J Biol Chem, Vol. 274, Issue 31, 21797-21803, July 30, 1999

Characterization of PECI, a Novel Monofunctional ³,²-Enoyl-CoA Isomerase of Mammalian Peroxisomes^*

Brian V. Geisbrecht, Dongyan Zhang^§, Horst Schulz^§, and Stephen J. Gould^¶

From the  Department of Biological Chemistry, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21205 and the ^§ Department of Chemistry, The City College of The City University of New York, New York, 10031

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

We report here the identification and characterization of human and mouse PECI, a novel gene that encodes a monofunctional peroxisomal ³,²-enoyl-CoA isomerase. Human and mouse PECI were identified on the basis of their sequence similarity to Eci1p, a recently characterized peroxisomal ³,²-enoyl-CoA isomerase from the yeast Saccharomyces cerevisiae. Cloning and sequencing of the human PECI cDNA revealed the presence of a 1077-base pair open reading frame predicted to encode a 359-amino acid protein with a mass of 39.6 kDa. The corresponding mouse cDNA contains a 1074-base pair open reading frame that encodes a 358-amino acid-long protein with a deduced mass of 39.4 kDa. Northern blot analysis demonstrated human PECI mRNA is expressed in all tissues. A bacterially expressed form of human PECI catalyzed the isomerization of 3-cis-octenoyl-CoA to 2-trans-octenoyl-CoA with a specific activity of 27 units/mg of protein. The human and mouse PECI proteins contain type-1 peroxisomal targeting signals, and human PECI was localized to peroxisomes by both subcellular fractionation and immunofluorescence microscopy techniques. The potential roles for this monofunctional ³,²-enoyl-CoA isomerase in peroxisomal metabolism are discussed.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The utilization of fatty acids as an energy source is characteristic of nearly all free-living organisms. The series of enzyme-catalyzed reactions required to degrade fatty acids is evolutionarily conserved and is accomplished primarily through the pathway of -oxidation. In lower eukaryotes such as yeasts and plants, -oxidation is localized exclusively to the peroxisome (1, 2). Studies of the model organism Saccharomyces cerevisiae have demonstrated that three genes encode the four core enzymes of the -oxidation pathway. This pathway proceeds sequentially through (i) the oxidation of a fatty acyl-CoA to a 2-enoyl-CoA in a reaction catalyzed by acyl-CoA oxidase (Pox1p) (3), (ii) the hydration of the 2-enoyl-CoA to a D-3-hydroxyacyl-CoA, (iii) the subsequent dehydrogenation of this intermediate to a 3-ketoacyl-CoA in reactions catalyzed by the bifunctional 2-enoyl-CoA hydratase/D-3-hydroxyacyl-CoA dehydrogenase enzyme (Fox2p) (4), and (iv) the cleavage of the 3-ketoacyl-CoA by thiolase (Pot1p) (5) to acetyl-CoA and a fatty acyl-CoA shortened by two carbon units. Although these enzymes are necessary for all fatty acid -oxidation, the complete metabolism of unsaturated fatty acids requires one or more additional auxiliary enzymes (6). These include ³,²-enoyl-CoA isomerase (Eci1p) (7, 8), a 2,4-dienoyl-CoA reductase (Sps19p) (9), an NADP⁺-dependent isocitrate dehydrogenase (Idp3p) (10, 11), and a ^3,5,^2,4-dienoyl-CoA isomerase (although this enzyme has yet to be identified in yeast). Among these auxiliary enzymes, ³,²-enoyl-CoA isomerase (Eci1p) is unique in that its activity is essential for the -oxidation of all unsaturated fatty acids (Fig. 1).

View larger version (27K): [in this window] [in a new window]   Fig. 1.   Pathways specific for unsaturated fatty acid metabolism. -Oxidation of unsaturated fatty acids with double bonds extending from even-numbered carbons yields 2,4-dienoyl-CoAs that are metabolized as shown at left (10, 11, 42). 2,5-Dienoyl-CoAs form during the -oxidation of unsaturated fatty acids with double bonds at odd-numbered carbons and return to the core spiral through either the NADP+-independent (center) (30) or the NADP+-dependent (right) (43, 44) pathway as shown. Although not depicted in this figure, an intraperoxisomal NADP+-dependent isocitrate dehydrogenase is required to regenerate the NADPH consumed by the 2,4-dienoyl-CoA reductase (10, 11). Note that the final step in each pathway is catalyzed by 3,2-enoyl-CoA isomerase.

In mammals, fatty acid -oxidation is considerably more complex than in yeast, primarily due to the existence of overlapping but distinct fatty acid -oxidation pathways. These include separate, complete fatty acid oxidation systems in both mitochondria and peroxisomes, as well as substrate specific sets of enzymes in each organelle. For instance, mammalian peroxisomes contain at least three fatty acyl-CoA oxidases (12, 13), both L-specific and D-specific 2-enoyl-CoA hydratase/3-hydroxyacyl-CoA dehydrogenase multifunctional proteins (13), and at least two thiolases (12, 13), all of which are encoded by separate genes. Although only single peroxisomal ³,²-enoyl-CoA (14) and ^3,5,^2,4-dienoyl-CoA isomerases (15) have been reported, the complex array of peroxisomal core -oxidation enzymes suggests that additional forms of these auxiliary enzymes may also be present in this organelle.

The ³,²-enoyl-CoA isomerase activity of mammalian peroxisomes has been ascribed previously to MFE1, the L-specific 2-enoyl-CoA hydratase/3-hydroxyacyl-CoA dehydrogenase multifunctional enzyme (14). However, we and others have recently identified and characterized the S. cerevisiae ³,²-enoyl-CoA isomerase (7, 8) and have found that this enzyme, Eci1p, shares few primary sequence features with MFE1. In this report we describe the identification and characterization of a novel, ubiquitously expressed mammalian peroxisomal ³,²-enoyl-CoA isomerase (PECI) that is homologous to yeast Eci1p. Human and mouse cDNAs for PECI were cloned, the human protein was found to have significant ³,²-enoyl-CoA isomerase activity, and the human PECI was localized to the peroxisome matrix by biochemical and immunofluorescence techniques.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

cDNA Cloning of Mammalian Eci1p Homologs-- A cDNA clone (GenBank^TM accession number AA188052) encoding a candidate human homolog (human PECI) of yeast Eci1p was obtained from Genome Systems, Inc. (St. Louis, MO). The 5'-untranslated region of the human PECI cDNA was extended by 5'-directed rapid amplification of cDNA ends (5'-RACE)¹ polymerase chain reaction using a human fetal liver Marathon RACE cDNA (CLONTECH). The oligonucleotide 5'-ATATAGCGCGTAGAGTTTTAGCTTCACTTCGTTTCCTGG-3' was used as a human PECI cDNA-specific primer, and all 5'-RACE reactions were performed in the presence of 6% Me₂SO according to the manufacturer's suggestions. A cDNA clone (GenBank^TM accession number AA030780) encoding a candidate mouse homolog (mouse PECI) of Eci1p was identified by essentially the same strategy as was the human cDNA, and this clone was obtained from Genome Systems, Inc.

Plasmids-- Two forms of the human PECI open reading frame (ORF) were amplified by polymerase chain reaction using the AA188052 plasmid as a template. The complete ORF was amplified using the oligonucleotides 5'-AAAGTCGACAATGAGAGCCAGTCAGAAGGACTTTG-3' and 5'-TTTGCGGCCGCTCATCACAGTTTTGATTTTCTGGATAAGAA-3'. A form of human PECI lacking the final three codons of the ORF (PECISKL) was amplified by using the oligonucleotide 5'-TTTGCGGCCGCTCATTTTCTGGATAAGAAGTTCACCAC-3'. Both sets of oligonucleotides append SalI and NotI sites (underlined sequences) at the 5'- and 3'-ends of the PECI ORF. All polymerase chain reactions were performed with a low error-rate mixture of polymerases (Expand, Roche Molecular Biochemicals). The polymerase chain reaction product from each reaction was digested with SalI and NotI and subcloned into the SalI and NotI sites of pMBP (7, 16). The sequence of each form of the human PECI ORF in pMBP was confirmed by automated fluorescent sequencing, and the resulting plasmids were denoted pMBP-PECI and pMBP-PECISKL. The SalI-NotI fragment of pMBP-PECI was excised and transferred to the plasmid pT7His. This plasmid is a modified form of pET28a (Novagen, Inc.) and contains additional XhoI and SalI sites in place of the NheI-HindIII fragment from the parental vector. The resulting plasmid, pT7His-PECI, allows for T7 polymerase-driven expression of an N-terminal His₆-PECI fusion protein. The SalI-NotI fragment of pMBP-PECI was also excised and transferred to the vector pT7, a modified version of pET28a that lacks sequences encoding the hexahistidinyl tag and instead contains XhoI, SalI, and NotI sites immediately downstream of the initiating ATG codon. The resulting plasmid, pT7-PECI, allows for the in vitro transcription and translation of unmodified PECI.

Strains and Culture-- For routine manipulations of cDNAs and plasmids, the Escherichia coli strain DH10B was used (17). DH10B cells were also used for the expression of all maltose-binding protein (MBP) fusion proteins. For T7 polymerase-driven expression of recombinant proteins, the E. coli strain BL21(DE3) was chosen (Novagen, Inc.). All media for bacterial culture have been described (18).

Northern Blot Analysis-- Human multi-tissue Northern blots were obtained from CLONTECH. Probes were generated by random primed labeling of PECI cDNA fragments in the presence of [-³²P]dATP using the Prime It kit (Stratagene, Inc.). Hybridizations and washing were carried out according to standard protocols (18).

Sequence Analysis and Alignments-- The Clustal algorithm and PAM250 substitution matrix were used in conjunction with DNASTAR (Madison, WI) software to perform all sequence alignments.

Expression and Purification of Recombinant PECI Proteins-- The plasmids pMBP-PECI and pMBP-PECISKL are designed to express the human PECI and human PECISKL proteins, respectively, in fusion with E. coli MBP. Induction of protein expression, cell growth and lysis, and amylose-affinity chromatography methods have been described in a previous report (7). For expression of His₆-PECI, freshly transformed BL21(DE3) cells harboring the pT7His-PECI plasmid were grown at 30 °C with vigorous shaking (275 rpm) for 12-14 h in 50 ml of LB media supplemented with 25 µg/ml kanamycin sulfate and sterile dextrose to 1%. After this incubation period, 7.5 ml of cell suspension from the preculture were diluted into 500 ml of 2YT media containing 25 µg/ml kanamycin sulfate and sterile dextrose to 0.2%. This culture was grown with vigorous shaking at 18 °C until the A₆₀₀ reached 0.5, at which time induction of protein expression was begun by the addition of isopropyl--D-thiogalactoside to a final concentration of 1 mM.

Following growth of the induced culture for 18 h, cells were harvested and a cleared, soluble protein lysate was prepared as described previously (7) in 50 ml of Column Buffer A (20 mM sodium-P_i (pH 7.8), 500 mM NaCl, and 5 mM 2-mercaptoethanol). The His₆-PECI present in the soluble protein lysate was purified by metal chelate affinity chromatography using Ni-NTA agarose (ProBond; Invitrogen, Inc.). Briefly, the soluble protein lysate was to diluted to a final volume of 200 ml in Column Buffer A and was applied at 2 ml/min to a 5 ml bed of ProBond agarose (prepared according to the manufacturer's suggestions) at 4 °C. The bed was washed with 5 volumes of Buffer A and then with 20 volumes of Buffer B (20 mM sodium-P_i (pH 6.0), 500 mM NaCl, and 5 mM 2-mercaptoethanol). Following these washing steps, the His₆-PECI was eluted from the bed using an imidazole step gradient in Buffer B. One column volume each of buffers containing 50, 250, 350, and 500 mM imidazole was applied sequentially to the resin bed, the eluants were collected, and each fraction was assayed for the presence of His₆-PECI by SDS-polyacrylamide gel electrophoresis (observed M_r  42,000). Fractions containing highly purified (>90% by Coomassie stain) His₆-PECI were pooled, and the His₆-PECI present was precipitated slowly by the addition of solid ammonium sulfate to 0.4 g/ml. The His₆-PECI precipitate was collected from this suspension and then stored at 70 °C until use.

Analytical Procedures-- ³,²-enoyl-CoA isomerase activity using 3-cis-octenoyl-CoA as a substrate was monitored spectrophotometrically at 340 nm according to the coupled assay originally described by Binstock and Schulz (19). One unit of ³,²-enoyl-CoA isomerase activity is defined as the amount of enzyme required to catalyze the isomerization of 1 µmol of 3-cis-octenoyl-CoA to 2-trans-octenoyl-CoA in 1 min under standard assay conditions. Measurements of 2-enoyl-CoA hydratase activity (using crotonyl-CoA) and ^3,5,^2,4-dienoyl-CoA isomerase activity (using 3,5-octadienoyl-CoA) were performed essentially as described (20). Assays for succinate dehydrogenase (a mitochondrial marker) (21) and for catalase (a peroxisomal marker) (22, 23) have been described. Total protein concentration was determined using the Bradford method (Bio-Rad) with bovine serum albumin as a reference.

In Vitro Translations and Preparation of Whole Cell Protein Lysates-- The plasmid pT7-PECI contains a T7 promoter 5' of the PECI cDNA and was used as a template for in vitro transcription and translation reactions. Reactions were carried out on a 25-µl scale using the T7-coupled TnT system (Promega, Inc.) according to the manufacturer's suggestions. Total cellular protein lysates were prepared from human skin fibroblasts (cell line GM5756) as described previously (24).

Mammalian Cell Culture and Antibodies-- Methods for the culture of skin fibroblasts and HepG2 cells have been described (25). The normal human fibroblast cell line GM5756 was purchased from the Coriell Cell Repository (Vineland, NJ). The pex10-deficient cell line PBD100 has been described and was a gift from A. Moser and H. Moser (The Kennedy-Krieger Institute, Baltimore, MD) (26). Methods for indirect immunofluorescence microscopy have been described (25).

The tissue culture supernatant from mouse hybridoma line 1-9E10 (Roche Molecular Biochemicals) was the source of the monoclonal anti-c-Myc antibody. The Binding Site (San Diego, CA) was the source of affinity-purified sheep antibodies recognizing human catalase. Affinity-purified fluorescein-anti-mouse, fluorescein-anti-sheep, and Texas red-anti-rabbit secondary antibodies were obtained from Kirkegaard and Perry Laboratories (Gaithersburg, MD) and were used according to the manufacturer's suggestions. Guinea pig polyclonal anti-PMP70 antibodies were raised against a synthetic peptide corresponding to the C-terminal 18 amino acids of human PMP70 (26). Bacterially expressed MBP-PECISKL was used to elicit the production of polyclonal anti-PECI antibodies in New Zealand White rabbits. Rabbits were purchased from, maintained at, and immunized according to the standard protocols of Cocalico Biologicals, Inc. (Reamstown, PA). Anti-PECI antibodies were purified on an antigen column consisting of purified MBP-PECISKL that was chemically coupled to a support of cyanogen bromide Sepharose (Sigma) according to the manufacturer's suggestions. Specifically, 3 ml of anti-PECI immune sera were diluted to 30 ml in phosphate buffered saline (18). This sample was applied to a 300-µl bed of MBP-PECISKL-Sepharose equilibrated in the same buffer at 18 °C. Binding, washing, elution, and storage of purified antibodies was as described (27). Immunoblotting was performed as described in Crane et al. (28).

Subcellular Fractionation of HepG2 Cells-- Preparation of a postnuclear supernatant from HepG2 cells and fractionation of this supernatant by ultracentrifugation on a 15-42% linear Nycodenz density gradient has been described (29). Following ultracentrifugation fractions (750 µl) were drawn from the bottom of the gradient and assayed for a peroxisomal marker enzyme, catalase, and for a mitochondrial marker enzyme, succinate dehydrogenase. Immediately following these assays, the proteins present in each fraction were precipitated by adding trichloroacetic acid to a final concentration of 15%. The precipitated samples were prepared for 10% SDS-polyacrylamide gel electrophoresis and immunoblotting as described (28, 29).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Identification of PECI, a Novel Mammalian ³,²-Enoyl-CoA Isomerase-- We recently reported the identification and characterization of a peroxisomal ³,²-enoyl-CoA isomerase, Eci1p, from the yeast S. cerevisiae (7). This apparently monofunctional enzyme is the sole physiologically relevant source of ³,²-enoyl-CoA isomerase activity in S. cerevisiae and is required to completely oxidize unsaturated fatty acids (7, 30, 31). Peroxisomes of mammalian cells also contain a ³,²-enoyl-CoA isomerase. Although this enzymatic activity has been attributed previously to the multifunctional enzyme (MFE1) that also catalyzes 2-enoyl-CoA hydratase and L-3-hydroxyacyl-CoA dehydrogenase reactions (14), we tested whether mammalian peroxisomes might contain a protein similar to Eci1p. The BLAST algorithm was used to search the data base of expressed sequence tags (dbEST) for mammalian cDNAs capable of encoding a protein with significant similarity to Eci1p, the deduced product of the yeast ECI1 gene. Iterative searches identified a human endothelial cell cDNA (clone AA188052) with the potential of encoding such a protein. This cDNA (denoted PECI) was sequenced in its entirety and the sequence of the 5'-untranslated region was extended by 5'-RACE. The PECI cDNA was found to contain a 1077-base pair open reading frame and was predicted to encode a 39.6-kDa protein of 359 amino acids (Fig. 2). These searches likewise identified a mouse cDNA (MmPECI, clone AA030780) with the potential to encode a protein with significant homology to both Eci1p and human PECI. The MmPECI cDNA was also sequenced in its entirety and was found to contain a 1074-base pair open reading frame. This open reading frame was predicted to encode a 358 amino acid polypeptide with a deduced molecular mass of 39.4 kDa (Fig. 3).

View larger version (71K): [in this window] [in a new window]   Fig. 2.   Identification of a human homolog of S. cerevisiae Eci1p. The nucleotide sequence of the human PECI cDNA is shown along with the deduced translation of its 1077-base pair open reading frame. The proposed initiating ATG was chosen because it is present in a more efficient Kozak context than the ATG at cDNA position 15 (underlined) and corresponds to the start of the MmPECI open reading frame (see Fig. 3). The deduced protein is basic (pI = 9.00), has a molecular mass of 39.6 kDa, and terminates in the consensus type-1 peroxisomal targeting signal Ser-Lys-Leu-COOH (underlined).

View larger version (70K): [in this window] [in a new window]   Fig. 3.   Identification of a mouse homolog of S. cerevisiae Eci1p. The nucleotide sequence of the mouse PECI cDNA is depicted with the deduced translation of the 1074-base pair open reading frame contained therein. Note the presence of a TTG codon at cDNA position 15 (underlined) in contrast to the ATG found in this position of human PECI. In frame, upstream stop codons (underlined) are found at positions 139, 333, and 348. MmPECI is basic (pI = 8.42), has a molecular mass of 39.4 kDa, and terminates in the atypical type-1 peroxisomal targeting signal, Pro-Lys-Leu-COOH (underlined).

Almost all human tissues are capable of peroxisomal fatty acid -oxidation. Northern blot analysis was used to determine whether the pattern of PECI expression was similarly broad. The 1.5-kilobase PECI mRNA was detected in all 16 human tissues examined (Fig. 4). Furthermore, the size of the PECI mRNA was consistent with the 1.3-kilobase size of the PECI cDNA and suggested that our cDNA may approach full length. PECI mRNA appeared to be most abundant in heart, skeletal muscle, and liver.

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Tissue-specific expression of human PECI mRNA. Multitissue Northern blots were probed with a 32P-labeled PECI cDNA. Lane 1, heart. Lane 2, brain. Lane 3, placenta. Lane 4, lung. Lane 5, liver. Lane 6, skeletal muscle. Lane 7, kidney. Lane 8, pancreas. Lane 9, spleen. Lane 10, thymus. Lane 11, prostate. Lane 12, testis. Lane 13, ovary. Lane 14, small intestine. Lane 15, colon. Lane 16, peripheral blood leukocyte. The position of the 1.35-kilobase marker is shown.

An examination of the PECI cDNA sequence demonstrated the presence of two potential initiating ATG codons in the same reading frame (Fig. 2). This raised questions as to the authentic translation start site. Initiation of translation on mammalian mRNAs is favored when purines are present at the highest priority positions, namely +4 and 3 relative to the A of the ATG (32, 33). Examination of the second ATG revealed a match for this consensus. In contrast, the first ATG (position 15 in Fig. 2) has a poor consensus match for high efficiency initiation because a pyrimidine is found at position 3 relative to this ATG (denoted 18 in Fig. 2). Thus, the second ATG appears favored for translation initiation on the basis of the context. An independent line of evidence supporting the assignment of the second ATG as the translation start site in human PECI is derived from the MmPECI cDNA sequence in which the upstream ATG is absent (Fig. 3).

A comparison of the deduced amino acid sequences of human and mouse PECI with yeast Eci1p revealed that these novel proteins share 22 and 19% identity to Eci1p, respectively (Fig. 5). The fact that both human and murine PECI are the most similar proteins in these species to Eci1p (the yeast peroxisomal ³,²-enoyl-CoA isomerase) suggested that they may also have ³,²-enoyl-CoA isomerase activity. To test this hypothesis, we expressed human PECI in E. coli with an N-terminal hexahistidinyl tag. Soluble His₆-PECI was purified by affinity chromatography on a nickel agarose column and was assayed for ³,²-enoyl-CoA isomerase activity (Table I). We observed that recombinant PECI catalyzed the isomerization 3-cis-octenoyl-CoA to 2-trans-enoyl-CoA with a specific activity of 27 units/mg of protein. This level of activity is similar to values reported for the rat short-chain mitochondrial ³,²-enoyl-CoA isomerase (38 units/mg) (34) and for bacterially expressed Eci1p (12-16 units/mg) (7, 8). We also assayed recombinant PECI for 2-enoyl-CoA hydratase and ^3,5,^2,4-dienoyl-CoA isomerase activities (Table I) but were unable to detect these activities using crotonyl-CoA and 3,5-octadienoyl-CoA as the respective substrates.

View larger version (67K): [in this window] [in a new window]   Fig. 5.   Amino acid alignment of the mammalian PECI proteins with yeast Eci1p. Clustal amino acid alignment of human and mouse PECI with S. cerevisiae Eci1p is shown with residues conserved in two of three sequences in reverse type. Note the presence of the NGPA(V/I)G(I/L)S motif (positions 208-215 of human PECI) that is shared between yeast Eci1p and its mammalian homologs.

PECI Is a Matrix Enzyme of Human Peroxisomes-- Human and murine PECI contained putative type-1 peroxisomal targeting signals, Ser-Lys-Leu-COOH and Pro-Lys-Leu-COOH, respectively. In order to assess the subcellular localization of this enzyme, we first raised polyclonal antibodies to recombinant PECI (a truncated derivative lacking the last three residues was used to avoid raising antibodies to the type-1 peroxisomal targeting signal (35)). Immunoblots of human fibroblast cell lysates demonstrated that immune sera detected a 39-kDa protein, whereas preimmune sera failed to detect this protein (Fig. 6). Affinity-purified anti-PECI antibodies were also tested against total human fibroblast protein as well as in vitro synthesized PECI that was encoded by the PECI cDNA clone. These experiments revealed that the affinity-purified anti-PECI antibodies specifically recognized a 39-kDa protein in whole cell extracts from fibroblasts and that this protein was indistinguishable in size from the product of the PECI cDNA (Fig. 6).

View larger version (33K): [in this window] [in a new window]   Fig. 6.   Affinity-purified polyclonal antibodies specifically recognize human PECI. Whole cell lysates from human skin fibroblasts (40 µg/lane) were separated by 15% SDS-polyacrylamide gel electrophoresis, transferred to a membrane, and blotted with either preimmune (lane 1) or immune (lane 2) sera at a dilution of 1:2500. An identical electrophoresis system was also used to separate 0.8 µl of a control rabbit reticulocyte lysate (luciferase) (lane 3), 0.8 µl of a rabbit reticulocyte lysate that had been charged with the human PECI cDNA (pT7-PECI) (lane 4), and 40 µg of total cellular protein from human fibroblasts (lane 5). The resolved protein samples were transferred to a membrane and were probed with affinity-purified anti-PECI antibodies. The immunoreactive band in each lane corresponds to the predicted mass of human PECI (39 kDa).

To determine the subcellular distribution of endogenous PECI, a postnuclear supernatant from human hepatocellular carcinoma (HepG2) cells was separated by ultracentrifugation on a linear Nycodenz density gradient. Fractions were assayed for enzyme markers of the peroxisome and the mitochondrion (Fig. 7A), as well as for PECI by immunoblot (Fig. 7B). PECI was localized predominantly to the peroxisomal fractions and appeared to leak from peroxisomes even less than catalase, the prototypical marker for peroxisomes. It should be noted that these experiments were performed with 10% SDS-polyacrylamide gels (a suboptimal resolving system for a protein of this size), which resulted in aberrant migration of PECI with the 35-kDa marker.

View larger version (29K): [in this window] [in a new window]   Fig. 7.   Endogenous PECI is present in the peroxisomes of human hepatocytes. A postnuclear supernatant from human HepG2 cells was fractionated by Nycodenz density gradient ultracentrifugation. A, equal portions of each fraction were assayed for a peroxisomal marker enzyme, catalase (filled bar), and a mitochondrial marker enzyme, succinate dehydrogenase (open bar). The bottom of the gradient is to the left. B, equal portions of each fraction were also separated by 10% SDS-polyacrylamide gel electrophoresis, transferred to membranes, and blotted with affinity-purified anti-human PECI antibodies.

We also examined the subcellular distribution of human PECI by immunofluorescence microscopy. A human skin fibroblast cell line was processed for double indirect immunofluorescence using affinity-purified rabbit anti-PECI antibodies and guinea pig anti-PMP70 antibodies that recognize the cytoplasmically exposed C-terminal tail of this integral peroxisomal membrane protein. PECI colocalized with PMP70 (Fig. 8, A and B) under conditions in which all cellular membranes were permeabilized, demonstrating that PECI is indeed a peroxisomal protein. However, when the same cells were processed so that only the plasma membrane was permeabilized (digitonin permeabilization (37)), the signal for PECI was lost, whereas the signal for PMP70 could still be detected (Fig. 8, C and D). This result was also observed for catalase, a well established marker of the peroxisome matrix (Fig. 8, E and F). Thus, PECI appeared to reside within the peroxisome lumen. As a final test of PECI localization, we examined its distribution in PBD100 cells, a human skin fibroblast cell line derived from a Zellweger syndrome patient. PBD100 is homozygous for inactivating mutations in the PEX10 gene (26), is unable to import peroxisomal matrix proteins, but does synthesize peroxisomes and import peroxisomal membrane proteins (26). As expected for a peroxisomal matrix protein, PECI accumulated in the cytosol of PBD100 cells (Fig. 8, G and H). Based on these results and those presented in Fig. 7, we conclude that PECI is a novel peroxisomal matrix enzyme.

View larger version (96K): [in this window] [in a new window]   Fig. 8.   PECI is a matrix protein of human peroxisomes. Human skin fibroblasts were fixed, permeabilized with 1% Triton X-100 for 5 min, and then processed for double indirect immunofluorescence using affinity-purified anti-PECI antibodies (A) and antibodies that recognize the C-terminal 18 amino acids of human PMP70 (B). Human skin fibroblasts were also fixed, permeabilized with 25 µg/ml digitonin for 2 min, and then processed for double indirect immunofluorescence using the same antibodies to PECI (C) and PMP70 (D). Cells processed in exactly the same manner were processed with antibodies specific for catalase (E) and PMP70 (F). Double indirect immunofluorescence was used to examine the distribution of PECI in pex10-deficient PBD100 cells (permeabilized with Triton X-100), again with the affinity-purified anti-PECI (G) and anti-PMP70 (H) antibodies. Bar, 25 µm.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The activity of ³,²-enoyl-CoA isomerase is essential to completely oxidize unsaturated fatty acids, and its presence has been demonstrated in bacteria, yeast peroxisomes, mammalian peroxisomes, and mammalian mitochondria. We have reported here the identification of cDNAs encoding a novel, ubiquitously expressed monofunctional peroxisomal ³,²-enoyl-CoA isomerase (PECI) of humans and mice. These genes were identified based on their potential to encode proteins homologous to Eci1p, the peroxisomal ³,²-enoyl-CoA isomerase of yeast. Human PECI was shown to have ³,²-enoyl-CoA isomerase activity of 27 units/mg, and the human form of PECI was localized to the peroxisomal matrix.

A previous study of ³,²-enoyl-CoA isomerase in rat liver peroxisomes suggested that this activity was an integral feature of MFE1, the L-specific multifunctional enzyme that also catalyzes the second and third steps of the core -oxidation spiral (14). Our results raise the question as to why there are two proteins with the same apparent activity in the peroxisomal lumen. The occurrence of substrate-specific core enzymes in the peroxisome suggests that two forms of ³,²-enoyl-CoA isomerase might exist to metabolize distinct classes of 3-cis-enoyl-CoAs generated in the organelle. It is interesting to note that whereas MFE1 has been reported to function as a ³,²-enoyl-CoA isomerase, no evidence exists that describes such an activity of the D-specific multifunctional enzyme, MFE2. Furthermore, the multifunctional enzyme of yeast peroxisomes (Fox2p) is also D-specific and lacks ³,²-enoyl-CoA isomerase activity (4). Thus, it may be possible that the active site organization of the L-specific MFE hydratase domain confers ³,²-enoyl-CoA isomerase activity to this enzyme, a hypothesis originally proposed by Palossari et al. (38).

Previously characterized ³,²-enoyl-CoA isomerases have been grouped into the hydratase/isomerase superfamily of acyl-CoA-binding proteins (31, 39) and contain the sequence fingerprint VSXINGX₃AGGXLX₄CDY (31). However, yeast Eci1p and the mammalian PECI proteins lack this motif (7). Furthermore, we find that these monofunctional peroxisomal ³,²-enoyl-CoA isomerases contain a conserved NGPA(V/I)G(I/L)S motif that is absent from the previously described members of the hydratase/isomerase superfamily. When this eight-residue sequence is used to search the nonredundant data base of protein sequences, three additional polypeptides are identified. Among these are Yor180Cp, an S. cerevisiae protein with a role in fatty acid oxidation,² a putative PECI homolog of C. elegans, and a human testis-specific protein of unknown function, CDY (40). These proteins appear to represent a previously unrecognized branch of the hydratase/isomerase superfamily and may lack the active site glutamate residue that has been proposed previously for these enzymes (31, 39). However, the significance of these structural differences remain to be determined.

The identification of mammalian forms of PECI expands our current understanding of the enzymology of peroxisomal fatty acid oxidation. In humans, it is clear that defects in peroxisomal fatty acid metabolism are associated with lethal inherited diseases (41). Although defects in peroxisomal fatty acid oxidation are rare, four complementation groups of these disorders have been described (41). Two complementation groups correspond to defects in peroxisomal acyl-CoA oxidase and the D-specific multifunctional enzyme (41), whereas the other two complementation groups are not defective in any of the known enzymes of fatty acid oxidation.³ Because ³,²-enoyl-CoA isomerase is essential for unsaturated fatty acid metabolism, human PECI should be considered a candidate gene for these disorders.

	ACKNOWLEDGEMENTS

We thank James Morrell for technical assistance with the Northern blotting. We also thank Stephanie Mihalik for generous assistance with the subcellular fractionation experiments.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grants DK45787 and HD10981 (to S. J. G.) and HL30847 (to H. S.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank^TM/EMBL Data Bank with accession number(s) AA188052.

^¶ To whom correspondence should be addressed: Dept. of Biological Chemistry, The Johns Hopkins University School of Medicine, 725 N. Wolfe St., Baltimore, MD 21205. Tel.: 410-955-3085; Fax: 410-955-0125.

² Geisbrecht, B. V., Schulz, K., Nau, K., Geraghty, M. T., Schulz, H., Erdmann, R., and Gould, S. J. (1999) Biochem. Biophys. Res. Commun., in press.

³ R. J. A. Wanders, personal communication.

	ABBREVIATIONS

The abbreviations used are: RACE, rapid amplification of cDNA ends; ORF, open reading frame; MBP, maltose-binding protein.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES