., Veenhuis, M., Kunau, W.H. and Erdmann, R. (2000) Pex8p, an intraperoxisomal peroxin of Saccharomyces cerevisiae required for protein transport into peroxisomes binds the PTS1 receptor Pex5p. J. Biol. Chem. 275, 3593-3602.

[116] Swinkels, B.W., Gould, S.J., Bodnar, A.G., Rachubinski, R.A. and Subramani, S. (1991) A novel, cleavable peroxisomal targeting signal at the amino-terminus of the rat 3-ketoacyl-CoA thiolase. EMBO J. 10, 3255-3262.

[117] Wiemer, E.A.C. and Subramani, S. (1994) Protein import deficiencies in human peroxisomal disorders. Mol. Genet. Med. 4, 119-152.

[118] Brosius, U. and Gartner, J. (2002) Cellular and molecular aspects of Zellweger syndrome and other peroxisome biogenesis disorders. Cell. Mol. Life Sci. 59, 1058-1069.

[119] Rehling, P., Marzioch, M., Niesen, F., Wittke, E., Veenhuis, M. and Kunau, W.H. (1996) The import receptor for the peroxisomal targeting signal 2 (PTS2) in Saccharomyces cerevisiae is encoded by the PAS7 gene. EMBO J. 15, 29012913.

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

[121] Glover, J.R., Andrews, D.W., Subramani, S. and Rachubinski, R.A. (1994) Mutagenesis of the amino targeting signal of Saccharomyces cerevisiae 3-ketoacyl-CoA thiolase reveals conserved amino acids required for import into peroxisomes in vivo. J. Biol. Chem. 269, 7558-7563.

[122] Tsukamoto, T., Hata, S., Yokota, S., Miura, S., Fujiki, Y., Hijikata, M., Miyazawa, S., Hashimoto, T. and Osumi, T. (1994) Characterization of the signal peptide at the amino-terminus of the rat peroxisomal 3-ketoacyl-coA thiolase precursor. J. Biol. Chem. 269, 6001-6010.

[123] Flynn, C R., Mullen, R.T. and Trelease, R.N. (1998) Mutational analyses of a type 2 peroxisomal targeting signal that is capable of directing oligomeric protein import into tobacco BY-2 glyoxysomes. Plant J. 16, 709-720.

[124] Marzioch, M., Erdmann, R., Veenhuis, M. and Kunau, W.H. (1994) PAS7 encodes a novel yeast member of the WD 40 protein family essential for import of 3-oxoacyl-CoA thiolase, a PTS2-containing protein, into peroxisomes. EMBO J. 13, 49084918.

[125] Zhang, J.W. and Lazarow, P.B. (1995) Peb1 (Pas7) in Saccha-romyces cerevisiae encodes a hydrophilic, intraperoxisomal protein that is a member of the WD repeat family and is essential for the import of thiolase into peroxisomes. J. Cell Biol. 129, 65-80.

[126] Braverman, N., Steel, G., Obie, C., Moser, A., Moser, H., Gould, S.J. and Valle, D. (1997) Human Pex7 encodes the peroxisomal PTS2 receptor and is responsible for Rhizomelic chondrodysplasia punctata. Nat. Genet. 15, 369-376.

[127] Motley, A.M., Hettema, E.H., Hogenhout, E.M., Brites, P., Ten Asbroek, A.L., Wijburg, F.A., Baas, F., Heijmans, H.S., Tabak, H.F., Wanders, R.J.A. and Distel, B. (1997) Rhizomelic chon-drodysplasia punctata is a peroxisomal protein targeting disease caused by a non-functional PTS2 receptor. Nat. Genet. 15, 377380.

[128] Purdue, P.E., Zhang, J.W., Skoneczny, M. and Lazarow, P.B. (1997) Rhizomelic chondrodysplasia punctata is caused by deficiency of human Pex7, a homologue of the yeast PTS2 receptor. Nat. Genet. 15, 381-384.

[129] Elgersma, Y., Elgersma-Hooisma, M., Wenzel, T., McCaffery, J.M., Farquhar, M.G. and Subramani, S. (1998) A mobile PTS2 receptor for peroxisomal protein import in Pichia pastoris. J. Cell Biol. 140, 807-820.

[130] Motley, A.M., Hettema, E.H., Ketting, R., Plasterk, R. and Tabak, H.F. (2000) Caenorhabditis elegans has a single pathway to target matrix proteins to peroxisomes. EMBO Rep. 1, 40-46.

[131] Dodt, G., Warren, D., Becker, E., Rehling, P. and Gould, S.J.

(2001) Domain mapping of human PEX5 reveals functional and structural similarities to Saccharomyces cerevisiae Pex18p and Pex21p. J. Biol. Chem. 276, 41769-41781.

[132] Ghys, K., Fransen, M., Mannaerts, G.P. and Van Veldhoven, P.P. (2002) Functional studies on human Pex7p: subcellular localization and interaction with proteins containing a perox-isome-targeting signal type 2 and other peroxins. Biochem. J. 365, 41-50.

[133] Mukai, S., Ghaedi, K. and Fujiki, Y. (2002) Intracellular localization, function, and dysfunction of the peroxisome-targeting signal type 2 receptor, Pex7p, in mammalian cells. J. Biol. Chem. 277, 9548-9561.

[134] Girzalsky, W., Rehling, P., Stein, K., Kipper, J., Blank, L., Kunau, W.H. and Erdmann, R. (1999) Involvement of Pex13p in Pex14p localization and peroxisomal targeting signal 2-depend-ent protein import into peroxisomes. J. Cell Biol. 144, 11511162.

[135] Stein, K., Schell-Steven, A., Erdmann, R. and Rottensteiner, H.

(2002) Interactions of Pex7p and Pex18p/Pex21p with the peroxisomal docking machinery: implications for the first steps in PTS2 protein import. Mol. Cell. Biol. 22, 6056-6069.

[136] Purdue, P.E., Yang, X.D. and Lazarow, P.B. (1998) Pex18p and Pex21p, a novel pair of related peroxins essential for peroxisomal targeting by the PTS2 pathway. J. Cell Biol. 143, 1859-1869.

[137] Nito, K., Hayashi, M. and Nishimura, M. (2002) Direct interaction and determination of binding domains among

peroxisomal import factors in Arabidopsis thaliana. Plant Cell Physiol. 43, 355-366.

[138] Titorenko, V.I., Smith, J.J., Szilard, R.K. and Rachubinski, R.A. (1998) Pex20p of the yeast Yarrowia lipolytica is required for the oligomerization of thiolase in the cytosol and for its targeting to the peroxisome. J. Cell Biol. 142, 403-420.

[139] Einwachter, H., Sowinski, S., Kunau, W.H. and Schliebs, W. (2001) Yarrowia lipolytica Pex20p, Saccharomyces cerevisiae Pex18p/Pex21p and mammalian Pex5pL fulfil a common function in the early steps of the peroxisomal PTS2 import pathway. EMBO Rep. 2, 1035-1039.

[140] Sichting, M., Schell-Steven, A., Prokisch, H., Erdmann, R. and Rottensteiner, H. (2003) Pex7p and Pex20p of Neurospora crassa function together in PTS2-dependent protein import into peroxisomes. Mol. Biol. Cell 14, 810-821.

[141] Small, G.M., Szabo, J. and Lazarow, P.B. (1988) Acyl-CoA oxidase contains two targeting sequences each of which can mediate protein import into peroxisomes. EMBO J. 7, 11671173.

[142] Klein, A.T.J., Van den Berg, M., Bottger, G., Tabak, H.F. and Distel, B. (2002) Saccharomyces cerevisiae acyl-CoA oxidase follows a novel, non-PTS1, import pathway into peroxisomes that is dependent on Pex5p. J. Biol. Chem. 277, 25011-25019.

[143] McNew, J.A. and Goodman, J.M. (1994) An oligomeric protein is imported into peroxisomes in vivo. J. Cell Biol. 127, 12451257.

[144] Lee, M.S., Mullen, R.T. and Trelease, R.N. (1997) Oilseed isocitrate lyases lacking their essential type 1 peroxisomal targeting signal are piggybacked to glyoxysomes. Plant Cell 9, 185-197.

[145] Rachubinski, R.A. and Subramani, S. (1995) How proteins penetrate peroxisomes. Cell 83, 525-528.

[146] Elgersma, Y. and Tabak, H.F. (1996) Proteins involved in peroxisome biogenesis and functioning. Biochim. Biophys. Acta 1286, 269-283.

[147] Yang, X.D., Purdue, P.E. and Lazarow, P.B. (2001) Eci1p uses a PTS1 to enter peroxisomes: either its own or that of a partner, Dci1p. Eur. J. Cell Biol. 80, 126-138.

[148] Komori, M., Rasmussen, S.W., Kiel, J.A.K.W., Baerends, R.J., Cregg, J.M., Van der Klei, I.J. and Veenhuis, M. (1997) The Hansenula polymorpha PEX14 gene encodes a novel peroxisomal membrane protein essential for peroxisome biogenesis. EMBO J. 16, 44-53.

[149] Johnson, M.A., Snyder, W.B., Cereghino, J.L., Veenhuis, M., Subramani, S. and Cregg, J.M. (2001) Pichia pastoris Pex14p, a phosphorylated peroxisomal membrane protein, is part of a PTS-receptor docking complex and interacts with many perox-ins. Yeast 18, 621-641.

[150] Lopez-Huertas, E., Oh, J.S. and Baker, A. (1999) Antibodies against Pex14p block ATP-independent binding of matrix proteins to peroxisomes in vitro. FEBS Lett. 459, 227-229.

[151] Hayashi, M., Nito, K., Toriyama-Kato, K., Kondo, M., Yamaya, T. and Nishimura, M. (2000) AtPex14p maintains peroxisomal functions by determining protein targeting to three kinds of plant peroxisomes. EMBO J. 19, 5701-5710.

[152] Oliveira, M.E.M., Reguenga, C., Gouveia, A.M.M., Guimaraes, C.P., Schliebs, W., Kunau, W.H., Silva, M.T., Sa-Miranda, C. and Azevedo, J.E. (2002) Mammalian Pex14p: membrane topology and characterisation of the Pex14p-Pex14p interaction. Biochim. Biophys. Acta 1567, 13-22.

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

[154] Snyder, W.B., Koller, A., Choy, A.J., Johnson, M.A., Cregg, J.M., Rangell, L., Keller, G.A. and Subramani, S. (1999) Pex17p is required for import of both peroxisome membrane and

lumenal protein and interacts with Pex19p and the peroxisome targeting signal-receptor docking complex in Pichia pastoris. Mol. Biol. Cell 10, 4005-4019.

[155] Fransen, M., Brees, C., Ghys, K., Amery, L., Mannaerts, G.P., Ladant, D. and Van Veldhoven, P.P. (2002) Analysis of mammalian peroxin interactions using a non-transcription-based bacterial two-hybrid assay. Mol. Cell. Proteomics 1, 243-252.

[156] Otera, H., Setoguchi, K., Hamasaki, M., Kumashiro, T., Shimizu, N. and Fujiki, Y. (2002) Peroxisomal targeting signal receptor Pex5p interacts with cargoes and import machinery components in a spatiotemporally differentiated manner: conserved Pex5p WXXXF/Y motifs are critical for matrix protein import. Mol. Cell. Biol. 22, 1639-1655.

[157] Pires, J.R., Hong, X.J., Brockmann, C., Volkmer-Engert, R., Schneider-Mergener, J., Oschkinat, H. and Erdmann, R. (2003) The Sc Pex13p SH3 domain exposes two distinct binding sites for Pex5p and Pex14p. J. Mol. Biol. 326, 1427-1435.

[158] Mayer, B.J. and Eck, M.J. (1995) SH3 domains: minding your PS and QS. Curr. Biol. 5, 364-367.

[159] Douangamath, A., Filipp, F.V., Klein, A.T.J., Barnett, P., Zou, P.J., Voorn-Brouwer, T., Vega, M.C., Mayans, O.M., Sattler, M., Distel, B. and Wilmanns, M. (2002) Topography for independent binding of a-helical and PPII-helical ligands to a peroxisomal SH3 domain. Mol. Cell 10, 1007-1017.

[160] Smith, J.J., Szilard, R.K., Marelli, M. and Rachubinski, R.A. (1997) The peroxin Pex17p of the yeast Yarrowia lipolytica is associated peripherally with the peroxisomal membrane and is required for the import of a subset of matrix proteins. Mol. Cell. Biol. 17, 2511-2520.

[161] Harper, C.C., South, S.T., McCaffery, J.M. and Gould, S.J. (2002) Peroxisomal membrane protein import does not require Pex17p. J. Biol. Chem. 277, 16498-16504.

[162] Hettema, E.H., Girzalsky, W., Van den Berg, M., Erdmann, R. and Distel, B. (2000) Saccharomyces cerevisiae Pex3p and Pex19p are required for proper localization and stability of peroxisomal membrane proteins. EMBO J. 19, 223-233.

[163] Chang, C.C., Lee, W.H., Moser, H., Valle, D. and Gould, S.J. (1997) Isolation of the human PEX12 gene, mutated in group 3 of the peroxisome biogenesis disorders. Nat. Genet. 15, 385-388.

[164] Okumoto, K., Shimozawa, N., Kawai, A., Tamura, S., Tsuka-moto, T., Osumi, T., Moser, H., Wanders, R.J.A., Suzuki, Y., Kondo, N. and Fujiki, Y. (1998) PEX12, the pathogenic gene of group III Zellweger syndrome: cDNA cloning by functional complementation on a CHO cell mutant, patient analysis, and characterization of PEX12p. Mol. Cell. Biol. 18, 4324-4336.

[165] Harano, T., Shimizu, N., Otera, H. and Fujiki, Y. (1999) Transmembrane topology of the peroxin, Pex2p, an essential component for the peroxisome assembly. J. Biochem. 125, 11681174.

[166] Huang, Y., Ito, R., Miura, S., Hashimoto, T. and Ito, M. (2000) A missense mutation in the RING finger motif of PEX2 protein disturbs the import of peroxisome targeting signal 1 (PTS1)-containing protein but not the PTS2-containing protein. Bio-chem. Biophys. Res. Commun. 270, 717-721.

[167] Wiebel, F.F. and Kunau, W.H. (1992) The Pas2 protein essential for peroxisome biogenesis is related to ubiquitin-conjugating enzymes. Nature 359, 73-76.

[168] Crane, D.I., Kalish, J.E. and Gould, S.J. (1994) The Pichia pastoris PAS4 gene encodes a ubiquitin-conjugating enzyme required for peroxisome assembly. J. Biol. Chem. 269, 2183521844.

[169] Gould, S.J., McCollum, D., Spong, A.P., Heyman, J.A. and Subramani, S. (1992) Development of the yeast Pichia pastoris as a model organism for a genetic and molecular analysis of peroxisome assembly. Yeast 8, 613-628.

[170] Koller, A., Snyder, W.B., Faber, K.N., Wenzel, T.J., Rangell, L., Keller, G.A. and Subramani, S. (1999) Pex22p of Pichia pastoris,

essential for peroxisomal matrix protein import, anchors the ubiquitin-conjugating enzyme, Pex4p, on the peroxisomal membrane. J. Cell Biol. 146, 99-112.

[171] Eckert, J.H. and Johnsson, N. (2003) Pex10p links the ubiquitin conjugating enzyme Pex4p to the protein import machinery of the peroxisome. J. Cell Science 116, 3623-3634.

[172] Collins, C.S., Kalish, J.E., Morrell, J.C., McCaffery, J.M. and Gould, S.J. (2000) The peroxisome biogenesis factors Pex4p, Pex22p, Pex1p, and Pex6p act in the terminal steps of peroxi-somal matrix protein import. Mol. Cell. Biol. 20, 7516-7526.

[173] Van der Klei, I.J., Hilbrands, R.E., Kiel, J.A.K.W., Rasmussen, S.W., Cregg, J.M. and Veenhuis, M. (1998) The ubiquitin-conjugating enzyme Pex4p of Hansenula polymorpha is required for efficient functioning of the PTS1 import machinery. EMBO J. 17, 3608-3618.

[174] Purdue, P.E. and Lazarow, P.B. (2001) Pex18p is constitutively degraded during peroxisome biogenesis. J. Biol. Chem. 276, 47684-47689.

[175] Smith, J.J. and Rachubinski, R.A. (2001) A role for the peroxin Pex8p in Pex20p-dependent thiolase import into peroxisomes of the yeast Yarrowia lipolytica. J. Biol. Chem. 276, 1618-1625.

[176] Agne, B., Meindl, N.M., Niederhoff, K., Einwachter, H., Rehling, P., Sickmann, A., Meyer, H.E., Girzalsky, W. and Kunau, W.H. (2003) Pex8p: an intraperoxisomal organizer of the peroxisomal import machinery. Mol. Cell 11, 635-646.

[177] Erdmann, R., Wiebel, F.F., Flesseau, A., Rytka, J., Beyer, A., Frohlich, K.U. and Kunau, W.H. (1991) PAS1, a yeast gene required for peroxisome biogenesis, encodes a member of a novel family of putative ATPases. Cell 64, 499-510.

[178] Kunau, W.H., Beyer, A., Franken, T., Gotte, K., Marzioch, M., Saidowsky, J., Skaletzrorowski, A. and Wiebel, F.F. (1993) Two complementary approaches to study peroxisome biogenesis in Saccharomyces cerevisiae: forward and reversed genetics. Biochimie 75, 209-224.

[179] Faber, K.N., Heyman, J.A. and Subramani, S. (1998) Two AAA family peroxins, PpPex1p and PpPex6p, interact with each other in an ATP-dependent manner and are associated with different subcellular membranous structures distinct from peroxisomes. Mol. Cell. Biol. 18, 936-943.

[180] Geisbrecht, B.V., Collins, C.S., Reuber, B.E. and Gould, S.J. (1998) Disruption of a PEX1-PEX6 interaction is the most common cause of the neurologic disorders Zellweger syndrome, neonatal adrenoleukodystrophy, and infantile refsum disease. Proc. Natl. Acad. Sci. USA 95, 8630-8635.

[181] Kiel, J.A.K.W., Hilbrands, R.E., Van der Klei, I.J., Rasmussen, S.W., Salomons, F.A., Van der Heide, M., Faber, K.N., Cregg, J.M. and Veenhuis, M. (1999) Hansenula polymorpha Pex1p and Pex6p are peroxisome-associated AAA proteins that functionally and physically interact. Yeast 15, 1059-1078.

[182] Tsukamoto, T., Miura, S., Nakai, T., Yokota, S., Shimozawa, N., Suzuki, Y., Orii, T., Fujiki, Y., Sakai, F., Bogaki, A., Yasumo, H. and Osumi, T. (1995) Peroxisome assembly factor-II, a putative ATPase cloned by functional complementation on a peroxisome-deficient mammalian cell mutant. Nat. Genet. 11, 395-401.

[183] Yahraus, T., Braverman, N., Dodt, G., Kalish, J.E., Morrell, J.C., Moser, H.W., Valle, D. and Gould, S.J. (1996) The peroxisome biogenesis disorder group 4 gene, PXAAA1, encodes a cytoplasmic ATPase required for stability of the PTS1 receptor. EMBO J. 15, 2914-2923.

[184] Tamura, S., Shimozawa, N., Suzuki, Y., Tsukamoto, T., Osumi, T. and Fujiki, Y. (1998) A cytoplasmic AAA family peroxin, Pex1p, interacts with Pex6p. Biochem. Biophys. Res. Commun. 245, 883-886.

[185] Titorenko, V.I., Chan, H. and Rachubinski, R.A. (2000) Fusion of small peroxisomal vesicles in vitro reconstructs an early step

in the in vivo multistep peroxisome assembly pathway of Yarrowia lipolytica. J. Cell Biol. 148, 29-43.

[186] Titorenko, V.I. and Rachubinski, R.A. (2000) Peroxisomal membrane fusion requires two AAA family ATPases, Pexlp and Pex6p. J. Cell Biol. 150, 881-886.

[187] Sacksteder, K.A. and Gould, S.J. (2000) The genetics of peroxisome biogenesis. Annu. Rev. Genet. 34, 623-652.

[188] Pool, M.R., Lopez-Huertas, E. and Baker, A. (1998) Characterization of intermediates in the process of plant peroxisomal protein import. EMBO J. 17, 6854-6862.

[189] Terlecky, S.R., Legakis, J.E., Hueni, S.E. and Subramani, S. (2001) Quantitative analysis of peroxisomal protein import in vitro. Exp. Cell Res. 263, 98-106.

[190] Oliveira, M.E., Gouveia, A.M., Pinto, R.A., Sa-Miranda, C. and Azevedo, J.E. (2003) The energetics of Pex5p-mediated peroxi-somal protein import. J. Biol. Chem. 278, 39483-39488.

[191] Titorenko, V.I. and Rachubinski, R.A. (2001) Dynamics of peroxisome assembly and function. Trends Cell Biol. 11, 22-29.

[192] Eitzen, G.A., Aitchison, J.D., Szilard, R.K., Veenhuis, M., Nuttley, W.M. and Rachubinski, R.A. (1995) The Yarrowia lipolytica gene PAY2 encodes a 42-kDa peroxisomal integral membrane-protein essential for matrix protein import and peroxisome enlargement but not for peroxisome proliferation. J. Biol. Chem. 270, 1429-1436.

[193] Brown, T.W., Titorenko, V.I. and Rachubinski, R.A. (2000) Mutants of the Yarrowia lipolytica PEX23 gene encoding an integral peroxisomal membrane peroxin mislocalize matrix proteins and accumulate vesicles containing peroxisomal matrix and membrane proteins. Mol. Biol. Cell 11, 141-152.

[194] Tam, Y.Y.C. and Rachubinski, R.A. (2002) Yarrowia lipoly-tica cells mutant for the PEX24 gene encoding a peroxisomal membrane peroxin mislocalize peroxisomal proteins and accumulate membrane structures containing both peroxisomal matrix and membrane proteins. Mol. Biol. Cell 13, 26812691.

[195] Matsumoto, N., Tamura, S., Furuki, S., Miyata, N., Moser, A., Shimozawa, N., Moser, H.W., Suzuki, Y., Kondo, N. and Fujiki, Y. (2003) Mutations in novel peroxin gene PEX26 that cause peroxisome-biogenesis disorders of complementation group 8 provide a genotype-phenotype correlation. Am. J. Hum. Genet. 73, 233-246.

[196] Matsumoto, N., Tamura, S. and Fujiki, Y. (2003) The pathogenic peroxin Pex26p recruits the Pex1p-Pex6p AAA ATPase complexes to peroxisomes. Nat. Cell Biol. 5, 454-460.

[197] Birschmann, I., Stroobants, A.K., Van den Berg, M., Schafer, A., Rosenkranz, K., Kunau, W.H. and Tabak, H.F. (2003) Pex15p of Saccharomyces cerevisiae provides a molecular basis for recruitment of the AAA peroxin Pex6p to peroxisomal membranes. Mol. Biol. Cell 14, 2226-2236.

[198] Dammai, V. and Subramani, S. (2001) The human peroxisomal targeting signal receptor, Pex5p, is translocated into the peroxi-somal matrix and recycled to the cytosol. Cell 105, 187-196.

[199] Van der Klei, I.J. and Veenhuis, M. (2002) Peroxisomes: flexible and dynamic organelles. Curr. Opin. Cell Biol. 14, 500-505.

[200] McNew, J.A. and Goodman, J.M. (1996) The targeting and assembly of peroxisomal proteins: some old rules do not apply. Trends Biochem. Sci. 21, 54-58.

[201] Hausler, T., Stierhof, Y.D., Wirtz, E. and Clayton, C. (1996) Import of a DHFR hybrid protein into glycosomes in vivo is not inhibited by the folate-analogue aminopterin. J. Cell Biol. 132, 311-324.

[202] Legakis, J.E. and Terlecky, S.R. (2001) PTS2 protein import into mammalian peroxisomes. Traffic 2, 252-260.

[203] Titorenko, V.I., Nicaud, J.M., Wang, H.J., Chan, H. and Rachubinski, R.A. (2002) Acyl-CoA oxidase is imported as a heteropentameric, cofactor-containing complex into peroxi-somes of Yarrowia lipolytica. J. Cell Biol. 156, 481-494.

[204] Faber, K.N., Van Dijk, R., Keizer-Gunnink, I., Koek, A., Van der Klei, I.J. and Veenhuis, M. (2002) Import of assembled PTS1 proteins into peroxisomes of the yeast Hansenula polymorpha: yes and no! Biochim. Biophys. Acta 1591, 157-162.

[205] Brocard, C.B., Jedeszko, C., Song, H., C, Terlecky, S.R. and Walton, P.A. (2003) Protein structure and import into the peroxisomal matrix. Traffic 4, 74-82.

[206] Walton, P.A., Hill, P.E. and Subramani, S. (1995) Import of stably folded proteins into peroxisomes. Mol. Biol. Cell 6, 675683.

[207] Hettema, E.H., Distel, B. and Tabak, H.F. (1999) Import of proteins into peroxisomes. Biochim. Biophys. Acta 1451, 17-34.

[208] Gould, S.J. and Collins, C.S. (2002) Peroxisomal-protein import: is it really that complex?. Nat. Rev. Mol. Cell. Biol. 3, 382-389.

[209] Bellion, E. and Goodman, J.M. (1987) Proton ionophores prevent assembly of a peroxisomal protein. Cell 48, 165-173.

[210] Santos, M.J., Imanaka, T., Shio, H. and Lazarow, P.B. (1988) Peroxisomal integral membrane proteins in control and Zellwe-ger fibroblasts. J. Biol. Chem. 263, 10502-10509.

[211] Santos, M.J., Imanaka, T., Shio, H., Small, G.M. and Lazarow, P.B. (1988) Peroxisomal membrane ghosts in Zellweger syndrome-aberrant organelle assembly. Science 239, 1536-1538.

[212] Yamasaki, M., Hashiguchi, N., Fujiwara, C., Imanaka, T., Tsukamoto, T. and Osumi, T. (1999) Formation of peroxisomes from peroxisomal ghosts in a peroxisome-deficient mammalian cell mutant upon complementation by protein microinjection. J. Biol. Chem. 274, 35293-35296.

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

[214] Salomons, F.A., Faber, K.N., Veenhuis, M. and Van der Klei, I.J. (2001) Peroxisomal remnant structures in Hansenula poly-morpha Pex5 cells can develop into normal peroxisomes upon induction of the PTS2 protein amine oxidase. J. Biol. Chem. 276, 4190-4198.

[215] Fujiki, Y., Rachubinski, R.A. and Lazarow, P.B. (1984) Synthesis of a major integral membrane polypeptide of rat liver peroxisomes on free polysomes. Proc. Natl. Acad. Sci. USA 81, 7127-7131.

[216] Suzuki, Y., Orii, T., Takiguchi, M., Mori, M., Hijikata, M. and Hashimoto, T. (1987) Biosynthesis of membrane polypeptides of rat liver peroxisomes. J. Biochem. 101, 491-496.

[217] Tsukamoto, T., Shimozawa, N. and Fujiki, Y. (1994) Perox-isome assembly factor 1: nonsense mutation in a peroxisome deficient Chinese hamster ovary cell mutant and deletion analysis. Mol. Cell. Biol. 14, 5458-5465.

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

[219] Tabak, H.F., Murk, J.L., Braakman, I. and Geuze, H.J. (2003) Peroxisomes start their life in the endoplasmic reticulum. Traffic 4, 512-518.

[220] Dyer, J.M., McNew, J.A. and Goodman, J.M. (1996) The sorting sequence of the peroxisomal integral membrane protein PMP47 is contained within a short hydrophilic loop. J. Cell Biol. 133, 269-280.

[221] Elgersma, Y., Kwast, L., Van den Berg, M., Snyder, W.B., Distel, B., Subramani, S. and Tabak, H.F. (1997) Overexpression of Pex15p, a phosphorylated peroxisomal integral membrane protein required for peroxisome assembly in S. cerevisiae, causes proliferation of the endoplasmic reticulum membrane. EMBO J. 16, 7326-7341.

[222] Höhfeld, J., Veenhuis, M. and Kunau, W.H. (1991) PAS3, a Saccharomyces cerevisiae gene encoding a peroxisomal integral membrane protein essential for peroxisome biogenesis. J. Cell Biol. 114, 1167-1178.

[223] Baerends, R.J.S., Rasmussen, S.W., Hilbrands, R.E., Van der Heide, M., Faber, K.N., Reuvekamp, P.T.W., Kiel, J.A.K.W.,

Cregg, M., Van der Klei, I.J. and Veenhuis, M. (1996) The Hansenula polymorpha PER9 gene encodes a peroxisomal membrane protein essential for peroxisome assembly and integrity. J. Biol. Chem. 271, 8887-8894.

[224] Wiemer, E.A.C., Luers, G.H., Faber, K.N., Wenzel, T., Veenhuis, M. and Subramani, S. (1996) Isolation and characterization of Pas2p, a peroxisomal membrane protein essential for perox-isome biogenesis in the methylotrophic yeast Pichia pastoris. J. Biol. Chem. 271, 18973-18980.

[225] Kammerer, S., Holzinger, A., Welsch, U. and Roscher, A.A. (1998) Cloning and characterization of the gene encoding the human peroxisomal assembly protein Pex3p. FEBS Lett. 429, 53-60.

[226] Soukupova, M., Sprenger, C., Gorgas, K., Kunau, W.H. and Dodt, G. (1999) Identification and characterization of the human peroxin PEX3. Eur. J. Cell Biol. 78, 357-374.

[227] Ghaedi, K., Tamura, S., Okumoto, K., Matsuzono, Y. and Fujiki, Y. (2000) The peroxin Pex3p initiates membrane assembly in peroxisome biogenesis. Mol. Biol. Cell 11, 2085-2102.

[228] Muntau, A.C., Mayerhofer, P.U., Paton, B.C., Kammerer, S. and Roscher, A.A. (2000) Defective peroxisome membrane synthesis due to mutations in human PEX3 causes Zellweger syndrome, complementation group G. Am. J. Hum. Genet. 67, 967-975.

[229] Hazra, P.P., Suriapranata, I., Snyder, W.B. and Subramani, S. (2002) Peroxisome remnants in pex3D cells and the requirement of Pex3p for interactions between the peroxisomal docking and translocation subcomplexes. Traffic 3, 560-574.

[230] Gotte, K., Girzalsky, W., Linkert, M., Baumgart, E., Kammerer, S., Kunau, W.H. and Erdmann, R. (1998) Pex19p, a farnesylated protein essential for peroxisome biogenesis. Mol. Cell. Biol. 18, 616-628.

[231] Snyder, W.B., Faber, K.N., Wenzel, T.J., Koller, A., Luers, G H., Rangell, L., Keller, G.A. and Subramani, S. (1999) Pex19p interacts with Pex3p and Pex10p and is essential for peroxisome biogenesis in Pichia pastoris. Mol. Biol. Cell 10, 1745-1761.

[232] Matsuzono, Y., Kinoshita, N., Tamura, S., Shimozawa, N., Hamasaki, M., Ghaed, K., Wanders, R.J.A., Suzuki, Y., Kondo, N. and Fujiki, Y. (1999) Human PEX19: cDNA cloning by functional complementation, mutation analysis in a patient with Zellweger syndrome, and potential role in peroxisomal membrane assembly. Proc. Natl. Acad. Sci. USA 96, 2116-2121.

[233] Sacksteder, K.A., Jones, J.M., South, S.T., Li, X.L., Liu, Y.F. and Gould, S.J. (2000) PEX19 binds multiple peroxisomal membrane proteins, is predominantly cytoplasmic, and is required for peroxisome membrane synthesis. J. Cell Biol. 148, 931-944.

[234] Snyder, W.B., Koller, A., Choy, A.J. and Subramani, S. (2000) The peroxin Pex19p interacts with multiple, integral membrane proteins at the peroxisomal membrane. J. Cell Biol. 149, 11711177.

[235] Fransen, M., Wylin, T., Brees, C., Mannaerts, G.P. and Van Veldhoven, P.P. (2001) Human Pex19p binds peroxisomal integral membrane proteins at regions distinct from their sorting sequences. Mol. Cell. Biol. 21, 4413-4424.

[236] Baerends, R.J.S., Faber, K.N., Kram, A.M., Kiel, J.A.K.W., Van der Klei, I.J. and Veenhuis, M. (2000) A stretch of positively charged amino acids at the N-terminus of Hansenula polymorpha Pex3p is involved in incorporation of the protein into the peroxisomal membrane. J. Biol. Chem. 275, 9986-9995.

[237] South, S.T., Sacksteder, K.A., Li, X.L., Liu, Y.F. and Gould, S.J. (2000) Inhibitors of COPI and COPII do not block PEX3-mediated peroxisome synthesis. J. Cell Biol. 149, 1345-1359.

[238] Honsho, M., Tamura, S., Shimozawa, N., Suzuki, Y., Kondo, N. and Fujiki, Y. (1998) Mutation in PEX16 is causal in the peroxisome-deficient Zellweger syndrome of complementation group D. Am. J. Hum. Genet. 63, 1622-1630.

[239] South, S.T. and Gould, S.J. (1999) Peroxisome synthesis in the absence of preexisting peroxisomes. J. Cell Biol. 144, 255-266.

[240] Honsho, M., Hiroshige, T. and Fujiki, Y. (2002) The membrane biogenesis peroxin Pex16p: topogenesis and functional roles in peroxisomal membrane assembly. J. Biol. Chem. 277, 4451344524.

[241] Eitzen, G.A., Szilard, R.K. and Rachubinski, R.A. (1997) Enlarged peroxisomes are present in oleic acid-grown Yarrowia lipolytica overexpressing the PEX16 gene encoding an intraper-oxisomal peripheral membrane peroxin. J. Cell Biol. 137, 12651278.

[242] Tabak, H.F., Braakman, I. and Distel, B. (1999) Peroxisomes: simple in function but complex in maintenance. Trends Cell Biol. 9, 447-453.

[243] Sakai, Y., Marshall, P.A., Saiganji, A., Takabe, K., Saiki, H., Kato, N. and Goodman, J.M. (1995) The Candida boidinii peroxisomal membrane-protein PMP30 has a role in peroxi-somal proliferation and is functionally homologous to PMP27 from Saccharomyces cerevisiae. J. Bacteriol. 177, 6773-6781.

[244] Marshall, P.A., Dyer, J.M., Quick, M.E. and Goodman, J.M. (1996) Redox-sensitive homodimerization of Pex11p: a proposed mechanism to regulate peroxisomal division. J. Cell Biol. 135, 123-137.

[245] Marshall, P.A., Krimkevich, Y.I., Lark, R.H., Dyer, J.M., Veenhuis, M. and Goodman, J.M. (1995) Pmp27 promotes peroxisomal proliferation. J. Cell Biol. 129, 345-355.

[246] Schrader, M., Reuber, B.E., Morrell, J.C., Jimenez-Sanchez, G., Obie, C., Stroh, T.A., Valle, D., Schroer, T.A. and Gould, S.J. (1998) Expression of PEX11ß mediates peroxisome proliferation in the absence of extracellular stimuli. J. Biol. Chem. 273, 2960729614.

[247] Van Roermund, C.W.T., Tabak, H.F., Van den Berg, M., Wanders, R.J.A. and Hettema, E.H. (2000) Pex11p plays a primary role in medium-chain fatty acid oxidation, a process that affects peroxisome number and size in Saccharomyces cerevisiae. J. Cell Biol. 150, 489-497.

[248] Li, X., Baumgart, E., Dong, G.X., Morrell, J.C., Jimenez-Sanchez, G., Valle, D., Smith, K.D. and Gould, S.J. (2002) PEX11a is required for peroxisome proliferation in response to 4-phenylbutyrate but is dispensable for peroxisome proliferator-activated receptor alpha-mediated peroxisome proliferation. Mol. Cell. Biol. 22, 8226-8240.

[249] Li, X., Baumgart, E., Morrell, J.C., Jimenez-Sanchez, G., Valle, D. and Gould, S.J. (2002) PEX11ß deficiency is lethal and impairs neuronal migration but does not abrogate peroxisome function. Mol. Cell. Biol. 22, 4358-4365.

[250] Smith, J.J., Marelli, M., Christmas, R.H., Vizeacoumar, F.J., Dilworth, D.J., Ideker, T., Galitski, T., Dimitrov, K., Rachubinski, R.A. and Aitchison, J.D. (2002) Transcriptome profiling to identify genes involved in peroxisome assembly and function. J. Cell Biol. 158, 259-271.

[251] Rottensteiner, H., Stein, K., Sonnenhol, E. and Erdmann, R. (2003) Conserved function of pex11p and the novel pex25p and pex27p in peroxisome biogenesis. Mol. Biol. Cell 14, 4316-4328.

[252] Tam, Y.Y., Torres-Guzman, J.C., Vizeacoumar, F.J., Smith, J.J., Marelli, M., Aitchison, J.D. and Rachubinski, R.A. (2003) Pex11-related proteins in peroxisome dynamics: a role for the novel peroxin Pex27p in controlling peroxisome size and number in Saccharomyces cerevisiae. Mol. Biol. Cell 14, 4089-4102.

[253] Vizeacoumar, F.J., Torres-Guzman, J.C., Tam, Y.Y., Aitchison, J.D. and Rachubinski, R.A. (2003) YHR150w and YDR479c encode peroxisomal integral membrane proteins involved in the regulation of peroxisome number, size, and distribution in Saccharomyces cerevisiae. J. Cell Biol. 161, 321-332.

[254] Vizeacoumar, F.J., Torres-Guzman, J.C., Bouard, D., Aitchison, J.D. and Rachubinski, R.A. (2004) Pex30p, Pex31p, and Pex32p form a family of peroxisomal integral membrane proteins

regulating peroxisome size and number in Saccharomyces cerevisiae. Mol. Biol. Cell 15, 665-677.

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

[256] Heinemann, P. and Just, W.W. (1992) Peroxisomal protein import. In vivo evidence for a novel translocation competent compartment. FEBS Lett. 300, 179-182.

[257] Hoepfner, D., Van den Berg, M., Philippsen, P., Tabak, H.F. and Hettema, E.H. (2001) A role for Vps1p, actin, and the Myo2p motor in peroxisome abundance and inheritance in Saccharomyces cerevisiae. J. Cell Biol. 155, 979-990.

[258] Ghaedi, K., Honsho, M., Shimozawa, N., Suzuki, Y., Kondo, N. and Fujiki, Y. (2000) PEX3 is the causal gene responsible for peroxisome membrane assembly-defective Zellweger syndrome of complementation group G. Am. J. Hum. Genet. 67, 976-981.

[259] Shimozawa, N., Suzuki, Y., Zhang, Z.Y., Imamura, A., Ghaedi, K., Fujiki, Y. and Kondo, N. (2000) Identification of PEX3 as the gene mutated in a Zellweger syndrome patient lacking peroxisomal remnant structures. Hum. Mol. Genet. 9, 19951999.

[260] Lambkin, G.R. and Rachubinski, R.A. (2001) Yarrowia lipoly-tica cells mutant for the peroxisomal peroxin Pex19p contain structures resembling wild-type peroxisomes. Mol. Biol. Cell 12, 3353-3364.

[261] Titorenko, V.I., Smith, J.J., Szilard, R.K. and Rachubinski, R.A. (2000) Peroxisome biogenesis in the yeast Yarrowia lipolytica. Cell. Biochem. Biophys. 32, 21-26.

[262] Titorenko, V.I., Eitzen, G.A. and Rachubinski, R.A. (1996) Mutations in the PAY5 gene of the yeast Yarrowia lipolytica cause the accumulation of multiple subpopulations of peroxi-somes. J. Biol. Chem. 271, 20307-20314.

[263] Titorenko, V.I. and Rachubinski, R.A. (2001) The life cycle of the peroxisome. Nat. Rev. Mol. Cell. Biol. 2, 357-368.

[264] Schneiter, R., Brugger, B., Sandhoff, R., Zellnig, G., Leber, A., Lampl, M., Athenstaedt, K., Hrastnik, C., Eder, S., Daum, G., Paltauf, F., Wieland, F.T. and Kohlwein, S.D. (1999) Electro-spray ionization tandem mass spectrometry (ESI-MS/MS) analysis of the lipid molecular species composition of yeast subcellular membranes reveals acyl chain-based sorting/remodeling of distinct molecular species en route to the plasma membrane. J. Cell Biol. 146, 741-754.

[265] Geuze, H.J., Murk, J.L., Stroobants, A.K., Griffith, J.M., Kleijmeer, M.J., Koster, A.J., Verkleij, A.J., Distel, B. and Tabak, H.F. (2003) Involvement of the endoplasmic reticulum in peroxisome formation. Mol. Biol. Cell 14, 2900-2907.

[266] Hart, D.T., Baudhuin, P., Opperdoes, F.R. and de Duve, C. (1987) Biogenesis of the glycosome in Trypanosoma brucei: the synthesis, translocation and turnover of glycosomal polypep-tides. EMBO J. 6, 1403-1411.

[267] Clayton, C.E. (1987) Import of fructose bisphosphate aldolase into the glycosomes of Trypanosoma brucei. J. Cell Biol. 105, 2649-2654.

[268] Clayton, C.E. (1988) Most proteins, including fructose bisphosphate aldolase, are stable in the procyclic trypomastigote form of Trypanosoma brucei. Mol. Biochem. Parasitol. 28, 43-46.

[269] Blattner, J., Swinkels, B., Dorsam, H., Prospero, T., Subramani, S. and Clayton, C. (1992) Glycosome assembly in trypanosomes: variations in the acceptable degeneracy of a COOH-terminal microbody targeting signal. J. Cell Biol. 119, 1129-1136.

[270] Sommer, J.M., Cheng, Q.L., Keller, G.A. and Wang, C.C. (1992) In vivo import of firefly luciferase into the glycosomes of Trypanosoma brucei and mutational analysis of the C-terminal targeting signal. Mol. Biol. Cell 3, 749-759.

[271] Blattner, J., Dorsam, H. and Clayton, C.E. (1995) Function of N-terminal import signals in trypanosome microbodies. FEBS Lett. 360, 310-314.

[272] Peterson, G.C., Sommer, J.M., Klosterman, S., Wang, C.C. and Parsons, M. (1997) Trypanosoma brucei: identification of an internal region of phosphoglycerate kinase required for targeting to glycosomal microbodies. Exp. Parasitol. 85, 16-23.

[273] Wierenga, R.K., Noble, M.E., Vriend, G., Nauche, S. and Hol, W.G.J. (1991) Refined 1.83 A structure of trypanosomal triosephosphate isomerase crystallized in the presence of 2.4 M-ammonium sulphate. A comparison with the structure of the trypanosomal triosephosphate isomerase-glycerol-3-phosphate complex. J. Mol. Biol. 220, 995-1015.

[274] Osinga, K.A., Swinkels, B.W., Gibson, W.C., Borst, P., Veen-eman, G.H., Van Boom, J.H., Michels, P.A.M. and Opperdoes, F.R. (1985) Topogenesis of microbody enzymes: a sequence comparison of the genes for the glycosomal (microbody) and cytosolic phosphoglycerate kinases of Trypanosoma brucei. EMBO J. 4, 3811-3817.

[275] Michels, P.A.M., Marchand, M., Kohl, L., Allert, S., Wierenga, R.K. and Opperdoes, F.R. (1991) The cytosolic and glycosomal isoenzymes of glyceraldehyde-3-phosphate dehydrogenase in Trypanosoma brucei have a distant evolutionary relationship. Eur. J. Biochem. 198, 421-428.

[276] Nyame, K., Do-Thi, C.D., Opperdoes, F.R. and Michels, P.A.M. (1994) Subcellular distribution and characterization of glucosephosphate isomerase in Leishmania mexicana mexicana. Mol. Biochem. Parasitol. 67, 269-279.

[277] Duffieux, F., Van Roy, J., Michels, P.A.M. and Opperdoes, F.R. (2000) Molecular characterization of the first two enzymes of the pentose-phosphate pathway of Trypanosoma brucei. Glucose-6-phosphate dehydrogenase and 6-phosphogluconolactonase. J. Biol. Chem. 275, 27559-27565.

[278] Eilers, M. and Schatz, G. (1986) Binding of a specific ligand inhibits import of a purified precursor protein into mitochondria. Nature 322, 228-232.

[279] Flaspohler, J.A., Rickoll, W.L., Beverley, S.M. and Parsons, M. (1997) Functional identification of a Leishmania gene related to the peroxin 2 gene reveals common ancestry of glycosomes and peroxisomes. Mol. Cell. Biol. 17, 1093-1101.

[280] Flaspohler, J.A., Lemley, K. and Parsons, M. (1999) A dominant negative mutation in the GIM1 gene of Leishmania donovani is responsible for defects in glycosomal protein localization. Mol. Biochem. Parasitol. 99, 117-128.

[281] Mannion-Henderson, J., Flaspohler, J.A., Lemley, K.R., Rickoll, W.L. and Parsons, M. (2000) Isolation and characterization of Leishmania mutants defective in glycosomal protein import. Mol. Biochem. Parasitol. 106, 225-237.

[282] Guerra-Giraldez, C., Quijada, L. and Clayton, C.E. (2002) Compartmentation of enzymes in a microbody, the glycosome, is essential in Trypanosoma brucei. J. Cell Sci. 115, 2651-2658.

[283] Aman, R.A. and Wang, C.C. (1987) Identification of two integral glycosomal membrane proteins in Trypanosoma brucei. Mol. Biochem. Parasitol. 25, 83-92.

[284] Lorenz, P., Maier, A.G., Baumgart, E., Erdmann, R. and Clayton, C. (1998) Elongation and clustering of glycosomes in Trypanosoma brucei overexpressing the glycosomal Pex11p. EMBO J. 17, 3542-3555.

[285] Maier, A., Lorenz, P., Voncken, F. and Clayton, C. (2001) An essential dimeric membrane protein of trypanosome glycosomes. Mol. Microbiol. 39, 1443-1451.

[286] Voncken, F., Van Hellemond, J.J., Pfisterer, I., Maier, A., Hillmer, S. and Clayton, C. (2003) Depletion of GIM5 causes cellular fragility, a decreased glycosome number, and reduced levels of ether-linked phospholipids in trypanosomes. J. Biol. Chem. 278, 35299-35310.

[287] de Walque, S., Kiel, J.A.K.W., Veenhuis, M., Opperdoes, F.R. and Michels, P.A.M. (1999) Cloning and analysis of the PTS-1 receptor in Trypanosoma brucei. Mol. Biochem. Parasitol. 104, 106-119.

[288] Jardim, A., Liu, W., Zheleznova, E. and Ullman, B. (2000) Peroxisomal targeting signal-1 receptor protein PEX5 from Leishmania donovani. Molecular, biochemical, and immunocyt-ochemical characterization. J. Biol. Chem. 275, 13637-13644.

[289] Kumar, A., Roach, C., Hirsh, I.S., Turley, S., de Walque, S., Michels, P.A.M. and Hol, W.G.J. (2001) An unexpected extended conformation for the third TPR motif of the peroxin PEX5 from Trypanosoma brucei. J. Mol. Biol. 307, 271-282.

[290] Furuya, T., Kessler, P., Jardim, A., Schnaufer, A., Crudder, C. and Parsons, M. (2002) Glucose is toxic to glycosome-deficient trypanosomes. Proc. Natl. Acad. Sci. USA 99, 14177-14182.

[291] Moyersoen, J., Choe, J., Kumar, A., Voncken, F.G.J., Hol, W.G.J. and Michels, P.A.M. (2003) Characterization of Try-panosoma brucei PEX14 and its role in the import of glycosomal matrix proteins. Eur. J. Biochem. 270, 2059-2067.

[292] Jardim, A., Rager, N., Liu, W. and Ullman, B. (2002) Peroxisomal targeting protein 14 (PEX14) from Leishmania donovani. Molecular, biochemical, and immunocytochemical characterization. Mol. Biochem. Parasitol. 124, 51-62.

[293] Heise, N. and Opperdoes, F.R. (1999) Purification, localisation and characterisation of glucose-6-phosphate dehydrogenase of Trypanosoma brucei. Mol. Biochem. Parasitol. 99, 21-32.

[294] Wall, M.A., Coleman, D.E., Lee, E., Iniguez-Lluhi, J.A., Posner, B.A., Gilman, A G. and Sprang, S R. (1995) The structure of the G protein heterotrimer Gi 1ß1y2. Cell 83, 1047-1058.

[295] Voegtli, W.C., Madrona, A.Y. and Wilson, D.K. (2003) The structure of Aip1p, a WD repeat protein that regulates Cofilin-mediated actin depolymerization. J. Biol. Chem. 278, 3437334379.

[296] Chudzik, D.M., Michels, P.A.M., de Walque, S. and Hol, W.G.J. (2000) Structures of type 2 peroxisomal targeting signals in two trypanosomatid aldolases. J. Mol. Biol. 300, 697-707.

[297] Choe, J., Moyersoen, J., Roach, C., Carter, T.L., Fan, E., Michels, P.A.M. and Hol, W.G.J. (2003) Analysis of the sequence motifs responsible for the interactions of peroxins 14 and 5, which are involved in glycosome biogenesis in Trypan-osoma brucei. Biochemistry 42, 10915-10922.

[298] Liuzzi, M., Deziel, R., Moss, N., Beaulieu, P., Bonneau, A.M., Bousquet, C., Chafouleas, J.G., Garneau, M., Jaramillo, J., Krogsrud, R.L., Lagace, L., McCollum, R.S., Nawoot, S. and Guindon, Y. (1994) A potent peptidomimetic inhibitor of HSV

ribonucleotide reductase with antiviral activity in vivo. Nature 372, 695-698.

[299] Klenchin, V.A., Allingham, J.S., King, R., Tanaka, J., Marriott, G. and Rayment, I. (2003) Trisoxazole macrolide toxins mimic the binding of actin-capping proteins to actin. Nat. Struct. Biol. 10, 1058-1063.

[300] Gadek, T.R. and Nicholas, J.B. (2003) Small molecule antagonists of proteins. Biochem. Pharmacol. 65, 1-8.

[301] Cochran, A.G. (2001) Protein-protein interfaces: mimics and inhibitors. Curr. Opin. Chem. Biol. 5, 654-659.

[302] Mullen, R.T., Flynn, C.R. and Trelease, R.N. (2001) How are peroxisomes formed. The role of the endoplasmic reticulum and peroxins. Trends Plant Sci. 6, 256-261.

[303] Wang, X.D., Unruh, M.J. and Goodman, J.M. (2001) Discrete targeting signals direct PMP47 to oleate-induced peroxisomes in Saccharomyces cerevisiae. J. Biol. Chem. 276, 10897-10905.

[304] Wang, X., McMahon, M.A., Shelton, S.N., Nampaisansuk, M., Ballard, J.L. and Goodman, J.M. (2004) Multiple targeting modules on peroxisomal proteins are not redundant: discrete functions of targeting signals within Pmp47 and Pex8p. Mol. Biol. Cell 15, 1702-1710.

[305] Honsho, M. and Fujiki, Y. (2001) Topogenesis of peroxisomal membrane protein requires a short, positively charged intervening-loop sequence and flanking hydrophobic segments. Study using human membrane protein PMP34. J. Biol. Chem. 276, 9375-9382.

[306] Jones, J.M., Morell, J.C. and Gould, S.J. (2001) Multiple distinct targeting signals in integral peroxisomal membrane proteins. J. Cell Biol. 153, 1141-1149.

[307] Pause, B., Saffrich, R., Hunziker, A., Ansorge, W. and Just, W.W. (2000) Targeting of the 22 kDa integral peroxisomal membrane protein. FEBS Lett. 471, 23-28.

[308] Brosius, U., Dehmel, T. and Gartner, J. (2002) Two different targeting signals direct human peroxisomal membrane protein 22 to peroxisomes. J. Biol. Chem. 277, 774-784.

[309] Huang, K. and Lazarow, P.B. (1996) Targeting of green fluorescent protein to peroxisomes and peroxisome membranes in S. cerevisiae. Mol. Biol. Cell 7, 2871.

[310] Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F. and Higgins, D.G. (1997) The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 24, 4876-4882.