STRUCTURE-FUNCTION STUDIES OF THE MAMMALIAN PEROXISOMAL MULTIFUNCTIONAL ENZYME TYPE 2 (MFE-2)

ANTTIHAAPALAINEN

Biocenter Oulu andDepartment of Biochemistry,

University of Oulu

OULU 2002

ANTTI HAAPALAINEN

STRUCTURE-FUNCTION STUDIES OF THE MAMMALIAN PEROXISOMAL MULTIFUNCTIONAL ENZYME TYPE 2 (MFE-2)

Academic Dissertation to be presented with the assent ofthe Faculty of Science, University of Oulu, for publicdiscussion in Raahensali (Auditorium L10), Linnanmaa, onNovember 8th, 2002, at 2 p.m.

OULUN YLIOPISTO, OULU 2002

Copyright © 2002University of Oulu, 2002

Supervised byDoctor Tuomo GlumoffProfessor Kalervo Hiltunen

Reviewed byDocent Sarah ButcherDocent Matti Poutanen

ISBN 951-42-6838-5 (URL: http://herkules.oulu.fi/isbn9514268385/)

ALSO AVAILABLE IN PRINTED FORMATActa Univ. Oul. A 389, 2002ISBN 951-42-6837-7ISSN 0355-3191 (URL: http://herkules.oulu.fi/issn03553191/)

OULU UNIVERSITY PRESSOULU 2002

Haapalainen, Antti, Structure-function studies of the mammalian peroxisomalmultifunctional enzyme type 2 (MFE-2) Biocenter Oulu, University of Oulu, P.O.Box 5000, FIN-90014 University of Oulu, Department ofBiochemistry, University of Oulu, P.O.Box 3000, FIN-90014 University of Oulu Oulu, Finland2002

Abstract

Mammalian peroxisomes contain two parallel multifunctional enzymes (MFE), MFE type 1 and MFEtype 2 (MFE-2), which are responsible for the degradation of fatty acids. They both catalyze thesecond and third reactions of the β-oxidation pathway, but through reciprocal stereochemical courses.MFE-2 possesses (2E)-enoyl-CoA hydratase-2 and (3R)-hydroxyacyl-CoA dehydrogenase activities.In addition, the carboxy-terminal part is similar to the sterol carrier protein type 2 (SCP-2).

The purpose of this work was to study the structure-function relationship of functional domainsof mammalian MFE-2 by recombinant DNA technology, enzyme kinetics and X-ray crystallography.The work started with the identification of conserved regions in MFE-2. This information was utilizedwhen dehydrogenase, hydratase-2 and/or SCP-2-like domain were produced as separate recombinantproteins. Subsequently, both dehydrogenase and SCP-2-like domains were crystallized and theircrystal structures were solved.

The structure of the dehydrogenase region of rat MFE-2 contains the basic α/β short-chain alcoholdehydrogenase/reductase (SDR) fold and the four-helix bundle at the dimer interface, which is typicalof dimeric SDR enzymes. However, the structure has a novel carboxy-terminal domain not seenamong the known structures. This domain lines the active site cavity of the neighbouring monomer,reflecting cooperative behaviour within a homodimer.

The monomeric SCP-2-like domain of human MFE-2 has the same fold as rabbit SCP-2. Thestructure includes a hydrophobic tunnel occupied by an ordered Triton X-100 molecule,demonstrating the ligand-binding site. Compared to the unliganded rabbit SCP-2 structure, theposition of the carboxy-terminal helix is different. The movement of this helix in the liganded humanSCP-2-like domain resulted in the exposure of a peroxisomal targeting signal, suggesting ligand-assisted protein import into peroxisomes.

The roles of conserved protic residues in the hydratase-2 region of human MFE-2 were studied bymutating them to alanine. In the first step, the ability of mutated variants to utilize oleic acid in vivowas tested with Saccharomyces cerevisiae fox-2 cells (devoid of endogenous MFE-2). Subsequently,in vitro characterization of the mutant enzymes revealed two amino acid residues, Glu366 andAsp510, vital for hydratase-2 activity. The results indicate that the acid-base catalysis is valid forhydratase-2.

Keywords: β-oxidation, dehydrogenase, hydratase, SCP-2, SDR

Acknowledgements

This work was carried out in the Department of Biochemistry and Biocenter Oulu, University of Oulu, during the years 1997-2002.

I wish to express my deepest gratitude to my supervisors, Dr. Tuomo Glumoff and Professor Kalervo Hiltunen. Their enormous knowledge, support and advice made it possible to complete this study. I am especially grateful that I was allowed to express myself and to make my own decisions whenever possible. I also thank Professor Rik Wierenga for his valuable comments and interest in my project over the years. I also wish to thank the Professors Raili Myllylä and Seppo Vainio and all the other group leaders at the Department of Biochemistry for providing excellent research facilities and a stimulating working atmosphere.

I am grateful to Docent Sarah Butcher and Docent Matti Poutanen for their valuable comments on the manuscript of the thesis, and Sirkka-Liisa Leinonen, Lic. Phil., for the careful revision of the language.

I am deeply indebted to my co-authors, Docent Jorma Jalonen, Dr. Daan van Aalten, Dr. Yong-Mei Qin, Dr. Päivi Pirilä, Dr. Dmitry Novikov, Kristian Koski, M.Sc., Mari Marttila, M.Sc., Gitte Meriläinen, M.Sc., and Seppo Kilpeläinen, M.Sc., for their contributions to our joint publications. Dr. Daan van Aalten deserves my warmest thanks for helping me whenever I was having a hard time with my crystallographic studies and Dr. Yong-Mei Qin is further acknowledged for her guidance at the beginning of my doctoral studies.

All the former and present members of the KH team deserve special thanks for their company over the years. My warmest thanks go to Tiila-Riikka Kiema, Jukka Taskinen, Kristian Koski, Mari Marttila, Tomi Airenne, Anu Mursula, Juha Torkko, Aare Rokka and Kalle Savolainen for their fruitful discussions and valuable help in everyday problems. Especially Jukka, Tomi and Kristian are thanked for their ability to convert “the synchrotron trips” into so much fun.

I wish to extend my sincere thanks to the whole staff of our department. Eeva-Liisa Stefanius, Marika Kamps and Ville Ratas are acknowledged for their excellent technical assistance and Seppo Kilpeläinen, Ari-Pekka Kvist, Jyrki Hänninen and Petri Kursula for setting up and maintaining the computer systems. Virpi Hannus, Helena Heikura, Anneli

Kaattari, Kyösti Keränen, Jaakko Keskitalo, Maila Konttila, Tuula Koret and Pirjo Mustaniemi are appreciated for their kind help with many practical tasks.

There are so many things I want to thank my mother and father for. Their wisdom, care and ever-lasting faith in me have helped me in each step of my life. I also want to thank my sisters and their families for their cheerful company and support. Finally, I owe my loving thanks to my dear Teija for her loving care and patience.

This work was financially supported by grants from the Sigrid Jusélius Foundation, Academy of Finland and Finnish Cultural Foundation.

Oulu, August 2002 Antti Haapalainen

Abbreviations

ABC ATP-binding cassette ACBP acyl-CoA-binding protein AMN adrenomyeloneuropathy ASP acylation-stimulating protein ATP adenosine triphosphate CCALD cerebral childhood adrenoleukodystrophy CD circular dichroism cDNA complementary DNA CG complementation group CoA coenzyme A COOH-terminus carboxyl terminus d h SCP-2L SCP-2-like domain of human MFE-2 DHA docosahexaenoic acid DHCA 3α,7α-dihydroxy-5β-cholestanoic acid dh SCP-2L dehydrogenase region of rat MFE-2 DNA deoxyribonucleic acid FABP fatty acid binding protein FAD flavine adenine dinucleotide FADH2 reduced form of flavine adenine dinucleotide 17β-HSD 17β-hydroxysteroid dehydrogenase HsMFE-2(dh ) human MFE-2 without dehydrogenase region IRD infantile Refsum disease kDa kilodalton(s) LCFA long-chain fatty acid MAD multiwavelength anomalous dispersion MFE multifunctional enzyme MFE-1 multifunctional enzyme type 1 MFE-2 multifunctional enzyme type 2 NAD+ nicotinamide adenine dinucleotide NADH reduced form of nicotinamide adenine dinucleotide NALD neonatal adrenoleukodystrophy

NH2-terminus amino terminus nsLTP non-specific lipid transfer protein PCR polymerase chain reaction PDB Protein Data Bank PMP peroxisomal membrane protein PPRE peroxisome proliferator response element PTS peroxisomal targeting signal PTS1 peroxisomal targeting signal type 1 PTS2 peroxisomal targeting signal type 2 RCDP rhizomelic chondrodysplasia punctata RNA ribonucleic acid SCP-2 sterol carrier protein type 2 SCP-2L sterol carrier protein type 2 like domain SCPx sterol carrier protein x SDR short-chain alcohol dehydrogenase/reductase SDS-PAGE sodium dodecyl sulphate polyacrylamide gel electrophoresis Se-Met selenomethionine THCA 3α,7α,12α-trihydroxy-5β-cholestanoic acid VLCAS very-long-chain acyl-CoA synthetase VLCFA very-long-chain fatty acid X-ALD X-linked adrenoleukodystrophy ZS Zellweger syndrome Å ångström, 10-10 m

List of original articles

This thesis is based on the original articles, which are referred to in the text by their Roman numerals: I Qin YM, Haapalainen AM, Kilpeläinen SH, Marttila MS, Koski MK, Glumoff T,

Novikov DK & Hiltunen JK (2000) Human peroxisomal multifunctional enzyme type 2: site-directed mutagenesis studies show the importance of two protic residues for 2-enoyl-CoA hydratase 2 activity. J Biol Chem 275: 4965-4972.

II Haapalainen AM, van Aalten DMF, Meriläinen G, Jalonen JE, Pirilä P, Wierenga

RK, Hiltunen JK & Glumoff T (2001) Crystal structure of the liganded SCP-2-like domain of human peroxisomal multifunctional enzyme type 2 at 1.75 Å resolution. J Mol Biol 313: 1127-1138.

III Haapalainen AM, Koski MK, Qin YM, Hiltunen JK & Glumoff T (200X) Binary

structure of the two-domain (3R)-hydroxyacyl-CoA dehydrogenase from Rattus norvegicus peroxisomal multifunctional enzyme type 2 at 2.38 Å resolution. (Under revision)

Contents

Acknowledgements Abbreviations List of original articles Contents 1 Introduction ...................................................................................................................13 2 Review of the literature..................................................................................................15

2.1 Fatty acids as an energy store .................................................................................15 2.1.1 Transport of fatty acids ....................................................................................16 2.1.2 Compartmentalization of fatty acid oxidation .................................................16 2.1.3 Activation of fatty acids prior to β-oxidation...................................................18

2.2 Peroxisomal β-oxidation system.............................................................................19 2.2.1 Peroxisomes.....................................................................................................19 2.2.2 Peroxisomal targeting signals ..........................................................................20

2.2.2.1 Additional targeting signals ......................................................................21 2.2.3 Protein translocation into peroxisomes............................................................22 2.2.4 Transport of fatty acyl-CoAs across peroxisomal membrane ..........................23 2.2.5 Peroxisomal β-oxidation cycle ........................................................................24

2.2.5.1 Acyl-CoA oxidases ...................................................................................24 2.2.5.2 Second and third steps are catalyzed by two multifunctional

enzymes....................................................................................................26 2.2.5.3 3-Ketoacyl-CoA thiolases .........................................................................28 2.2.5.4 Auxiliary enzymes ....................................................................................29

2.3 Multifunctionality of MFE-2 ..................................................................................30 2.3.1 Domain structure of various MFE-2s...............................................................30 2.3.2 Catalytic properties of MFE-2 .........................................................................31 2.3.3 Tissue distribution............................................................................................34 2.3.4 (3R)-Hydroxyacyl-CoA dehydrogenase, a member of the

short-chain alcohol dehydrogenase/reductase superfamily.............................35 2.3.5 Sterol carrier protein type 2 .............................................................................38

2.3.5.1 Other fatty acid/fatty acyl-CoA-binding proteins .....................................39 2.4 Peroxisomal β-oxidation deficiencies.....................................................................41

3 Outlines of the present study .........................................................................................43 4 Materials and methods...................................................................................................44

4.1 cDNA cloning .........................................................................................................44 4.1.1 Cloning of the (3R)-hydroxyacyl-CoA dehydrogenase region

from rat MFE-2 (III) .......................................................................................44 4.1.2 Cloning of human MFE-2 (HsMFE-2) and its variants,

HsMFE-2(dh∆) and d∆h∆SCP-2L (I, II) ........................................................45 4.2 Site-directed mutagenesis (I) ..................................................................................46 4.3 Complementation of Saccharomyces cerevisiae fox-2 with human

MFE-2 and its mutated variants (I) ........................................................................46 4.4 Production of recombinant proteins (I, II) ..............................................................46

4.4.1 Production of selenomethionine-labelled recombinant (3R)-hydroxyacyl-CoA dehydrogenase (III) ...........................................................47

4.5 Protein purification (I-III) .......................................................................................47 4.6 Enzyme assays (I, III) .............................................................................................47 4.7 Dynamic light scattering and CD-spectroscopic measurements (I) ........................48 4.8 Surface plasmon resonance measurements (II).......................................................48 4.9 Crystallization, data collection and processing (II, III)...........................................48 4.10 Structural determination, model building and refinement (II, III) ........................49 4.11 Other methods.......................................................................................................49

5 Results ...........................................................................................................................51 5.1 Binary structure of the two-domain (3R)-hydroxyacyl-CoA

dehydrogenase region of rat MFE-2 (III) ...............................................................51 5.2 Site-directed mutagenesis studies on (2E)-enoyl-CoA hydratase-2

derived from human MFE-2 (I)..............................................................................55 5.3 Crystal structure of the liganded SCP-2-like domain of human

MFE-2 (II)..............................................................................................................56 6 Discussion......................................................................................................................59

6.1 Structural studies on the (3R)-hydroxyacyl-CoA dehydrogenase region of rat MFE-2 (III)........................................................................................59

6.2 Importance of two protic residues for (2E)-enoyl-CoA hydratase-2 activity (I)............................................................................................................60

6.3 Ligand binding to the SCP-2-like domain of human MFE-2 (II)............................61 7 Conclusions ...................................................................................................................63 8 References .....................................................................................................................65

1 Introduction

Fatty acids, as such, are energy-rich molecules, from which energy is withdrawn during periods of fasting and starvation. In addition to being structural components of biomembranes and serving as energy stores, fatty acids and their derivatives act as ligands for a number of transcription factors and modifiers of enzymatic activity and other protein functions to sustain vital biochemical processes. During energy conversion, fatty acids are degraded by the oxidative pathways, such as α-oxidation, β-oxidation and -oxidation, which take place in peroxisomes, mitochondria and/or endoplasmic reticulum. However, β-oxidation is a major process by which fatty acids are oxidized, and this oxidation occurs in mammals both in mitochondria and in peroxisomes (Lazarow & de Duve 1976). Prior to oxidation, free fatty acids need to be activated to their CoA esters. Even though the chemical modifications of fatty acyl-CoAs during peroxisomal and mitochondrial β-oxidations are somewhat similar, the enzymes involved are not. Another difference is that peroxisomal β-oxidation can operate in relative independence of the cellular energy status. On the contrary, mitochondrial β-oxidation is directly linked with the synthesis of ATP.

Saturated fatty acyl-CoAs are β-oxidized by four sequential reactions. This cycle includes oxidation/dehydrogenation, hydration, dehydrogenation and, finally, thiolytic cleavage. To overcome double bonds in unsaturated fatty acids, additional auxiliary enzymes are needed prior to the β-oxidation cycle. Each turn of the cycle shortens the fatty acid chain by two carbons. Unlike the mitochondrial system, the peroxisomal β-oxidation pathway does not proceed to completion in mammals, and the chain-shortened acyl-CoA, similarly to acetyl-CoA, is transported to mitochondria to be further oxidized (for reviews, see Osmundsen et al. 1991, Eaton et al. 1996, Hiltunen & Qin 2000). In mitochondria, acetyl-CoAs are fed into a citric acid cycle, from which the high electrons pass via NADH and FADH2 to the respiratory chain, eventually forming ATP and H2O. Alternatively, in liver, acetyl-CoA may condense with itself to form ketone bodies, which serve as fuel for other tissues, such as the skeletal and cardiac muscle and brain, during starvation.

All the characterized β-oxidation pathways appear to have a multifunctional enzyme (MFE). Although all of them catalyze the second and third reaction of the pathway, the subunit composition and the associated enzymatic activities vary, depending on the

14

species and organelle. The mammalian peroxisome contains two MFEs, MFE type 1 (MFE-1) and MFE type 2 (MFE-2), with different stereochemical specificities (Qin et al. 1997a). MFE-2 acts on ∆2 carboxylates, resulting in an intermediate with the 3-hydroxyl group in the R-configuration, which is then dehydrogenated by the same polypeptide (Hiltunen et al. 1992). This study aimed to reveal the structural basis of mammalian MFE-2. The NH2-terminal (3R)-hydroxyacyl-CoA dehydrogenase region and the COOH-terminal SCP-2-like domain were crystallized and their crystal structures were solved. Furthermore, the function of the central (2E)-enoyl-CoA hydratase-2 region was investigated by applying recombinant DNA technology and enzyme kinetics.

2 Review of the literature

2.1 Fatty acids as an energy store

Fatty acids have a central role in a variety of cellular processes, including energy storage and the synthesis of cellular membranes. In addition, long-chain fatty acids, through their metabolites (e.g., pheromones, prostaglandins, leukotrienes, thromboxanes, platelet activating factor), serve as intracellular signalling molecules. Furthermore, fatty acids have been shown to regulate the expression of genes involved in lipid metabolism and cell differentiation (Distel et al. 1992a, Grimaldi et al. 1992, for a review, see Grimaldi 2001). Disturbances in fatty acid metabolism and regulation, especially those involving fatty acid synthesis and oxidation, may contribute to hyperlipidemia, obesity, insulin resistance and atherosclerosis (Spiegelman & Flier 1996, Unger et al. 1999).

Higher animals, under fed conditions, preferentially burn carbohydrate to generate ATP. When the energy (glucose) intake is abundant, fatty acid synthesis is enhanced and fatty acid oxidation reduced. As a consequence, excess carbohydrate is converted into fatty acids, which are then stored as a triacylglycerol, in which fatty acids are connected to glycerol via ester linkage. These hydrophobic triacylglycerols account for most of the volume of the adipocytes, the large cells specialized for fat storage in adipose tissues (for a review, see Ramsay 1996). When glucose availability is low during periods of starvation, the triacylglycerols stored in adipose tissue are hydrolyzed to free fatty acids and mobilized into plasma. The rate at which triacylglycerides are hydrolysed is determined by lipoprotein lipase and insulin, which is the principal hormone that regulates lipoprotein lipase (Seitz et al. 1977, Eaton et al. 1996, Hashimoto et al. 1999a, Kersten et al. 1999, Leone et al. 1999, Hashimoto et al. 2000). By contrast, the acylation-stimulating protein (ASP) pathway modulates the rate at which fatty acids are taken up and converted to triacylglycerides by adipocytes. Under certain circumstances, however, reduction of the activity of the ASP pathway may negatively affect the first step of the process. Consequently, ASP also influences the rate at which fatty acids are released by adipocytes (for a review, see Sniderman et al. 2000).

16

2.1.1 Transport of fatty acids

Since free fatty acids, above their solubility limit, form insoluble acid soaps (Hicks & Gebecki 1977, Cistola et al. 1986), the transport of fatty acids in the body must interfere with other biological molecules. For this reason, in the blood, free fatty acids are transported mainly in complexes with albumin, and a only very small fraction is bound to plasma lipoproteins (for a review, see Spector 1984, Hamilton & Kamp 1999).

The mechanisms by which fatty acids cross the plasma membrane of mammalian cells are a topic of active investigation and debate. According to one model, the movement of most fatty acids through the plasma membrane is rapid and spontaneous, suggesting that fatty acids can enter cells by free diffusion rather than by protein-mediated mechanisms (DeGrella & Light 1980, Kamp & Hamilton 1992, Veerkamp 1995, Ek-Von Mentzer et al. 2001, Hamilton et al. 2001). Other studies have described the presence of a protein-mediated translocation system, where the transport of fatty acid across the plasma membrane is facilitated by a fatty acid transport protein, a plasma membrane protein, possibly requiring energy or the co-transportation of Na+ (Berk et al. 1997, Stuhlsatz-Krouper et al. 1998). These putative fatty acid transport proteins comprise a large evolutionarily conserved family of proteins present in organisms ranging from mycobacterium to man (Hirsch et al. 1998).

Once free fatty acid has reached the inner leaflet of the plasma membrane, it must move to intracellular sites for utilization. Cytosolic fatty acid-binding proteins (FABP) (Ockner et al. 1972), which are 14-15 kDa in size, shuttle the intracellular aqueous insoluble fatty acids to different cellular compartments (Tipping & Ketterer 1981, Weisiger 1996, Storch & Thumser 2000). To date, 14 members of the FABP gene family have been identified and divided into two main classes according to their ability to interact with membranes. The members of one class transfer fatty acids to and from membranes involving interaction between a positively charged region of the binding protein and negative charges on the membrane surface (Herr et al. 1996, Hsu & Storch 1996, Gericke et al. 1997, Corsico et al. 1998, Thumser & Storch 2000). The second class of FABP molecules lack the ability to interact with membranes and are therefore only able to bind fatty acid monomers already in aqueous solution. So far, eight different three-dimensional structures of FABPs have been determined from human, bovine, rat and mouse, revealing highly conserved tertiary structures; ten antiparallel β-strands form a flattened barrel with a hydrophobic pocket in the centre that accommodates hydrophobic ligands.

2.1.2 Compartmentalization of fatty acid oxidation

Oxidation of fatty acids occurs in three subcellular organelles via α-, β- and -oxidation, depending on the species. In mammals, both mitochondria and peroxisomes carry out fatty acid β-oxidation, which is the dominant oxidative pathway (Fig. 1). The β-oxidation of fatty acids in mammals will be discussed in more detail later. Yeast degrades fatty acids via β-oxidation only in peroxisomes (Kunau et al. 1988). Peroxisomal β-oxidation

17

Fig. 1. Simplified scheme of the different roles of peroxisomes and mitochondria in cellular fatty acid β-oxidation in mammals. (Modified from Wanders et al. 2001)

is also the primary pathway of fatty acid catabolism in plants. There is also evidence that β-oxidation of short-branched-chain 2-oxo acids takes place in plant mitochondria. In this pathway, the branched-chain amino acids are broken down and the end products of the pathway are converted to acyl-CoA esters. The degradation of these esters requires β-oxidation (for a review, see Graham & Eastmond 2002).

Mammalian peroxisomes degrade fatty carboxylates not only via β-oxidation but also by α-oxidation (Fig. 2), since fatty acids with a methyl group at the β-position cannot undergo direct β-oxidation. This applies to, for instance, phytanic acid (3,7,11,15-tetramethylhexadecanoic acid), which, in the human body, is mainly derived from dairy products. Phytanic acid is converted from phytol, the side chain of chlorophyll. The α-oxidation of phytanic acid includes activation of phytanic acid to phytanoyl-CoA by very-long-chain acyl-CoA synthetase, which is localized on the peroxisomal membrane (Watkins et al. 1994, Pahan & Singh 1995, Watkins et al. 1996). Once inside the peroxisome, phytanoyl-CoA is converted into 2-hydroxyphytanoyl-CoA by phytanoyl-CoA hydroxylase (Mihalik et al. 1995, Croes et al. 1996, Jansen et al. 1996, 1997, Mihalik et al. 1997). Subsequent studies resolved a pathway in which 2-hydroxyphytanoyl-CoA first undergoes cleavage to pristanal and formyl-CoA (Verhoeven et al. 1997, Croes et al. 1997, Jansen et al. 1999, Foulon et al. 1999), after which pristanal is oxidized to pristanic acid via an as yet undefined aldehyde dehydrogenase. Prior to β-oxidation, the 2-methyl-branched fatty acid, pristanic acid, is activated to pristanoyl-CoA. The α-oxidation of phytanic acid is not stereoselective. Consequently, two isomers are the end products of the pathway. However, peroxisomal oxidases accept only 2S-isomers, and racemase is therefore needed to convert 2R- into 2S-isomer (see the chapter titled “Auxiliary enzymes”).

18

Fig. 2. α-Oxidation of phytanic acid. The numbers refer to the following enzymes: 1, very-long-chain acyl-CoA synthetase; 2, phytanoyl-CoA hydroxylase; 3, 2-hydroxyphytanoyl-CoA lyase; 4, long-chain aldehyde dehydrogenase (not properly characterized); 5, very-long-chain acyl-CoA synthetase; 6, 2-methylacyl-CoA racemase. 2S-pristanoyl-CoA is further oxidized via the peroxisomal β-oxidation pathway. (Modified from Van Veldhoven et al. 2001)

Long-chain fatty acids are also ligands for -oxidation. The initial step in -oxidation is -hydroxylation of fatty acid in smooth endoplasmic reticulum catalyzed by lauric acid

(C12:0) -hydroxylase (Hardwick et al. 1987). Rat and human -hydroxylases do not show preference for longer-chain fatty acids, such as oleic (C18:1) and arachidonic acid (C20:4) (Dierks et al. 1998). In contrast, rabbit hydroxylases appear to be highly active towards arachidonic acid (Johnson et al. 1996). In cytosol, the -hydroxy fatty acid produced by the initial hydroxylation reaction is dehydrogenated to dicarboxylic acid by alcohol dehydrogenase and aldehyde dehydrogenase. Especially long-chain dicarboxylic acids formed via the -oxidation pathway regulate peroxisomal β-oxidation enzymes (Kaikaus et al. 1993); it has been shown that long-chain dicarboxyl-CoAs are almost exclusively further oxidized by the peroxisomal alcohol oxidase of the classical β-oxidation pathway in human and, to some extent, via pristanoyl-CoA oxidase in rat (for reviews, see Reddy & Mannaerts 1994, Mannaerts et al. 2000). The recent finding on long-chain fatty acid oxidase genes in Candida species led to the identification of a family of genes involved in lipid -oxidation in yeast with homologues in plants and bacteria (Vanhanen et al. 2000). The identified enzymes were membrane-associated with unknown subcellular location.

2.1.3 Activation of fatty acids prior to β-oxidation

The enzymes of β-oxidation all act on CoA esters, which means that prerequisite for β-oxidation is the ATP-dependent thioesterification of fatty acids to their fatty acyl-CoAs,

19

catalysed by acyl-CoA synthetases (Aas & Bremer 1968, Aas 1971). There are specific acyl-CoA synthetase isoenzymes for long-chain and very-long-chain fatty acids (VLCFA), which are localized in endoplasmic reticulum, peroxisomal membrane and/or outer mitochondrial membrane (Kornberg & Pricer 1953, Aas 1971, Singh & Poulos 1988, Lazo et al. 1990, Suzuki et al. 1990, Subramani 1998). Short- and medium-chain fatty acids are activated only intramitochondrially by two different acyl-CoA synthetases. One synthetase utilizes preferentially octanoate, whereas butyrate is the fatty acid most preferred by the other (Scaife & Tichivangana 1980, Fujino et al. 2001). In bacteria, the activation process requires the cooperative action of the outer membrane-bound fatty acid transport protein FadL and the inner membrane-associated fatty acyl-CoA synthetase (for a review, see DiRusso & Black 1999).

Peroxisomal membrane contains long-chain acyl-CoA synthetase. The long-chain acyl-CoA synthetase is also localized in the mitochondrial outer membrane and in the endoplasmic reticulum, and the catalytic site of these synthetases is exposed to cytosol (Hesler et al. 1990, Lazo et al. 1990, Suzuki et al. 1990, Subramani 1998). Very-long-chain acyl-CoA synthetase (VLCAS), which activates VLCFAs and has sequence similarity to the fatty acid transport protein (Schaffer & Lodish 1994), is present in peroxisomes and in the endoplasmic reticulum but is absent from mitochondria (Uchiyama et al. 1996). VLCAS and long-chain acyl-CoA synthetase differ with respect to their molecular and catalytic properties, and the absence of the former synthetase in mitochondria may partly explain why VLCFAs are β-oxidized in peroxisomes. Indirect immunofluorescent and electron microscopic studies showed that peroxisomal VLCAS is localized on the matrix side rather than the cytosolic face of peroxisomal membrane (Smith et al. 2000). Furthermore, the human variant of VLCAS was described to activate not only VLCFAs but also long- and branched-chain fatty acids (Steinberg et al. 1999). In rat, bile acid intermediates, which also have a branched methyl group, were shown to be activated by a separate enzyme, trihydroxycoprostanoyl-CoA synthetase (Van Veldhoven et al. 1996).

2.2 Peroxisomal β-oxidation system

2.2.1 Peroxisomes

Peroxisomes, which are single-membrane organelles, are found in yeast, plants and mammals, and they are present within almost all cell types. Most commonly, peroxisomes appear spherical in shape in electron micrographs and have a mean diameter of 0.2 to 1.0 µm with a crystalloid core largely composed of urate oxidase (Volkl et al. 1988). The size and shape depend on the tissue studied and the metabolic circumstances of the species. For instance, in fibroblasts and glucose-grown yeast cells, peroxisomes are smaller. In contrast, larger and spherical peroxisomes frequently interconnected by tubular elements as well as convoluted cup-shaped structures are found in liver and preputial glands,

20

respectively. Some peroxisomes are interconnected and clustered (Mendis-Handagama et al. 1990) in a way that supports the concept of a peroxisomal reticulum.

The functions of peroxisomes are often specialized by organism and cell type. In mammals, two major functions of peroxisomes are part of lipid metabolism. Namely, peroxisomes catalyse the β-oxidation of fatty acids and the initial steps in the biosynthesis of plasmalogens, which are phospholipids present in myelin. Especially, peroxisomal β-oxidation is essential for the catabolism of a variety of substrates that are not oxidised by mitochondria, such as VLCFAs, dicarboxylic and branched-chain fatty acids and cholesterol, whose side chain undergoes obligatory β-oxidative cleavage upon conversion to bile acids (Lazarow & de Duve 1976, for reviews, see Wanders & Tager 1998, Wanders et al. 2001). Although eukaryotic cells can survive without peroxisomes, defects in peroxisome biogenesis have significant metabolic and developmental consequences (Subramani 1993). Peroxisomes are essential for normal human development, since defects in peroxisome assembly cause disorders of peroxisome biogenesis (Lazarow & Moser 1994). These diseases are characterized by severe developmental abnormalities, particularly within the nervous system, and usually result in death in early infancy. In yeast, such defects eliminate growth on fatty acids, an important

carbon and energy source in natural environments (Hettema & Tabak 2000). A third important peroxisomal function is respiration, which involves the metabolism of hydrogen peroxide. Hydrogen peroxide, which is formed in reactions catalyzed by various oxidases, is efficiently decomposed by catalase. The energy released in this reaction is dissipated as heat (de Duve & Baudhuin 1966).

2.2.2 Peroxisomal targeting signals

Since peroxisomes have no genome, the proteins destined for peroxisomes are synthesized on free ribosomes in cytosol and contain a peroxisomal targeting signal (PTS), which directs them to peroxisomes; this applies to both membrane and internal matrix proteins (Fujiki et al. 1984, Miura et al. 1984, Rachubinski et al. 1984, Lazarow & Fujiki 1985, Borst 1989, Subramani 1993). Targeting to the peroxisomal matrix is mediated by two cis-acting sequences, PTS1 (Gould et al. 1987, 1988, 1989, 1990a, 1990b, Aitchison et al. 1991, Keller et al. 1991, Aitchison et al. 1992, Blattner et al. 1992, Didion & Roggenkamp 1992, Distel et al. 1992b, Hansen et al. 1992, Miura et al. 1992, Swinkels et al. 1992) and PTS2 (Tsukamoto et al. 1994). At first, the existence of a tripeptide in the COOH-terminus with the consensus (S/A/C)-(K/R/H)-L was characterized and named PTS1 (Gould et al. 1989). Subsequent studies revealed a range of additional functional sequences of PTS1 much larger than expected as well as significant differences between species (Kragler et al. 1998). For example, in rat trihydroxycoprostanoyl-CoA and pristanoyl-CoA oxidases the COOH-terminal –HKM and –SQL sequences are used as PTS1s, respectively (Baumgart et al. 1996a, Vanhooren et al. 1996). The first example of tetrapeptide PTS1 was that of human catalase (-KANL) (Purdue et al. 1996), where lysine has the greatest functional importance. Subsequently,

21

mouse and human 2-methylacyl-CoA racemases were found to contain a critical basic residue at position -4 as well (Amery et al. 2000, Kotti et al. 2000).

The nonapeptides of the PTS2 signal have a broad consensus sequence of (R/K)-(L/V/I)-(XXXXX)-(H/Q)-(L/A/F) in the NH2-terminal part of some peroxisomal matrix proteins. Despite interspecific variations, the five-amino-acid linker between the conserved dipeptides seems to be critical (Tsukamoto et al. 1994, Glover et al. 1994, Flynn et al. 1998). Several studies have pointed out the special importance of Arg1, His8 and Leu9 for targeting into peroxisomes (Flynn et al. 1998). In the crystal structures of peroxisomal fructose-1,6-bisphosphate aldolases from Leishmania mexicana [Protein Data Bank (PDB) accession number 1EPX] and Trypanosoma brucei (PDB accession number 1F27), the PTS2 signal sequences from two monomers interact with each other, so that there are two “PTS2 dimers” per aldolase tetramer (Chudzik et al. 2000). The structures indicated that the exposed Arg1 may play an important role in interacting with Pex7p, which is the cytosolic receptor for peroxisomal proteins targeted by PTS2s (see the chapter “Protein translocation into peroxisomes”), and Leu9 is needed for the dimerization of the signalling peptide. Since several other proteins with the PTS2 signal are dimers or higher multimers, the dimeric PTS2 suggests functional importance.

2.2.2.1 Additional targeting signals

There is also evidence of additional PTS signals for membrane and matrix proteins. Similarly to matrix proteins, integral peroxisomal membrane proteins (PMP) are synthesized on free polyribosomes and imported posttranslationally from the cytosol

(Fujiki et al. 1984, Diestelkötter & Just 1993, Imanaka et al. 1996), though their import does not require the hydrolysis of ATP (Diestelkötter & Just 1993). It has also been speculated that some of the peroxisomal membrane proteins are transported via the endoplasmic reticulum (Bodnar & Rachubinski 1991, Titorenko & Rachubinski 1998). In any case, integral PMPs lack functional PTS1 or PTS2, and their import is hence independent of the PTS1 and PTS2 receptors (Chang et al. 1999, Hettema et al. 2000). Integral PMPs are therefore thought to be imported into peroxisomes by a targeting mechanism different from that used by peroxisomal matrix proteins.

The studies done with Candida boidinii (C. boidinii) PMP47 (Dyer et al. 1996) revealed that the targeting information lies between the transmembrane domains four and five in a positively charged intraperoxisomal loop. The loop is necessary and sufficient for targeting to peroxisomes. The human homologue of C. boidinii PMP47, PMP34, was shown to contain two additional transmembrane regions, either of which was sufficient for targeting to the peroxisome membrane (Honsho & Fujiki 2001, Jones et al. 2001). The hydrophilic peptide, XX(K/R)(K/R)3-7X(T/S)XX(D/E)X (Dyer et al. 1996), adjacent to at least one transmembrane domain, was observed with other proteins, such as PMP70, Pex3p and PMP22 (Kammerer et al. 1998, Ghaedi et al. 2000, Pause et al. 2000, Brosius et al. 2002).

Acyl-CoA oxidases from Saccharomyces cerevisiae (S. cerevisiae) and Candida tropicalis (C. tropicalis) possess another uncharacterized targeting phenomenon. Both of

22

these oxidases lack PTS1 and PTS2, but are still located in the peroxisomal matrix (Small et al. 1988, Skoneczny & Lazarow 1998). This might indicate a third branch, PTS3, for peroxisomal matrix proteins. Nevertheless, the peroxisomal import of S. cerevisiae acyl-CoA oxidase was shown to be dependent on binding to Pex5p, the PTS1 import receptor (Klein et al. 2002).

2.2.3 Protein translocation into peroxisomes

Several studies have shown the importance of peroxins, which are responsible for peroxisome biogenesis. Currently, more than twenty complementation groups of yeast mutants (pex mutants) have been produced, and twelve complementation groups of human peroxisome biogenesis disease patients have been identified (for a review, see Sacksteder & Gould 2000). Because empty peroxisomes, called peroxisomal ghosts, exist in the cells of almost all of these groups, most of the identified peroxins are probably involved in the import of soluble peroxisomal enzymes from cytosol into the peroxisomal matrix (Shimozawa et al. 1998, South & Gould 1999, Hettema et al. 2000). Regarding the import process itself, it has been shown that ATP hydrolysis (Imanaka et al. 1987, Wendland & Subramani 1993) and cytosolic factors (Wendland & Subramani 1993), including heat shock proteins (Walton et al. 1994), are required. In this respect, it has been suggested that the hydrolysis of ATP is carried out by the ATP-binding proteins, such as PMP70 (Kamijo et al. 1990), or a newly identified ATPase (Koenig et al. 2002) in the peroxisomal membranes.

The two targeting signals, PTS1 and PTS2, are recognised by different PTS-specific receptors. PTS1 interacts directly with its cytosolic receptor, Pex5p. In mammals, including humans, two isoforms of Pex5p, termed Pex5pS and Pex5pL, with an internal 37 amino acid insertion have been identified (Braverman et al. 1998, Otera et al. 1998). Pex5pS and Pex5pL form homomeric as well as heteromeric dimers (Otera et al. 2000). The importance of the interaction between PTS1 and tetratricopeptide repeats in Pex5p has been characterized by site-directed mutagenesis (Otera et al. 2000, Szilard & Rachubinski 2000, Klein et al. 2001) and crystallographic (Gatto et al. 2000a,b) studies. The interaction between Pex5p and S. cerevisiae acyl-CoA oxidase, however, involves novel contact sites in both proteins. The interaction region in Pex5p is located in a defined area of the NH2-terminal part of the protein outside the tetratricopeptide repeat domain involved in PTS1 recognition; the interaction site in S. cerevisiae acyl-CoA oxidase is located internally and not at the COOH-terminus, where PTS1 is normally found (Klein et al. 2002). Pex7p, on the other hand, interacts with the PTS2 nonapeptide and is an important cytosolic receptor for peroxisomal proteins targeted by PTS2s (Zhang & Lazarow 1996, Rehling et al. 1996, Elgersma et al. 1998).

Both PTS receptors, Pex5p and Pex7p, are recognised by two peroxisomal integral membrane proteins called Pex13p and Pex14p. Pex13p binds to the PTS1 receptor, Pex5p, with its cytosolic domain (Gould et al. 1996, Elgersma et al. 1996, Erdmann & Blobel 1996), and Pex14p binds to the PTS2 receptor, Pex7p (Albertini et al. 1997). Since Pex5p additionally interacts with other membrane peroxins, such as Pex14p,

23

Pex10p, Pex12p and Pex8p, and Pex7p remains cytosolic in mutants lacking Pex14p and/or Pex13p, the way in which the translocation machinery acts is not fully understood (for a review, see Kiel & Veenhuis 2000). Moreover, Pex18p and Pex21p were identified as key components in the targeting of the Pex7p-cargo complex to peroxisomes (Purdue et al. 1998). Strong evidence suggests that both cytosolic receptors, Pex5p and Pex7p, enter the peroxisomal matrix with their cargo protein. In this respect, the interaction with Pex5pL, but not with Pex5pS, is crucial for the Pex7p-cargo complex to be translocated inside peroxisomes (Matsumura et al. 2000, Otera et al. 2000). Furthermore, Pex5p and Pex7p not only bind cargo and deliver it to the peroxisome membrane, but also participate in multiple rounds of entry into the peroxisome matrix and export to cytosol, indicating receptor recycling (Braverman et al. 1998, Rehling et al. 2000, Dammai & Subramani 2001).

So far, four peroxins have been implicated in the process of proper PMP import: Pex3p (Shimozawa et al. 2000), Pex16p (South & Gould 1999), Pex17p and Pex19p (Matsuzono et al. 1999), although the role of Pex17p is still controversial (Hettema et al. 2000, Harper et al. 2002). Moreover, the NH2-terminal 50 amino acid residues of Pex3p were associated with the formation of vesicular membrane structures, pre-peroxisomes. These structures also contained Pex14p, and the membrane of the vesicles was obviously donated by the nuclear membrane (Faber et al. 2002).

Peroxisomes are capable of importing remarkably large structures, even oligomers of folded proteins. More interestingly, gold beads 4-9 mm in diameter, coated with PTS1 and then injected into cultured mammalian cells, were taken up by peroxisomes (Walton et al. 1995). Some multimeric proteins appear to oligomerize after import (Evers et al. 1994, 1996), and other proteins appear to be imported into peroxisomes only when they first form dimers in the cytosol (Leiper et al. 1996, Titorenko et al. 1998).

2.2.4 Transport of fatty acyl-CoAs across peroxisomal membrane

Continuous β-oxidation depends on the availability of acyl-CoA, NAD+, NADPH, free CoA and export of acetyl groups. This implies the existence of specialized metabolite transport systems in the peroxisomal membrane. In the mitochondria, the activated long-chain fatty acid esters, long-chain acyl-CoAs, are transported by a carnitine-dependent mechanism through the inner membrane into the matrix, where β-oxidation takes place (for a review, see Kerner & Hoppel 2000). In contrast, the role of the carnitine-associated pathway in the peroxisomal membrane is not completely known, and it has been speculated that fatty acyl-CoAs diffuse freely across the membrane. However, two carnitine acyltransferases, one soluble and the other membrane-associated, were isolated from rat liver peroxisomes (Singh et al. 1996). Moreover, the carnitine:acylcarnitine exchange carrier, first characterized in mitochondria, was demonstrated immunologically in peroxisomal membranes as well (for a review, see Ramsay 2000). Besides this, inside peroxisomes, human carnitine octanoyltransferase (COT) was found to convert one of the end products of the peroxisomal β-oxidation of pristanic acid, 4,8-dimethylnonanoyl-CoA, to its corresponding carnitine ester, which is required for its transport into the

24

mitochondrion, where further β-oxidation takes place (Ferdinandusse et al. 1999). The permeability barrier of the peroxisomal membrane is further supported by the identification of the Pxa1p-Pxa2p dimer, Pex11p, and YPR128cp in S. cerevisiae responsible for the transport of long-chain acyl-CoAs, medium-chain acyl-CoAs and ATP, respectively (van Roermund et al. 2000, Nakagawa et al. 2000, van Roermund et al. 2001). In humans, one of the four peroxisomal membrane half ABC transporters was characterized to be vital for the peroxisomal translocation of VLCFAs because non-functionality of this transporter resulted in elevated plasma levels of VLCFAs in the patients (for a review, see Smith et al. 1999). However, in view of the finding that the mammalian peroxisomal membrane is highly permeable to H+, it can be concluded that peroxisomes do not regulate their pH independently (Jankowski et al. 2001).

2.2.5 Peroxisomal β-oxidation cycle

2.2.5.1 Acyl-CoA oxidases

Acyl-CoA oxidase (EC 1.3.3.6) is a flavoenzyme that contains one molecule of noncovalently bound flavine adenine dinucleotide (FAD) per subunit as the prosthetic group and catalyzes the initial and rate-determining step of the peroxisomal β-oxidation pathway (Fig. 3). In the reductive half-reaction, the substrate acyl-CoA is α,β-dehydrogenated into the corresponding trans-2-enoyl-CoA and FAD becomes reduced. In the oxidative half-reaction, FADH2 donates electrons directly to molecular oxygen, thereby generating H2O2. In mammals, altogether four acyl-CoA oxidases have been described, which differ in respect to substrate specificity, immunological reactivity and subunit organization. These oxidases are palmitoyl-CoA oxidase, branched-chain acyl-CoA oxidase, pristanoyl-CoA oxidase and trihydroxycoprostanoyl-CoA oxidase.

Palmitoyl-CoA oxidase acts on straight-chain fatty acyl-CoAs, and it has been cloned from rat (Osumi et al. 1984), human (Aoyama et al. 1994a) and mouse (Nöhammer et al. 2000). The upstream regions of rat (Osumi et al. 1987) and human (Varanasi et al. 1994, 1996) genes revealed peroxisome proliferator response elements (PPREs) with direct repeats of the core motif TGACCT, which is required for proliferator-receptor binding. Two splicing variants of the rat gene were identified, resulting in two oxidases, oxidase I and oxidase II, slightly differing from the amino acid sequences (Osumi et al. 1980a, Setoyama et al. 1995). Oxidase I showed optimal activity towards a shorter acyl chain length with a maximum at C10 compared to oxidase II with a maximum at C14. The recombinant human palmitoyl-CoA oxidase showed the highest catalytic rate with

25

Fig. 3. Peroxisomal β-oxidation cycle of straight-chain fatty acyl-CoA. The numbers refer to the following enzymes: 1, palmitoyl-CoA oxidase; 2, (2E)-enoyl-CoA hydratase-1 (MFE-1); 3, (2E)-enoyl-CoA hydratase-2 (MFE-2); 4, (3S)-hydroxyacyl-CoA dehydrogenase (MFE-1); 5, (3R)-hydroxyacyl-CoA dehydrogenase (MFE-2); 6, 3-ketoacyl-CoA thiolase. The fatty acyl-CoA shortened by two carbon atoms re-enters the β-oxidation cycle. The figure was prepared with ChemDraw (CambridgeSoft Corporation, MA, USA).

26

saturated fatty acids of 12 to 18 carbon atoms (Chu et al. 1995). Palmitoyl-CoA oxidase can, however, oxidase longer acyl tails than these, since other studies indicated that the rat enzyme, for instance, was able to oxidase lignoceroyl-CoA (C24:0) as well (Van Veldhoven et al. 1992, Wanders et al. 1993). Furthermore, in mice, a knock-out in the palmitoyl-CoA oxidase-encoding gene resulted in the accumulation of both saturated and polyunsaturated (5 to 6 double bonds) VLCFAs of 24 to 30 carbon atoms (Fan et al. 1996, Infante et al. 2002). The acceptance of VLCFA-CoAs is further supported by the recently solved crystal structure of rat palmitoyl-CoA oxidase II. The structure showed a hydrophobic crevice for binding fatty acyl chains of up to 23 carbon atoms (Nakajima et al. 2002). All the characterized palmitoyl-CoA oxidases are homodimers composed of two subunits approximately 70 kDa in size with Ser-Lys-Leu (PTS1) as the three COOH-terminal amino acid residues, and they can additionally metabolize mono- and dicarboxylyl-CoAs and prostaglandin E2-CoA but are inactive towards short-chain substrates (Van Veldhoven et al. 1992).

In humans, the branched-chain acyl-CoA oxidase functions as the first and rate-limiting step of the β-oxidation pathway for 2-methyl-branched-chain fatty acyl-CoAs, such as pristanoyl-CoA, and bile acid intermediates of di- and trihydroxycoprostanoyl-CoA (DHCA-CoA and THCA-CoA), which also contain a 2-methyl substitution in their side chain (Vanhove et al. 1993). Molecular characterization of the human peroxisomal branched-chain acyl-CoA oxidase revealed a cDNA coding for a protein with Ser-Lys-Leu peptide in the COOH-terminus (Baumgart et al. 1996b). The purified enzyme appeared to be a 70 kDa monomeric protein that did not cross-react with antisera raised against rat palmitoyl-CoA oxidase and pristanoyl-CoA oxidase (see below). Unlike in humans, the branched-chain β-oxidation spiral in rats can be initiated by two different noninducible oxidases, namely pristanoyl-CoA oxidase, which acts on CoA esters of 2-methyl-branched-chain fatty acids, and trihydroxycoprostanoyl-CoA oxidase, which uses the bile acid intermediate as substrate (Schepers et al. 1990). Both cDNAs for pristanoyl-CoA oxidase (Vanhooren et al. 1996) and trihydroxycoprostanoyl-CoA oxidase (Baumgart et al. 1996a) had tripeptides of the conserved PTS1 variants in the corresponding enzymes. Rat pristanoyl-CoA oxidase is 420 kDa in size and consists of identical subunits of approximately 70 kDa (Van Veldhoven et al. 1991). In contrast, trihydroxycoprostanoyl-CoA oxidase is a homodimer of 139 kDa (Baumgart et al. 1996a). More interestingly, the gene coding for the human variant of pristanoyl-CoA oxidase is present, but its expression is undetectable (Vanhooren et al. 1997).

2.2.5.2 Second and third steps are catalyzed by two multifunctional enzymes

The second and third reactions of the β-oxidation cycle in mammals are catalysed by two unrelated peroxisomal MFEs. They both catalyse sequential hydratase and dehydrogenase reactions but through reciprocal stereochemical courses (Fig. 3). In addition to isomerization of cis-3-enoyl-CoA to trans-2-enoyl-CoA (Palosaari & Hiltunen 1990), MFE-1 hydrates (2E)-enoyl-CoA to (3S)-hydroxyacyl-CoA and dehydrogenates (3S)-

27

hydroxyacyl-CoA to 3-ketoacyl-CoA (Osumi & Hashimoto 1979). The other multifunctional enzyme, MFE-2, proceeds via a (3R)-hydroxy intermediate (Fig. 3), and thus possesses (2E)-enoyl-CoA hydratase-2 and (3R)-hydroxyacyl-CoA dehydrogenase activities (Dieuaide-Noubhani et al. 1997, Qin et al. 1997a).

MFE-1 has been cloned from various species, including rat (Osumi et al. 1985), human (Hoefler et al. 1994), guinea pig (Caira et al. 1996) and mouse (Kawai et al. 2001). The rat gene encoding MFE-1 shared similar structural features in the 5'-flanking region with the 5'-upstream sequence of the gene for acyl-CoA oxidase, which results in parallel induction of both genes (Ishii et al. 1987). The purified proteins from rat (Osumi & Hashimoto 1979), mouse (Lalwani et al. 1981) and human (Reddy et al. 1987) showed a monomeric protein 79 kDa in size. The structures of the NH2-terminal and COOH-terminal sides of MFE-1 are similar to those of (2E)-enoyl-CoA hydratase-1 and (3S)-hydroxyacyl-CoA dehydrogenase, respectively, of the mitochondrial matrix-bound β-oxidation enzymes (Ishii et al. 1987, Minami-Ishii et al. 1989). Further alignment of the homologous proteins indicated that ∆3-∆2-enoyl-CoA isomeration occurs in the same NH2-terminal catalytic domain (Palosaari et al. 1991). The reported kinetic properties indicate that MFE-1 has catalytic preference for medium straight-chain fatty acyl-CoAs (Furuta et al. 1980, Osumi & Hashimoto 1980b, Palosaari & Hiltunen 1990). However, functional MFE-1 is not vital because mutant mice lacking the gene for MFE-1 were viable and fertile and exhibited no detectable gross phenotypic defects (Qi et al. 1999). Even though in vitro measurements indicated the capability of rat liver MFE-1 to dehydrogenate not only straight-chain fatty acyl-CoA but also CoA derivatives of 2-methyl-branched-chain fatty acids and bile acid intermediates (Dieuaide-Noubhani et al. 1996), full-length MFE-1 is not sufficient for the formation of a physiological intermediate in bile acid biosynthesis (Xu & Cuebas 1996).

The second multifunctional enzyme, MFE-2 (EC 1.1.1.62), which degrades (2E)-enoyl-CoA via a (3R)-hydroxyl intermediate to 3-ketoacyl-CoA (Fig. 3), has been cloned from pig (Leenders et al. 1994), human (Adamski et al. 1995), mouse (Normand et al. 1995), rat (Dieuaide-Noubhani et al. 1996, Qin et al. 1997a), chicken (Kobayashi et al. 1997) and guinea pig (Caira et al. 1998). The chromosomal assignment and the gene structure of MFE-2 have been determined, and it was localized to chromosome 5q2 (Leenders et al. 1996a, Novikov et al. 1997, Möller et al. 1999). A comparison of the MFE-1 and MFE-2 cDNAs revealed that they have hardly any sequence homology. MFE-2, unlike MFE-1, is a homodimer of two approximately 79 kDa subunits (Jiang et al. 1996), but it also has a COOH-terminal PTS1. Moreover, the domain order in MFE-2 is reversed compared to MFE-1. (3R)-Hydroxyacyl-CoA dehydrogenase, which catalyzes the third reaction of the β-oxidation pathway, is located at the NH2-terminus of MFE-2 and followed by the (2E)-enoyl-CoA hydratase-2 domain, which is responsible for the second reaction. The hydratase-2 domain in mammalian MFE-2s is followed by a sterol carrier protein type 2 (SCP-2)-like domain. In vitro characterizations indicated that MFE-2 can accept variable substrates, such as medium-chain fatty acyl-CoA (Li et al. 1990, Jiang et al. 1996, Adamski et al. 1997, Dieuaide-Noubhani et al. 1997, Qin et al. 1997a), 2-methyl-branched-chain fatty acyl-CoA (Dieuaide-Noubhani et al. 1996, Qin et al. 1997a), ∆24-THCA-CoA intermediate in bile acid synthesis (Dieuaide-Noubhani et al. 1996, 1997, Kurosawa et al. 1997, Qin et al. 1997b), long-chain fatty acyl-CoA (Dieuaide-Noubhani et al. 1996, Jiang et al. 1997), 17β-estradiol (Adamski et al. 1995,

28

Fig. 4. β-Oxidation of branched-chain fatty acyl-CoAs. In addition to very-long-chain fatty acyl-CoAs, mammalian MFE-2 participates in the degradation of (2E)-pristenoyl-CoA and (24E)-THCA-CoA, the intermediate in bile acid synthesis. The numbers refer to the following enzymes: 1, very-long-chain acyl-CoA synthetase (human) / trihydroxycoprostanoyl-CoA synthetase (rat); 2, methylacyl-CoA racemase; 3, branched-chain acyl-CoA oxidase (human) / pristanoyl-CoA oxidase (rat); 4, branched-chain acyl-CoA oxidase (human) / trihydroxycoprostanoyl-CoA oxidase (rat); 5, MFE-2; 6, SCPx.

1997) and 5-androstene-3β,17β-diol (Leenders et al. 1994, Adamski et al. 1995). However, the inactivation of MFE-2 in mice led to the conclusion that MFE-2 is not only essential for the degradation of CoA derivatives of 2-methyl-branched-chain fatty acids and the bile acid intermediates of DHCA and THCA (Fig. 4) but also for the catabolism of VLCFA-CoAs (Baes et al. 2000).

2.2.5.3 3-Ketoacyl-CoA thiolases

The last step in the peroxisomal β-oxidation process, the thiolytical cleavage of 3-ketoacyl-CoA into chain-shortened acyl-CoA and acetyl-CoA or propionyl-CoA (Fig. 3 and Fig. 4), is catalyzed by distinct thiolases (EC 2.3.1.16). In rat, two closely related homodimeric 3-ketoacyl-CoA thiolases, thiolase A and thiolase B, are solely responsible for the chain shortening of straight-chain fatty acyl-CoAs. The cDNAs for straight-chain thiolases are 95 % identical, as are the proteins at the amino acid sequence level (Hijikata et al. 1987, Bodnar & Rachubinski 1990, Hijikata et al. 1990). The major difference lies in the NH2-terminus. The constitutively expressed thiolase A has an NH2-terminal PTS2 ten amino acid residues longer than thiolase B, whose gene is markedly activated by peroxisome proliferators (Hijikata et al. 1990, Nicolas-Frances et al. 2000). The specific activities and substrate specificities of thiolase A and thiolase B isolated from rat livers

29

were virtually identical, and they both carried out thiolytic cleavage of short, medium and long straight-chain 3-ketoacyl-CoAs with medium chain being the best substrate (Antonenkov et al. 1997, 1999). Thus far, only one human thiolase has been identified with similarity to rat straight-chain 3-ketoacyl-CoA thiolases (Bout et al. 1988, Fairbairn & Tanner 1989, Bout et al. 1991).

In mammalian peroxisomes, the thiolytic function is also performed by the sterol carrier protein x (SCPx). SCPx is a homodimeric protein composed of 58 kDa subunits with 3-ketoacyl-CoA thiolase activity in the NH2-terminal domain and SCP-2 function in the COOH-terminal domain (Mori et al. 1991, Pfeifer et al. 1993, Seedorf et al. 1993, Ohba et al. 1994). SCPx has broad substrate specificity because, in addition to medium and long straight-chain 3-ketoacyl-CoAs, it cleaves 2-methyl-branched 3-ketoacyl-CoAs and the bile acid intermediate 24-keto-THCA-CoA in vitro (Antonenkov et al. 1997). Mice inactivated for the gene encoding SCPx showed impaired catabolism of 2-methyl-branched-chain fatty acyl CoAs. On the other hand, the serum concentrations of cholesterol, steroids, VLCFA and long-chain fatty acids were not affected. (Seedorf et al. 1998)

In successive cycles of peroxisomal β-oxidation, propionyl-CoA, acetyl-CoA and acyl-CoAs shortened by two carbon atoms are released (Fig. 3). Propionyl-CoA and 4,8-dimethyl-nonanoyl-CoA, which are products of the branched-chain acyl-CoA β-oxidation pathway (Fig. 4), are converted into their carnitine esters within the peroxisome and transported into mitochondria for further oxidation (Jakobs & Wanders 1995, Verhoeven et al. 1998, Fredinandusse et al. 1999). The peroxisomal palmitoyl-CoA oxidase acts on very-long and long-chain fatty acyl-CoAs. Therefore, the fatty acyl-CoAs with a chain length of approximately eight carbons are exported to mitochondria, where their β-oxidation continues. The fate of acetyl units produced by peroxisomal β-oxidation is not clear (Chu et al. 1995, Eaton et al. 1996, Hashimoto 1999b).

2.2.5.4 Auxiliary enzymes

Double bonds in naturally occurring unsaturated fatty acids are generally in a cis configuration and can be found in both even- and odd-numbered positions. Since (2E)-enoyl-CoA, with a double bond at the ∆2-position and in a trans configuration, is the only unsaturated intermediate of the β-oxidation cycle, metabolism beyond the pre-existing double bonds illustrates the requirement for accessory or auxiliary enzymes, such as ∆3,5-∆2,4-dienoyl-CoA isomerase, 2,4-dienoyl-CoA reductase and ∆3-∆2-enoyl-CoA isomerase (for reviews, see Osmundsen et al. 1991, Hiltunen et al. 1996, Hiltunen & Qin 2000). In the oxidation of an odd-numbered double bond, stepwise β-oxidation leads to cis-5-enoyl-CoA, which is then dehydrogenated and isomerized to trans-3,cis-5-dienoyl-CoA. This is followed by the conversion of ∆3,∆5-dienoyl-CoA to trans-2,trans-4-dienoyl-CoA catalyzed by ∆3,5-∆2,4-dienoyl-CoA isomerase. Subsequently, trans-2,trans-4-dienoyl-CoA undergoes NADPH-dependent reduction, which is catalyzed by 2,4-dienoyl-CoA reductase, to ∆3-enoyl-CoA with the double bond in a trans configuration. ∆3-Enoyl-CoA is subsequently converted to trans-2-enoyl-CoA by ∆3,∆2-enoyl-CoA isomerase. Fatty

30

acyl-CoAs with double bonds at even-numbered positions requires subsequent reactions catalyzed first by 2,4-dienoyl-CoA reductase and then by ∆3,∆2-enoyl-CoA isomerase.

In the metabolism of 2-methyl-branched-chain acyl-CoAs, the peroxisomal β-oxidation system is stereoselective because only a (2S)-isomer is accepted by the oxidases. α-2-Methylacyl-CoA racemase is an auxiliary enzyme that catalyzes the conversion of (2R)-pristanoyl-CoA and the bile acid intermediate, (25R)-isomer of THCA-CoA, into their corresponding S-stereoisomers (Schmitz et al. 1994, 1995, Schmitz & Conzelmann 1997, Ferdinandusse et al. 2002). α-2-Methylacyl-CoA racemase is present not only in peroxisomes but also in mitochondria, and it is encoded by a single gene (Amery et al. 2000, Kotti et al. 2000).

2.3 Multifunctionality of MFE-2

2.3.1 Domain structure of various MFE-2s

The cDNA coding for MFE-2 was first identified in C. tropicalis. It showed that a partial gene duplication event has occurred during the evolution of the gene. Namely, the translated polypeptide contained a duplicate stretch in the NH2-terminus with approximately 44 % amino acid sequence similarity (Nuttley et al. 1988). However, the physiological activity of yeast MFE-2 was first identified with S. cerevisiae (Hiltunen et al. 1992). It was shown that the β-oxidation of fatty acids in yeast proceeds only via a (3R)-hydroxyl intermediate. Additional deletion studies localized (2E)-enoyl-CoA hydratase-2 into the COOH-terminus and the two copies of (3R)-hydroxyacyl-CoA dehydrogenases into the NH2-terminal part of MFE-2 (Hiltunen et al. 1992). Subsequently, this domain order was shown to be characteristic of other yeast MFE-2s (Fig. 5) (Fosså et al. 1995).

Characterization of porcine MFE-2 showed that the 32 kDa NH2-terminal peptide was able to perform dehydrogenation via the same enantiomer as yeast MFE-2, and the central part possessed (2E)-enoyl-CoA hydratase-2 activity (Leenders et al. 1996b). Moreover, the COOH-terminal extension showed 40 % identity to SCP-2. The intron-exon structure of the gene encoding the (3R)-hydroxyacyl-CoA dehydrogenase and (2E)-enoyl-CoA hydratase-2 regions of mammalian MFE-2 is unique, but the gene structure of the last three exons is similar to the gene coding for SCP-2 (Ohba et al. 1994, Möller et al. 1999). This observation supports the hypothesis that the mammalian MFE-2 gene is the result of gene fusion.

The domain compositions of other MFE-2-related proteins are different (Fig. 5). The mycorrhizal fungus Glomus mosseae has the same domain structure as the yeast MFE-2, with the exception that the SCP-2-like domain is fused to the hydratase-2 region (Requena et al. 1999), as it is with mammalian MFE-2. In Drosophila melanogaster and Caenohabditis elegans (C. elegans), the domains have been rearranged, so that either the hydratase (C. elegans) or the SCP-2-like domain (Drosophila) is present as a separate polypeptide (Breitling et al. 2001). All the variants have a PTS1 at the COOH-terminus.

31

Fig. 5. Domain organization of several MFE-2s. (3R)-hydroxyacyl-CoA dehydrogenase, (2E)-enoyl-CoA hydratase-2 and SCP-2-like domain are represented as dark grey, white and light grey, respectively. Every variant has a peroxisomal targeting signal, PTS1, at its COOH-terminus. Domain sizes are not drawn to scale. (Modified from Breitling et al. 2001)

2.3.2 Catalytic properties of MFE-2

Mammalian MFE-2 was first identified as 17β-hydroxysteroid dehydrogenase (17β-HSD) converting 17β-estradiol and ∆5-androstene-3β,17β-diol into their less active keto-forms. The enzyme performed oxidation 360-fold more efficiently than reduction. Therefore, MFE-2 was first named 17β-HSD type 4 (Leendrs et al. 1994, Adamski et al. 1995). In view of the observation that yeast MFE-2 is also in vitro a 17β-HSD, the potential physiological role of MFE-2 as a 17β-HSD is unclear (Adamski et al. 1995, Leenders et al. 1996b, Qin et al. 2000). Furthermore, the catalytic efficiency (kcat/Km) of mammalian and yeast MFE-2s with 17β-estradiol was 1000 times lower than that of fatty acyl-CoAs. Moreover, the observation that 17β-HSD activity and the MFE-2 expression pattern do not overlap in breast tissue supports the minor role of MFE-2 in steroid metabolism (Miettinen et al. 1999).

The kinetic properties of the (3R)-hydroxyacyl-CoA dehydrogenase and (2E)-enoyl-CoA hydratase-2 activities of MFE-2 have been investigated as well and they are listed in Tables 1 and 2, respectively. The maximal velocities of the dehydrogenase reaction of mammalian MFE-2 doubled when the chain length of the substrate increased from C4 to C16 (Dieuaide-Noubhani et al. 1996, Qin et al. 1997a). However, the dehydrogenase region of yeast MFE-2 showed maximal velocity with 3-hydroxydecanoyl-CoA as well as the lowest Km, indicating the highest catalytic efficiency with a fatty acyl tail of ten carbon atoms (Qin et al. 1999). This result is, however, due to the combined action of two dehydrogenases in the same polypeptide. Namely, further characterization of yeast MFE-2 showed that the most NH2-terminal dehydrogenase, dehydrogenase A, is active towards medium- and long-chain (3R)-hydroxyacyl-CoAs, while dehydrogenase B shows the highest catalytic rate with short-chain (C4) substrates (Qin et al. 1999). The NH2-terminal dehydrogenase region does not interfere with the hydratase-2 reaction, as demonstrated by the kinetic properties of both full-length MFE-2 and the 46 kDa hydratase-2 fragment isolated from human liver (Jiang et al. 1996). With both enzymes, the maximal velocities

Caenohabditis elegans

Mammalian MFE-2

Yeast MFE-2

Glomus mosseae MFE-2

Drosophila melanogaster

PTS1

PTS1 PTS1

PTS1 PTS1

PTS1

PTS1

32

Table 1. Kinetic properties of the dehydrogenase region reported for purified MFE-2 or its variants from various species.

Species kcat

a (s-1) Km (µM) Reference

Human

MFE-2 (liver)

17β-Estradiol 0.2x10-3 0.2 Adamski et al. 1995

MFE-2b

17β-Estradiol 0.8 Adamski et al. 1995

∆5-Androstene-3β,17β-diol 0.9 Adamski et al. 1995

Porcine

MFE-2 (kidney)

17β-Estradiol 0.3x10-3 0.3 Leenders et al. 1996b

Acetoacetyl-CoA 4.4 34.8 Leenders et al. 1996b

MFE-2b

17β-Estradiol 0.2x10-3 0.4 Leenders et al. 1996b

Acetoacetyl-CoA 3.9 35.3 Leenders et al. 1996b

Dehydrogenase fragmentb

17β-Estradiol 0.1x10-3 0.3 Leenders et al. 1996b

Acetoacetyl-CoA 0.7 31.1 Leenders et al. 1996b

Rat

MFE-2b

3-Hydroxybutyroyl-CoA 0.6 Qin et al. 1997a

3-Hydroxydecanoyl-CoA 0.7 Qin et al. 1997a

3-Hydroxyhexadecanoyl-CoA 1.2 Dieuaide-Noubhani et al. 1996

17β-Estradiol 2.1 x10-6 Qin et al. 1997a

24R,25S-THCA-CoA 0.9 3.3 Dieuaide-Noubhani et al. 1996

Candida tropicalis

Dehydrogenase fragment (A+/B+)b

3-Hydroxybutyroyl-CoA 31 55 Qin et al. 1999

3-Hydroxydecanoyl-CoA 53 5.4 Qin et al. 1999

3-Hydroxyhexadecanoyl-CoA 41 24 Qin et al. 1999

17β-Estradiol 6.2x10-3 370 Qin et al. 2000

24R,25S-THCA-CoA 0.1 Qin et al. 2000

24R,25R-THCA-CoA 0.3 Qin et al. 2000

Dehydrogenase fragment (A+/B-)b

3-Hydroxybutyroyl-CoA ND Qin et al. 1999

3-Hydroxydecanoyl-CoA 33 5.4 Qin et al. 1999

3-Hydroxyhexadecanoyl-CoA 36 24 Qin et al. 1999

Dehydrogenase fragment (A-/B+)b

3-Hydroxybutyroyl-CoA 29 55 Qin et al. 1999

3-Hydroxydecanoyl-CoA 17 5.8 Qin et al. 1999

3-Hydroxyhexadecanoyl-CoA 12 25 Qin et al. 1999 aEnzyme activities are converted to catalytic activities by dividing the published values by the molecular weight

of the enzyme. bRecombinant protein. ND means not detected. A+ or B+ indicates a functional dehydrogenase

A or dehydrogenase B domain, respectively.

33

Table 2. Kinetic properties of the hydratase-2 region reported for purified MFE-2 or its variants from various species.

Species kcat

a (s-1) Km (µM) Reference

Human

MFE-2 (liver)

trans-2-Hexenoyl-CoA 447 100 Jiang et al. 1996

trans-2-Octenoyl-CoA 603 40 Jiang et al. 1996

trans-2-Decenoyl-CoA 253 9 Jiang et al. 1996

trans-2-Dodecenoyl-CoA 199 5 Jiang et al. 1996

Hydratase+SCP fragment (liver)

trans-2-Hexenoyl-CoA 571 100 Jiang et al. 1996

trans-2-Octenoyl-CoA 697 30 Jiang et al. 1996

trans-2-Decenoyl-CoA 279 12 Jiang et al. 1996

trans-2-Dodecenoyl-CoA 216 4 Jiang et al. 1996

Porcine

MFE-2 (kidney)

Crotonyl-CoA 2.4 34.0 Leenders et al. 1996b

MFE-2b

Crotonyl-CoA 2.1 37.1 Leenders et al. 1996b

Hydratase fragmentb

Crotonyl-CoA 0.7 34.7 Leenders et al. 1996b

Rat

MFE-2b

Crotonyl-CoA 1.2 Qin et al. 1997a

Crotonyl-CoA 1.3 Dieuaide-Noubhani et al. 1996

trans-2-Decenoyl-CoA 8.2 Qin et al. 1997a

trans-2-Decenoyl-CoA 161 Dieuaide-Noubhani et al. 1996

trans-2-Hexadecenoyl-CoA 8.2 Dieuaide-Noubhani et al. 1996

24R,25S-THCA-CoA 54 Dieuaide-Noubhani et al. 1996

3-hydroxy-2-methyl-hexadecanoyl-CoA 9.2 Dieuaide-Noubhani et al. 1996

Hydratase+SCP fragment (liver)

3-hydroxyoctanoyl-CoA 49 71 Li et al. 1990

trans-2-Octenoyl-CoA 42 22.5 Li et al. 1990

Hydratase+SCP fragmentb

Crotonyl-CoA 2.3 60 Qin et al. 1997b

trans-2-Hexenoyl-CoA 23 8.7 Qin et al. 1997b

trans-2-Decenoyl-CoA 26 4.6 Qin et al. 1997b aEnzyme activities are converted to catalytic activities by dividing the published values by the molecular weight

of the enzyme. bRecombinant protein.

34

increased with an increase in chain length up to eight carbon atoms and then started to decrease gradually. Since the Km values decreased markedly with an increase in chain length, the catalytic efficiency (kcat/Km) was highest with the C12 substrate. The rat counterpart showed a similar profile, with the exception that the catalytic efficiencies with longer substrates were ten times lower than that reported for the human variant (Li et al. 1990, Qin et al. 1997b).

2.3.3 Tissue distribution

The multifunctionality of MFE-2 does not only lie in its multidomain structure and broad substrate specificity but also in its tissue distribution and expression levels. The highest MFE-2 activities and the strongest immunoreactions in human were found in liver (hepatocytes) followed by heart, prostate and testis. Moderate expression occurred in lung, skeletal muscle, kidney, pancreas, thymus, ovary, intestine and placenta. Weak signals were observed in brain, spleen, colon and lymphocytes. (Adamski et al. 1995, Carstensen et al. 1996, Fan et al. 1998). In human brain, MFE-2 appeared already in the 13th gestational week. Each neuron exhibited increased immunoreactivity along with growth in size as age increased. Glial cells in white matter showed immunoreactivity after the 30th gestational week (Itoh et al. 1999). In the developing mouse brain, the mRNA levels of peroxisomal β-oxidation enzymes, including MFE-2, reached a maximum on postnatal day 5. Thereafter, the levels of mRNA studied decreased progressively, reaching their steady-state levels on postnatal day 40 (Knoll et al. 2000). More interestingly, palmitoyl-CoA oxidase and MFE-2 were shown to be essential for docosahexaenoic acid (DHA, C22:6n-3) synthesis (Su et al. 2001). DHA is needed for normal brain and retinal development. In the DHA synthetic pathway, C22:5n-3 is elongated to C24:5n-3, desaturated to C24:6n-3 in microsomes and then retroconverted to docosahexaenoic acid by oxidase and MFE-2 in peroxisomes.

Normal colonic mucosa has a high level of 17β-estradiol metabolism, and the expression of mRNA for MFE-2 was shown to be significantly decreased in tumours compared to normal mucosa. Loss of expression of MFE-2 has also been demonstrated in breast cancer (Krazeisen et al. 1999). Similarly, studies of the loss of heterozygosity at tumour suppressor loci involved in sporadic breast cancer as well as in colorectal cancer (Nishisho et al. 1991, Thomson et al. 1993) revealed allele loss at 5q21, which is the location to which the MFE-2 gene has been assigned (Leenders et al. 1996a, Novikov et al. 1997, Möller et al. 1999). On the contrary, high expression levels of MFE-2 were consistently observed in prostate cancer cells, but it is not completely certain whether the β-oxidation of 17β-estradiol driven by MFE-2 associates with hormone-refractory prostate cancer (Castagnetta et al. 1997).

The full-length MFE-2 polypeptide is partially cleaved in vivo into separate (3R)-hydroxyacyl-CoA dehydrogenase (34 kDa) and (2E)-enoyl-CoA hydratase-2 (45 kDa) components (Adamski et al. 1992, Adamski et al. 1995, Dieuaide-Noubhani et al. 1997, Qin et al. 1997a). The cleavage takes place only in peroxisomes (Markus et al. 1995). There are different cleavage efficiencies: high in hormone target organs, such as uterus

35

and breast epithelium, but low in non-target tissues participating in the β-oxidation of fatty acids, such as liver and kidney (Kaufmann et al. 1995). It is not yet known whether the release of a 45 kDa fragment is advantageous in hormone inactivation catalyzed by the NH2-terminal 34 kDa fragment.

2.3.4 (3R)-Hydroxyacyl-CoA dehydrogenase, a member of the short-

chain alcohol dehydrogenase/reductase superfamily

The (3R)-hydroxyacyl-CoA dehydrogenase region of MFE-2 belongs to the short-chain alcohol dehydrogenase/reductase (SDR) superfamily. This is a large and diverse group of NAD(H) or NADP(H)-dependent oxidoreductases found in bacteria, plants and animals (Baker 1991, Krozowski 1992, Jörnvall et al. 1995, Bailey et al. 1997, Oppermann et al. 1997b, Jörnvall et al. 1999, Oppermann et al. 2001). At the present, more than 2000 sequences of the SDR family can be found in protein and nucleic acid databases, and the crystal structures of close to twenty members of the family have been determined. Although residue identity between the different forms usually does not amount to a level of more than 15-30 %, all the known three-dimensional structures show a notably similar one-domain α/β pattern, the SDR fold (Fig. 6).

Members of the SDR superfamily are identified through the occurrence of the several distinct sequence motifs that they have in common (Persson et al. 1991, Jörnvall et al. 1995, Oppermann et al. 1997b, Oppermann et al. 2001). These amino acid elements are restricted to certain segments in the sequence, indicating common fold, active site, reaction mechanism and coenzyme and substrate-binding regions (Persson et al. 1991, Filling et al. 2001, Filling et al. 2002). Essential parts of the coenzyme-binding site, as determined by X-ray crystallographic studies (Ghosh et al. 1994, 1995, Tanaka et al. 1996) and amino acid sequence comparisons (Persson et al. 1991, Jörnvall et al. 1995), are located in the NH2-terminal part, comprising a Rossmann fold structure (Rossmann et al. 1974, Wierenga et al. 1985) with a conserved Gly-X-X-X-Gly-X-Gly pattern. This dinucleotide-binding region is composed of βαβ units that create a parallel β-sheet sandwiched between two arrays of parallel α-helices (Fig. 6). Additionally, crystallographic and site-directed mutagenesis studies of various SDR enzymes revealed another conserved motif, Tyr-X-X-X-Lys, essential for catalysis (see below). Significantly, half of all the conserved residues are glycines (Borrás et al. 1989), which is typical of distantly related proteins with a conserved fold.

Most SDRs are homodimeric or homotetrameric by quaternary structure. The dimeric interface in homodimeric SDRs consists of a four-helix bundle, which is composed of long parallel α-helices, αE and αF, of two neighbouring subunits (Benach et al. 1998). All the structures of homotetrameric SDRs determined so far exhibit two main subunit interfaces; the four-helix bundle is responsible for constituting a dimer, and two dimers simultaneously interact with each other via their COOH-terminal regions (Ghosh et al. 1994, Tanaka et al. 1996). However, these associations are not vital for a functional SDR enzyme. Namely, porcine carbonyl reductase is a monomeric enzyme (Ghosh et al. 2001), whereas 3α-hydroxysteroid dehydrogenase/carbonyl reductase is a homodimer,

36

but the dimeric interactions are solely dominated by the COOH-terminal regions (Grimm et al. 2000), which is usually characteristic of homotetramers. Both enzymes have a basic α/β SDR fold, but the unique insertions before the α-helix, αF, comprise a subdomain that packs against interfacial helices, eliminating the four-helix bundle interface conserved in the superfamily.

Fig. 6. Crystal structure of rat type II hydroxyacyl-CoA dehydrogenase/amyloid- -binding alcohol dehydrogenase (PDB code 1E6W, Powell et al. 2000). The conserved SDR fold consists of the central β-sheet, which is surrounded by two arrays of three α-helices. The two long α-helices on the right side of the figure comprise a four-helix bundle with the neighbouring monomer at the dimer interface. The molecule bound to the protein is NAD+. N and C indicate NH2-terminus and COOH-terminus, respectively. The figure was prepared with the Swiss-PDBViewer (Guex & Peitsch 1997).

Probably all SDRs share a common reaction mechanism (Fig. 7), which follows a compulsory ordered pathway with the coenzyme binding first. In some cases, the binding of the coenzyme induces changes in the structure (Benach et al. 1999), but sometimes there is no influence on the protein conformation (Grimm et al. 2000). Upon binding of the hydroxyl or carbonyl substrate, a “substrate-binding loop”, which is highly flexible in most apo-SDRs, becomes well ordered and covers the substrate as well as the catalytic

centre from an aqueous environment (Chapman et al. 1999). In addition to the tyrosine and lysine in the common motif, a conserved serine also

plays an important role in the catalysis (Fig. 7) (Ensor & Tai 1991, Albalat et al. 1992, Obeid & White 1992, Chen et al. 1993, Cols et al. 1993, Tanaka et al. 1996, Benach et al. 1998, 1999, Powell et al. 2000). In the first step, the deprotonated phenolic group of tyrosine forms a hydrogen bond with the hydroxyl group attached to position 3 of a fatty acyl-CoA (Rozwarski et al. 1999) or an alcohol. In the second step, the deprotonated tyrosine residue acts as a catalytic base to extract hydrogen from the hydroxyl group of the substrate. Since the substrate is oriented in such a way that its α-face is facing tyrosine with its β-face oriented toward the nicotinamide ring of NAD+, the β-hydride ion, liberated from position 3 of the substrate, can be directly transferred to position 4 of

37

Fig. 7. Proposed catalytic reaction mechanism of Drosophila alcohol dehydrogenase. (I) Apo state: three water molecules hydrogen-bonded to a catalytic triad. (II) Binary state: NAD+ at the active site and one water molecule hydrogen-bonded to a catalytic triad. (III) Ternary state: binding of alcohol, proton transfer and hydride transfer. (IV) Ternary state: ketone and NADH at the active site. Finally, the ketone and NADH leave the active site. (Modified from Benach et al. 1999). The figure was prepared with ChemDraw (CambridgeSoft Corporation, MA, USA).

the nicotinamide ring. In addition to the interaction with the nicotinamide ring of NAD+, the serine residue stabilizes the reaction intermediate via its hydroxyl group (Oppermann et al. 1997a). Since the phenolic group of the tyrosine side chain has a pKa value close to 10, the ε-amino group of lysine is needed to convert tyrosine to tyrosinate (pKa 7.6), to facilitate catalysis at neutral pH (Chen et al. 1993). Additionally, lysine orients the nicotinamide ring in the syn conformation through a bifurcated hydrogen bond to both the 2´- and the 3´-hydroxyl groups of the nicotinamide ribose moiety to allow B-face 4-pro-S hydride transfer. Due to their important roles in catalysis, serine, lysine and tyrosine are called the Ser-Tyr-Lys catalytic triad. Recently, the asparagine, located close to the active site, was shown to be important in stabilizing the position of the catalytic lysine via a conserved water molecule, extending the previously recognised catalytic triad to form a tetrad of Asn-Ser-Tyr-Lys (Filling et al. 2002).

Alternatively, site-directed mutagenesis and modelling studies proposed that the deprotonated serine may act as a catalytic base, with tyrosine serving as an electrophilic catalyst, stabilizing the transition state for hydride transfer by hydrogen bonding to the

38

substrate carbonyl. Even in this case, the dual role of the lysine in the catalytic triad is to properly orient the cofactor by forming hydrogen bonds to the nicotinamide ribose moiety and to lower the pKa not only of the tyrosine but also of the serine via electrostatic interaction. (Parikh et al. 1999, Winberg et al. 1999)

2.3.5 Sterol carrier protein type 2

The COOH-terminal domain of mammalian MFE-2, which consists of approximately 120 amino acid residues, is 40 % identical to SCP-2. SCP-2 is also called non-specific lipid transfer protein (nsLTP), and its proposed multiple functions arise from its complex gene structure, post-translational processing, intracellular localization and ligand specificity (for reviews, see Seedorf et al. 2000, Gallegos et al. 2001). The SCP-2 gene has two initiation sites coding for proteins that share a common 13 kDa SCP-2 COOH-terminus; one site codes for the 58 kDa SCPx, which is partially post-translationally cleaved to 13 kDa SCP-2 and 45 kDa thiolase. The other site codes for the 15 kDa pro-SCP-2, which is completely post-translationally cleaved to 13 kDa SCP-2. SCP-2 appears to be localized in subcellular organelles intimately involved with fatty acid oxidation, peroxisomes and mitochondria, with lesser amounts in the cytosol and near to the endoplasmic reticulum (Keller et al. 1989), even though all the translated variants share identical PTS1 in their COOH-terminus.

In vitro studies have shown that SCP-2 binds sterols, phospholipids, long-chain isoprenoids (Schröder et al. 1991, Schröder et al. 1996), long-chain fatty acids (Schröder et al. 1995), long-chain fatty acyl-CoAs (Frolov et al. 1996c) and branched-chain fatty acyl-CoAs (Frolov et at. 1997). Additionally, SCP-2 promotes the exchange of a wide variety of lipids and sterols between membranes (Frolov et al. 1996a, 1996b). The Kd values of SCP-2 for saturated and unsaturated fatty acyl-CoAs, with the length of the acyl tail between C10 and C20, were the same as for cholesterol, i.e. close to 2-4 nM. The binding of fatty acids was 100 times weaker. The introduction of a 3-hydroxy or 3-keto group into fatty acyl-CoA decreased the affinity 5-fold (Dansen et al. 1999). Interestingly, upon interacting with long-chain fatty acyl-CoA, SCP-2 forms specific complexes with acyl-CoA oxidase, the first enzyme of the peroxisomal β-oxidation cycle (Wouters et al. 1998). Therefore, SCP-2 may facilitate peroxisomal β-oxidation by supplying fatty acyl-CoA esters. Furthermore, human 13 kDa SCP-2 isolated from recombinant bacteria contained a small amount of endogenously bound fatty acid (Stolowich et al. 1997). However, complete deficiency of SCP-2 and SCPx in mice was associated with the impaired catabolism of methyl-branched fatty acyl-CoAs (Seedorf et al. 1998). On the other hand, the serum concentrations of VLCFAs and long-chain fatty acids were not affected, which might be due to up-regulation of the liver fatty acid-binding protein (Fuchs et al. 2001).

39

2.3.5.1 Other fatty acid/fatty acyl-CoA-binding proteins

Recent data show that long-chain fatty acids and their CoA derivatives directly or indirectly regulate the activity of many cellular components and processes, such as membrane receptors, enzymes, ion channels, cell differentiation, cellular development, gene expression and cellular energy conversion (for reviews, see Glatz et al. 1995, Ek et al. 1997, Hertz et al. 1998, McArthur et al. 1999). Therefore, the cellular pool of fatty acids and fatty acyl-CoAs is restricted and their distribution and trafficking are modulated by fatty acid-binding proteins (FABP), fatty acyl-CoA-binding proteins (ACBP) and obviously also by non-specific lipid transfer protein (nsLTP, SCP-2). The small sizes of all of these proteins, i.e. less than 15 kDa, indicate that their only function is to provide a proper environment for the binding of hydrophobic substrates.

Most unesterified fatty acids in the cytoplasm are bound to cellular membranes. Especially long-chain fatty acids are minimally soluble in water (Vorum et al. 1992) and have a strong affinity for membranes. These properties restrict their movement through the cytoplasm. Consequently, the transport and solubilization of fatty acids are facilitated by FABPs. Members of the FABP family are approximately 14-15 kDa in size and monomeric (Glatz et al. 1996, Weisiger 1996). The rate of fatty acid diffusion within liver cells is directly proportional to the concentration of liver FABP (Luxon & Weisiger 1993, Luxon 1996, Luxon et al. 1998). The fatty acid uptake from membranes involves interaction between a positively charged α-helical region of the FABP and negative charges on the membrane surface, resulting in a conformational change that permits direct transfer of fatty acid between the protein-binding site and the membrane (Herr et al. 1995, 1996, Gericke et al. 1997, Corsico et al. 1998). Such interaction was also demonstrated for SCP-2 to accept ligands from membranes (Stolowich et al. 1997, Huang et al. 1999).

ACBP is a cytosolic protein with 86–103 residues and a highly conserved amino acid sequence. A total of 30 sequences have been recognized either as proteins or as gene products (Kragelund et al. 1999). In mammals, the highest concentration of ACBP is found in liver (Bovolin et al. 1990). In other tissues, ACBP is reported to be abundant in

specialized cells, such as the steroid-producing cells of adrenal cortex and testis. The main function of ACBP appears to be the modulation of acyl-CoA concentrations within the cell. By its ability to bind acyl-CoA molecules, ACBP mediates the product feedback inhibition of long-chain acyl-CoA synthetase. ACBP binds medium- and long-chain acyl-CoA esters with very high affinity, with a preference for acyl-CoA esters of 14-22 carbons atoms (Rasmussen et al. 1990, Færgeman et al. 1996, Rosendal et al. 1993), but it does not bind fatty acids, acyl carnitines and cholesterol (Rosendal et al. 1993).

Despite their capability to bind fatty acids or their CoA derivatives, SCP-2, FABP and ACBP are not sequence- or structure-related (Fig. 8), which indicates evolutionarily different origins. ACBPs are α-helical proteins with either four or seven helices arranged differently with respect to each other (Andersen & Poulsen 1993, Lerche et al. 1997, van Aalten et al. 2000). The acyl chain, which is buried in the hydrophobic groove, is completely protected from the aqueous solvent by the acyl-CoA head group. The CoA head forms a lid on the binding pocket by interacting with specific residues on the rim of the binding cavity. In contrast to ACBP, FABP and SCP-2 are α/β proteins. FABPs are

40

Fig. 8. Various fatty acid/fatty acyl-CoA-binding proteins. Top panel: human FABP with oleic acid (PDB code 1HMS, Young et al. 1994). Central panel: ACBP from Plasmodium falciparum in complex with myristoyl-CoA (PDB code 1HBK, van Aalten et al. 2001). The lowest panel: rabbit SCP-2 (PDB code 1C44, Choinowski et al. 2000). The figure was prepared with the Swiss-PDBViewer (Guex & Peitsch 1997).

composed of ten antiparallel β-strands, which form a barrel-like structure capped by two short α-helical segments (Sacchettini et al. 1989, Zanotti et al. 1992). The flattened barrel provides a hydrophobic pocket, where the carboxylate of the fatty acid is located. Even though the tertiary structures for all FABPs solved are conserved, the bound fatty acids are very different in their conformations and interactions. In SCP-2 proteins, the putative ligand-binding site is built up of both β-sheet and α-helices; the β-sheet with a mixed sheet structure is flanked on one side by α-helices. This arrangement forms a hydrophobic tunnel, providing an environment for apolar ligands, such as fatty acids and fatty acyl-CoAs (Choinowski et al. 2000, García et al. 2000).

41

2.4 Peroxisomal β-oxidation deficiencies

The inherited peroxisomal disorders in human are divided into two categories. One group consists of peroxisome biogenesis disorders involving an absence of peroxisome formation (Braverman et al. 1995, Lazarow & Moser 1995), while the other group is composed of single peroxisomal protein deficiencies (Wanders & Tager 1998). Human fatal genetic peroxisomal biogenesis disorders (PBDs) include the Zellweger syndrome

(ZS), neonatal adrenoleukodystrophy (NALD), infantile Refsum disease (IRD) and rhizomelic chondrodysplasia punctata (RCDP) (for a review, see Gould & Valle 2000). ZS is characterized by an array of neural, hepatic and renal defects. Individuals with ZS exhibit severe neurological dysfunction and rarely survive their first year. By contrast, NALD patients display similar but less severe phenotypes than ZS patients and can survive for up to a decade. IRD patients are even more mildly affected, with some surviving beyond their third decade (Moser et al. 1995). RCDP, by contrast, is characterized by unique skeletal abnormalities (Moser et al. 1995).

Several disorders relevant to the peroxisomal β-oxidation cycle include X-linked adrenoleukodystrophy, palmitoyl-CoA oxidase deficiency, MFE-2 deficiency and thiolase deficiency, and they all are associated with the accumulation of VLCFAs, resulting in defects in the nervous system. Adrenomyeloneuropathy (AMN) and cerebral childhood adrenoleukodystrophy (CCALD) are the main phenotypic variants of an X-linked inherited metabolic disorder, X-linked adrenoleukodystrophy (X-ALD). Neuropathological and electrophysiological studies have shown that X-ALD is a disease with mixed features of axonal degeneration, leading to myeloneuropathy, and a severe inflammatory reaction in the cerebral white matter, resulting in demyelination of the peripheral and central nervous system, respectively (for a review, see McGuinness & Smith 1999). Other tissues, such as adrenal gland and testis, are also affected. X-ALD is caused by mutations in the ABCD1 (ALD) gene encoding a peroxisomal ABC transporter (Mosser et al. 1993), an adrenoleukodystrophy protein, which in the peroxisomal membrane participates in the transport of VLCFAs across the peroxisome membrane into the matrix (Smith et al. 2000). Therefore, the principal biochemical abnormality in X-ALD patients is the accumulation of VLCFAs in tissues and body fluids (Moser et al. 1995). To date, 406 X-ALD mutations have been identified with no apparent correlation between genotype and phenotype (for a review, see Kemp et al. 2001).

So far, palmitoyl-CoA oxidase deficiency (pseudoneonatal adrenoleukodystrophy) has been reported in only a few patients with a large deletion in the corresponding gene (Fournier et al. 1994). All the patients show severe neurological abnormalities, including hypotonia, hearing impairment and visual failure. In this respect, the absence of functional oxidase leads to deficient oxidation of especially the CoA derivatives of VLCFAs. This explains the elevated VLCFA levels in plasma and tissues from these patients.

Among the single-enzyme deficiencies, MFE-2 deficiency is very common (Paton & Pollard 2000). The clinical presentation of MFE-2 deficiency is severe and resembles Zellweger syndrome in many respects. The MFE-2 disorder, however, can be distinguished from ZS by its normal plasmalogen levels. Abnormalities include neonatal hypotonia, craniofacial dysmorphia, developmental delay, neuronal migration defects or

42

demyelination, which result in early death, mostly within the first year of life. Mutations are found in both the dehydrogenase and the hydratase-2-coding regions of the MFE-2 gene, but not in the SCP-2-like domain coding region. The defects include point mutations and nucleotide deletions of variable size (Möller et al. 2001). The patients show elevated plasma levels of VLCFAs and branched-chain fatty acids and in most cases, accumulate bile acid intermediates. Consequently, the reduced ability of pristanic acid β-oxidation can also be used as one of the indicators of MFE-2 deficiency.

The situation with 3-ketoacyl-CoA thiolase deficiency is less clear. Only in one case was peroxisomal thiolase shown to be absent according to immunoblot analyses (Schram et al. 1987). Moreover, the finding indicated that not only VLCFAs but also abnormal bile acids accumulated in this patient, which is reminiscent of SCPx deficiency (Antonenkov et al. 1997, Seedorf et al. 1998) rather than straight-chain 3-ketoacyl-CoA thiolase deficiency, as postulated here.

3 Outlines of the present study

Mammalian peroxisome contains two multifunctional enzymes, MFE-1 and MFE-2. Both of them catalyze the second and third reaction of the β-oxidation cycle, but via reciprocal stereochemical routes. When this work was started, MFE-2 was cloned from various species, including human (Adamski et al. 1995) and rat (Dieuaide-Noubhani et al. 1996, Qin et al. 1997a). According to sequence alignments, MFE-2 consists of multiple domains. The domain order was first resolved by deletion studies done with S. cerevisiae MFE-2 (Hiltunen et al. 1992). Subsequent experiments with mammalian MFE-2 revealed that the NH2-terminal (3R)-hydroxyacyl-CoA dehydrogenase region is followed by (2E)-enoyl-CoA hydratase-2, and the SCP-2-like domain is located in the COOH-terminus. Despite the fact that MFE-2 was shown to participate in hormone inactivation and in the degradation of straight-chain and 2-methyl-branched-chain fatty acyl-CoAs in vitro, nothing was known about the mode of action of any of the domains at the structural level. Moreover, conserved amino acid residues in the NH2-terminal part of MFE-2 had been shown to be critical for catalysis in the enzymes belonging to the SDR superfamily, but no experiments had identified the residues needed in the water-adding reaction catalyzed by hydratase-2. Furthermore, the role of the SCP-2-like domain for the functionality of MFE-2 remained to be investigated; at least it was not needed for the oxidation of medium straight-chain fatty acyl-CoAs catalyzed by the recombinantly produced dehydrogenase or hydratase-2 region of porcine MFE-2 (Leenders et al. 1996b). Therefore, in this work, the structure-function relationship of mammalian MFE-2 was studied in terms of the following topics:

1. Crystallization and structural determination of the NH2-terminal (3R)-hydroxyacyl-CoA dehydrogenase region of rat MFE-2 (III),

2. Identification and characterization of amino acid residues critical for (2E)-enoyl-CoA hydratase-2 activity in human MFE-2 (I),

3. Crystallization and structural determination of the SCP-2-like domain to reveal its putative role in human MFE-2 (II).

4 Materials and methods

The materials and methods have been described in more detail in the original articles referred to by their Roman numerals (I-III).

4.1 cDNA cloning

4.1.1 Cloning of the (3R)-hydroxyacyl-CoA dehydrogenase region from

rat MFE-2 (III)

Total RNA from livers of Wistar rats (Laboratory Animal Centre of the University of Oulu) was isolated using the REX total RNA isolation kit. The cDNA encoding rat MFE-2 was obtained from the isolated total RNA by reverse transcription with Moloney murine leukemia virus reverse transcriptase (Gibco BRL, Gaithersburg, MD, USA). The coding sequence for the (3R)-hydroxyacyl-CoA dehydrogenase region was amplified by PCR using 5´-primer (cacttcc ATG GCT TCG CCT CTG AGG TTC GAC G, where a lower case sequence indicates mismatches to the MFE-2 sequence) and 3´-primer (catctggatcctca ATC TGC AGA TGC CAC TTG ACC G). After subcloning into the pUC18 vector with a SureClone ligation kit (Amersham Pharmacia Biotech AB, Uppsala, Sweden) for the verification of the nucleotide sequence, the insert was released by using NcoI and BamHI restriction endonucleases (the restriction sites are underlined in the primers) and subsequently subcloned into the pET3d expression vector (Novagen, Madison, WI, USA), resulting in a plasmid pET3d::dh∆SCP-2L∆ encoding dehydrogenase region of rat MFE-2 (Fig. 9).

45

Fig. 9. The variants of MFE-2 used in this study. All the other proteins, except dh SCP-2L , which is from rat MFE-2, are from human protein.

4.1.2 Cloning of human MFE-2 (HsMFE-2) and its variants, HsMFE-

2(dh∆) and d∆h∆SCP-2L (I, II)

The open reading frame of human MFE-2 cDNA was obtained from total RNA isolated from human fibroblasts (Chomczynski & Sacchi 1987) using Moloney murine leukemia virus reverse transcriptase (Gibco BRL). The resulting cDNA was used as a template in PCR reactions in order to amplify DNA fragments encoding full-length human MFE-2 (HsMFE-2), the residues 319-736 (lacking the coding region for the dehydrogenase region, HsMFE-2(dh∆)) and the residues 618-736 (lacking the coding regions for the dehydrogenase and hydratase-2 regions, d∆h∆SCP-2L) (Fig. 9). The primers in the PCR reactions were as follows: 5´-primer (cgcaggagctctagaagATG GGC TCA CCG CTG AGG TTC GA) and 3´-primer (cctagctcgagTCA GAG CTT GGC GTA GTC TTT AAG AA) for HsMFE-2, 5´-primer (catatgACA GCA ACA TCA GGA TTT GCT) and 3´-primer (ggatccTCA GAG CTT GGC GTA GTC TTT AAG) for HsMFE-2(dh∆), and 5´-primer (cacttccatg GAG GGC GGG AAG CTT CAG AGT) and 3´-primer (cacctggatcc TCA GAG CTT GGC GTA GTC TTT AA) for d∆h∆SCP-2L. The PCR products were subsequently subcloned into the pUC18 vector using the SureClone Ligation kit (Amersham Pharmacia Biotech AB), and the nucleotide sequences of the inserted DNAs were verified to ensure that no mutations had been introduced. The SacI- and XhoI-digested fragment consisting of HsMFE-2 cDNA was isolated and subsequently cloned into similarly digested plasmid pYE352 (Filppula et al. 1995). NdeI and BamHI as well as NcoI and BamHI were used to release the HsMFE-2(dh∆) and d∆h∆SCP-2L inserts, which were subcloned into the pET3a and pET3d expression vectors (Novagen), respectively. This gave the following plasmids: pYE352::HsMFE-2, pET3a::HsMFE-2(dh∆) and pET3d::d∆h∆SCP-2L.

46

4.2 Site-directed mutagenesis (I)

A QuickChange mutagenesis kit (Stratagene, La Jolla, CA, USA) was applied to introduce point mutations in the pYE352::HsMFE-2 and pET3a::HsMFE-2(dh∆) plasmids to identify the amino acid residues critical for (2E)-enoyl-CoA hydratase-2 activity. The mutated variants of pYE352::HsMFE-2 that were not able use oleic acid in vivo (see below) but were able to dehydrogenate (3R)-hydroxyacyl-CoA substrates in vitro were subjected to further characterization by introducing the same mutations in pET3a::HsMFE-2(dh∆). This resulted in the following plasmids: pET3a::HsMFE-2(dh∆Y347A), pET3a::HsMFE-2(dh∆E366A) and pET3a::HsMFE-2(dh∆D510A) (see Fig. 9). The mutations in the constructs were confirmed by nucleotide sequencing.

4.3 Complementation of Saccharomyces cerevisiae fox-2 with human

MFE-2 and its mutated variants (I)

The pYE352::HsMFE-2 and its mutated variants were transformed into S. cerevisiae fox-2 cells, which lacked the endogenous gene for MFE-2 (Hiltunen et al. 1992), by the lithium acetate method (Gietz et al. 1995) and selected on ura- plates. The transforms growing on the ura- plates were streaked onto oleic plates. Utilization of oleic acid by the transformed fox-2 cells was assessed by observing the clearing zones on the oleic acid plates. For enzyme assays, transformed yeast cells were cultured at 30°C on either a rich or a synthetic medium as described (Qin et al. 1999). The harvested cells were stored frozen at -70°C until used.

4.4 Production of recombinant proteins (I, II)

The recombinant HsMFE-2(dh∆), HsMFE-2(dh∆Y347A), HsMFE-2(dh∆E366A), HsMFE-2(dh∆D510A) and d∆h∆SCP-2L (Fig. 9) were expressed in Escherichia coli (E. coli) BL21(DE3)plysS cells following the supplier´s instructions (Novagen). M9ZB medium supplemented with 50 µg/ml carbenicillin and 34 µg/ml chloramphenicol was used in the expression experiments. E. coli cells containing the plasmid of interest were used to inoculate one litre of culture. The cells were allowed to grow under aerobic conditions at 37°C until A600 of 0.6 was reached. The expression of recombinant protein was induced by adding IPTG to a final concentration of 0.4 mM. After two hours of additional incubation, the bacterial cells were harvested, washed with cold buffer (120 mM NaCl, 16 mM potassium phosphate, pH 7.4, 0.02 % NaN3) and stored at -70°C until used.

47

4.4.1 Production of selenomethionine-labelled recombinant (3R)-

hydroxyacyl-CoA dehydrogenase (III)

E. coli strain of B834(DE3) was transformed with pET3d::dh∆SCP-2L∆. After the transformation, a single, freshly grown colony of E. coli B834(DE3)(pET3d::dh∆SCP-2L∆) was picked from a LB plate and grown in LB medium overnight at 37°C. The cells from this overnight culture were collected by centrifugation, washed with 20 ml of selenomethionine (Se-Met) minimal medium [Se-Met-MM, 48 mM Na2HPO4, 22 mM KH2PO4, 9 mM NaCl, 0.4 % (w/v) glucose, 2 mM MgSO4, 0.1 mM CaCl2, 8 mM (NH4)2SO4, 1x BME vitamin solution (Gibco BRL), 0.5 mM L-Se-Met (CalBiochem-NovaBiochem Corp., San Diego, CA, USA)], centrifuged once again and resuspended in 20 ml of Se-Met-MM. This 20 ml culture was then used as an inoculum for 2 liters of prewarmed (30°C) Se-Met-MM medium in the presence of 50 µg/ml carbenicillin. The cells were allowed to grow under aerobic conditions until A600 of 0.9 was reached. The temperature was shifted from 37°C to 20°C, and the expression of dh∆SCP-2L∆ was induced by the addition of IPTG to a final concentration of 1 mM. After 20 hours of additional incubation, the bacterial cells were harvested, washed with cold buffer (120 mM NaCl, 16 mM potassium phosphate, pH 7.4, 0.02 % NaN3) and stored at -70°C until used.

4.5 Protein purification (I-III)

The cell walls of E. coli cells were digested with lysozyme (100 µg/ml), and the viscosity of the cellular lysate was reduced by using DNaseI (25 µg/ml) and RNaseA (25 µg/ml). Subsequently, the cell debris was sedimented by centrifugation, and the soluble protein fraction was applied to the first column of the purification protocol. HsMFE-2(dh∆), HsMFE-2(dh∆E366A), HsMFE-2(dh∆D510A), d∆h∆SCP-2L and dh∆SCP-2L∆ (Fig. 9) were purified using anion exchange, cation exchange and size exclusion chromatography columns, including Q-Sepharose (II & III, Amersham Pharmacia Biotech AB), DEAE-Sephacel (I, Amersham Pharmacia Biotech AB), Poros 10 SP (I, PerSeptive Biosystems, Cambridge, MA, USA), Resource S (II & III, Amersham Pharmacia Biotech AB), SuperdexTM 200 10/30 (I & III, Amersham Pharmacia Biotech AB) and SuperdexTM 75 10/30 (II, Amersham Pharmacia Biotech AB). Detailed descriptions of the purification protocols are given in the original articles.

4.6 Enzyme assays (I, III)

The activities of (2E)-enoyl-CoA hydratase-2 and (3R)-hydroxyacyl-CoA dehydrogenase or their combined activity were measured spectrophotometrically using a Shimadzu UV 3000 spectrophotometer. In the reaction mixture, the formation of the magnesium

48

complex of 3-ketodecanoyl-CoA was monitored at 303 nm (Hiltunen et al. 1989). The kinetic constants (kcat and Km) for hydratase-2 were measured via the reduction of the double bond in (2E)-decenoyl-CoA at 263 nm (Binstock & Schulz 1981) and using variable concentrations (4-60 µM) of the substrate. The pH dependence experiment was carried out in 200 mM potassium phosphate buffer at pH values ranging from five to 10. The kinetic data, the pKa values and their errors were calculated with the GraFit computer software (Sigma, St. Louis, MO, USA).

4.7 Dynamic light scattering and CD-spectroscopic measurements (I)

The monodispersity and hydrodynamic radii of purified HsMFE-2(dh∆) and its mutated variants were studied with a DynaPro-MSTC dynamic light scattering device (Protein solution, Charlottsville, VA, USA) at 6°C. The secondary structural elements were analyzed using circular dichroism (CD) spectroscopy with a Jasco JHO spectropolarimeter at 22°C. The far-UV spectra of the recombinant proteins were measured from 200 to 250 nm in 80 mM potassium phosphate buffer, pH 7.0.

4.8 Surface plasmon resonance measurements (II)

d∆h∆SCP-2L was immobilized on a carboxymethylated dextran matrix (CM5) chip with standard amine coupling chemistry by flowing a protein solution in 10 mM phosphate buffer (pH 4.8) over the chip in a Biacore 3000 instrument (Biacore AB, Uppsala, Sweden). The amount of immobilized protein was 1900 resonance units. Kd values were estimated with the steady-state affinity model according to the manufacturer’s procedure implemented in BIA evaluation 3.1.

4.9 Crystallization, data collection and processing (II, III)

Initial crystallization screening was performed using the sparse matrix screening method described by Jancarik & Kim (Jancarik & Kim 1991) and the hanging-drop and sitting-drop vapour diffusion methods. The crystallization conditions of crystals suitable for structural determination have been described in more detail in the original articles.

d∆h∆SCP-2L was crystallized at 4°C, and a data set was collected from a single crystal using the oscillation method with 0.5° rotations per frame at a wavelength of 0.9785 Å and a MAR CCD detector at beamline BW7A at the EMBL outstation at the DESY synchrotron, Hamburg, Germany. A multiwavelength anomalous dispersion (MAD) experiment, with a Se-Met-labelled dh∆SCP-2L∆ crystal grown at 22°C, was performed on beam line BM30, ESRF, Grenoble, France. The X-ray fluorescence from

49

the crystal was measured as a function of the incident X-ray energy near the Se-κ edge. The wavelengths chosen for data collection were λ2 = 0.97927 Å and λ3 = 0.97243 Å, corresponding to a minimum of f´ and a maximum of f´´, respectively. A third, remote energy was selected at 12653 eV (λ1 = 0.97958 Å). The MAD data was collected from a single crystal with a strategy of strict inverse beam settings, 1° rotation per frame and a MAR345 detector. All three wavelengths were collected successively in portions of forty frames that consisted of 20 frames from ø and 20 frames from ø + 180º. In both cases, d∆h∆SCP-2L and dh∆SCP-2L∆, the images were processed using DENZO and the reflections merged by applying SCALEPACK of the HKL suite (Otwinowski & Minor 1997).

4.10 Structural determination, model building and refinement (II, III)

For d∆h∆SCP-2L, the initial phases were determined from a molecular replacement solution obtained with AMoRe (Nawaza 1994) using rabbit SCP-2 (PDB code 1C44, Choinowski et al. 2000) as the search model. Subsequently, the phases were used as an input for the autobuilding procedure in warpNtrace (Perrakis et al. 1999). The resulting structure was further refined and model built using CNS (Brünger 1998) and O (Jones et al. 1991).

In the case of dh∆SCP-2L∆, the initial experimental phases were obtained and refined by the program SOLVE (Telwilliger & Berendzen 1999). The phases were then improved by the application of phase extension, 2-fold averaging and solvent flattening, in a single run with the program DM (Cowtan & Zhang 1999) using the AUTOMASK option. This allowed the calculation of better-quality maps, which were used as a starting point for autobuilding with MAID (Levitt 2001). The resulting model was further improved with model building in O (Jones et al. 1991) and refinement with CNS (Brünger 1998).

Throughout the refinement, structural integrity was monitored by WHATIF (Vriend 1991) and PROCHECK (Laskowski et al. 1993). The locations and types of structural motifs were defined by the PROMOTIF (Hutchinson & Thornton 1996) and DSSP (Kabsch & Sander 1983) programs. Interatomic contacts were analyzed by the LPC software (Sobolev et al. 1999).

4.11 Other methods

Oligonucleotides were prepared with an Applied Biosystems DNA Synthesizer (Perkin-Elmer, Norwalk, CT, USA) or purchased from Amersham Pharmacia Biotech. Nucleotide sequences were verified by the dideoxynucleotide or the dye terminator cycle sequencing methods.

Proteins were separated on a SDS-polyacrylamide slab gel according to their molecular weights (Laemmli 1970) in the presence of 1.8 M urea and using low molecular weight standards (Bio-Rad Laboratories, Hercules, CA, USA) for calibration.

50

The protein concentrations of the samples, which were subjected to SDS-polyacrylamide gel electrophoresis (SDS-PAGE), were determined according to Bradford (Bradford 1976) with the Bradford reagent (Bio-Rad Laboratories). For the purified proteins, absorption at 280 nm was measured and used for fine adjustment of the protein concentration.

To find out if the mutant variants of human MFE-2 were expressed in S. cerevisiae, the proteins from SDS-PAGE were blotted electrophoretically onto a nitrocellulose filter (Towbin et al. 1979). The transferred proteins were identified by using first anti-hydratase-2 antibody (1:1000 dilution) (Malila et al. 1993) and then affinity-purified goat anti-rabbit IgG with horseradish peroxidase conjugate (Bio-Rad Laboratories) as the secondary antibody. As a visual indicator, 4-chloro-1-naphthol (Sigma) was used as the electron acceptor.

To identify the degree of Se-Met labelling mass spectrometric analyses were carried out for the purified dh∆SCP-2L∆ and Se-Met labelled dh∆SCP-2L∆. The experiments were

performed using a LCT (Micromass LTD) orthogonal time-of-flight mass spectrometer with OpenLynx3 Data system. The diluted samples in acetonitrile-water solution (50:50) containing 0.1% formic acid solution were directed by a Harward 5311 syringe pump into the Z-spray electrospray source at a flow rate of 10 µl/minute. The vaporizer temperature was 140°C and N2 was used both as nebulizer (80 l/hour) gas and desolvation (400 l/hour)

gas. The electrospray capillary was kept at 3.8 kV and the sample cone voltage at 35 V.

5 Results

5.1 Binary structure of the two-domain (3R)-hydroxyacyl-CoA

dehydrogenase region of rat MFE-2 (III)

Multiple-sequence alignment of several mammalian peroxisomal MFE-2s revealed nonconserved amino acid residues (His311-Ser321, in the rat sequence) between the dehydrogenase and hydratase-2 regions (Fig. 1, original article III). Based on this information, plasmid pET3d::dh∆SCP-2L∆ was constructed. This plasmid encodes the Met1-Asp319 of rat MFE-2 and therefore lacks the coding regions for the hydratase-2 and SCP-2-like domains. This was in line with the results of the study, where the NH2-terminal fragments of MFE-2, including Met1-Thr312 and Met1-Ala316, were isolated from rat liver with (3R)-hydroxyacyl-CoA dehydrogenase activity (Dieuaide-Noubhani et al. 1996, 1997). The purified dh∆SCP-2L∆ showed similar catalytic efficiency (kcat of 3.7 s-1) as that obtained for the full-length recombinant rat MFE-2 (in this study). When applied to a size exclusion column, the elution volume of dh∆SCP-2L∆ corresponded to the size of a homodimer.

For structural determination, dh∆SCP-2L∆ was produced as a Se-Met-labelled protein and purified. The mass-spectrometric analysis of native and Se-Met dh∆SCP-2L∆ indicated complete labelling, where all of the five methionines were replaced by Se-Met. Subsequently, thin trigonal prismatic crystals 0.4 x 0.2 x 0.05 mm3 in size were obtained from Se-Met dh∆SCP-2L∆ within a few weeks.

The MAD dataset on the frozen Se-Met-labelled dh∆SCP-2L∆ crystal consisted of three different wavelengths with 5-fold redundancy and 92.1-99.2 % completeness, when 2.38 Å was used as the diffraction limit (Table 1, original article III). The crystal belonged to the space group P21 with unit cell dimensions of a = 89.466, b = 82.748, c = 95.745 and β = 94.201. Assuming four monomers per asymmetric unit, the Vm value of the crystal was 2.55 Å3/Da, which is within the most probable range of values for proteins (Matthews 1968). The initial experimental phases were obtained and refined by the program SOLVE (Telwilliger & Berendzen 1999) from twelve of the twenty expected selenium atoms per asymmetric unit using 2.6 Å resolution as the diffraction limit. With

52

Table 3. MAD-data collection statistics of dh∆SCP-2L∆ with resolution shells.

Data set Resolution shell

Rmerge

(%) Completeness (%)

I/ I Unique reflections

Remote 24.00 – 5.11 7.8 86.8 18.2 4995

5.11 – 4.06 7.3 89.4 17.8 5057

4.06 – 3.55 9.1 90.7 14.9 5106

3.55 – 3.23 9.8 91.5 14.2 5136

3.23 – 3.00 9.7 92.8 13.4 5202

3.00 – 2.82 11.4 93.2 11.5 5212

2.82 – 2.68 14.6 93.9 8.9 5259

2.68 – 2.56 20.0 93.7 6.9 5242

2.56 – 2.46 19.6 94.8 7.1 5292

2.46 – 2.38 21.2 94.4 6.1 5251

Inflection point 24.00 – 5.11 8.3 99.9 15.4 5744

5.11 – 4.06 7.7 99.6 16.4 5622

4.06 – 3.55 9.8 99.5 13.7 5606

3.55 – 3.23 10.2 99.0 13.5 5560

3.23 – 3.00 10.4 99.4 12.6 5563

3.00 – 2.82 12.2 99.2 11.1 5552

2.82 – 2.68 15.4 99.0 9.3 5545

2.68 – 2.56 19.9 98.8 7.4 5523

2.56 – 2.46 19.6 99.1 7.3 5536

2.46 – 2.38 21.1 98.4 6.6 5466

Peak 24.00 – 5.11 10.2 99.8 13.5 6294

5.11 – 4.06 9.9 99.7 13.5 6167

4.06 – 3.55 11.8 99.6 11.4 6148

3.55 – 3.23 12.4 99.6 10.9 6133

3.23 – 3.00 13.3 99.7 10.2 6130

3.00 – 2.82 15.7 99.4 9.3 6077

2.82 – 2.68 21.3 99.3 7.5 6092

2.68 – 2.56 22.3 99.3 6.3 6088

2.56 – 2.46 23.3 97.1 6.1 5936

2.46 – 2.38 17.8 30.0 4.7 1841

this resolution limit the completenesses were 91.5 %, 99.3 % and 99.6 % for the remote, inflection point and peak data sets, respectively (Table 3). This, however, did not result in interpretable maps and clear solvent boundaries. The figure-of-merit of the experimental phases was 0.34. Since dh∆SCP-2L∆ was shown to exist as a dimer in aqueous solution, a search for non-crystallographic symmetry was initiated. Indeed, three solutions of the self-rotation function, calculated with GLRF (Tong & Rossmann 1997), were found at κ = 180°, φ = 0°, ψ = 86° and κ = 180°, φ = 27°, ψ = 23° and κ = 180°, φ = 160°, ψ = 31° with peak heights of 10.2 σ, 2.1 σ and 2.0 σ, respectively. When applied to the list of

53

heavy atom sites obtained from SOLVE, the program FINDNCS (Lu 1999) was used to both confirm the orientation of the non-crystallographic 2-fold axis (κ = 180°, φ = 0°, ψ = 86°) found from the self-rotation function and to provide the translational component for the DM run (Cowtan & Zhang 1999). When the local 2-fold axis, which connected two dimers and therefore covered the whole asymmetric unit, was used in line with solvent flattening and phase extension in a single DM run, the correlation coefficient between masked map areas improved from 0.49 to 0.95. This allowed the calculation of better maps, which were used as a starting point for autobuilding with MAID (Levitt 2001). Eight selenium sites out of those 12 found by SOLVE were in the electron density of the experimental maps and they corresponded to selenomethionines in the αE and βE of each monomer (Fig. 1, original article III). The other four selenium sites, which were a little bit off the density, belonged to Se-Mets preceding αCT1 (Fig. 1, original article III). Further iterations of refinement and model building with O (Jones et al. 1991) resulted in a final model with R and free R-factors of 19.7 % and 23.6 %, respectively (Table 1, original article III), with good stereochemistry, as judged from the Ramachandran plot.

The crystal structure reveals two identical homodimers per asymmetric unit with no conformational changes. The two monomers of a dimer superimpose well with an rms deviation of 0.21 Å using Cα-backbone. The main body of the dimer can be described as being oval-shaped (Fig. 2, original article III) and consisting of monomers with an α/β doubly wound structure (Fig. 10). Each monomer is built of a core of a β-sheet of six parallel β-strands (βA to βF) with the following topology: -1×, -1×, 3×, 1×, 1×, 1×, where × represents a crossover connection between the strands (Fig. 3, original article III). The central β-sheet is sandwiched between two arrays of three parallel α-helices (αB, αC, αG and αD, αE, αF) on both sides (Fig. 10). This folding pattern is conserved in the enzymes of the SDR superfamily, and it is usually referred to as the basic α/β SDR fold that mainly covers the cofactor-binding site. The association of monomers into a dimer is partly accomplished by the long helices, αE and αF, comprising the interfacial four-helix

Fig. 10. The overall fold of the two-domain rat (3R)-hydroxyacyl-CoA dehydrogenase monomer. The bound NAD+ sits on the surface created by the COOH-terminal ends of the β-strands of the central β-sheet. The α-helices, αE and αF, form a four-helix bundle with the neighbouring monomer at the dimer interface. The novel COOH-terminal (C) domain is not seen in the other enzymes of the SDR superfamily. The figure was prepared with the Swiss-PDBViewer (Guex & Peitsch 1997).

54

bundle typical of dimeric and tetrameric SDR proteins. In dh∆SCP-2L∆, the αE and αF helices from one monomer are in contact with the corresponding helices from the other monomer in such a way that the helices from different monomers run nearly antiparallel. In the αE-αE interface, large hydrophobic and polar side chains face each other, and the backbone carbonyls are approximately 8 Å apart. In the αF-αF interface, a few large hydrophobic side chains do not point towards the dimer interface. Consequently, the two αF helices are closer to each other (4 Å) compared to the αE-αE interface.

All the SDR protein structures available from the PDB are one-domain enzymes and the dimeric interactions are dominated solely by the four-helix bundle. In this respect, the crystal structure of dh∆SCP-2L∆ is unique because it is the first two-domain enzyme (Fig. 10) that belongs to the SDR superfamily with a more advanced quaternary association. The antiparallel β-strands, βG and βH, act as a linker between the NH2-terminal NAD(H)-binding domain (Fig. 2, original article III) and the novel COOH-terminal domain. Because of the hairpin structure formed by the β-strands βG and βH, the COOH-terminal domain of monomer A continues to the other side of a dimer, where it closely interacts with monomer B, and vice versa (Fig. 2, original article III). The point at which the COOH-terminal domains of both monomers intersect is Trp249 in the β-strand βH. The β-strand βI and three α-helices, αCT1, αCT2 and αH, in the COOH-terminal domain are in close proximity with the active site cavity of the neighbouring monomer (Fig. 5, original article III). This arrangement further strengthens the dimeric interactions.

Fig. 11. The NAD+ and catalytic triad. The conformation of the bound NAD+ and the distances of Ser151, Tyr164 and Lys168 are identical to those reported for other SDR enzymes. This indicates that the hydride transfer catalyzed by the Ser-Tyr-Lys catalytic triad is valid for the (3R)-hydroxyacyl-CoA dehydrogenase region of rat MFE-2 as well. The close contact between Asp40 and the hydroxyl groups of the adenine ribose does not leave enough space for the phosphate group in NADP(H). This guides specificity towards NAD(H). The figure was prepared with the Swiss-PDBViewer (Guex & Peitsch 1997).

55

In each monomer, the bound NAD+ molecule is in an extended conformation and sits on a surface created by the COOH-terminal ends of the β-strands in the NH2-terminal cofactor-binding domain (Fig. 3, original article III). The adenine and nicotinamide rings are oriented roughly perpendicular to the planes of the respective riboses, resulting in an anti conformation for the adenine ring and a syn conformation for the nicotinamide ring (Fig. 11). Both ribose rings reveal 2E (C2´-endo) puckering. The hydroxyl group of Tyr164 is hydrogen-bonded (2.6 Å) to the 2´-hydroxyl group on the nicotinamide ribose. Thus, the tyrosine is in close proximity with the nicotinamide ring (3.0 Å) (Fig. 11). The ε-amino group of Lys168 forms a hydrogen bond (3.0 Å) with the 3´-hydroxyl of the nicotinamide ribose (Fig. 11), an interaction needed to orient the nicotinamide ring in the syn conformation. The arrangement of Ser151, Tyr164 and Lys168 around NAD+ and the syn conformation of the nicotinamide ring strongly support the B-face 4-pro-S hydride transfer reaction, which is preserved within the enzymes in the SDR superfamily.

5.2 Site-directed mutagenesis studies on (2E)-enoyl-CoA hydratase-2

derived from human MFE-2 (I)

The amino acid sequence alignment of various hydratase-2 domains was used to identify the conserved protic residues putatively responsible for (2E)-enoyl-CoA hydratase-2 activity (Fig. 2, original article I). The comparison revealed that the hydratase-2 region contains 12 conserved protic amino acid residues. Additionally, the most conserved part of hydratase-2 (residues 452-612 in human MFE-2) yielded significant similarity to five dehydratases of the β-subunit of fungal fatty acid synthetase complexes and three domains of the fatty acid synthetases of prokaryotic microorganisms. Further characterization of the conserved region allowed the identification of a motif [YF]-X(1,2)-[LVIG]-[STGC]-G-D-X-N-P-[LIV]-H-X(5)-[AS], which was named the hydratese-2 motif (Fig. 2, original article I).

Because the amino acid residues crucial for catalysis are generally highly conserved and the dehydration-hydration of ∆2 carboxylates has been shown to take place through acid-base catalysis, the conserved protic residues in the human MFE-2 were mutated to alanine and tested in vivo in yeast. Both fox-2 cells transformed with pYE352::HsMFE-2 and wild-type S. cerevisiae UTL-7A cells were able to use oleic acid as a carbon source (see “Materials and methods”). However, five mutated variants of the 12 identified protic residues failed to utilize oleic acid (Fig. 3, original article I), even though all of them were properly translated as judged by the immunoblotting experiments. In vitro characterization revealed that the three noncomplementing variants, Y347A, E366A and D510A, had lost only their hydratase-2 but not dehydrogenase activity (Table 3, original article I). Therefore, these three mutants were subjected to a further exploration of the catalytic properties of the hydratase-2 reaction. For this purpose, deleted variants, lacking the dehydrogenase region, of human MFE-2 were produced as recombinant proteins and purified. HsMFE-2(dh∆), HsMFE-2(dh∆E366A) and HsMFE-2(dh∆D510A) showed similar chromatographic properties, but the HsMFE-2(dh∆Y347A) variant could not be purified with any of the different approaches applied. The composition of secondary

56

structures and the quaternary association of HsMFE-2(dh∆) and its variants were shown to be practically identical, as demonstrated by the light scattering and CD measurements (Fig. 5, original article I).

The kinetic variables were measured in the direction of hydration using (2E)-decenoyl-CoA as a substrate. The Km and kcat of HsMFE-2(dh∆) were 8.1 ± 1 µM and 196 ± 7 s-1, which agree with the values estimated for HsMFE-2 isolated from human liver (Jiang et al. 1996). The kcat value of HsMFE-2(dh∆E366A) was nearly 100-fold lower than that of HsMFE-2(dh∆) at pH 5, but the Km values were practically the same (Table 4, original article I). When the kcat values were plotted against pH, the curves were bell-shaped (Fig. 6, original article I), with apparent values of pKa1 of 8.2 ± 0.1 (6.5 ± 0.1) and pKa2 of 9.7 ± 0.2 (9.6 ± 0.2). The values in parenthesis are for HsMFE-2(dh∆E366A). The pH dependence of catalytic efficiency (kcat/Km) showed different profiles for HsMFE-2(dh∆) and HsMFE-2(dh∆E366A), the catalytic efficiency of HsMFE-2(dh∆E366A) being 10 (pH 7.0) to 100 (pH 5.0) times lower than that of HsMFE-2(dh∆). HsMFE-2(dh∆D510A) was enzymatically inactive (Table 4, original article I). This demonstrates the importance of Glu366 and Asp510 for catalysis.

5.3 Crystal structure of the liganded SCP-2-like domain of human

MFE-2 (II)

Multiple amino acid sequence alignment of MFE-2s was used to identify the COOH-terminal SCP-2-like domain (SCP-2L) (data not shown). Subsequently, the fragment Glu618-Leu736 (d∆h∆SCP-2L) of human MFE-2 (Fig. 1, original article II) was expressed in E. coli, purified and crystallized. A dataset of frozen trigonal prismatic crystals 0.25 × 0.20 × 0.40 mm3 in size was collected using synchrotron radiation. The data was 98.3 % complete with 3.6-fold redundancy to 1.75 Å resolution (Table 2, original article II). The data collection statistics are given in Table 4. The liganded structure of d∆h∆SCP-2L was solved at 1.75 Å resolution by applying molecular replacement techniques. Table 4. Data collection statistics of d h SCP-2L with resolution shells.

Resolution shell Rmerge

(%) Completeness (%) I/ I Unique reflections 20.00 – 3.76 2.3 95.1 42.3 1238

3.76 – 2.99 2.2 98.0 40.1 1173

2.99 – 2.61 3.0 97.9 29.5 1176

2.61 – 2.37 4.3 99.1 20.4 1148

2.37 – 2.20 6.7 98.6 9.4 1179

2.20 – 2.07 6.5 99.1 13.3 1156

2.07 – 1.97 9.1 99.5 9.0 1154

1.97 – 1.89 15.2 99.6 4.6 1167

1.89 – 1.81 19.1 99.4 4.2 1136

1.81 – 1.75 25.9 97.1 3.4 1132

57

Fig. 12. Stereo view of d∆h∆SCP-2L. The spatial arrangement of the central β-sheet and the α-helices results in a hydrophobic tunnel that traverses the protein. The ligand bound to the protein is Triton X-100. The arrow indicates a peroxisomal targeting signal, which became solvent-exposed upon ligand binding and movement of the COOH-terminal helix, helix E. The figure was prepared with the Swiss-PDBViewer (Guex & Peitsch 1997).

d∆h∆SCP-2L has an α/β-fold composed of five α-helices and β-strands (Fig. 12). The overall dimensions of monomeric d∆h∆SCP-2L are 39 × 32 × 32 Å3, and it is thus more spherical in shape than rabbit SCP-2 (Choinowski et al. 2000), which was used as a search model in the molecular replacement experiment. Five β-strands with a mixed sheet structure of -1, -1, +3×, +1 topology form the core of the protein. The solvent-exposed side of the β-sheet layer is mainly populated by polar residues, while the side facing the interior of the protein is composed of hydrophobic amino acids. Moreover, the central β-sheet is covered on one side by five α-helices (Fig. 12), which are all amphipathic, with the polar residues facing the solvent and the apolar ones facing the interior of the protein. The last helix, helix E (Met105-Tyr117), is functionally in a central position with respect to both ligand binding (see “Discussion”) and PTS1, which is located at its COOH-terminal end (Fig. 12). In contrast to human (García et al. 2000) and rabbit (Choinowski et al. 2000) SCP-2 structures, the peroxisomal targeting signal (-Ala-Lys-LeuOH) is ordered and fully solvent-exposed in d∆h∆SCP-2L.

The hydrophobic tunnel formed by the spatial arrangement of the central β-sheet and the α-helices is occupied by an ordered triton molecule. Although Triton X-100 is not a natural ligand of d∆h∆SCP-2L, it could be considered a model molecule for lipids binding to SCP-2 or SCP-2L proteins. Triton X-100 is positioned at the binding site in such a way that the hydrophobic end lies deep inside the hydrophobic tunnel, while the hydrophilic end extends outside the protein (Fig. 6, original article II). Most of the contacts between the triton molecule and d∆h∆SCP-2L are hydrophobic or aromatic (Table 1, original article II), and the surface plasmon resonance measurements gave a Kd value of 400 µM for Triton X-100. Only one hydrogen bond is formed by the Gln108 and O55 of the triton molecule near the entrance of the tunnel. The additional cavities close to the exit of the tunnel and between the helices C and D suggest that molecules larger than Triton X-100 could bind in the tunnel. However, the tunnel as such, with a diameter of 9

58

Å and a length of 18 Å, is large enough to accommodate ligands of the size of various fatty acids or their CoA derivatives, such as straight-chain and branched-chain fatty acyl groups.

6 Discussion

6.1 Structural studies on the (3R)-hydroxyacyl-CoA dehydrogenase

region of rat MFE-2 (III)

The NH2-terminal (3R)-hydroxyacyl-CoA dehydrogenase fragment of MFE-2 proteins belongs to the SDR superfamily. This relationship was already demonstrated by the amino acid sequence analysis with porcine (Leenders et al. 1994) and human (Adamski et al. 1995) MFE-2s. The most conserved motifs in this family are Gly-X-X-X-Gly-X-Gly in the cofactor-binding site and Tyr-X-X-X-Lys, which is part of the active site. Conservation of these critical residues suggests that the short-chain alcohol dehydrogenases are functionally related, especially in respect to the stereochemical course of the catalyzed reaction and the chirality of the accepted substrates. Crystallographic studies showed the architecture of the active site to be consistent with the observed stereochemistry of the enzymes (Ghosh et al. 1991, 1994, Tanaka et al. 1996). Additionally, yeast MFE-2 was shown to follow the same stereochemical course by accepting only the R-isomer of the hydroxyl group of a substrate in vitro (Hiltunen et al. 1992).

To study the structural basis of the rat (3R)-hydroxyacyl-CoA dehydrogenase region (dh∆SCP-2L∆) of MFE-2, the corresponding fragment was expressed in E. coli. Purified dh∆SCP-2L∆ had a kcat value of 3.7 s-1, which is comparable to the kcat value of 3.4 s-1 obtained for full-length recombinant rat MFE-2 (this study). Similar results have been found for porcine MFE-2 (Leenders et al. 1996b). This showed that the dehydrogenase produced as itself had similar catalytic properties as when it was part of full-length MFE-2.

The crystal structure of dh∆SCP-2L∆ showed structural similarities with the other SDR enzymes. Especially the central β-sheet (βA to βF) and the two arrays of three parallel α-helices (αB, αC, αG and αD, αE, αF) surrounding it are virtually identical in spatial arrangement. This conserved SDR fold covers the regions for cofactor binding and contains the catalytic triad and α-helices, αE and αF, for dimeric interactions. To date, all the members of the SDR superfamily have been one-domain enzymes with the conserved

60

SDR fold and, sometimes, short variable amino acid stretches, approximately ten residues long, in the COOH-terminus. The COOH-terminal extension in dh∆SCP-2L∆ is much longer, including about 60 residues with a folding pattern found neither in previously characterized SDR, nor in other proteins, giving rise to the first two-domain member of the SDR protein family. The characterized COOH-terminal domain is involved in lining the active site cavity of the neighbouring monomer. This kind of quaternary association makes the dimeric interactions more extensive when compared to other dimeric SDR enzymes. Furthermore, the COOH-terminal domain folds into a circle and provides an entrance towards the amino acid residues putatively critical for catalysis. Additionally, the positive clusters in this domain demonstrate interaction sites for the negatively charged phosphates of the fatty acyl-CoA molecule.

The bound NAD+ is in an extended conformation, as judged from the distance (14.3 Å) between the C6 of the adenine and the C2 of the nicotinamide in the syn conformation. This value agrees with the distances reported for the other crystal structures of SDR proteins. Furthermore, when compared to other SDR enzymes, the positions of Ser151, Tyr164 and Lys168 in respect to extended nicotinamide dinucleotide suggest the occurrence of hydride transfer on the si face of the nicotinamide ring. In this reaction mechanism, the tyrosine residue acts as a catalytic base to extract hydrogen from the hydroxyl group of the substrate, while the serine stabilizes the reaction intermediate via its hydroxyl group. Subsequently, the β-hydride ion, liberated from position 3 of the substrate, is directly transferred to position 4 of the nicotinamide ring.

In dh∆SCP-2L∆, the glycine-rich loop of 12 residues between the β-strand βB and the α-helix αC could be linked to substrate preference. This glycine-rich loop, located at the bottom of the active site cavity, is present among the (3R)-hydroxyacyl-CoA dehydrogenases accepting substrates of long chain lengths. However, dehydrogenases acting on short-chain substrates and the non-peroxisomal SDR dehydrogenases lack this loop region (Qin et al. 2000), suggesting a role in the accommodation of long-chain substrates.

6.2 Importance of two protic residues for (2E)-enoyl-CoA hydratase-2

activity (I)

The physiological importance of MFE-2 is well demonstrated by the incapability of yeast fox-2 cells to utilize fatty acids, especially long-chain fatty acids, such as oleic acid (cis-C18:1(9)), as a carbon source (Hiltunen et al. 1992). An even more striking feature of the patients with MFE-2 deficiency is their failure to thrive, resulting in death in early infancy. A similar phenotype compared to patients with MFE-2 deficiency was observed with MFE-2 “knock-out” mice (Baes et al. 2000). In both cases, increased plasma levels of VLCFAs and branched-chain fatty acids were demonstrated. Despite the extensive characterization of MFE-2 genetic errors, surprisingly little was known about the catalytic properties or mechanism of the hydratase-2 reaction when catalyzed by MFE-2. Since the hydratase-1 reaction catalyzed by either the multifunctional β-oxidation complex of E. coli (He & Yang 1997) or mitochondrial hydratase-1 from rat liver (Kiema

61

et al. 1999) follows a two-residue mechanism involving protic residues, it was hypothesized that acid-base catalysis is valid for hydratase-2. Furthermore, in the biosynthesis of unsaturated fatty acids in E. coli, the reaction catalyzed by 3-hydroxydecanoyl thiol ester dehydratase was shown to involve the participation of both a histidine and an aspartic acid residue (Leesong et al. 1996).

In the first approach, multiple amino acid sequence analysis was used to identify the conserved protic residues within hydratase-2 proteins. Subsequently, these residues were mutated to alanine, and the ability of the mutated variants to utilize oleic acid in vivo was tested. Three amino acid residues, Tyr347, Glu366 and Asp510, were identified as critical only for the hydratase-2 activity of MFE-2. In the second step, the recombinant variants affecting only hydratase-2 activity were subjected to further characterization in vitro. Purified HsMFE-2(dh∆), HsMFE-2(dh∆E366A) and HsMFE-2(dh∆D510A) exhibited dimeric subunit composition similar to the 46 kDa degradation product of MFE-2 from either human (Jiang et al. 1996) or rat (Li et al. 1990) liver. HsMFE-2(dh∆Y347A) behaved chromatographically differently and was therefore left uncharacterized. The enzymatic inactivation of D510A and the decrease in the catalytic efficiency of E366A resulting from the unfolding of the polypeptide were ruled out by investigating secondary structural elements with a CD spectropolarimeter using HsMFE-2 as a reference. The pH dependence curve of kcat/Km of HsMFE-2(dh∆) was shown to be bell-shaped, suggesting that two protic residues participate in the catalysis. The estimated pKa2 values were about the same for both HsMFE-2(dh∆) and HsMFE-2(dh∆E366A), whereas the apparent pKa1 of HsMFE-2(dh∆E366A) dropped from 8.2 to 6.5. Furthermore, the 100-fold decrease in catalytic efficiency at low pH was not caused by a change in substrate binding because the Km values remained unchanged over the pH range tested. This suggests that the negatively charged side-chain of Glu366 acts as a base in catalysis. Additionally, alignment of HsMFE-2(dh∆) with the other known hydratase-2 proteins of the β-oxidation pathway revealed a new fingerprint of hydratase-2. The identified novel hydratase-2 motif is also conserved in the R-specific dehydratases of fatty acid de novo synthesis in fungi and some prokaryotes. Even though the catalytic properties of HsMFE-2(dh∆D510A) could not be analyzed because of the overall inactivity, the location of Asp510 within the conserved hydratase-2 motif indicates its important role in the hydratase-2 reaction. As a consequence, Asp510 either acts as an acid in the catalysis or affects the binding of the substrate.

6.3 Ligand binding to the SCP-2-like domain of human MFE-2 (II)

The structure of d∆h∆SCP-2L complexed with Triton X-100 demonstrates a hydrophobic tunnel that traverses the protein molecule and accommodates the bound ligand. The hydrophobic tunnel was also described in the unliganded rabbit SCP-2 crystal structure (Choinowski et al. 2000), and the authors hypothesized the putative entrance and exit of the tunnel. In d∆h∆SCP-2L, the triton molecule enters the tunnel through the entrance built up by the helices D and E and the β-strand V. Compared to rabbit SCP-2, the tunnel is similar in length but wider. Moreover, the positions of the loop between the β-strands I

62

and II and the COOH-terminal helix are different in d∆h∆SCP-2L and rabbit SCP-2 (Fig. 3, original article II), indicating conformational changes upon ligand binding. Furthermore, the amino acid residues that form the entrance are conserved in SCP-2 or SCP-2-like proteins (Fig. 1, original article II). This conservation might reflect functional similarity within this family.

Human SCP-2 was shown to interact strongly with acyl-CoAs with carbon chain length of 24 or 26 with Kd of 2-4 nM (Frolov et al. 1996c, Dansen et al. 1999). Similarly, the surface plasmon resonance binding experiments of various fatty acyl-CoAs demonstrated the strongest interaction between d∆h∆SCP-2L and long- and very-long-chain fatty acyl-CoAs (unpublished results). In the mode of binding, the positive cluster (Arg97-Arg101) near the entrance of the tunnel could serve as a potential interaction site for the phosphate groups of the CoA molecule. This would indicate that both the pantetheine group and the fatty acyl tail of a fatty acyl-CoA molecule are localized in the tunnel. Due to the limited space of the tunnel, the -end of an acyl tail longer than eight carbons should protrude out of the tunnel. The hydrophobic face of d∆h∆SCP-2L, which also contains the exit of the tunnel, is the potential interaction site with the hydratase-2 domain in full-length MFE-2. This interface could provide a hydrophobic environment for the longer fatty acyl tails. Such a mode of binding has been observed in (2E)-enoyl-CoA hydratase-1 (Engel et al. 1998), where the -end of the fatty acid group longer than C6 enters the hydrophobic region between the subunits of the enzyme. Unfortunately, attemps to incorporate fatty acyl-CoAs into the crystal of d∆h∆SCP-2L to test this hypothesis, have so far failed.

d∆h∆SCP-2L has PTS1 in the extreme COOH-terminus. Proteins with this targeting signal mediate with the soluble receptor protein, Pex5p, in cytosol. Subsequently, upon translocation into peroxisome, the Pex5p-cargo protein complex interacts with the peroxins in peroxisomal membrane (Elgersma et al. 1996, Albertini et al. 1997). So far, a number of structures of proteins containing PTS1 have been reported, but the structure of PTS1 has only been defined in rat ∆3,5-∆2,4-dienoyl-CoA isomerase (PDB code 1dci, Modis et al. 1998) and rabbit SCP-2 (PDB code 1c44, Choinowski et al. 2000). However, both structures are unliganded proteins, and PTS1 is buried in a pocket between two dimers in dienoyl-CoA isomerase or bound to the surface of the bulk of SCP-2. In the liganded structure of d∆h∆SCP-2L, the COOH-terminal residues have adopted different conformations. Namely, PTS1 (-Ala-Lys-LeuOH) is ordered and fully solvent-exposed, being thus able to interact with a receptor molecule of the peroxisomal import system. Compared to unliganded SCP-2, the exposure of PTS1 results from the movement of the COOH-terminal helix upon ligand binding. This observation supports the hypothesis about ligand-assisted protein import into peroxisomes.

7 Conclusions

Mammalian MFE-2 consists of three functional regions: an NH2-terminal (3R)-hydroxyacyl-CoA dehydrogenase region followed by an (2E)-enoyl-CoA hydratase-2 domain and a COOH-terminal SCP-2-like domain. The full-length MFE-2 is partially cleaved in vivo into separate dehydrogenase (34 kDa) and hydratase-2 (45 kDa) components. As the catalytic properties are similar before and after cleavage, the domains in MFE-2, obviously do not need structural support from each other to be functional. Consequently, in this study, both the dehydrogenase region and the SCP-2-like domain were able to produce as separate recombinant proteins for crystallization experiments. Additionally, the recombinant 45-kDa hydratase-2 fragment was used to study the catalytic properties of hydratase-2 reaction.

The crystal structures of both the rat MFE-2 dehydrogenase region and the human MFE-2 SCP-2-like domain were solved. The dehydrogenase region of MFE-2 belongs to the SDR superfamily and the structure of rat dehydrogenase revealed the basic α/β SDR fold. This fold is composed of a central β-sheet of six parallel β-strands sandwiched between two arrays of parallel α-helices on both sides, and it is mainly responsible for cofactor binding. The long α-helices, αE and αF, of one subunit were shown to comprise a four-helix bundle with the neighbouring subunit at the dimer interface, which is typical of dimeric SDR enzymes. However, the dimeric interactions are more extensive in rat dehydrogenase. Namely, the separate COOH-terminal domain, which is not seen among the known SDR structures, lines the active site cavity of the neighbouring monomer. The structure revealed a glycine-rich loop that could be linked to substrate preference. This loop, which is located at the bottom of the active site cavity in rat dehydrogenase, is absent in dehydrogenases acting on short-chain substrates, suggesting a role for it in the accommodation of long-chain fatty acyl-CoAs. Additionally, the conformation of the bound NAD+ in relation to conserved Ser151, Tyr164 and Lys168 found in other SDR enzymes allows interpretation of the reaction mechanism, which involves B-face 4-pro-S hydride transfer.

The SCP-2-like domain was shown to be monomeric and to have the same fold as the previously determined crystal structure of rabbit SCP-2. Both proteins are composed of a β-sheet covered on one side by α-helices. The structure showed a hydrophobic tunnel traversing the protein similar to rabbit SCP-2. Nevertheless, the tunnel was, this time,

64

occupied by an ordered Triton X-100 molecule, an analogue of a lipid molecule. This demonstrated the ligand-binding site and, consequently, the entrance and exit of the tunnel. The ligand binding to the human SCP-2-like domain caused changes in the structure. Especially, the COOH-terminal helix, helix E, had moved relative to its position in the unliganded rabbit SCP-2 structure. Therefore, the PTS1 located after helix E became solvent-exposed and thus able to interact with Pex5p, the soluble receptor molecule of the peroxisomal import system. This suggests ligand-assisted protein import into peroxisomes.

The role of the SCP-2-like domain in MFE-2, however, remains to be resolved. SCP-2 and SCP-2-like domain with similar folds may associate with the degradation of long-chain and very-long-chain fatty acids. When occupied by long-chain fatty acyl-CoA, SCP-2 formed specific complexes with acyl-CoA oxidase, the first enzyme of the peroxisomal β-oxidation cycle. Furthermore, the peroxisomal thiolase, SCPx, which has SCP-2 as a COOH-terminal domain, cleaves long-chain 3-ketoacyl-CoAs. Consequently, the SCP-2-like domain, in MFE-2, could be needed to provide additional space for binding longer substrates, as it has been shown that one of the physiological roles of MFE-2 is the degradation of VLCFAs.

Site-directed mutagenesis studies indicated the importance of conserved protic residues for the hydration reaction catalyzed by the central hydratase-2 region or the folding of MFE-2. In vitro characterization of the enzymatic properties of human recombinant hydratase-2 revealed that the hydratase-2 reaction involves an acid and a base, as demonstrated by the bell-shaped curve of the pH dependence of catalytic efficiency. Glu366 and Asp510 were shown to be the amino acid residues essential for catalysis, with Glu366 putatively acting as a base. The importance of Asp510 for the hydratase-2 reaction was further supported by its localization in a novel conserved hydratase-2 motif. This motif was also found in the dehydratase region of the β-subunit of the fungal fatty acid synthetase complexes. Similar observations are often made regarding the amino acid residues around an active site indicating conservation of a reaction mechanism in otherwise distantly related enzymes. A deeper understanding of the mechanism of the hydratase-2 reaction could be inferred if the high-resolution structure of hydratase-2 were available.

8 References

van Aalten DM, DiRusso CC, Knudsen J & Wierenga RK (2000) Crystal structure of FadR, a fatty acid-responsive transcription factor with a novel acyl coenzyme A-binding fold. EMBO J 19: 5167-5177.

van Aalten DM, Milne KG, Zou JY, Kleywegt GJ, Bergfors T, Ferguson MA, Knudsen J & Jones TA (2001) Binding site differences revealed by crystal structures of Plasmodium falciparum and bovine acyl-CoA binding protein. J Mol Biol 309: 181-192.

Aas M (1971) Organ and subcellular distribution of fatty acid activating enzymes in the rat. Biochim Biophys Acta 231: 32-47.

Aas M & Bremer J (1968) Short-chain fatty acid activation in rat liver. A new assay procedure for the enzymes and studies on their intracellular localisation. Biochim Biophys Acta 164: 157-166.

Adamski J, Husen B, Marks F & Jungblut PW (1992) Purification and properties of oestradiol 17β-dehydrogenase extracted from cytoplasmic vesicles of porcine endometrial cells. Biochem. J 288: 375-381.

Adamski J, Leenders F, Carstensen JF, Kaufmann M, Markus MM, Husen B, Tesdorpf JG, Seedorf U, de Launoit Y & Jakob F (1997) Steroids, fatty acyl-CoA, and sterols are substrates of 80-kDa multifunctional protein. Steroids 62: 159-163.

Adamski J, Normand T, Leenders F, Monte D, Begue A, Stehelin D, Jungblut PW & de Launoit Y (1995) Molecular cloning of a novel widely expressed human 80 kDa 17β-hydroxysteroid dehydrogenase IV. Biochem J 311: 437-443.

Aitchison JD, Murray WW & Rachubinski R (1991) The carboxyterminal ala-lys-leu is essential for targeting Candida tropicalis trifunctional enzyme to the yeast peroxisomes. J Biol Chem 266: 23197-23203.

Aitchison JD, Szilard RK, Nuttley WM & Rachubinski RA (1992) Antibodies directed against a yeast carboxyl-terminal peroxisomal targeting signal specifically recognize peroxisomal proteins from various yeasts. Yeast 8: 721-734.

Albalat R, Gonzalez-Duarte R & Atrian S (1992) Protein engineering of Drosophila alcohol dehydrogenase. The hydroxyl group of Tyr152 is involved in the active site of the enzyme. FEBS Lett 308: 235-239.

Albertini M, Rehling P, Erdmann R, Girzalsky W, Kiel JAKW, Veenhuis M & Kunau W-H (1997) Pex14p, a peroxisomal membrane protein binding both receptors of the two PTS-dependent import pathways. Cell 89: 83-92.

Amery L, Fransen M, De Nys K, Mannaerts GP & Van Veldhoven PP (2000) Mitochondrial and peroxisomal targeting of 2-methylacyl-CoA racemase in humans. J Lipid Res 41: 1752-1759.

Andersen KV & Poulsen FM (1993) The three-dimensional structure of acyl-coenzyme A binding protein from bovine liver: structural refinement using heteronuclear multidimensional NMR spectroscopy. J Biomol NMR 3: 271-284.

66

Antonenkov VD, Van Veldhoven PP, Waelkens E & Mannaerts GP (1997) Substrate specificities of 3-oxoacyl-CoA thiolase A and sterol carrier protein 2/3-oxoacyl-CoA thiolase purified from normal rat liver peroxisomes. Sterol carrier protein 2/3-oxoacyl-CoA thiolase is involved in the metabolism of 2-methyl-branched fatty acids and bile acid intermediates. J Biol Chem 272: 26023-26031.

Antonenkov VD, Van Veldhoven PP, Waelkens E & Mannaerts GP (1999) Comparison of the stability and substrate specificity of purified peroxisomal 3-oxoacyl-CoA thiolases A and B from rat liver. Biochim Biophys Acta 1437: 136-141.

Aoyama T, Tsushima K, Souri M, Kamijo T, Suzuki Y, Shimozawa N, Orii T & Hashimoto T (1994a) Molecular cloning and functional expression of a human peroxisomal acyl-coenzyme A oxidase. Biochem Biophys Res Commun 198: 1113-1118.

Baes M, Huyghe S, Carmeliet P, Declercq PE, Collen D, Mannaerts GP & Van Veldhoven PP (2000) Inactivation of the peroxisomal multifunctional protein-2 in mice impedes the degradation of not only 2-methyl-branched fatty acids and bile acid intermediates but also of very long chain fatty acids. J Biol Chem 275: 16329-16336.

Bailey TL, Baker ME & Elkan CP (1997) An artificial intelligence approach to motif discovery in protein sequences: application to steroid dehydrogenases. J Steroid Biochem Mol Biol 62: 29-44.

Baker ME (1991) Genealogy of regulation of human sex and adrenal function, prostaglandin action, snapdragon and petunia flower colors, antibiotics, and nitrogen fixation: functional diversity from two ancestral dehydrogenases. Steroids 56: 354-360.

Baumgart E, Vanhooren JC, Fransen M, Van Leuven F, Fahimi HD, Van Veldhoven PP & Mannaerts GP (1996a) Molecular cloning and further characterization of rat peroxisomal trihydroxycoprostanoyl-CoA oxidase. Biochem J 320: 115-121.

Baumgart E, Vanhooren JC, Fransen M, Marynen P, Puype M, Vandekerckhove J, Leunissen JA, Fahimi HD, Mannaerts GP & Van Veldhoven PP (1996b) Molecular characterization of the human peroxisomal branched-chain acyl-CoA oxidase: cDNA cloning, chromosomal assignment, tissue distribution, and evidence for the absence of the protein in Zellweger syndrome. Proc Natl Acad Sci USA 93: 13748-13753.

Benach J, Atrian S, Gonzàlez-Duarte R & Ladenstein R (1998) The refined crystal structure of Drosophila lebanonensis alcohol dehydrogenase at 1.9 Å resolution. J Mol Biol 282: 383-399.

Benach J, Atrian S, Gonzàlez-Duarte R & Ladenstein R (1999) The catalytic reaction and inhibition mechanism of Drosophila alcohol dehydrogenase: observation of an enzyme-bound NAD-ketone adduct at 1.4 Å resolution by X-ray crystallography. J Mol Biol 289: 335-355.

Berk PD, Zhou SL, Kiang CL, Stump D, Bradbury M & Isola LM (1997) Uptake of long chain free fatty acids is selectively up-regulated in adipocytes of Zucker rats with genetic obesity and non-insulin-dependent diabetes mellitus. J Biol Chem 272: 8830-8835.

Binstock JF & Schulz H (1981) Fatty acid oxidation complex from Escherichia coli. Methods Enzymol 71: 403-411.

Blattner J, Swinkels B, Dorsam H, Prospers T, Subramani S & 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.

Bodnar AG & Rachubinski RA (1990) Cloning and sequence determination of cDNA encoding a second rat liver peroxisomal 3-ketoacyl-CoA thiolase. Gene 91: 193-199.

Bodnar AG & Rachubinski RA (1991) Characterization of the integral membrane polypeptides of rat liver peroxisomes isolated from untreated and clofibrate-treated rats. Biochem Cell Biol 69: 499-508.

Borrás T, Persson B & Jörnvall H (1989) Eye lens zeta-crystallin relationships to the family of "long-chain" alcohol/polyol dehydrogenases. Protein trimming and conservation of stable parts. Biochemistry 28: 6133-6139.

Borst P (1989) Peroxisome biogenesis revisited. Biochim Biophys Acta 1008: 1-13. Bout A, Franse MM, Collins J, Blonden L, Tager JM & Benne R (1991) Characterization of the

gene encoding human peroxisomal 3-oxoacyl-CoA thiolase (ACAA). No large DNA rearrangement in a thiolase-deficient patient. Biochim Biophys Acta 1090: 43-51.

67

Bout A, Teunissen Y, Hashimoto T, Benne R & Tager JM (1988) Nucleotide sequence of human peroxisomal 3-oxoacyl-CoA thiolase. Nucleic Acids Res 16: 10369.

Bovolin P, Schlichting J, Miyata M, Ferrarese C, Guidotti A & Alho H (1990) Distribution and characterization of diazepam binding inhibitor (DBI) in peripheral tissues of rat. Regul Pept 29: 267-281.

Bradford MM (1976) A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248-254.

Braverman N, Dodt G, Gould SJ & Valle D (1995) Disorders of peroxisome biogenesis. Hum Mol Genet 4: 1791-1798.

Braverman N, Dodt G, Gould SJ & 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.

Breitling R, Marijanović Z, Perović D & Adamski J (2001) Evolution of 17β-HSD type 4, a multifunctional protein of β-oxidation. Mol Cell Endocrinol 171: 205-210.

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

Brünger AT (1998) Crystallography, and NMR system: a new software system for macromolecular structure determination. Acta Crystallogr 54: 905-921.

Caira F, Cherkaoui-Malki M, Hoefler G & Latruffe N (1996) Cloning and tissue expression of two cDNAs encoding the peroxisomal 2-enoyl-CoA hydratase/3-hydroxyacyl-CoA dehydrogenase in the guinea pig liver. FEBS Lett 378: 57-60.

Caira F, Clemencet MC, Cherkaoui-Malki M, Dieuaide-Noubhani M, Pacot C, Van Veldhoven PP & Latruffe N (1998) Differential regulation by a peroxisome proliferator of the different multifunctional proteins in guinea pig: cDNA cloning of the guinea pig D-specific multifunctional protein 2. Biochem J 330: 1361-1368.

Carstensen JF, Tesdorpf JG, Kaufmann M, Markus MM, Husen B, Leenders F, Jakob F, de Launoit Y & Adamski J (1996) Characterization of 17β-hydroxysteroid dehydrogenase IV. J Endocrinol 150: 3-12.

Castagnetta LA, Carruba G, Traina A, Granata OM, Markus M, Pavone-Macaluso M, Blomquist CH & Adamski J (1997) Expression of different 17β-hydroxysteroid dehydrogenase types and their activities in human prostate cancer cells. Endocrinology 38: 4876-4882.

Chang CC, Warren DS, Sacksteder KA & Gould SJ (1999) PEX12 interacts with PEX5 and PEX10 and acts downstream of receptor docking in peroxisomal matrix protein import. J Cell Biol 147: 761-774.

Chapman AD, Cortes A, Dafforn TR, Clarke AR & Brady RL (1999) Structural basis of substrate specificity in malate dehydrogenases: crystal structure of a ternary complex of porcine cytoplasmic malate dehydrogenase, alpha-ketomalonate and tetrahydoNAD. J Mol Biol 285: 703-712.

Chen Z, Jiang JC, Lin ZG, Lee WR, Baker ME & Chang SH (1993) Site-specific mutagenesis of Drosophila alcohol dehydrogenase: evidence for involvement of tyrosine-152 and lysine-156 in catalysis. Biochemistry 32: 3342-3346.

Choinowski T, Hauser H & Piontek K (2000) Structure of sterol carrier protein 2 at 1.8 Å resolution reveals a hydrophobic tunnel suitable for lipid binding. Biochemistry 39: 1897-1902.

Chomczynski P & Sacchi N (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162: 156-159.

Chu R, Varanasi U, Chu S, Lin Y, Usuda N, Rao MS & Reddy JK (1995) Overexpression and characterization of the human peroxisomal acyl-CoA oxidase in insect cells. J Biol Chem 270: 4908-4915.

Chudzik DM, Michels PA, de Walque S & Hol WG (2000) Structures of type 2 peroxisomal targeting signals in two trypanosomatid aldolases. J Mol Biol 300: 697-707.

Cistola DP, Atkinson D, Hamilton JA & Small DM (1986) Phase behavior and bilayer properties of fatty acids: hydrated 1:1 acid-soaps. Biochemistry 25: 2804-2812.

Cols N, Marfany G, Atrian S & Gonzalez-Duarte R (1993) Effect of site-directed mutagenesis on conserved positions of Drosophila alcohol dehydrogenase. FEBS Lett 319: 90-94.

68

Corsico B, Cistola DP, Frieden C & Storch J (1998) The helical domain of intestinal fatty acid binding protein is critical for collisional transfer of fatty acids to phospholipid membranes. Proc Natl Acad Sci USA 95: 12174-12178.

Cowtan KD & Zhang KY (1999) Density modification for macromolecular phase improvement. Prog Biophys Mol Biol 72: 245-270.

Croes K, Casteels M, Asselberghs S, Herdewijn P, Mannaerts GP & Van Veldhoven PP (1997) Formation of a 2-methyl-branched fatty aldehyde during peroxisomal α-oxidation. FEBS Lett 412: 643-645.

Croes K, Casteels M, De Hoffmann E, Mannaerts GP & Van Veldhoven PP (1996) α-Oxidation of 3-methyl-substituted fatty acids in rat liver. Production of formic acid instead of CO

2, cofactor

requirements, subcellular localization and formation of a 2-hydroxy-3-methylacyl-CoA intermediate. Eur J Biochem 240: 674-683.

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

Dansen TB, Westerman J, Wouters FS, Wanders RJ, van Hoek A, Gadella TW Jr & Wirtz KW (1999) High-affinity binding of very-long-chain fatty acyl-CoA esters to the peroxisomal non-specific lipid-transfer protein (sterol carrier protein-2). Biochem J 339: 193-199.

DeGrella RF & Light RJ (1980) Uptake and metabolism of fatty acids by dispersed adult rat heart myocytes. Inhibition by albumin and fatty acid homologues, and the effect of temperature and metabolic reagents. J Biol Chem 255: 9739-9745.

Didion T & Roggenkamp R (1992) Targeting signal of the peroxisomal catalase in the methylotrophic yeast Hansulena polymorpha. FEBS Lett 303: 113-116.

Dierks EA, Davis SC & Ortiz de Montellano PR (1998) Glu-320 and Asp-323 are determinants of the CYP4A1 hydroxylation regiospecificity and resistance to inactivation by 1-aminobenzotriazole. Biochemistry 37: 1839-1847.

Diestelkötter P & Just WW (1993) In vitro insertion of the 22-kD peroxisomal membrane protein into isolated rat liver peroxisomes. J Cell Biol 123: 1717-1725.

Dieuaide-Noubhani M, Novikov D, Baumgart E, Vanhooren JC, Fransen M, Goethals M, Vandekerckhove J, Van Veldhoven PP & Mannaerts GP (1996) Further characterization of the peroxisomal 3-hydroxyacyl-CoA dehydrogenases from rat liver. Relationship between the different dehydrogenases and evidence that fatty acids and the C27 bile acids di- and tri-hydroxycoprostanic acids are metabolized by separate multifunctional proteins. Eur J Biochem 240: 660-666.

Dieuaide-Noubhani M, Novikov D, Vandekerckhove J, Van Veldhoven PP & Mannaerts GP (1997) Identification and characterization of the 2-enoyl-CoA hydratases involved in peroxisomal β-oxidation in rat liver. Biochem J 321: 253-259.

DiRusso CC & Black PN (1999) Long-chain fatty acid transport in bacteria and yeast. Paradigms for defining the mechanism underlying this protein-mediated process. Mol Cell Biochem 192: 41-52.

Distel B, Gould SJ, Voorn-Brouwer T, van der Berg M, Tabak HF & Subramani S (1992b) The carboxyl-terminal tripeptide serine-lysine-leucine of firefly luciferase is necessary but not sufficient for peroxisomal import in yeast. New Biol 4: 157-165.

Distel RJ, Robinson GS & Spiegelman BM (1992a) Fatty acid regulation of gene expression. Transcriptional and post-transcriptional mechanisms. J Biol Chem 267: 5937-5941.

de Duve C & Baudhuin P (1966) Peroxisomes (microbodies and related particles). Physiol Rev 46: 323-357.

Dyer JM, McNew JA & Goodman JM (1996) The sorting sequence of the peroxisomal integral membrane protein PMP47 is contained within a short hydrophilic loop. J Cell Biol 133: 269-280.

Eaton S, Bartlett K & Pourfarzam M (1996) Mammalian mitochondrial β-oxidation. Biochem J 320: 345-357.

Ek BA, Cistola DP, Hamilton JA, Kaduce TL & Spector AA (1997) Fatty acid binding proteins reduce 15-lipoxygenase-induced oxygenation of linoleic acid and arachidonic acid. Biochim Biophys Acta 1346: 75-85.

69

Ek-Von Mentzer BA, Zhang F & Hamilton JA (2001) Binding of 13-HODE and 15-HETE to phospholipid bilayers, albumin, and intracellular fatty acid binding proteins. Implications for transmembrane and intracellular transport and for protection from lipid peroxidation. J Biol Chem 276: 15575-15580.

Elgersma Y, Elgersma-Hooisma M, Wenzel T, McCaffery JM, Farquhar MG & Subramani S (1998) A mobile PTS2 receptor for peroxisomal protein import in Pichia pastoris. J Cell Biol 140: 807-820.

Elgersma Y, Kwast L, Klein A, Voom-Brower T, Van den Berg M, Metzig B, America T, Tabak HF & Distel B (1996) The SH3 somain of the Saccharomyces cerevisiae peroxisomal membrane protein Pex13p functions as a docking site for Pex5p, a mobile receptor for the import of PTS1 containing protein. J Cell Biol 135: 97-109.

Engel CK, Kiema TR, Hiltunen JK & Wierenga RK (1998) The crystal structure of enoyl-CoA hydratase complexed with octanoyl-CoA reveals the structural adaptations required for binding of a long chain fatty acid-CoA molecule. J Mol Biol 275: 847-859.

Ensor CM & Tai HH (1991) Site-directed mutagenesis of the conserved tyrosine 151 of human placental NAD(+)-dependent 15-hydroxyprostaglandin dehydrogenase yields a catalytically inactive enzyme. Biochem Biophys Res Commun 176: 840-845.

Erdmann R & Blobel G (1996) Identification of a peroxisomal membrane receptor for the C-terminal tripeptide signal recognition factor. J Cell Biol 135: 111-121.

Evers ME, Titorenko V, Harder W, ven der Klei I & Veenhuis M (1996) Flavin adenine dinucleotide binding is the crucial step in alcohol oxidase assembly in the yeast Hansenula polymorpha. Yeast 12: 917-923.

Evers ME, Titorenko VI, van der Klei IJ, Harder W & Veenhuis M (1994) Assembly of alcohol oxidase in peroxisomes of the yeast Hansenula polymorpha requires the cofactor flavin adenine dinucleotide. Mol Biol Cell 5: 829-837.

Faber KN, Haan GJ, Baerends RJ, Kram AM & Veenhuis M (2002) Normal peroxisome development from vesicles induced by truncated Hansenula polymorpha Pex3p. J Biol Chem 277: 11026-11033.

Færgeman NJ, Sigurskjold BW, Kragelund BB, Andersen KV & Knudsen J (1996) Thermodynamics of ligand binding to acyl-coenzyme A binding protein studied by titration calorimetry. Biochemistry 35: 14118-14126.

Fairbairn LJ & Tanner MJ (1989) Complete cDNA sequence of human foetal liver peroxisomal 3-oxoacyl-CoA thiolase. Nucleic Acids Res 17: 3588.

Fan CY, Pan J, Chu R, Lee D, Kluckman KD, Usuda N, Singh I, Yeldandi AV, Rao MS, Maeda N & Reddy JK (1996) Hepatocellular and hepatic peroxisomal alterations in mice with a disrupted peroxisomal fatty acyl-coenzyme A oxidase gene. J Biol Chem 271: 24698-24710.

Fan LQ, Cattley RC & Corton JC (1998) Tissue-specific induction of 17β-hydroxysteroid dehydrogenase type IV by peroxisome proliferator chemicals is dependent on the peroxisome proliferator-activated receptor α. J Endocrinol 158: 237-246.

Ferdinandusse S, Mulders J, Ijlst L, Denis S, Dacremont G, Waterham HR & Wanders RJ (1999) Molecular cloning and expression of human carnitine octanoyltransferase: evidence for its role in the peroxisomal β-oxidation of branched-chain fatty acids. Biochem Biophys Res Commun 263: 213-218.

Ferdinandusse S, Rusch H, van Lint AE, Dacremont G, Wanders RJ & Vreken P (2002) Stereochemistry of the peroxisomal branched-chain fatty acid α- and β-oxidation systems in patients suffering from different peroxisomal disorders. J Lipid Res 43: 438-444.

Filling C, Berndt KD, Benach J, Knapp S, Prozorovski T, Nordling E, Ladenstein R, Jörnvall H & Oppermann U (2002) Critical residues for structure and catalysis in short-chain dehydrogenases/reductases (SDR). J Biol Chem 277: 25677-25684.

Filling C, Nordling E, Benach J, Berndt KD, Ladenstein R, Jörnvall H & Oppermann U (2001) Structural role of conserved Asn179 in the short-chain dehydrogenase/reductase scaffold. Biochem Biophys Res Commun 289: 712-717.

Filppula SA, Sormunen RT, Hartig A, Kunau WH & Hiltunen JK (1995) Changing stereochemistry for a metabolic pathway in vivo. Experiments with the peroxisomal β-oxidation in yeast. J Biol Chem 270: 27453-27457.

70

Flynn CR, Mullen RT & Trelease RN (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.

Fosså A, Beyer A, Pfitzner E, Wenzel B & Kunau WH (1995) Molecular cloning, sequencing and sequence analysis of the fox-2 gene of Neurospora crassa encoding the multifunctional β-oxidation protein. Mol Gen Genet 247: 95-104.

Foulon V, Antonenkov VD, Croes K, Waelkens E, Mannaerts GP, Van Veldhoven PP & Casteels M (1999) Purification, molecular cloning, and expression of 2-hydroxyphytanoyl-CoA lyase, a peroxisomal thiamine pyrophosphate-dependent enzyme that catalyzes the carbon-carbon bond cleavage during α-oxidation of 3-methyl-branched fatty acids. Proc Natl Acad Sci USA 96: 10039-10044.

Fournier B, Saudubray JM, Benichou B, Lyonnet S, Munnich A, Clevers H & Poll-The BT (1994) Large deletion of the peroxisomal acyl-CoA oxidase gene in pseudoneonatal adrenoleukodystrophy. J Clin Invest 94: 526-531.

Frolov A, Cho TH, Billheimer JT & Schröder F (1996c) Sterol carrier protein-2, a new fatty acyl coenzyme A-binding protein. J Biol Chem 271: 31878-31884.

Frolov A, Miller K, Billheimer JT, Cho TH & Schröder F (1997) Lipid specificity and location of the sterol carrier protein-2 fatty acid-binding site: a fluorescence displacement and energy transfer study. Lipids 32: 1201-1209.

Frolov A, Woodford JK, Murphy EJ, Billheimer JT & Schröder F (1996a) Spontaneous and protein-mediated sterol transfer between intracellular membranes. J Biol Chem 271: 16075-16083.

Frolov A, Woodford JK, Murphy EJ, Billheimer JT & Schröder F (1996b) Fibroblast membrane sterol kinetic domains: modulation by sterol carrier protein-2 and liver fatty acid binding protein. J Lipid Res 37: 1862-1874.

Fuchs M, Hafer A, Munch C, Kannenberg F, Teichmann S, Scheibner J, Stange EF & Seedorf U (2001) Disruption of the sterol carrier protein 2 gene in mice impairs biliary lipid and hepatic cholesterol metabolism. J Biol Chem 276: 48058-48065.

Fujiki Y, Rachubinski RA & Lazarow PB (1984) Synthesis of a major integral membrane polypeptide of rat liver peroxisomes on free polysomes. Proc Natl Acad Sci USA 81: 7127-7131.

Fujino T, Takei YA, Sone H, Ioka RX, Kamataki A, Magoori K, Takahashi S, Sakai J & Yamamoto TT (2001) Molecular identification and characterization of two medium-chain acyl-CoA synthetases, MACS1 and the Sa gene product. J Biol Chem 276: 35961-35966.

Furuta S, Miyazawa S, Osumi T, Hashimoto T & Ui N (1980) Properties of mitochondria and peroxisomal enoyl-CoA hydratases from rat liver. J Biochem (Tokyo) 88: 1059-1070.

Gallegos AM, Atshaves BP, Storey SM, Starodub O, Petrescu AD, Huang H, McIntosh AL, Martin GG, Chao H, Kier AB & Schröder F (2001) Gene structure, intracellular localization, and functional roles of sterol carrier protein-2. Prog Lipid Res 40: 498-563.

García FL, Szyperski T, Dyer JH, Choinowski T, Seedorf U, Hauser H & Wüthrich K (2000) NMR structure of the sterol carrier protein-2: implications for the biological role. J Mol Biol 295: 595-603.

Gatto GJ Jr, Geisbrecht BV, Gould SJ & Berg JM (2000a) Peroxisomal targeting signal-1 recognition by the TPR domains of human PEX5. Nat Struct Biol 7: 1091-1095.

Gatto GJ Jr, Geisbrecht BV, Gould SJ & Berg JM (2000b) A proposed model for the PEX5-peroxisomal targeting signal-1 recognition complex. Proteins 38: 241-246.

Gericke A, Smith ER, Moore DJ, Mendelsohn R & Storch J (1997) Adipocyte fatty acid-binding protein: interaction with phospholipid membranes and thermal stability studied by FTIR spectroscopy. Biochemistry 36: 8311-8317.

Ghaedi K, Tamura S, Okumoto K, Matsuzono Y & Fujiki Y (2000) The peroxin pex3p initiates membrane assembly in peroxisome biogenesis. Mol Biol Cell 11: 2085-2102.

Ghosh D, Pletnev VZ, Zhu DW, Wawrzak Z, Duax WL, Pangborn W, Labrie F & Lin SX (1995) Structure of human estrogenic 17β-hydroxysteroid dehydrogenase at 2.20Å resolution. Structure 3: 503-513.

71

Ghosh D, Sawicki M, Pletnev V, Erman M, Ohno S, Nakajin S & Duax WL (2001) Porcine carbonyl reductase. structural basis for a functional monomer in short chain dehydrogenases/reductases. J Biol Chem 276: 18457-18463.

Ghosh D, Wawrzak Z, Weeks CM, Duax WL & Erman M (1994) The refined three-dimensional structure of 3α,20β-hydroxysteroid dehydrogenase and possible roles of the residues conserved in short-chain dehydrogenases. Structure 2: 629-640.

Ghosh D, Weeks CM, Grochulski P, Duax WL, Erman M, Rimsay RL & Orr JC (1991) Three-dimensional structure of holo 3α,20β-hydroxysteroid dehydrogenase: a member of a short-chain dehydrogenase family. Proc Natl Acad Sci USA 88: 10064-10068.

Gietz RD, Schiestl RH, Willems AR & Woods RA (1995) Studies on the transformation of intact yeast cells by the LiAc/SS-DNA/PEG procedure. Yeast 11: 355-360.

Glatz JF, Borchers T, Spener F & van der Vusse GJ (1995) Fatty acids in cell signalling: modulation by lipid binding proteins. Prostaglandins Leukot Essent Fatty Acids 52: 121-127.

Glatz JF & van der Vusse GJ (1996) Cellular fatty acid-binding proteins: their function and physiological significance. Prog Lipid Res 35: 243-282.

Glover JR, Andrews DW, Subramani S & Rachubinski RA (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.

Gould SJ, Kalish JE, Morrell JE, Bjorkman J, Urquhart AJ & Crane DI (1996) An SH3 protein in the peroxisome membrane is a docking factor for the PTS1 receptor. J Cell Biol 135: 85-95.

Gould SJ, Keller GA, Hosken N, Wilkinson J & Subramani S (1989) A conserved tripeptide sorts proteins to peroxisomes. J Cell Biol 108: 1657-1664.

Gould SJ, Keller GA, Schneider M, Howell SH, Garrad LJ, Goodman JM, Distel B, Tabak H & Subramani S (1990a) Peroxisomal protein import is conserved between yeast, plants, insects and mammals. EMBO J 9: 85-90.

Gould SJ, Keller GA & Subramani S (1987) Identification of a peroxisomal targeting signal at the carboxy terminus of firefly luciferase. J Cell Biol 105: 2923-2931.

Gould SJ, Keller GA & Subramani S (1988) Identification of peroxisomal targeting signals at the carboxy terminus of four peroxisomal proteins. J Cell Biol 107: 897-905.

Gould SJ, Krisans S, Keller GA & Subramani S (1990b) Antibodies directed against the peroxisomal targeting signal of firefly luciferase recognize multiple mammalian peroxisomal proteins. J Cell Biol 110: 27-34.

Gould SJ & Valle D (2000) Peroxisome biogenesis disorders: genetics and cell biology. Trends Genet 16: 340-345.

Graham IA & Eastmond PJ (2002) Pathways of straight and branched chain fatty acid catabolism in higher plants. Prog Lipid Res 41: 156-181.

Grimaldi PA (2001) Fatty acid regulation of gene expression. Curr Opin Clin Nutr Metab Care 4: 433-437.

Grimaldi PA, Knobel SM, Whitesell RR & Abumrad NA (1992) Induction of aP2 gene expression by nonmetabolized long-chain fatty acids. Proc Natl Acad Sci USA 89: 10930-10934.

Grimm C, Maser E, Möbus E, Klebs G, Reuter K & Ficner R (2000) The crystal structure of 3α-hydroxysteroid dehydrogenase/carbonyl reductase from Comamonas testosterone shows a novel oligomerization pattern within the short chain dehydrogenase/reductase family. J Biol Chem 275: 41333-41339.

Guex N & Peitsch MC (1997) SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling. Electrophoresis 18: 2714-2723.

Hamilton JA, Johnson RA, Corkey B & Kamp F (2001) Fatty acid transport: the diffusion mechanism in model and biological membranes. J Mol Neurosci 16: 99-108.

Hamilton JA & Kamp F (1999) How are free fatty acids transported in membranes? Is it by proteins or by free diffusion through the lipids? Diabetes 48: 2255-2269.

Hansen H, Didion T, Thieman A, Veenhuis M & Roggenkamp R (1992) Targeting signals of the major peroxisomal proteins in the methylotrophic yeast Hansula polymorpha. Mol Gen Genet 235: 269-278.

72

Hardwick JP, Song BJ, Huberman E & Gonzalez FJ (1987) Isolation, complementary DNA sequence, and regulation of rat hepatic lauric acid -hydroxylase (cytochrome P-450LA omega). Identification of a new cytochrome P-450 gene family. J Biol Chem 262: 801-810.

Harper CC, South ST, McCaffery JM & Gould SJ (2002) Peroxisomal membrane protein import does not require Pex17p. J Biol Chem 277: 16498-16504.

Hashimoto T (1999b) Peroxisomal β-oxidation enzymes. Neurochem Res 24: 551-563. Hashimoto T, Cook WS, Qi C, Yeldandi AV, Reddy JK & Rao MS (2000) Defect in peroxisome

proliferator-activated receptor α-inducible fatty acid oxidation determines the severity of hepatic steatosis in response to fasting. J Biol Chem 275: 28918-28928.

Hashimoto T, Fujita T, Usuda N, Cook W, Qi C, Peters JM, Gonzalez FJ, Yeldandi AV, Rao MS & Reddy JK (1999a) Peroxisomal and mitochondrial fatty acid β-oxidation in mice nullizygous for both peroxisome proliferator-activated receptor α and peroxisomal fatty acyl-CoA oxidase. Genotype correlation with fatty liver phenotype. J Biol Chem 274: 19228-19236.

He XY & Yang SY (1997) Glutamate-119 of the large α-subunit is the catalytic base in the hydration of 2-trans-enoyl-coenzyme A catalyzed by the multienzyme complex of fatty acid oxidation from Escherichia coli. Biochemistry 36: 11044-11049.

Herr FM, Aronson J & Storch J (1996) Role of portal region lysine residues in electrostatic interactions between heart fatty acid binding protein and phospholipid membranes. Biochemistry 35: 1296-1303.

Herr FM, Matarese V, Bernlohr DA & Storch J (1995) Surface lysine residues modulate the collisional transfer of fatty acid from adipocyte fatty acid binding protein to membranes. Biochemistry 34: 11840-11845.

Hertz R, Magenheim J, Berman I & Bar-Tana J (1998) Fatty acyl-CoA thioesters are ligands of hepatic nuclear factor-4α. Nature 392: 512-516.

Hesler CB, Olymbios C & Haldar D (1990) Transverse-plane topography of long-chain acyl-CoA synthetase in the mitochondrial outer membrane. J Biol Chem 265: 6600-6605.

Hettema EH, Girzalsky W, van den Berg M, Erdmann R & Distel B (2000) Saccharomyces cerevisiae pex3p and pex19p are required for proper localization and stability of peroxisomal membrane proteins. EMBO J 19: 223-233.

Hettema EH & Tabak HF (2000) Transport of fatty acids and metabolites across the peroxisomal membrane. Biochim Biophys Acta 1486: 18-27.

Hicks M & Gebecki JM (1977) Microscopic studies of fatty acid vesicles. Chem Phys Lipids 20: 143-252.

Hijikata M, Ishii N, Kagamiyama H, Osumi T & Hashimoto T (1987) Structural analysis of cDNA for rat peroxisomal 3-ketoacyl-CoA thiolase. J Biol Chem 262: 8151-8158.

Hijikata M, Wen JK, Osumi T & Hashimoto T (1990) Rat peroxisomal 3-ketoacyl-CoA thiolase gene. Occurrence of two closely related but differentially regulated genes. J Biol Chem 265: 4600-4606.

Hiltunen JK, Filppula SA, Koivuranta KT, Siivari K, Qin YM & Häyrinen HM (1996) Peroxisomal β-oxidation and polyunsaturated fatty acids. Ann N Y Acad Sci 804: 116-128.

Hiltunen JK, Palosaari PM & Kunau WH (1989) Epimerization of 3-hydroxyacyl-CoA esters in rat liver. Involvement of two 2-enoyl-CoA hydratases. J Biol Chem 264: 13536-13540.

Hiltunen JK & Qin Y (2000) β-Oxidation - strategies for the metabolism of a wide variety of acyl-CoA esters. Biochim Biophys Acta 1484: 117-128.

Hiltunen JK, Wenzel B, Beyer A, Erdmann R, Fosså A & Kunau WH (1992) Peroxisomal multifunctional β-oxidation protein of Saccharomyces cerevisiae. Molecular analysis of the fox2 gene and gene product. J Biol Chem 267: 6646-6653.

Hirsch D, Stahl A & Lodish HF (1998) A family of fatty acid transporters conserved from mycobacterium to man. Proc Natl Acad Sci USA 95: 8625-8629.

Hoefler G, Forstner M, McGuinness MC, Hulla W, Hiden M, Krisper P, Kenner L, Ried T, Lengauer C, Zechner R, Moser HW & Chen GL (1994) cDNA cloning of the human peroxisomal enoyl-CoA hydratase: 3-hydroxyacyl-CoA dehydrogenase bifunctional enzyme and localization to chromosome 3q26.3-3q28: a free left Alu arm is inserted in the 3' noncoding region. Genomics 19: 60-67.

73

Honsho M & 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.

Hsu KT & Storch J (1996) Fatty acid transfer from liver and intestinal fatty acid-binding proteins to membranes occurs by different mechanisms. J Biol Chem 271: 13317-13323.

Huang H, Ball JM, Billheimer JT & Schröder F (1999) The sterol carrier protein-2 amino terminus: a membrane interaction domain. Biochemistry 38: 13231-13243.

Hutchinson EG & Thornton JM (1996) PROMOTIF -a program to identify and analyze structural motifs in proteins. Protein Sci 5: 212-220.

Imanaka T, Shiina Y, Takano T, Hashimoto T & Osumi T (1996) Insertion of the 70-kDa peroxisomal membrane protein into peroxisomal membranes in vivo and in vitro. J Biol Chem 271: 3706-3713.

Imanaka T, Small GM & Lazarow PB (1987) Translocation of acyl-CoA oxidase into peroxisomes requires ATP hydrolysis but not a membrane potential. J Cell Biol 105: 2915-2922.

Infante JP, Tschanz CL, Shaw N, Michaud AL, Lawrence P & Brenna JT (2002) Straight-chain acyl-CoA oxidase knockout mouse accumulates extremely long chain fatty acids from α-linolenic acid: evidence for runaway carousel-type enzyme kinetics in peroxisomal β-oxidation diseases. Mol Genet Metab 75: 108-119.

Ishii N, Hijikata M, Osumi T & Hashimoto T (1987) Structural organization of the gene for rat enoyl-CoA hydratase:3-hydroxyacyl-CoA dehydrogenase bifunctional enzyme. J Biol Chem 262: 8144-8150.

Itoh M, Suzuki Y & Takashima S (1999) A novel peroxisomal enzyme, D-3-hydroxyacyl-CoA dehydratase/D-3-hydroxyacyl-CoA dehydrogenase bifunctional protein: its expression in the developing human brain. Microsc Res Tech 45: 383-388.

Jakobs BS & Wanders RJ (1995) Fatty acid β-oxidation in peroxisomes and mitochondria: the first, unequivocal evidence for the involvement of carnitine in shuttling propionyl-CoA from peroxisomes to mitochondria. Biochem Biophys Res Commun 213: 1035-1041.

Jancarik J & Kim SH (1991) Sparse matrix sampling: Screening method for crystallization of proteins. J Appl Cryst 24: 409-411.

Jankowski A, Kim JH, Collins RF, Daneman R, Walton P & Grinstein S (2001) In situ measurements of the pH of mammalian peroxisomes using the fluorescent protein pHluorin. J Biol Chem 276: 48748-48753.

Jansen GA, Mihalik SJ, Watkins PA, Moser HW, Jakobs C, Denis S & Wanders RJ (1996) Phytanoyl-CoA hydroxylase is present in human liver, located in peroxisomes, and deficient in Zellweger syndrome: direct, unequivocal evidence for the new, revised pathway of phytanic acid α-oxidation in humans. Biochem Biophys Res Commun 229: 205-210.

Jansen GA, Ofman R, Ferdinandusse S, Ijlst L, Muijsers AO, Skjeldal OH, Stokke O, Jakobs C, Besley GT, Wraith JE & Wanders RJ (1997) Refsum disease is caused by mutations in the phytanoyl-CoA hydroxylase gene. Nat Genet 17: 190-193.

Jansen GA, Verhoeven NM, Denis S, Romeijn G, Jakobs C, ten Brink HJ & Wanders RJ (1999) Phytanic acid α-oxidation: identification of 2-hydroxyphytanoyl-CoA lyase in rat liver and its localisation in peroxisomes. Biochim Biophys Acta 1440: 176-182.

Jiang LL, Kobayashi A, Matsuura H, Fukushima H & Hashimoto T (1996) Purification and properties of human D-3-hydroxyacyl-CoA dehydratase: medium-chain enoyl-CoA hydratase is D-3-hydroxyacyl-CoA dehydratase. J Biochem (Tokyo) 120: 624-632.

Jiang LL, Kurosawa T, Sato M, Suzuki Y & Hashimoto T (1997) Physiological role of D-3-hydroxyacyl-CoA dehydratase/D-3-hydroxyacyl-CoA dehydrogenase bifunctional protein. J Biochem (Tokyo) 121: 506-513.

Johnson EF, Palmer CN, Griffin KJ & Hsu MH (1996) Role of the peroxisome proliferator-activated receptor in cytochrome P450 4A gene regulation. FASEB J 10: 1241-1248.

Jones JM, Morrell JC & Gould SJ (2001) Multiple distinct targeting signals in integral peroxisomal membrane proteins. J Cell Biol 153: 1141-1150.

Jones TA, Zou JY, Cowan SW & Kjeldgaard M (1991) Improved methods for binding protein models in electron density maps and the location of errors in these models. Acta Crystallogr A 47: 110-119.

74

Jörnvall H, Hoog JO & Persson B (1999) SDR and MDR: Completed genome sequences show these protein families to be large, of old origin, and of complex nature. FEBS Lett 445: 261-264.

Jörnvall H, Persson B, Krook M, Atrian S, Gonzalez-Duarte R, Jeffery J & Ghosh D (1995) Short-chain dehydrogenases/reductases (SDR). Biochemistry 34: 6003-6013.

Kabsch W & Sander C (1983) Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features. Biopolymers 22: 2577-2637.

Kaikaus RM, Chan WK, Lysenko N, Ray R, Ortiz de Montellano PR & Bass NM (1993) Induction of peroxisomal fatty acid β-oxidation and liver fatty acid-binding protein by peroxisome proliferators. Mediation via the cytochrome P-450IVA1 -hydroxylase pathway. J Biol Chem 268: 9593-9603.

Kamijo K, Taketani S, Yokota S, Osumi T & Hashimoto T (1990) The 70-kDa peroxisomal membrane protein is a member of the Mdr (P-glycoprotein)-related ATP-binding protein superfamily. J Biol Chem 265: 4534-4540.

Kammerer S, Holzinger A, Welsch U & Roscher AA (1998) Cloning and characterization of the gene encoding the human peroxisomal assembly protein Pex3p. FEBS Lett 29: 53-60.

Kamp F & Hamilton JA (1992) pH gradients across phospholipid membranes caused by fast flip-flop of un-ionized fatty acids. Proc Natl Acad Sci USA 89: 11367-11370.

Kaufmann M, Carstensen J, Husen B & Adamski J (1995) The tissue distribution of porcine 17β-estradiol dehydrogenase and its induction by progesterone. J Steroid Biochem Mol Biol 55: 535-539.

Kawai J, Shinagawa A, Shibata K, Yoshino M, Itoh M, Ishii Y, Arakawa T, Hara A, Fukunishi Y, Konno H, Adachi J, Fukuda S, Aizawa K, Izawa M, Nishi K, Kiyosawa H, Kondo S, Yamanaka I, Saito T, Okazaki Y, Gojobori T, Bono H, Kasukawa T, Saito R, Kadota K, Matsuda H, Ashburner M, Batalov S, Casavant T, Fleischmann W, Gaasterland T, Gissi C, King B, Kochiwa H, Kuehl P, Lewis S, Matsuo Y, Nikaido I, Pesole G, Quackenbush J, Schriml LM, Staubli F, Suzuki R, Tomita M, Wagner L, Washio T, Sakai K, Okido T, Furuno M, Aono H, Baldarelli R, Barsh G, Blake J, Boffelli D, Bojunga N, Carninci P, de Bonaldo MF, Brownstein MJ, Bult C, Fletcher C, Fujita M, Gariboldi M, Gustincich S, Hill D, Hofmann M, Hume DA, Kamiya M, Lee NH, Lyons P, Marchionni L, Mashima J, Mazzarelli J, Mombaerts P, Nordone P, Ring B, Ringwald M, Rodriguez I, Sakamoto N, Sasaki H, Sato K, Schonbach C, Seya T, Shibata Y, Storch KF, Suzuki H, Toyo-oka K, Wang KH, Weitz C, Whittaker C, Wilming L, Wynshaw-Boris A, Yoshida K, Hasegawa Y, Kawaji H, Kohtsuki S & Hayashizaki Y (2001) Functional annotation of a full-length mouse cDNA collection. Nature 409: 685-690.

Keller GA, Krisans S, Gould SJ, Sommer JM, Schleibs W, Kunau WH, Brody S & Subramani S (1991) Evolutionary conservation of a microbody targeting signal that targets proteins to peroxisomes, glyoxysomes and glycosomes. J Cell Biol 114: 893-904.

Keller GA, Scallen TJ, Clarke D, Maher PA, Krisans SK & Singer SJ (1989) Subcellular localization of sterol carrier protein-2 in rat hepatocytes: its primary localization to peroxisomes. J Cell Biol 108: 1353-1361.

Kemp S, Pujol A, Waterham HR, van Geel BM, Boehm CD, Raymond GV, Cutting GR, Wanders RJ & Moser HW (2001) ABCD1 mutations and the X-linked adrenoleukodystrophy mutation database: role in diagnosis and clinical correlations. Hum Mutat 18: 499-515.

Kerner J & Hoppel C (2000) Fatty acid import into mitochondria. Biochim Biophys Acta 1486: 1-17.

Kersten S, Seydoux J, Peters JM, Gonzalez FJ, Desvergne B & Wahli W (1999) Peroxisome proliferator-activated receptor α mediates the adaptive response to fasting. J Clin Invest 103: 1489-1498.

Kiel JA & Veenhuis M (2000) Peroxisomal matrix protein import. Suppression of protein import defects in Hansenula polymorpha pex mutants by overproduction of the PTS1 receptor Pex5p. Cell Biochem Biophys 32: 9-19.

Kiema TR, Engel CK, Schmitz W, Filppula SA, Wierenga RK & Hiltunen JK (1999) Mutagenic and enzymological studies of the hydratase and isomerase activities of 2-enoyl-CoA hydratase-1. Biochemistry 38: 2991-2999.

75

Klein AT, Barnett P, Bottger G, Konings D, Tabak HF & Distel B (2001) Recognition of peroxisomal targeting signal type 1 by the import receptor Pex5p. J Biol Chem 276: 15034-15041.

Klein AT, van den Berg M, Bottger G, Tabak HF & 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.

Knoll A, Salles J, Sargueil F, Cassagne C & Garbay B (2000) Peroxisomal β-oxidation enzyme gene expression in the developing mouse brain. Neurosci Lett 285: 201-204.

Kobayashi K, Kobayashi H, Ueda M & Honda Y (1997) Expression of 17β-hydroxysteroid dehydrogenase type IV in chick retinal pigment epithelium. Exp Eye Res 64: 719-726.

Koenig C, Araya C, Skorin C, Valencia C, Toro A, Leighton F & Santos MJ (2002) Cytochemical and biochemical demonstration of an ATPase in membranes of human peroxisomes. J Histochem Cytochem 50: 405-414.

Kornberg A & Pricer WE Jr. (1953) Enzymatic synthesis of the coenzyme A derivatives of long chain fatty acids. J Biol Chem 204: 329-343.

Kotti TJ, Savolainen K, Helander HM, Yagi A, Novikov DK, Kalkkinen N, Conzelmann E, Hiltunen JK & Schmitz W (2000) In mouse α-methylacyl-CoA racemase, the same gene product is simultaneously located in mitochondria and peroxisomes. J Biol Chem 275: 20887-20895.

Kragelund BB, Knudsen J & Poulsen FM (1999) Acyl-coenzyme A binding protein (ACBP). Biochim Biophys Acta 1441: 150-161.

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

Krazeisen A, Möller G, Richter T, Kremmer E & Adamski J (1999) Human 17β-hydroxysteroid dehydrogenase type IV: inhibition by phytoestrogens and loss of expression in breast cancer. J Mol Med 77: 8.

Krozowski Z (1992) 11β-Hydroxysteroid dehydrogenase and the short-chain alcohol dehydrogenase (SCAD) superfamily. Mol Cell Endocrinol 84: 25-31.

Kunau WH, Bühne S, de la Garza M, Kionka C, Mateblowski M, Schultz-Borchard U & Thieringer R (1988) Comparative enzymology of beta-oxidation. Biochem Soc Trans 16: 418-420.

Kurosawa T, Sato M, Yoshimura T, Jiang LL, Hashimoto T & Tohma M (1997) Stereospecific formation of (24R,25R)-3α,7α,12α,24-tetrahydroxy-5β-cholestan-26-oic acid catalyzed with a peroxisomal bifunctional D-3-hydroxyacyl-CoA dehydratase/D-3-hydroxyacyl-CoA dehydrogenase. Biol Pharm Bull 20: 295-297.

Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T. Nature 227: 680-685.

Lalwani ND, Reddy MK, Mangkornkanok-Mark M & Reddy JK (1981) Induction, immunochemical identity and immunofluorescence localization of an 80 000-molecular-weight peroxisome-proliferation-associated polypeptide (polypeptide PPA-80) and peroxisomal enoyl-CoA hydratase of mouse liver and renal cortex. Biochem J 198: 177-186.

Laskowski RA, McArthur MW, Moss DS & Thornton JM (1993) PROCHECK: a program to check the stereochemical quality of protein structures. J Appl Crystallogr 26: 283-291.

Lazarow PB & de Duve C (1976) A fatty acyl-CoA oxidizing system in rat liver peroxisomes; enhancement by clofibrate, a hypolipidemic drug. Proc Natl Acad Sci USA 73: 2043-2046.

Lazarow PB & Fujiki Y (1985) Biogenesis of peroxisomes. Annu Rev Cell Biol 1: 489-530. Lazarow PB & Moser HW (1994) Disordres of peroxisome biogenesis. The metabolic and

molecular basis of inherited disease. McGraw Hill Book Co, New York, p 2287-2324. Lazarow PB & Moser HW (1995) Disorders of peroxisome biogenesis. In: Scriver CR, Beaudet AI,

Sly WS & Valle D (eds) The metabolic basis of inherited disease. McGraw-Hill, New York, p 2287-2324.

Lazo O, Contreras M & Singh I (1990) Topographical localization of peroxisomal acy-CoA ligases: differential localization of palmitoyl-CoA and lignoceroyl-CoA ligases. Biochemistry 29: 3981-3986.

76

Leenders F, Adamski J, Husen B, Thole HH & Jungblut PW (1994) Molecular cloning and amino acid sequence of the porcine 17β-estradiol dehydrogenase. Eur J Biochem 222: 221-227.

Leenders F, Prescher G, Dolez V, Begue A, de Launoit Y & Adamski J (1996a) Assignment of human 17β-hydroxysteroid dehydrogenase IV to chromosome 5q2 by fluorescence in situ hybridization. Genomics 37: 403-404.

Leenders F, Tesdorpf JG, Markus M, Engel T, Seedorf U & Adamski J (1996b) Porcine 80-kDa protein reveals intrinsic 17β-hydroxysteroid dehydrogenase, fatty acyl-CoA-hydratase/dehydrogenase, and sterol transfer activities. J Biol Chem 271: 5438-5442.

Leesong M, Henderson BS, Gillig JR, Schwab JM & Smith JL (1996) Structure of a dehydratase-isomerase from the bacterial pathway for biosynthesis of unsaturated fatty acids: two catalytic activities in one active site. Structure 4: 253-264.

Leiper JM, Oatey PB Danpure CJ (1996) Inhibition of alanine:glyoxylate aminotransferase 1 dimerization is a prerequisite for its peroxisome-to-mitochondrion mistargeting in primary hyperoxaluria type 1. J Cell Biol 135: 939-951.

Leone TC, Weinheimer CJ & Kelly DP (1999) A critical role for the peroxisome proliferator-activated receptor α (PPARα) in the cellular fasting response: the PPARα-null mouse as a model of fatty acid oxidation disorders. Proc Natl Acad Sci USA 96: 7473-7478.

Lerche MH, Kragelund BB, Bech LM & Poulsen FM (1997) Barley lipid-transfer protein complexed with palmitoyl CoA: the structure reveals a hydrophobic binding site that can expand to fit both large and small lipid-like ligands. Structure 5: 291-306.

Levitt DG (2001) A new software routine that automates the fitting of protein X-ray crystallographic electron-density maps. Acta Crystallogr D 57: 1013-1019.

Li JX, Smeland TE & Schulz H (1990) D-3-hydroxyacyl coenzyme A dehydratase from rat liver peroxisomes. Purification and characterization of a novel enzyme necessary for the epimerization of 3-hydroxyacyl-CoA thioesters. J Biol Chem 265: 13629-13634.

Lu G (1999) FINDNCS: A program to detect non-crystallography symmetries in protein crystals from heavy atoms sites. J Appl Crystallogr 32: 365-368.

Luxon BA (1996) Inhibition of binding to fatty acid binding protein reduces the intracellular transport of fatty acids. Am J Physiol 271: 113-120.

Luxon BA, Holly DC, Milliano MC & Weisiger RA (1998) Sex differences in multiple steps in the hepatic transport of palmitate support a balanced uptake mechanism. Am J Physiol Gastrointest Liver Physiol 274: 52-61.

Luxon BA & Weisiger RA (1993) Sex differences in intracellular fatty acid transport: role of cytoplasmic binding proteins. Am J Physiol 265: 831-841.

Malila LH, Siivari KM, Mäkelä MJ, Jalonen JE, Latipää PM, Kunau WH & Hiltunen JK (1993) Enzymes converting D-3-hydroxyacyl-CoA to trans-2-enoyl-CoA. Microsomal and peroxisomal isoenzymes in rat liver. J Biol Chem 268: 21578-21585.

Mannaerts GP, Van Veldhoven PP & Casteels M (2000) Peroxisomal lipid degradation via β- and α-oxidation in mammals. Cell Biochem Biophys 32: 73-87.

Markus M, Husen B, Leenders F, Jungblut PW, Hall PF & Adamski J (1995) The organelles containing porcine 17 β-estradiol dehydrogenase are peroxisomes. Eur J Cell Biol 68: 263-267.

Matsumura T, Otera H & Fujiki Y (2000) Disruption of the interaction of the longer isoform of Pex5p, Pex5pL, with Pex7p abolishes peroxisome targeting signal type 2 protein import in mammals. Study with a novel Pex5-impaired Chinese hamster ovary cell mutant. J Biol Chem 275: 21715-21721.

Matsuzono Y, Kinoshita N, Tamura S, Shimozawa N, Hamasaki M, Ghaedi K, Wanders RJ, Suzuki Y, Kondo N & 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.

Matthews BW (1968) Solvent content of crystals. J Mol Biol 33: 491-497. McArthur MJ, Atshaves BP, Frolov A, Foxworth WD, Kier AB & Schröder F (1999) Cellular

uptake and intracellular trafficking of long chain fatty acids. J Lipid Res 40: 1371-1383. McGuinness MC & Smith KD (1999) Cerebral inflammation in X-linked adrenoleukodystrophy.

Arch Immunol Ther Exp 47: 281-287.

77

Mendis-Handagama SM, Zirkin BR, Scallen TJ & Ewing LL (1990) Studies on peroxisomes of the adult rat Leydig cell. J Androl 11: 270-278.

Miettinen M, Mustonen M, Poutanen M, Isomaa V, Wickman M, Soderqvist G, Vihko R & Vihko P (1999) 17β-hydroxysteroid dehydrogenases in normal human mammary epithelial cells and breast tissue. Breast Cancer Res Treat 57: 175-182.

Mihalik SJ, Morrell JC, Kim D, Sacksteder KA, Watkins PA & Gould SJ (1997) Identification of PAHX, a Refsum disease gene. Nat Genet 17: 185-189.

Mihalik SJ, Rainville AM & Watkins PA (1995) Phytanic acid α-oxidation in rat liver peroxisomes. Production of α-hydroxyphytanoyl-CoA and formate is enhanced by dioxygenase cofactors. Eur J Biochem 232: 545-551.

Minami-Ishii N, Taketani S, Osumi T & Hashimoto T (1989) Molecular cloning and sequence analysis of the cDNA for rat mitochondrial enoyl-CoA hydratase. Structural and evolutionary ralationships linked to the bifunctional enzyme of the peroxisomal β-oxidation system. Eur J Biochem 185: 73-78.

Miura S, Kasuya-Arai I, Mori H, Miyazawa S, Osumi T, Hashimoto T & Fujiki Y (1992) Carboxyl-terminal consensus Ser-Lys-Leu-related tripeptide of peroxisomal proteins functions in vitro as a minimal peroxisomal targeting signal. J Biol Chem 267: 14405-14411.

Miura S, Mori M, Takiguchi M, Tatibana M, Furuta S, Miyazawa S & Hashimoto T (1984) Biosynthesis and intracellular transport of enzymes of peroxisomal β-oxidation. J Biol Chem 259: 6397-6402.

Modis Y, Filppula SA, Novikov DK, Norledge B, Hiltunen JK & Wierenga RK (1998) The crystal structure of dienoyl-CoA isomerase at 1.5 Å resolution reveals the importance of aspartate and glutamate sidechains for catalysis. Structure 6: 957-970.

Mori T, Tsukamoto T, Mori H, Tashiro Y & Fujiki Y (1991) Molecular cloning and deduced amino acid sequence of nonspecific lipid transfer protein (sterol carrier protein 2) of rat liver: a higher molecular mass (60 kDa) protein contains the primary sequence of nonspecific lipid transfer protein as its C-terminal part. Proc Natl Acad Sci USA 88: 4338-4342.

Moser AB, Rasmussen M, Naidu S, Watkins PA, McGuinness M, Hajra AK, Chen G, Raymond G, Liu A, Gordon D, Garnaas K, Walton DS, Skjeldal OH, Guggenhelm MA, Jackson LG, Ellas ER & Moser HW (1995) Phenotype of patients with peroxisomal disorders subdivided into sixteen complementation groups. J Pediatr 127: 13-22.

Mosser J, Douar AM, Sarde CO, Kioschis P, Feil R, Moser H, Poustka AM, Mandel JL & Aubourg P (1993) Putative X-linked adrenoleukodystrophy gene shares unexpected homology with ABC transporters. Nature 361: 726-730.

Möller G, van Grunsven EG, Wanders RJ & Adamski J (2001) Molecular basis of D-bifunctional protein deficiency. Mol Cell Endocrinol 171: 61-70.

Möller G, Leenders F, van Grunsven EG, Dolez V, Qualmann B, Kessels MM, Markus M, Krazeisen A, Husen B, Wanders RJ, de Launoit Y & Adamski J (1999) Characterization of the HSD17B4 gene: D-specific multifunctional protein 2/17β-hydroxysteroid dehydrogenase IV. J Steroid Biochem Mol Biol 69: 441-446.

Nakagawa T, Imanaka T, Morita M, Ishiguro K, Yurimoto H, Yamashita A, Kato N & Sakai Y (2000) Peroxisomal membrane protein PMP47 is essential in the metabolism of middle-chain fatty acid in yeast peroxisomes and is associated with peroxisome proliferation. J Biol Chem 275: 3455-3461.

Nakajima Y, Miyahara I, Hirotsu K, Nishina Y, Shiga K, Setoyama C, Tamaoki H & Miura R (2002) Three-Dimensional Structure of the Flavoenzyme Acyl-CoA Oxidase-II from Rat Liver, the Peroxisomal Counterpart of Mitochondrial Acyl-CoA Dehydrogenase. J Biochem (Tokyo) 131: 365-374.

Nawaza J (1994) AMoRe: an automated package for molecular replacement. Acta Crystallogr A 50: 157-163.

Nicolas-Frances V, Dasari VK, Abruzzi E, Osumi T & Latruffe N (2000) The peroxisome proliferator response element (PPRE) present at positions -681/-669 in the rat liver 3-ketoacyl-CoA thiolase B gene functionally interacts differently with PPARα and HNF-4. Biochem Biophys Res Commun 269: 347-351.

78

Nishisho I, Nakamura Y, Miyoshi Y, Miki Y, Ando H, Horii A, Koyama K, Utsunomiya J, Baba S & Hedge P (1991) Mutations of chromosome 5q21 genes in FAP and colorectal cancer patients. Science 253: 665-669.

Normand T, Husen B, Leenders F, Pelczar H, Baert JL, Begue A, Flourens AC, Adamski J & de Launoit Y (1995) Molecular characterization of mouse 17 β-hydroxysteroid dehydrogenase IV. J Steroid Biochem Mol Biol 55: 541-548.

Novikov D, Dieuaide-Noubhani M, Vermeesch JR, Fournier B, Mannaerts GP & Van Veldhoven PP (1997) The human peroxisomal multifunctional protein involved in bile acid synthesis: activity measurement, deficiency in Zellweger syndrome and chromosome mapping. Biochim Biophys Acta 1360: 229-240.

Nuttley WM, Aitchison JD & Rachubinski RA (1988) cDNA cloning and primary structure determination of the peroxisomal trifunctional enzyme hydratase-dehydrogenase-epimerase from the yeast Candida tropicalis pK233. Gene 69: 171-180.

Nöhammer C, El-Shabrawi Y, Schauer S, Hiden M, Berger J, Forss-Petter S, Winter E, Eferl R, Zechner R & Hoefler G (2000) cDNA cloning and analysis of tissue-specific expression of mouse peroxisomal straight-chain acyl-CoA oxidase. Eur J Biochem 267: 1254-1260.

Obeid J & White PC (1992) Tyr-179 and Lys-183 are essential for enzymatic activity of 11β-hydroxysteroid dehydrogenase. Biochem Biophys Res Commun 188: 222-227.

Ockner RK, Manning JA, Poppenhausen RB & Ho WKL (1972) A binding protein for fatty acids in cytosol of intestinal mucosa, liver, myocardium, and other tissues. Science 177: 56

¯58.

Ohba T, Rennert H, Pfeifer SM, He Z, Yamamoto R, Holt JA, Billheimer JT & Strauss JF 3rd (1994) The structure of the human sterol carrier protein X/sterol carrier protein 2 gene (SCP2). Genomics 24: 370-374.

Oppermann UC, Filling C, Berndt KD, Persson B, Benach J, Ladenstein R & Jörnvall H (1997a) Active site directed mutagenesis of 3β/17β-hydroxysteroid dehydrogenase establishes differential effects on short-chain dehydrogenase/reductase reactions. Biochemistry 36: 34-40.

Oppermann UC, Filling C & Jörnvall H (2001) Forms and functions of human SDR enzymes. Chem-Biol Interact 130-132: 699-705.

Oppermann UC, Persson B, Filling C & Jörnvall H (1997b) Structure-function relationship of SDR hydroxysteroid dehydrogenases. Adv Exp Med Biol 414: 403-415.

Osmundsen H, Bremer J & Pedersen JI (1991) Metabolic aspects of peroxisomal β-oxidation. Biochim Biophys Acta 1085: 141-158.

Osumi T & Hashimoto T (1979) Peroxisomal β-oxidation system of rat liver. Copurification of enoyl-CoA hydratase and 3-hydroxyacyl-CoA dehydrogenase. Biochem Biophys Res Commun 89: 580-584.

Osumi T & Hashimoto T (1980b) Purification and properties of mitochondrial and peroxisomal 3-hydroxyacyl-CoA dehydrogenase from rat liver. Arch Biochem Biophys 203: 372-383.

Osumi T, Hashimoto T & Ui N (1980a) Purification and properties of acyl-CoA oxidase from rat liver. J Biochem (Tokyo) 87: 1735-1746.

Osumi T, Ishii N, Hijikata M, Kamijo K, Ozasa H, Furuta S, Miyazawa S, Kondo K, Inoue K, Kagamiyama H & Hashimoto T (1985) Molecular cloning and nucleotide sequence of the cDNA for rat peroxisomal enoyl-CoA: hydratase-3-hydroxyacyl-CoA dehydrogenase bifunctional enzyme. J Biol Chem 260: 8905-8910.

Osumi T, Ishii N, Miyazawa S & Hashimoto T (1987) Isolation and structural characterization of the rat acyl-CoA oxidase gene. J Biol Chem 262: 8138-8143.

Osumi T, Ozasa H & Hashimoto T (1984) Molecular cloning of cDNA for rat acyl-CoA oxidase. J Biol Chem 259: 2031-2034.

Otera H, Harano T, Honsho M, Ghaedi K, Mukai S, Tanaka A, Kawai A, Shimizu N & 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.

Otera H, Okumoto K, Tateishi K, Ikoma Y, Matsuda E, Nishimura M, Tsukamoto T, Osumi T, Ohashi K, Higuchi O & 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.

79

Otwinowski Z & Minor W (1997) Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol 276: 307-326.

Pahan K & Singh I (1995) Phytanic acid oxidation: topographical localization of phytanoyl-CoA ligase and transport of phytanic acid into human peroxisomes. J Lipid Res 36: 986-997.

Palosaari PM & Hiltunen JK (1990) Peroxisomal bifunctional protein from rat liver is a trifunctional enzyme possessing 2-enoyl-CoA hydratase, 3-hydroxyacyl-CoA dehydrogenase, and ∆3,∆2-enoyl-CoA isomerase activities. J Biol Chem 265: 2446-2449.

Palosaari PM, Vihinen M, Mäntsälä PI, Alexson SEH, Pihlajaniemi T & Hiltunen JK (1991) Amino acid sequence similarities of the mitochondrial short chain ∆3,∆2-enoyl-CoA isomerase and peroxisomal multifunctional ∆3,∆2-enoyl-CoA isomerase, 2-enoyl-CoA hydratase, 3-hydroxyacyl-CoA dehydrogenase enzyme in rat liver. J Biol Chem 266: 10750-10753.

Parikh S, Moynihan DP, Xiao G & Tonge PJ (1999) Roles of tyrosine 158 and lysine 165 in the catalytic mechanism of InhA, the enoyl-ACP reductase from Mycobacterium tuberculosis. Biochemistry 38: 13623-13634.

Paton BC & Pollard AN (2000) Molecular changes in the D-bifunctional protein cDNA sequence in Australasian patients belonging to the bifunctional protein complementation group. Cell Biochem Biophys 32: 247-251.

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

Perrakis A, Morris R & Lamzin VS (1999) Automated protein model building combined with iterative structure refinement. Nat Struct Biol 6: 458-463.

Persson B, Krook M & Jörnvall H (1991) Characteristics of short-chain alcohol dehydrogenases and related enzymes. Eur J Biochem 200: 537-543.

Pfeifer SM, Sakuragi N, Ryan A, Johnson AL, Deeley RG, Billheimer JT, Baker ME & Strauss JF 3rd (1993) Chicken sterol carrier protein 2/sterol carrier protein x: cDNA cloning reveals evolutionary conservation of structure and regulated expression. Arch Biochem Biophys 304: 287-293.

Powell AJ, Read JA, Banfield MJ, Gunn-Moore F, Yan SD, Lustbader J, Stern AR, Stern DM & Brady RL (2000) Recognition of structurally diverse substrates by type II 3-hydroxyacyl-CoA dehydrogenase (HADH II)/amyloid-beta binding alcohol dehydrogenase (ABAD). J Mol Biol 303: 311-327.

Purdue PE, Castro SM, Protopopov V & Lazarow PB (1996) Targeting of human catalase to peroxisomes is dependent upon a novel C-terminal peroxisomal targeting sequence. Ann N Y Acad Sci 804: 775-776.

Purdue PE, Yang X & Lazarow PB (1998) Pex18p and Pex21p, a novel pair of related peroxins essential for peroxisomal targeting by the PTS2 pathway. J Cell Biol 143: 1859-1869.

Qi C, Zhu Y, Pan J, Usuda N, Maeda N, Yeldandi AV, Rao MS, Hashimoto T & Reddy JK (1999) Absence of spontaneous peroxisome proliferation in enoyl-CoA hydratase/L-3-hydroxyacyl-CoA dehydrogenase-deficient mouse liver. Further support for the role of fatty acyl CoA oxidase in PPARα ligand metabolism. J Biol Chem 274: 15775-15780.

Qin YM, Haapalainen AM, Conry D, Cuebas DA, Hiltunen JK & Novikov DK (1997b) Recombinant 2-enoyl-CoA hydratase derived from rat peroxisomal multifunctional enzyme 2: role of the hydratase reaction in bile acid synthesis. Biochem J 328: 377-382.

Qin YM, Marttila MS, Haapalainen AM, Siivari KM, Glumoff T & Hiltunen JK (1999) Yeast peroxisomal multifunctional enzyme: (3R)-hydroxyacyl-CoA dehydrogenase domains A and B are required for optimal growth on oleic acid. J Biol Chem 274: 28619-28625.

Qin YM, Poutanen MH, Helander HM, Kvist AP, Siivari KM, Schmitz W, Conzelmann E, Hellman U & Hiltunen JK (1997a) Peroxisomal multifunctional enzyme of β-oxidation metabolizing D-3-hydroxyacyl-CoA esters in rat liver: molecular cloning, expression and characterization. Biochem J 321: 21-28.

Qin YM, Poutanen MH & Novikov DK (2000) Substrate specificities of peroxisomal members of short-chain alcohol dehydrogenase superfamily: expression and characterization of dehydrogenase part of Candida tropicalis multifunctional enzyme. J Lipid Res 41: 93-98.

80

Rachubinski RA, Fujiki Y, Mortensen RM & Lazarow PB (1984) Acyl-CoA oxidase and hydratase-dehydrogenase, two enzymes of the peroxisomal β-oxidation system, are synthesized on free polysomes of clofibrate-treated rat liver. J Cell Biol 99: 2241-2246.

Ramsay RR (2000) The carnitine acyltransferases: modulators of acyl-CoA-dependent reactions. Biochem Soc Trans 28: 182-186.

Ramsay TG (1996) Fat cells. Endocrinol Metab Clin 25: 847-870. Rasmussen JT, Borchers T & Knudsen J (1990) Comparison of the binding affinities of acyl-CoA-

binding protein and fatty-acid-binding protein for long-chain acyl-CoA esters. Biochem J 265: 849-855.

Reddy JK & Mannaerts GP (1994) Peroxisomal lipid metabolism. Annu Rev Nutr 14: 343-370. Reddy MK, Usuda N, Reddy MN, Kuczmarski ER, Rao MS & Reddy JK (1987) Purification,

properties, and immunocytochemical localization of human liver peroxisomal enoyl-CoA hydratase/3-hydroxyacyl-CoA dehydrogenase. Proc Natl Acad Sci USA 84: 3214-3218.

Rehling P, Marzioch M, Niesen F, Wittke E, Veenhuis M & Kunau WH (1996) The import receptor for the peroxisomal targeting signal 2 (PTS2) in Saccharomyces cerevisiae is encoded by the PAS7 gene. EMBO J 15: 2901-2913.

Rehling P, Skaletz-Rorowski A, Girzalsky W, Voorn-Brouwer T, Franse MM, Distel B, Veenhuis M, Kunau WH & 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.

Requena N, Fuller P & Franken P (1999) Molecular characterization of GmFOX2, an evolutionarily highly conserved gene from the mycorrhizal fungus Glomus mosseae, down-regulated during interaction with rhizobacteria. Mol Plant Microbe Interact 12: 934-942.

van Roermund CW, Drissen R, van den Berg M, Ijlst L, Hettema EH, Tabak HF, Waterham HR & Wanders RJ (2001) Identification of a peroxisomal ATP carrier required for medium-chain fatty acid β-oxidation and normal peroxisome proliferation in Saccharomyces cerevisiae. Mol Cell Biol 21: 4321-4329.

van Roermund CW, Tabak HF, van den Berg M, Wanders RJ & Hettema EH (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: 489-498.

Rosendal J, Ertbjerg P & Knudsen J (1993) Characterization of ligand binding to acyl-CoA-binding protein. Biochem J 290: 321-326.

Rossmann MG, Moras D & Olsen KW (1974) Chemical and biological evolution of a nucleotide-binding protein. Nature 250: 194-199.

Rozwarski DA, Vilcheze C, Sugantino M, Bittman R & Sacchettini JC (1999) Crystal structure of the Mycobacterium tuberculosis enoyl-ACP reductase, InhA, in complex with NAD+ and a C16 fatty acyl substrate. J Biol Chem 274: 15582-15589.

Sacchettini JC, Gordon JI & Banaszak LJ (1989) Crystal structure of rat intestinal fatty-acid-binding protein. Refinement and analysis of the Escherichia coli-derived protein with bound palmitate. J Mol Biol 208: 327-339.

Sacksteder KA & Gould SJ (2000) The genetics of peroxisome biogenesis. Annu Rev Genet 34: 623-652.

Scaife JR & Tichivangana JZ (1980) Short chain acyl-CoA synthetases in ovine rumen epithelium. Biochim Biophys Acta 619: 445-450.

Schaffer JE & Lodish HF (1994) Expression, cloning and characterization of a novel adipocyte long chain fatty acid transport protein. Cell 79: 427-436.

Schepers L, Van Veldhoven PP, Casteels M, Eyssen HJ & Mannaerts GP (1990) Presence of three acyl-CoA oxidases in rat liver peroxisomes. An inducible fatty acyl-CoA oxidase, a noninducible fatty acyl-CoA oxidase, and a noninducible trihydroxycoprostanoyl-CoA oxidase. J Biol Chem 265: 5242-5246.

Schmitz W, Albers C, Fingerhut R & Conzelmann E (1995) Purification and characterization of an α-methylacyl-CoA racemase from human liver. Eur J Biochem 231: 815-822.

Schmitz W & Conzelmann E (1997) Stereochemistry of peroxisomal and mitochondrial β-oxidation of α-methylacyl-CoAs. Eur J Biochem 244: 434-440.

81

Schmitz W, Fingerhut R & Conzelmann E (1994) Purification and properties of an α-methylacyl-CoA racemase from rat liver. Eur J Biochem 222: 313-323.

Schram AW, Goldfischer S, van Roermund CW, Brouwer-Kelder EM, Collins J, Hashimoto T, Heymans HS, van den Bosch H, Schutgens RB, Tager JM & Wanders RJA (1987) Human peroxisomal 3-oxoacyl-coenzyme A thiolase deficiency. Proc Natl Acad Sci USA 84: 2494-2496.

Schröder F, Frolov AA, Murphy EJ, Atshaves BP, Jefferson JR, Pu L, Wood WG, Foxworth WB & Kier AB (1996) Recent advances in membrane cholesterol domain dynamics and intracellular cholesterol trafficking. Proc Soc Exp Biol Med 213: 150-177.

Schröder F, Jefferson JR, Kier AB, Knittel J, Scallen TJ, Wood WG & Hapala I (1991) Membrane cholesterol dynamics: cholesterol domains and kinetic pools. Proc Soc Exp Biol Med 196: 235-252.

Schröder F, Myers-Payne SC, Billheimer JT & Wood WG (1995) Probing the ligand binding sites of fatty acid and sterol carrier proteins: effects of ethanol. Biochemistry 34: 11919-11927.

Seedorf U, Ellinghaus P & Roch Nofer J (2000) Sterol carrier protein-2. Biochim Biophys Acta 1486: 45-54.

Seedorf U, Raabe M & Assmann G (1993) Cloning, expression and sequences of mouse sterol-carrier protein-x-encoding cDNAs and a related pseudogene. Gene 123: 165-172.

Seedorf U, Raabe M, Ellinghaus P, Kannenberg F, Fobker M, Engel T, Denis S, Wouters F, Wirtz KW, Wanders RJ, Maeda N & Assmann G (1998) Defective peroxisomal catabolism of branched fatty acyl coenzyme A in mice lacking the sterol carrier protein-2/sterol carrier protein-x gene function. Genes Dev 12: 1189-1201.

Seitz HJ, Muller MJ, Krone W & Tarnowski W (1977) Coordinate control of intermediary metabolism in rat liver by the insulin/glucagon ratio during starvation and after glucose refeeding. Regulatory significance of long-chain acyl-CoA and cyclic AMP. Arch Biochem Biophys 183: 647-663.

Setoyama C, Tamaoki H, Nishina Y, Shiga K & Miura R (1995) Functional expression of two forms of rat acyl-CoA oxidase and their substrate specificities. Biochem Biophys Res Commun 217: 482-487.

Shimozawa N, Suzuki Y, Zhang Z, Imamura A, Ghaedi K, Fujiki Y & Kondo N (2000) Identification of PEX3 as the gene mutated in a Zellweger syndrome patient lacking peroxisomal remnant structures. Hum Mol Genet 2000 9: 1995-1999.

Shimozawa N, Suzuki Y, Zhang Z, Imamura A, Kondo N, Kinoshita N, Fujiki Y, Tsukamoto T, Osumi T, Imanaka T, Orii T, Beemer F, Mooijer P, Dekker C & Wanders RJ (1998) Genetic basis of peroxisome-assembly mutants of humans, Chinese hamster ovary cells, and yeast: identification of a new complementation group of peroxisome-biogenesis disorders apparently lacking peroxisomal-membrane ghosts. Am J Hum Genet 63: 1898-1903.

Singh H, Beckman K & Poulos A (1996) Evidence of two catalytically active carnitine medium/long chain acyltransferases in rat liver peroxisomes. J Lipid Res 37: 2616-2626.

Singh H & Poulos A (1988) Distinct long chain and very long chain fatty acyl-CoA synthetases in rat liver peroxisomes and microsomes. Arch Biochem Biophys 266: 486-495.

Skoneczny M & Lazarow PB (1998) A novel, non-PTS1, peroxisomal import route dependent on the PTS1 receptor Pex5p. Mol Biol Cell 9: 348a.

Small GM, Szabo LJ & Lazarow PB (1988) Acyl-CoA oxidase contains two targeting sequences each of which can mediate protein import into peroxisomes. EMBO J 7: 1167-1173.

Smith BT, Sengupta TK & Singh I (2000) Intraperoxisomal localization of very-long-chain fatty acyl-CoA synthetase: implication in X-adrenoleukodystrophy. Exp Cell Res 254: 309-320.

Smith KD, Kemp S, Braiterman LT, Lu JF, Wei HM, Geraghty M, Stetten G, Bergin JS, Pevsner J & Watkins PA (1999) X-linked adrenoleukodystrophy: genes, mutations, and phenotypes. Neurochem Res 24: 521-535.

Sniderman AD, Maslowska M & Cianflone K (2000) Of mice and men (and women) and the acylation-stimulating protein pathway. Curr Opin Lipidol 11: 291-296.

Sobolev V, Sorokine A, Prilusky J, Abola EE & Edelman M (1999) Automated analysis of interatomic contacts in proteins. Bioinformatics 15: 327-332.

82

South ST & Gould SJ (1999) Peroxisome synthesis in the absence of preexisting peroxisomes. J Cell Biol 144: 255-266.

Spector AA (1984) Plasma lipid transport. Clin Physiol Biochem 2: 123-134. Spiegelman BM & Flier JS (1996) Adibogenesis and obesity: rounding out the big picture. Cell 87:

377-389. Steinberg SJ, Wang SJ, Kim DG, Mihalik SJ & Watkins PA (1999) Human very-long-chain acyl-

CoA synthetase: cloning, topography, and relevance to branched-chain fatty acid metabolism. Biochem Biophys Res Commun 257: 615-621.

Stolowich NJ, Frolov A, Atshaves B, Murphy EJ, Jolly CA, Billheimer JT, Scott AI & Schröder F (1997) The sterol carrier protein-2 fatty acid binding site: an NMR, circular dichroic, and fluorescence spectroscopic determination. Biochemistry 36: 1719-1729.

Storch J & Thumser AE (2000) The fatty acid transport function of fatty acid-binding proteins. Biochim Biophys Acta 1486: 28-44.

Stuhlsatz-Krouper SM, Bennett NE & Schaffer JE (1998) Substitution of alanine for serine 250 in the murine fatty acid transport protein inhibits long chain fatty acid transport. J Biol Chem 273: 28642-28650.

Su HM, Moser AB, Moser HW & Watkins PA (2001) Peroxisomal straight-chain Acyl-CoA oxidase and D-bifunctional protein are essential for the retroconversion step in docosahexaenoic acid synthesis. J Biol Chem 276: 38115-38120.

Subramani S (1993) Protein import into peroxisomes and biogenesis of the organelle. Annu Rev Cell Biol 9: 445-478.

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

Suzuki H, Kawarabayasi Y, Kondo J, Abe T, Nishikawa K, Kimura S, Hashimoto T & Yamamoto T (1990) Structure and regulation of rat long-chain acyl-CoA synthetase. J Biol Chem 265: 8681-8685.

Swinkels BW, Gould SJ & Subramani S (1992) Targeting efficiencies of various permutations of the consensus C-terminal tripeptide peroxisomal targeting signal. FEBS Lett 305: 133-136.

Szilard RK & Rachubinski RA (2000) Tetratricopeptide repeat domain of Yarrowia lipolytica Pex5p is essential for recognition of the type 1 peroxisomal targeting signal but does not confer full biological activity on Pex5p. Biochem J 346: 177-184.

Tanaka N, Nonaka T, Nakanishi M, Deyashiki Y, Hara A & Mitsui Y (1996) Crystal structure of the ternary complex of mouse lung carbonyl reductase at 1.8 Å resolution: the structural origin of coenzyme specificity in the short-chain dehydrogenase/reductase family. Structure 4: 33-45.

Telwilliger TC & Berendzen J (1999) Automated MAD and MIR structure solution. Acta Crystallogr D 55: 849-861.

Thomson AM, Morris R-G, Wallace M, Wyllie AH, Steel CM & Carter DC (1993) Allele loss from 5q21 (APC/MCC) and 18q21 (DCC) and DCC mRNA expression in breast cancer. Br J Cancer 68: 61-68.

Thumser AE & Storch J (2000) Liver and intestinal fatty acid-binding proteins obtain fatty acids from phospholipid membranes by different mechanisms. J Lipid Res 41: 647-656.

Tipping E & Ketterer B (1981) The influence of soluble binding proteins on lipophile transport and metabolism in hepatocytes. Biochem J 195: 441-452.

Titorenko VI & Rachubinski RA (1998) The endoplasmic reticulum plays an essential role in peroxisome biogenesis. Trends Biochem Sci 23: 231-233.

Tong L & Rossmann MG (1997) Rotation function calculations with GLRF program. Methods Enzymol 276: 594-611.

Towbin H, Staehlin T & Gordon J (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: prodecure and some applications. Proc Natl Acad Sci USA 76: 4350-4354.

Tsukamoto T, Hata S, Yokota S, Miura S, Fujiki Y, Hijikata M, Miyazawa S, Hashimoto T & 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.

83

Uchiyama A, Aoyama T, Kamijo K, Uchida Y, Kondo N, Orii T & Hashimoto T (1996) Molecular cloning of cDNA encoding rat very long-chain acyl-CoA synthetase. J Biol Chem 271: 30360-30365.

Unger RH, Zhou YT & Orci L (1999) Regulation of fatty acid homeostasis in cells: novel role of leptin. Proc Natl Acad Sci USA 96: 2327-2332.

Vanhanen S, West M, Kroon JT, Lindner N, Casey J, Cheng Q, Elborough KM & Slabas AR (2000) A consensus sequence for long-chain fatty-acid alcohol oxidases from Candida identifies a family of genes involved in lipid -oxidation in yeast with homologues in plants and bacteria. J Biol Chem 275: 4445-4452.

Vanhooren JC, Fransen M, de Bethune B, Baumgart E, Baes M, Torrekens S, Van Leuven F, Mannaerts GP & Van Veldhoven PP (1996) Rat pristanoyl-CoA oxidase. cDNA cloning and recognition of its C-terminal (SQL) by the peroxisomal-targeting signal 1 receptor. Eur J Biochem 239: 302-309.

Vanhooren JC, Marynen P, Mannaerts GP & Van Veldhoven PP (1997) Evidence for the existence of a pristanoyl-CoA oxidase gene in man. Biochem J 325: 593-599.

Vanhove GF, Van Veldhoven PP, Fransen M, Denis S, Eyssen HJ, Wanders RJ & Mannaerts GP (1993) The CoA esters of 2-methyl-branched chain fatty acids and of the bile acid intermediates di- and trihydroxycoprostanic acids are oxidized by one single peroxisomal branched chain acyl-CoA oxidase in human liver and kidney. J Biol Chem 268: 10335-10344.

Varanasi U, Chu R, Chu S, Espinosa R, LeBeau MM & Reddy JK (1994) Isolation of the human peroxisomal acyl-CoA oxidase gene: organization, promoter analysis, and chromosomal localization. Proc Natl Acad Sci USA 91: 3107-3111.

Varanasi U, Chu R, Huang Q, Castellon R, Yeldandi AV & Reddy JK (1996) Identification of a peroxisome proliferator-responsive element upstream of the human peroxisomal fatty acyl coenzyme A oxidase gene. J Biol Chem 271: 2147-2155.

Veerkamp JH (1995) Fatty acid transport and fatty acid-binding proteins. Proc Nutr Soc 54: 23-37. Van Veldhoven PP, Asselberghs S, Eyssen HJ & Mannaerts GP (1996) Stabilisation and partial

purification of Triton X-100 solubilised trihydroxycoprostanoyl-CoA synthetase from rat liver. Biochem Mol Biol Int 40: 447-457.

Van Veldhoven PP, Casteels M, Mannaerts GP & Baes M (2001) Further insights into peroxisomal lipid breakdown via α- and β-oxidation. Biochem Soc Trans 29: 292-298.

Van Veldhoven PP, Vanhove G, Assselberghs S, Eyssen HJ & Mannaerts GP (1992) Substrate specificities of rat liver peroxisomal acyl-CoA oxidases: palmitoyl-CoA oxidase (inducible acyl-CoA oxidase), pristanoyl-CoA oxidase (non-inducible acyl-CoA oxidase), and trihydroxycoprostanoyl-CoA oxidase. J Biol Chem 267: 20065-20074.

Van Veldhoven PP, Vanhove G, Vanhoutte F, Dacremont G, Parmentier G, Eyssen HJ & Mannaerts GP (1991) Identification and purification of a peroxisomal branched chain fatty acyl-CoA oxidase. J Biol Chem 266: 24676-24683.

Verhoeven NM, Roe DS, Kok RM, Wanders RJ, Jakobs C & Roe CR (1998) Phytanic acid and pristanic acid are oxidized by sequential peroxisomal and mitochondrial reactions in cultured fibroblasts. J Lipid Res 39: 66-74.

Verhoeven NM, Schor DS, ten Brink HJ, Wanders RJ & Jakobs C (1997) Resolution of the phytanic acid α-oxidation pathway: identification of pristanal as product of the decarboxylation of 2-hydroxyphytanoyl-CoA. Biochem Biophys Res Commun 237: 33-36.

Volkl A, Baumgart E & Fahimi HD (1988) Localization of urate oxidase in the crystalline cores of rat liver peroxisomes by immunocytochemistry and immunoblotting. J Histochem Cytochem 36: 329-336.

Vorum H, Brodersen R, Kragh-Hansen U & Pedersen AO (1992) Solubility of long-chain fatty acids in phosphate buffer at pH 7.4. Biochim Biophys Acta 1126: 135-142.

Vriend G (1990) WHAT IF: a molecular modeling and drug design program. J Mol Graph 8: 52-56. Walton PA, Hill PE & Subramani S (1995) Import of stably folded proteins into peroxisomes. Mol

Biol Cell 6: 675-683. Walton PA, Wendland M, Subramani S, Rachubinski RA & Welch WJ (1994) Involvement of 70-

kDa heat-shock proteins in peroxisomal import. J Cell Biol 125: 1037-1046.

84

Wanders BJ, Denis SW & Dacremont G (1993) Studies on the substrate specificity of the inducible and non-inducible acyl-CoA oxidases from rat kidney peroxisomes. J Biochem (Tokyo) 113: 577-582.

Wanders RJ & Tager JM (1998) Lipid metabolism in peroxisomes in relation to human disease. Mol Aspects Med 19: 69-154.

Wanders RJ, Vreken P, Ferdinandusse S, Jansen GA, Waterham HR, van Roermund CW & van Grunsven EG (2001) Peroxisomal fatty acid α- and β-oxidation in humans: enzymology, peroxisomal metabolite transporters and peroxisomal diseases. Biochem Soc Trans 29: 250-267.

Watkins PA, Howard AE, Gould SJ, Avigan J & Mihalik SJ (1996) Phytanic acid activation in rat liver peroxisomes is catalyzed by long-chain acyl-CoA synthetase. J Lipid Res 37: 2288-2295.

Watkins PA, Howard AE & Mihalik SJ (1994) Phytanic acid must be activated to phytanoyl-CoA prior to its α-oxidation in rat liver peroxisomes. Biochim Biophys Acta 1214: 288-294.

Weisiger RA (1996) When is a carrier not a membrane carrier? The cytoplasmic transport of amphipathic molecules. Hepatology 24: 1288-1295.

Wendland M & Subramani S (1993) Cytosol-dependent peroxisomal protein import in a permeabilized cell system. J Cell Biol 120: 675-685.

Wierenga RK, Maeyer MCH & Hol WGJ (1985) Interaction of pyrophosphate moieties with α-helices in dinucleotide binding protein. Biochemistry 24: 1346-1357.

Winberg JO, Brendskag MK, Sylte I, Lindstad RI & McKinley-McKee JS (1999) The catalytic triad in Drosophila alcohol dehydrogenase: pH, temperature and molecular modelling studies. J Mol Biol 294: 601-616.

Wouters FS, Bastiaens PI, Wirtz KW & Jovin TM (1998) FRET microscopy demonstrates molecular association of non-specific lipid transfer protein (nsL-TP) with fatty acid oxidation enzymes in peroxisomes. EMBO J 17: 7179-7189.

Xu R & Cuebas DA (1996) The reactions catalyzed by the inducible bifunctional enzyme of rat liver peroxisomes cannot lead to the formation of bile acids. Biochem Biophys Res Commun 221: 271-278.

Young AC, Scapin G, Kromminga A, Patel SB, Veerkamp JH & Sacchettini JC (1994) Structural studies on human muscle fatty acid binding protein at 1.4 Å resolution: binding interactions with three C18 fatty acids. Structure 2: 523-534.

Zanotti G, Scapin G, Spadon P, Veerkamp JH & Sacchettini JC (1992) Three-dimensional structure of recombinant human muscle fatty acid-binding protein. J Biol Chem 267: 18541-18550.

Zhang JW & Lazarow PB (1996) Peb1p (Pas7p) is an intraperoxisomal receptor for the NH2-

terminal, type 2, peroxisomal targeting sequence of thiolase: Peb1p itself is targeted to peroxisomes by an NH

2-terminal peptide. J Cell Biol 132: 325-334.