Advertisement

Proteomic Analysis of Rat Liver Peroxisome

PRESENCE OF PEROXISOME-SPECIFIC ISOZYME OF LON PROTEASE*
  • Miki Kikuchi
    Affiliations
    Harima Institute at SPring-8, RIKEN, Mikazuki, Sayo, Hyogo 679-5148
    Search for articles by this author
  • Naoya Hatano
    Affiliations
    Harima Institute at SPring-8, RIKEN, Mikazuki, Sayo, Hyogo 679-5148
    Search for articles by this author
  • Sadaki Yokota
    Affiliations
    Yamanashi Medical University, Tamaho, Yamanashi, 409-3898
    Search for articles by this author
  • Nobuyuki Shimozawa
    Affiliations
    School of Medicine, Gifu University, Gifu 500-8705
    Search for articles by this author
  • Tsuneo Imanaka
    Affiliations
    Faculty of Pharmaceutical Sciences, Toyama Medical and Pharmaceutical University, Toyama 930-0194
    Search for articles by this author
  • Hisaaki Taniguchi
    Correspondence
    To whom correspondence should be addressed: Harima Institute at Spring-8, RIKEN, 1-1-1 Kouto, Mikazuki, Sayo, Hyogo 679-5148, Japan. Tel.: 81-791-58-1825; Fax: 81-791-58-1826
    Affiliations
    Harima Institute at SPring-8, RIKEN, Mikazuki, Sayo, Hyogo 679-5148

    Institute for Enzyme Research, The University of Tokushima, Tokushima 770-8503, Japan
    Search for articles by this author
  • Author Footnotes
    * This work was supported in part by grants-in-aid for scientific research and Special Coordination Funds for promoting Science and Technology (Genome Frontier) from the Ministry of Education, Culture, Sports, Science and Technology, and a grant from the Uehara Memorial Foundation. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Open AccessPublished:October 15, 2003DOI:https://doi.org/10.1074/jbc.M305623200
      Subcellular proteomics, which includes isolation of subcellular components prior to a proteomic analysis, is advantageous not only in characterizing large macro-molecular complexes such as organelles but also in elucidating mechanisms of protein transport and organelle biosynthesis. Because of the high sensitivity achieved by the present proteomics technology, the purity of samples to be analyzed is important for the interpretation of the results obtained. In the present study, peroxisomes isolated from rat liver by usual cell fractionation were further purified by immunoisolation using a specific antibody raised against a peroxisomal membrane protein, PMP70. The isolated peroxisomes were analyzed by SDS-PAGE combined with liquid chromatography/mass spectrometry. Altogether 34 known peroxisomal proteins were identified in addition to several mitochondrial and microsomal proteins. Some of the latter may reside in the peroxisomes as well. Analysis of membrane fractions identified all known peroxins except for Pex7. Two new peroxisomal proteins of unknown function were of high abundance. One is a bi-functional protein consisting of an aminoglycoside phosphotransferase-domain and an acyl-CoA dehydrogenase domain. The other is a newly identified peroxisome-specific isoform of Lon protease, an ATP-dependent protease with chaperone-like activity. The peroxisomal localization of the protein was confirmed by immunological techniques. The peroxisome-type Lon protease, which is distinct from the mitochondrial isoform, may play an important role in the peroxisomal biogenesis.
      Peroxisome is a ubiquitous organelle present in nearly all eukaryotic cells and involved in various oxidative enzymatic reactions such as the α- and β-oxidation of fatty acids.
      The peroxisome website, http://www.peroxisome.org.
      1The peroxisome website, http://www.peroxisome.org.
      Enzymes catalyzing these reactions are well characterized and found in the matrix of the organelle. These matrix proteins are synthesized in the cytoplasm and imported across its membrane. Proteins localized in the peroxisomal membranes are also known, and some of them are involved in the transport of small metabolites and peroxisomal proteins. Biogenesis of the organelle proceeds via a cellular machinery consisting of more than 20 proteins (peroxins) that effect the recognition, targeting and import of peroxisomal proteins (

      Deleted in proof

      ). In human, dysfunction of the machinery results in inherited metabolic disorders and peroxisome biogenesis disorders such as Zellweger syndrome. Some of the genes responsible for the disease have yet to be identified. To characterize the protein composition of the organelle, therefore, is important not only in studying the proteins involved in the functions of the organelle, but also in elucidating the molecules responsible for the organelle's biogenesis.
      Because of the rapid expansion of protein sequence databases, mass spectrometry (MS)
      The abbreviations used are: MS, mass spectrometry; ER, endoplasmic reticulum; LC, liquid chromatography; PMP70, 70-kDa peroxisomal membrane protein; PMP22, 22-kDa peroxisomal membrane protein; CBB, Coomassie Brilliant Blue; AAA, ATPases associated with diverse cellular activities; PTS-1, peroxisomal targeting signal-1; ACDH, acyl-CoA dehydrogenase; ACBP, acyl-CoA-binding protein.
      2The abbreviations used are: MS, mass spectrometry; ER, endoplasmic reticulum; LC, liquid chromatography; PMP70, 70-kDa peroxisomal membrane protein; PMP22, 22-kDa peroxisomal membrane protein; CBB, Coomassie Brilliant Blue; AAA, ATPases associated with diverse cellular activities; PTS-1, peroxisomal targeting signal-1; ACDH, acyl-CoA dehydrogenase; ACBP, acyl-CoA-binding protein.
      -based identification of proteins, which relies on the sequence data base, is becoming more realistic and more reliable (
      • O'Donovan C.
      • Bairoch A.
      • Apweiler R.
      ,
      • Shevchenko A.
      • Jensen O.N.
      • Podtelejnikov A.V.
      • Sagliocco F.
      • Wilm M.
      • Vorm O.
      • Mortensen P.
      • Shevchenko A.
      • Boucherie H.
      • Mann M.
      ,
      • Pandey A.
      • Mann M.
      ). One important field of application of this emerging technique is to analyze proteins in whole cells and tissues, and thousands of proteins expressed can be detected and identified (
      • Washburn M.P.
      • Wolters D.
      • Yates J.R.
      ,
      • Wu C.C.
      • Yates J.R.
      ). The same technique can be used to analyze subcellular fractions ranging from organelles to large macromolecular complexes such as ribosomes. Analyzing subcellular fractions reduces the complexity of the samples to be analyzed and should yield more information on the minor protein components that would otherwise escape detection (
      • Jung E.
      • Heller M.
      • Sanchez J.C.
      • Hochstrasser D.F.
      ,
      • Bell A.W.
      • Ward M.A.
      • Blackstock W.P.
      • Freeman H.N.M.
      • Choudhary J.S.
      • Lewis A.P.
      • Chotai D.
      • Fazel A.
      • Gushue J.N.
      • Paiement J.
      • Palcy S.
      • Chevet E.
      • Lafrenière-Roula M.
      • Solari R.
      • Thomas D.Y.
      • Rowley A.
      • Bergeron J.J.M.
      ,
      • Andersen J.S.
      • Lyon C.E.
      • Fox A.H.
      • Leung A.K.L.
      • Lam Y.W.
      • Steen H.
      • Mann M.
      • Lamond A.I.
      ,
      • Hanson B.J.
      • Schulenberg B.
      • Patton W.F.
      • Capaldi R.A.
      ). A classical way of cell fractionation, i.e. differential centrifugation after homogenization, is clearly not adequate for such analyses, since the analytical methods employed may detect minor contaminants. Methods are necessary that allow purification of such macromolecular complexes without losing the structural or functional integrity of the complex. Peroxisomes are usually purified by combining the classical differential centrifugation and density gradient centrifugation. These fractions still contain other contaminating organelles such as mitochondria.
      In the present study, we combined the Nycodenz density gradient centrifugation with immunoisolation using anti-PMP70 (70-kDa peroxisomal membrane protein) antibody. The fractions obtained were practically free from contaminating mitochondria. MS-based proteomic analyses (
      • Gygi S.P.
      • Rist B.
      • Gerber S.A.
      • Turecek F.
      • Gelb M.H.
      • Aebersold R.
      ) gave an image of the “proteome” of the liver peroxisomes with the composition and the relative abundance of the protein components. In addition to the known peroxisomal proteins, several endoplasmic reticulum (ER) and mitochondrial proteins, which seem to exist in the organelle, and a few new proteins of unknown function were identified. One of them is a peroxisome-specific isozyme of Lon protease, which acts as molecular chaperone. The protein may play an important role in the biogenesis of the peroxisome.

      EXPERIMENTAL PROCEDURES

      Materials—Polyclonal antibody against rat PMP70 was raised using a synthetic peptide corresponding to the C-terminal 15 amino acids (
      • Imanaka T.
      • Shiina Y.
      • Takano T.
      • Hashimoto T.
      • Osumi T.
      ). Specific antibody was purified from rabbit antiserum by affinity chromatography using an antigen-bound column (SulfoLink kit, Pierce). Goat anti-rabbit IgG (Fc fragment-specific) was obtained from the same source. Nycodenz was obtained from Daiichi Pure Chemicals Co., Ltd. (Tokyo, Japan), while Dynabeads M-500 Subcellular were from DYNAL (Oslo, Norway). Protease inhibitor mixture tablets were purchased from Roche Diagnostics (Mannheim, Germany). Other chemicals and biochemicals were from Wako Pure Chemical Industries (Osaka, Japan).
      Preparation of Rat Liver Peroxisomes—The peroxisomal fraction from rat liver was prepared by conventional differential centrifugation in sucrose followed by isopycnic centrifugation in Nycodenz (
      • Imanaka T.
      • Lazarow P.B.
      • Takano T.
      ) with some modifications. About 0.6 ml of a light mitochondrial fraction-containing peroxisomes was layered onto 10 ml of linear Nycodenz gradient (density span from 1.15 to 1.25 g/ml) in a P40ST rotor (Hitachi, Tokyo, Japan). The gradient rested on a 0.9-ml cushion of 1.30 g/ml Nycodenz. All Nycodenz solutions contained 0.25 m sucrose, 1 mm EDTA, 0.1% (v/v) ethanol in 5 mm Hepes-KOH buffer (pH 7.4). Centrifugation was carried out at 77,000 × g for 180 min at 4 °C. Fractions of ∼1.0 ml were collected, and the peroxisome-containing fraction was determined by the distribution of catalase activity (
      • Imanaka T.
      • Lazarow P.B.
      • Takano T.
      ).
      Immunopurification of Peroxisomes—Anti-PMP70 specific antibody was bound to magnetic beads (1×108), using an anti-rabbit Fc fragment antibody as a linker antibody according to the manufacturer's instruction. The peroxisome-enriched fractions from the Nycodenz density gradient (about 500 μg of protein) were incubated with the magnetic beads suspended in phosphate-buffered saline containing 2 mm EDTA at 4 °C for 6 h. After incubation, bead complexes were collected by placing the reaction tube in a magnet stand. The complex was washed three times with phosphate-buffered saline containing 2 mm EDTA. Peroxisomes were eluted by boiling in 200 μl of 2× SDS sample buffer for SDS-PAGE (80 mm Tris, 4% SDS, 10% sucrose, 0.002% bromphenol blue, and 20 mm dithiothreitol, pH 6.8). The supernatant separated by placing in a magnet stand was subjected to SDS-PAGE. Peroxisomal membranes were obtained by incubating the bead-bound peroxisomes with 0.1 m Na2CO3 for 30 min at 4 °C. Bead complexes were washed twice with 0.25 m sucrose, 2 mm EDTA, and 0.1% ethanol, and eluted with the SDS sample buffer as above.
      In-gel Digestion with Trypsin—The purified peroxisomal fraction was subjected to SDS-PAGE and stained with MS-compatible silver staining (1% of the sample obtained) (
      • Shevchenko A.
      • Wilm M.
      • Vorm O.
      • Mann M.
      ) or with Coomassie Brilliant Blue (CBB) (10% of the sample). For the proteomic analysis a single lane from CBB-stained gel was divided equally into 64 slices of 2-mm width. Each gel slice was cut into small pieces and put into the wells of a 96-well PCR plate (BM Bio, Tokyo, Japan). The gel pieces were destained by rinsing in 50% acetonitrile containing 25 mm NH4HCO3. Disulfide bonds of cysteines were reduced by incubating with 10 mm dithiothreitol in 25 mm NH4HCO3 at 56 °C for 1 h and alkylated with 55 mm iodoacetamide in 25 mm NH4HCO3 at room temperature for 45 min in the dark. Gel pieces were washed with 50% acetonitrile and dried in a vacuum concentrator (SpeedVac, Thermo Savant, NY). Gel pieces were re-hydrated with 15 μl of trypsin solution (10 ng/μl in 50 mm NH4HCO3 containing 5 mm CaCl2). In-gel digestion was performed at 37 °C overnight. Resulting peptides were extracted with 50% acetonitrile containing 5% trifluoroacetic acid. Extracts were dried in Speed-Vac and re-dissolved in 0.1% (v/v) formic acid. The peptide mixtures thus obtained were subjected to LC/MS and data-dependent tandem mass (LC/MS/MS) analyses.
      Mass Spectrometric Analysis—LC/MS/MS analysis was carried out in a Q-Tof-type hybrid mass spectrometer (Micromass, Manchester, UK) interfaced on-line with a capillary HPLC (Waters-Micromass modular CapLC, Micromass) (
      • Taniguchi H.
      • Suzuki M.
      • Manenti S.
      • Titani K.
      ,
      • Konishi H.
      • Yamauchi E.
      • Taniguchi H.
      • Yamamoto T.
      • Matsuzaki H.
      • Takemura Y.
      • Ohmae K.
      • Kikkawa U.
      • Nishizuka Y.
      ). A linear gradient of acetonitrile produced was split at a 1:10 ratio, and the slow flow at around 100 nl/min was injected into a nanoLC column (PepMap C18, 75 μm × 150 mm, LC Packings, Amsterdam, Holland). The column eluate was directly injected into a self-constructed nano-spray ion source with a tapered stainless capillary. The LC/MS/MS analysis was carried out in a data-dependent mode; peptide peaks above a certain threshold were subjected automatically to the collision-induced dissociation. Peak lists obtained from the MS/MS spectra were used to identify proteins using Mascot search engine (Matrixscience, London, UK).
      Immunological Analyses—Antibody against the C-terminal 16 amino acids of RIKEN cDNA 1300002A08 clone, GGFPVKTRPGLIDSKL, was raised in rabbits (MBL, Nagoya, Japan). The distribution of the protein among subcellular fractions was determined by immunoblotting. Each 50 μg of protein of differential centrifugation fractions; nuclear (N), cytoplasmic (S), and light mitochondrial (ML) fractions, were subjected to SDS-PAGE and transferred onto a nitrocellulose membrane. The blot membrane was incubated with the antibody and visualized by ECL Plus (Amersham Biosciences).
      Immunoelectron microscopy of rat liver cells with anti-peroxisomal Lon protease antibody was performed as described previously (
      • Yoshihara T.
      • Hamamoto T.
      • Munakata R.
      • Tajiri R.
      • Ohsumi M.
      • Yokota S.
      ). Immunofluorescent microscopy of human fibroblasts was previously described (
      • Shimozawa N.
      • Tsukamoto T.
      • Suzuki Y.
      • Orii T.
      • Shirayoshi Y.
      • Mori T.
      • Fujiki Y.
      ). Human fibroblasts used in this study were from a normal, peroxisomal D-bifunctional protein deficiency patient (
      • Suzuki Y.
      • Jiang L.L.
      • Souri M.
      • Miyazawa S.
      • Fukuda S.
      • Zhang Z.
      • Une M.
      • Shimozawa N.
      • Kondo N.
      • Orii T.
      • Hashimoto T.
      ) and a Zellweger syndrome patient (
      • Shimozawa N.
      • Nagase T.
      • Takemoto Y.
      • Ohura T.
      • Suzuki Y.
      • Kondo N.
      ).

      RESULTS

      Purification of Peroxisomes by Antibody-conjugated Magnetic Beads—Peroxisomal fractions obtained from the conventional differential centrifugation were further enriched by Nycodenz gradient centrifugation. Fractions containing peroxisomes were determined based on the enrichment of a marker enzyme, catalase. Peroxisome was further purified by immunoisolation with magnetic beads coated with monospecific antibodies against PMP70 as described under “Experimental Procedures.” When two samples obtained before and after the immunopurification were compared, a few changes were noticed (Fig. 1A). Several bands disappeared during the purification, while some bands were clearly concentrated. Some of the protein bands that disappeared were cut out and subjected to the in-gel digestion/MS analysis. One protein was found of mitochondrial origin, carboamyl phosphate synthase (Fig. 1A). A few proteins of small molecular weight were from the cytoplasm (data not shown). It should be noted that major mitochondrial and ER proteins such as cytochrome c oxidase and cytochromes P-450 were not detected. Therefore, contaminating mitochondrial and cytoplasmic proteins seem to be effectively removed by the immunopurification. On the other hand, one of the typical peroxisomal membrane proteins, PMP22, was clearly concentrated during the purification. Interestingly, peroxisomal matrix proteins were not uniformly concentrated. The amount of the most abundant matrix protein, catalase, seemed to decrease, while the band corresponding to another major matrix protein, uricase, clearly increased in its intensity (Fig. 1A). These results suggest that there seems to be some leakage of soluble matrix proteins during the purification process, although no drastic change of the composition was noted.
      Figure thumbnail gr1
      Fig. 1Silver-stained gels of peroxisomal fractions.A, immunopurification of peroxisome. Peroxisomal fractions before and after immunopurification were subjected to SDS-PAGE and stained with silver staining. Lane 1, rat liver peroxisomal fraction (before purification); lane 2, purified peroxisome. Carb, carboamylphosphate synthase; RIK, RIKEN cDNA 1300002A08; Cat, catalase; P70, PMP70; H, IgG heavy chain; L, IgG light chain; Uri, uricase; P22, PMP22. B, summary of identified proteins. Major proteins corresponding to prominent bands are listed on the left side of the silver-stained gel, while less abundant proteins are shown on the right side.
      Proteins Identified in the Purified Peroxisomes—Each lane in the SDS-PAGE obtained from the purified peroxisomal fractions was evenly cut into 64 slices, regardless of the presence of stained bands, and subjected to the in-gel digestion. The resulting peptide mixtures were analyzed by LC-MS/MS as described under “Experimental Procedures.” Mascot search engine was used to identify proteins in the MS/MS ion search mode. Up to 10 proteins were identified in a single gel slice. Proteins thus identified are summarized in Table I.
      Table IList of identified proteins
      Protein nameNCBI accession numberM.W.Top score
      Proteins showing top scores above 100 (score for mitochondrial carboamyl phosphate synthase) were listed. Proteins known to be localized in peroxisomes are listed even if they show scores less than 100. Top score refers to the highest score returned from the Mascot search engine for the individual protein in each gel slice. Scores from the neighboring gel slices were added to give total score for each protein
      Total scoreNormalized score
      The Mascot score was normalized for molecular weight by dividing with the molecular weight and multiplying with 50,000, a typical molecular weight of proteins
      TaxonomyLocationC-term.
      PEXs (proteins involved in peroxisome biogenesis)
      Homolog to peroxisome biogenesis factor 1 (Pex1)gi 1285648553,726188610567.70mousePx
      Peroxisome biogenesis factor 3 (Pex3)gi 1378619442,582145315369.87ratPx
      Peroxisomal membrane anchor protein (Pex14)gi 412696240,944131301367.58ratPx
      Peroxisomal membrane protein 3, 35 kDa (Pex2)gi 839412135,427545476.21ratPx
      Peroxin 13p protein (Pex13)gi 1296377144,588404044.86mousePx
      PMPs (peroxisomal membrane proteins)
      Peroxisomal membrane protein 2, 22 kDa (PMP22)gi 1392879622,5621865851296.43ratPx
      ATP-binding cassette, subfamily D, member 3 (PMP70)gi 698145075,780312908599.10ratPx
      ATP-binding cassette, subfamily D, member 1 (ALDP)gi 667149782,320686841.30mousePx
      Lipid metabolism
      α-Oxidation
         Hydroxyacid oxidase 3 (medium-chain)gi 1409177539,63341611431441.98ratPxSRL
         Hydroxyacid oxidase 1, liver; glycolate oxidase 1gi 675415641,26032510161231.22mousePxSKI
         Phytanoyl-CoA α-hydroxylase (phytanic acid oxidase)gi 1264348239,077145346442.72ratPxINL
         2-Hydroxyphytanoyl-CoA lyasegi 691241864,42564209162.20humanPxSNM
         Peroxisomal long chain acyl-CoA thioesterase lbgi 647848546,694878793.16mousePxCRL
         Peroxisomal acyl-CoA thioester hydrolase 1gi 1454800736,146505069.16mousePxSKL
      β-Oxidation
         Catalasegi 697860760,06285433662802.10ratPxANL
         Acyl-CoA oxidase (EC 1.3.3.6) chain A, peroxisomal splice form IIgi 6606375,04244727931860.96ratPxSKL
         Peroxisomal multifunctional enzyme type IIgi 1324230379,87723420881307.01ratPxAKL
         Peroxisomal bifunctional enzymegi 11912079,17972018991199.18ratPxSKL
         Long-chain acyl-CoA synthetase 2gi 12601179,15544918301155.96ratPxIKI
         Trihydroxycoprostanoyl-CoA oxidasegi 168474777,54840717241111.57ratPxHKN
         Pristanoyl-CoA oxidasegi 249820879,08042017431102.05ratPxSQL
         Sterol carrier protein-2 (SCP-2)gi 20687229,25298431736.70ratPxAKL
         Acetyl-CoA acyltransferase, 3-oxo acyl-CoA thiolase A, peroxisomalgi 697842945,269164508561.09ratPxPGN
      Oxidation of polyunsaturated fatty acid
         Putative peroxisomal 2,4-dienoyl-CoA reductasegi 410526931,614233529836.65ratPxAKL
         Putative short-chain dehydrogenase/reductasegi 505220432,698255430657.53ratPxARL
         Isocitrate dehydrogenase I, solublegi 1392869047,047777781.83ratPxAKL
         Peroxisomal δ3, δ2-enoyl-Coenzyme A isomerasegi 675502639,739595974.23mousePxPKL
      Bile acid synthesis
         Bile acid-CoA dehydrogenase: amino acid n-acyltransferasegi 839296246,80952918932022.05ratPxSQL
      Glycerolipid synthesis
         Alkyl-dihydroxyacetonephosphate synthase precursorgi 1200220372,283505034.59ratPxNLL
         Dihydroxyacetone phosphate acyltransferasegi 1428541277,710525233.46ratPxAKL
      Amino acid metabolism
      Serine-pyruvate aminotransferase; alanine-glyoxylate aminotransferasegi 1347009643,771115206235.32ratPx, Mt MatrixARL
      NKL
      Peroxisomal sarcosine oxidasegi 667950744,454454550.61mousePxAHL
      Purine metabolism
      Uricase (urate oxidase)gi 13710835,14032014372044.68ratPxSRL
      Antioxidant enzyme B166gi 610372622,50760156346.56ratPx, MtSQL
      2-Arylpropionyl-CoA epimerasegi 155237440,0259696119.93ratPxANL
      Mitochondrial proteins
      Glutathione S-transferase, mitochondrial (GST13—13)gi 304168025,590129225439.62ratMt MatrixARL
      Carboamyl-phosphate synthetase 1gi 8393186165,67310010030.18ratMtKAA
      ER proteins
      Cytochrome b5gi 1156004615,346176262853.64ratERAED
      Aldehyde dehydrogenase 3A2gi 1392902854,503320648594.46ratERDQL
      Microsomal glutathione S-transferase 1 (microsomal GST-1)gi 12174117,517108108308.27ratERLYL
      Cytochrome-b5 reductase (EC 1.6.2.2), microsomal formgi 107044434,347133178259.12ratER MemFTT
      Protein-disulfide isomerase A6 precursor, microsomalgi 250120647,590133133139.74ratER LumDEL
      UDP-glucuronosyltransferase 1—1 precursor, microsomalgi 250147360,422163163134.88ratERKTH
      Membrane proteins
      RAB2, member RAS oncogene familygi 1392900623,692113173365.10ratMemGCC
      Vesicle-associated membrane protein, associated protein A (VAP-A)gi 1392887027,434107179326.24ratMemFIS
      Vesicle-associated membrane protein, associated protein B and C (VAP-B)gi 979028327,157119119219.10mouseMemIAL
      Cytoplasmic proteins
      β-Actingi 7162042,066177177210.38ratKCF
      Putative proteins
      RIKEN cDNA 1300002A08gi 1338529894,9806041234649.61mouseSKL
      RIKEN cDNA 2010003O14 genegi 1338499817,055175206603.93mouseSKS
      Similar to expressed sequence AI987948gi 2828002387,341360723413.89mouseARM
      Homolog to endozepine-related protein precursorgi 1283661656,615128335295.86mouseKLN
      Tumor-related proteingi 1457252632,297144190294.14mouseSKV
      a Proteins showing top scores above 100 (score for mitochondrial carboamyl phosphate synthase) were listed. Proteins known to be localized in peroxisomes are listed even if they show scores less than 100. Top score refers to the highest score returned from the Mascot search engine for the individual protein in each gel slice. Scores from the neighboring gel slices were added to give total score for each protein
      b The Mascot score was normalized for molecular weight by dividing with the molecular weight and multiplying with 50,000, a typical molecular weight of proteins
      We could identify 34 known peroxisomal proteins, and five proteins were those of unknown function. Furthermore, two mitochondrial proteins and six ER proteins were identified in addition to the four proteins from other subcellular fractions. Of the known peroxisomal proteins, seven proteins were peroxisomal membrane proteins. In addition to the two major matrix proteins, catalase, and uricase, as mentioned above, 25 proteins of the peroxisomal matrix were found in the list. Most of them are enzymes involved in lipid metabolism. Five proteins were peroxins (Pex1, -2, -3, -13, -14) involved in the various steps of the peroxisome biogenesis. Several mitochondrial and ER proteins identified in the present study can be considered to be authentic peroxisomal proteins rather than contaminating proteins from other organelles. Reasoning for this conclusion will be discussed below in detail.
      Since we could identify only 5 of 13 peroxins, and three peroxisomal membrane proteins, we have isolated peroxisomal membranes by treating the beads-bound peroxisome with mild alkaline treatment (Na2CO3). The washed membranes were eluted from the magnetic beads, and subjected to the same SDS gel electrophoresis/mass spectrometry-based analysis. The peroxins and peroxisomal membrane proteins thus identified are listed in Table II. We could identify 12 peroxins, and only two peroxins, Pex7 and Pex19, are missing in the list. This is not surprising, since these two proteins are predominantly cytoplasmic. It is interesting to compare the relative amounts of these proteins found in the whole and membrane fractions of the organelle. Pex1 is the most abundant peroxins in the whole fraction, but the amount of the protein relative to other peroxins decreased drastically. Pex1 has been shown to be cytoplasmic when examined by immunofluorescence microscopy. Our results suggest that part of the protein is associated with the peroxisome and can be washed away with alkaline treatment. On the other hand, Pex5, which have been predominantly cytoplasmic with a small amount associated with the peroxisome, were found in the washed membrane fractions. The association of the protein, therefore, is stronger to that of Pex1 proteins. All the peroxins that are membrane-bound (Pex2, Pex3, Pex6, Pex10, Pex11, Pex12, Pex13, Pex14, and Pex16) were concentrated in the membrane fraction, demonstrating the efficiency of the isolation method. Similarly, other membrane proteins of the organelle, which were not detected in the whole peroxisomal fraction, were found in the fraction. Judging from the normalized score, the relative abundance of ALDP and ALDPR seems very low compared with that of the abundant peroxisomal membrane proteins such as PMP70 and PMP22.
      Table IIList of proteins identified in peroxisomal membranes
      Protein nameNCBI accession numberM.W.Top scoreTotal scoreNormalized scoreTaxonomyLocationComment
      PEXs (proteins involved in peroxisome biogenesis)
      Peroxisomal membrane protein Pmp26p (Peroxin-11)gi 1675824627,9053809711740ratPxPex11
      Similar to pex5 proteingi:3485842969,51585116391179ratPxPex5
      Peroxisomal biogenesis factor 3gi 1378619442,1835029781159ratPxPex3
      Similar to peroxisomal biogenesis factor 11bgi 2767368328,6242046451127ratPxPex11b
      Peroxisomal membrane anchor proteingi 2545337440,9444409071108ratPxPex14
      Similar to Peroxisome assembly protein 10 (Peroxin-10)gi 2766338235,122407598851ratPxPex10
      Peroxisome biogenesis factor 16; peroxin-16gi 2155306938,625235626810mousePxPex16
      Peroxisomal biogenesis factor 12gi 1675880240,655274469577ratPxPex12
      Peroxisome assembly factor-1 (PAF-1) (Peroxin-2)gi 12955534,745186345496ratPxPex2
      Similar to Peroxisomal membrane protein PEX13 (Peroxin-13)gi 2769815745,097168350388ratPxPex13
      Peroxisome assembly factor-2gi 2911135104,250327608292ratPxPex6
      Similar to peroxisome biogenesis factor 1gi 2770842019,010110110289ratPxPex1
      PMPs (peroxisomal membrane proteins)
      ATP-binding cassette, subfamily D (ALD), member 3; Peroxisomal membrane protein 1gi 698145075,26791565664362ratPx, MemPMP70
      Fatty acid Coenzyme A ligase, long chain 2; Acyl CoA synthetase, long chaingi 2574273978,128132536162314ratPx, MemLACS
      Peroxisomal membrane protein 2; peroxisomal membrane protein 2, 22 kDagi 1392879622,5623348701928ratPx, MemPMP22
      Similar to solute carrier family 25 (mitochondrial carrier; Peroxisomal membrane protein, 34 kDa), member 17gi 2766294634,319558558813ratPx, MemPMP34
      Peroxisomal membrane protein 4gi 2644916424,121122195404ratPx, MemPMP24
      Solute carrier family 27 (fatty acid transporter), member 32gi 1392903470,648417417295ratPx, MemVLACS
      ATP-binding cassette, subfamily D (ALD), member 2gi 1537532483,315234234140ratPx, MemALDRP
      ATP-binding cassette, subfamily D, member 1; X-linked Adrenoleukodystrophy (ALD) gene homologgi 667149781,807127180110mousePx, MemALDP
      Proteome of Rat Liver Peroxisome—The positions of protein in SDS-PAGE are shown together with the image of a silverstained gel (Fig. 1B). On the left side of the gel image, abundant proteins whose stained bands can be easily recognized are listed. Less abundant proteins whose bands are either faint or invisible are listed on the right side of the gel. In some cases, more than one protein of similar amounts were identified in a single gel slice, making the relationship between the stained bands and proteins obscure. Many of the abundant proteins were found not only in a single gel slice but also in the neighboring slices. Furthermore, proteins were detected in positions where the apparent molecular mass differs appreciably from the theoretical mass calculated from the known sequence. Major proteins like catalase were, in fact, detected throughout many gel slices well above or below the theoretical mass range. They originated either from complexes that remained undissociated even in the presence of SDS or from proteolytic fragments. In these cases, only the gel slices in which most of the proteins were detected were marked.
      For the analysis of proteomes, regardless of whether it is a proteome from a cell or that of an organelle, it is important to know the abundance of each protein in that specific sample. Although this is not an easy task, we noticed that the probability score produced by the Mascot search engine is correlated with the amount of protein in that specific sample. The score presented for each protein identified is a sum of peptide scores detected in the LC/MS analysis. It should be noted that the more abundant a protein is, the more peptides that originated from the protein are likely to be observed. In other words, when one uses a large amount of protein for the analysis, the sequence coverage becomes high, and the number of peptides detected is large. On the other hand, when only a small amount of protein, such as the amount just above the detection limit, is subjected to analysis, very few peptides are observed. Furthermore, the probability score for each peptide is a function of its amount; more sequence-specific fragment ions of high S/N are observed from peptides from abundant proteins.
      To elucidate the relationship between the Mascot score and the relative abundance, we compared the Mascot score with the band intensity of the silver-stained gel calculated by image analysis (NIH Image, rsb.info.nih.gov/nih-image/). The Mascot score was normalized for molecular weight by dividing by the molecular weight and multiplying with 50,000, a typical molecular weight of proteins. Scores of abundant proteins that are found in successive slices were added. Only proteins whose bands are unambiguously identified were included in the comparison (Fig. 2). As is clear from the figure, there is a good correlation between the two parameters. Other less abundant proteins, which are not included in the figure and show low Mascot scores and band intensities, should be localized near the origin. It should be noted that although the silver staining is a clear indication of the protein amount, the linearity is very limited, and the staining varies appreciably depending on the protein. Therefore, both staining intensities and normalized Mascot score can be considered as semi-quantitative indicators of proteins. Judging from the correlation, whether a protein is highly or less abundant from the normalized Mascot score can be determined.
      Figure thumbnail gr2
      Fig. 2Correlation between Mascot scores and band intensities of silver staining. The vertical axis shows protein band intensities measured by NIH image, while the horizontal axis shows normalized Mascot scores. The correlation coefficient was 0.8028.
      We have included these scores in the list of the peroxisomal proteins (Table I). The proteins are listed in decreasing order of the score. By comparing the list and the gel image, a rough image of the proteome of liver peroxisomes can be obtained. In addition to the two major matrix proteins, catalase, and uricase, several enzymes in the lipid metabolism are of high abundance. Some of the membrane proteins such as PMP22 and PMP70 are abundant proteins, comparable to the average matrix proteins. On the other hand, the peroxins are mostly minor components in the organelle.
      In this context, proteins known to be localized in other subcellular fractions should be reexamined. Some of them have high scores compared with the abundant authentic peroxisomal proteins, suggesting that these proteins are intrinsic peroxisomal proteins rather than contaminating ones. Of these, aldehyde dehydrogenase and cytochrome b5 belong to the abundant proteins in the organelle. It may not be surprising to find such enzymes in peroxisomes. Proteins involved in the vesicle trafficking such as rab family proteins were also identified in the preparation.
      Identification of a Peroxisome-specific Isozyme of Lon Protease—One of the five proteins of unknown function identified has homology to Lon protease and has an AAA (ATPases associated with diverse cellular activities) domain (Fig. 3A). A close examination of sequences of Lon proteases of various origins indicated that at least two isoforms exist in mammalian cells. One isoform has been characterized as mitochondrial Lon protease, which is homologous to plant Lon2 sequences (
      • Sarria R.
      • Lyznik A.
      • Vallejos C.E.
      • Mackenzie S.A.
      ,
      • Adam Z.
      • Adamska I.
      • Nakabayashi K.
      • Ostersetzer O.
      • Haussuhl K.
      • Manuell A.
      • Zheng B.
      • Vallon O.
      • Rodermel S.R.
      • Shinozaki K.
      • Clarke A.K.
      ). The present isoform is likely an orthologue of the plant Lon1. Both types share similar structural characteristics with an AAA domain and the C-terminal region containing a protease active site. The N-terminal parts are more diverse compared with the C-terminal half. It should be noted that the putative peroxisomal isozymes (
      • Adam Z.
      • Adamska I.
      • Nakabayashi K.
      • Ostersetzer O.
      • Haussuhl K.
      • Manuell A.
      • Zheng B.
      • Vallon O.
      • Rodermel S.R.
      • Shinozaki K.
      • Clarke A.K.
      ,
      • Barakat S.
      • Pearce D.A.
      • Sherman F.
      • Rapp W.D.
      ) end with Ser-Lys-Leu, a typical peroxisomal targeting signal-1 (PTS-1). Interestingly, the bacterial protease shows structural characteristics that are intermediate between the two mammalian and plant isoforms. This suggests that the bacterial protein is ancestral to the two eukaryote isoforms.
      Figure thumbnail gr3
      Fig. 3Comparison of amino acid sequences.Acc. number, accession number in NCBI protein data base (www.ncbi.nlm.nih.gov/). A, eukaryotic Lon proteases are compared with a bacterial protease. Shown are mouse putative RIKEN cDNA 1300002A08, human putative, Arabidopsis thaliana Lon1, ATP-dependent protease of Myxococcus xanthus, mitochondrial Lon proteases of rat and human, A. thaliana Lon2 and Lon3. Bold boxes are regions showing homologies among all 8 proteins. Closed areas are similar in three peroxisomal isozymes and bacterial protease, the shaded areas are found in four mitochondrial Lon proteases, the hatched areas are common in bacterial proteases and mitochondrial proteases. Lon and AAA indicate the Lon domain and AAA domain, respectively. The asterisk indicates a serine residue, which is considered as an active site of Lon proteases. B, proteins containing ACDH domain. Shown are mouse putative (similar to AI987948), human putative, (FLJ12592), and medium-chain ACDH. Bold boxes are homologous regions among the three proteins, hatched areas are similar between the two putative proteins. APH and ACDH are conserved domains of known functions.
      To confirm the peroxisomal localization of the newly-identified Lon isozyme, we have studied the cellular localization of the protein in detail. For that purpose we have raised a specific antibody against a synthetic peptide based on the mouse C-terminal sequence. It differs from the human sequence at only two positions, and crossreacts with the human protein. The localization of the protease was studied by subcellular fractionation, immunocytochemistry, and immunoelectron microscopy (Fig. 4). In the immunoblotting, single band corresponding to expected size (about 95-kDa) was detected only in the ML fraction (Fig. 4A). When human fibroblasts were stained with the antibody, the immunofluorescence was found in a punctate pattern typical for peroxisome-associated proteins (Fig. 4B). A similar punctate pattern was observed when cells were stained with anti-catalase antibody (data not shown). These results established that the hypothetical proteins identified in the human and mouse genomic or cDNA sequences are the peroxisomal-specific isozyme of Lon protease, which is distinct from the known mitochondrial isozyme. The same antibody was used for immunoelectron microscopy. As shown in Fig. 4C, immunogold particles were associated mainly with the peroxisome core, an electron dense structure inside of the peroxisomal membranes. For a more quantitative analysis, labeling densities of several subcellular components were calculated. The particle density in the cytoplasmic area was 1.66 ± 0.44 particles/μm2, while that in the peroxisomes was 28.57 ± 7.67 particles/μm2. Within the peroxisome area, the labeling density in the peroxisomal core (123.90 ± 24.79 particles/μm2) was much higher than that in the matrix (6.84 ± 1.14 particles/μm2). This indicates that the protein has specific localization in the peroxisomal core. Judging from the band intensity and the Mascot score, the present peroxisomal isozyme is an abundant peroxisomal proteins.
      Figure thumbnail gr4
      Fig. 4Immunological analyses of localization of peroxisomal Lon protease.A, immunoblotting of rat liver subcellular fractions with anti-peroxisomal Lon protease antibody. Arrowhead indicates peroxisomal Lon protease. A minor band may be a truncated protein. B, immunofluorescence microscopy of human fibroblasts. Fibroblasts from control (a), patient of D-bifinctional protein deficiency (b), and Zellweger syndrome patient (c) were immunostained with the antibody. No punctate pattern was observed in fibroblast from Zellweger syndrome patient. C, immunocytochemistry of rat liver. Section of rat liver was incubated with the antibody. Immunogold particles showing peroxisomal Lon protease antigenic sites are observed in peroxisomes (P). Bar, 0.5 μm.
      Presence of a Novel Bi-functional Enzyme Containing Acyl-CoA Dehydrogenase Domain—Another protein of unknown function identified in the present study contains a typical acyl-CoA dehydrogenase (ACDH) domain (Fig. 3B). Two sequences are reported for the protein; human FLJ12592 and a longer sequence annotated as similar to AI987948. The N-terminal half of the longer molecule shows homology to the aminoglycoside phosphotransferase (APH) family. In the present MS-based analysis, peptides derived from both N- and C-terminal regions of the longer sequence were identified. Therefore, the shorter sequence seems to be a truncated form. These proteins end with three amino acids, Ala-Arg-Met or Ala-Lys-Ile, a signal reported to work as PTS-1 (
      • Mohan K.V.K.
      • Som I.
      • Atreya C.D.
      ). These results suggest the presence of a novel bi-functional enzyme with an ACDH domain in the peroxisomes. This is the first demonstration that an ACDH is present not only in the mitochondria but also in the peroxisomes. Only acyl-CoA oxidase activity has previously been thought to be associated with the organelle (
      • Lazarow P.B.
      • de Duve C.
      ).
      Other Proteins of Unknown Function—Three other proteins are listed under “putative proteins” in Table I. “Homolog to endozepine-related protein precursor” (accession number in NCBI, gi 12836616) has a high similarity to bovine endozepine-related protein. It has an acyl-CoA binding protein (ACBP) domain in the N-terminal that has been characterized as a cytoplasmic protein involved in the acyl-CoA transport. In the present analysis, this putative protein was detected in gel slices of low molecular weight under 20 kDa, and only peptides derived from the C-terminal region were detected. Therefore, only the C-terminal domain, which lacks the ACBP domain, seems to be localized in the peroxisome.
      RIKEN cDNA 2010003O14 (gi 13384998) is homologous to CGI-135. CGI proteins are proteins of unknown function, but are evolutionarily conserved between human and Caenorhabditis elegans. CGI-135 shows similarity to Fis1p, which is required for Dnm1p GTPase-mediated mitochondrial fission (
      • Mozdy A.D.
      • McCaffery J.M.
      • Shaw J.M.
      ). Tumor-related protein (gi 14572526) is homologous to CGI-111, which shows some similarity to MAR1, a mitochondrial-associated ribonuclease (
      • Alfonzo J.D.
      • Thiemann O.H.
      • Simpson L.
      ). In Leishmania tarentolae in which the enzyme was characterized, it is an endoribonuclease and the 18 amino acid sequence of N-terminal shows characteristics of an uncleaved mitochondrial sequence. However, the C-terminal sequence of the tumor-related protein, found in this study, is Ser-Lys-Val and may function as a PTS-1. Because of the low abundance, the peroxisomal localization of these proteins should be determined by other techniques before conducting functional analysis.

      DISCUSSION

      Organelle proteomics, which is becoming a popular way to analyze cellular fractions, is advantageous in two respects; one is the decreased complexity of the samples to be analyzed, and the other is the information on the subcellular localization of protein components provided. Unfortunately, such analyses have been conducted often without much consideration of the methods and purity of samples subjected to analysis (
      • Jung E.
      • Heller M.
      • Sanchez J.C.
      • Hochstrasser D.F.
      ,
      • Bell A.W.
      • Ward M.A.
      • Blackstock W.P.
      • Freeman H.N.M.
      • Choudhary J.S.
      • Lewis A.P.
      • Chotai D.
      • Fazel A.
      • Gushue J.N.
      • Paiement J.
      • Palcy S.
      • Chevet E.
      • Lafrenière-Roula M.
      • Solari R.
      • Thomas D.Y.
      • Rowley A.
      • Bergeron J.J.M.
      ,
      • Andersen J.S.
      • Lyon C.E.
      • Fox A.H.
      • Leung A.K.L.
      • Lam Y.W.
      • Steen H.
      • Mann M.
      • Lamond A.I.
      ). On the other hand, the analytical methods employed are becoming more and more sensitive, enabling detection of less abundant components. In the present study, peroxisomal fractions obtained by differential centrifugation were further purified by density gradient and by immunopurification. Judging from the behavior of several mitochondrial and cytoplasmic proteins, the final preparation seems to be practically free from the contamination of the two subcellular fractions. Interestingly, several ER proteins, especially those present in the lumen of ER, have been detected. Since major ER proteins such as cytochrome P-450 isoforms were not detected, it is possible that some of these proteins are also localized in the peroxisomes. One example is aldehyde dehydrogenase, which is an endoplasmic reticulum protein. Judging from the abundance of the protein, this protein may well be an intrinsic peroxisomal protein involved in the peroxisomal oxidative pathways. An immunoreactive protein has, in fact, been observed in the peroxisomal membranes (
      • Miyauchi K.
      • Masaki R.
      • Taketani S.
      • Yamamoto A.
      • Akayama M.
      • Tashiro Y.
      ). Cytochrome b5 and NADH-cytochrome b5 reductase, which form part of the microsomal electron transfer pathways, have been shown to be localized both to the endoplasmic reticulum and the outer membrane of mitochondria. The present study suggests that these proteins may also be localized to the outer surface of the peroxisomal membranes (
      • Fowler S.
      • Remacle J.
      • Trouet A.
      • Beaufay H.
      • Berthet J.
      • Wibo M.
      • Hauser P.
      ).
      Many peroxisome proteins were identified that are involved in important functions of peroxisome. However, the number of identified peroxins and that of peroxisomal membrane proteins was relatively small. Several proteins that are known to be localized to the peroxisomal matrix were not identified either, suggesting that their amounts in the sample analyzed might be small. The limitation of the present analytical methods comes mainly from the huge difference in the relative abundance of the component proteins in a sample. This is demonstrated in our study of the peroxisomal membranes. By treating the peroxisomal membranes with mild alkaline, we could remove most of the abundant matrix proteins and concentrate the membrane-bound proteins. In addition to all the membrane-bound peroxins, we could detect Pex1 and Pex5 that are loosely associated with the membrane. Pex5 seems to be more tightly bound to the membrane than Pex1. These results again demonstrate the importance of the sample purification in organelle proteomics.
      Five putative proteins were identified in the present study. The most abundant one was identified as the peroxisomal isozyme of Lon protease. Lon protease is an ATP-dependent protease with an AAA domain. It is distributed widely among various species including Archaea, Eubacteria, and Eukaryotes (
      • Goldberg A.L.
      ). Together with other proteases such as Clps, Lon protease is considered to function as molecular chaperone in bacteria (
      • Goldberg A.L.
      • Moerschell R.P.
      • Chung C.H.
      • Maurizi M.R.
      ). In eukaryotes these functions are affected by more complex proteosome machineries, but a Lon protease homolog has been found in mitochondria of yeast and mammals (
      • Van Dyck L.
      • Pearce D.A.
      • Sherman F.
      ,
      • Wang N.
      • Gottesman S.
      • Willingham M.C.
      • Gottesman M.M.
      • Maurizi M.R.
      ). Lon proteases in plants are now classified into three isoforms; Lon1, Lon2, and Lon3 (
      • Sarria R.
      • Lyznik A.
      • Vallejos C.E.
      • Mackenzie S.A.
      ). Sequence comparison revealed the relationship among various Lon isoforms (Fig. 3A). The overall structure is well conserved, especially in the C-terminal half of the molecule that contains the AAA domain. The main differences are found in the N-terminal half, where the Lon protease domain localizes. The plant Lon1 isoforms share high homology with the peroxisomal isoform, whereas Lon2 isoforms are homologous to the mammalian mitochondrial Lon protease. The plant Lon1 isoforms contain the C-terminal sequence, Ser-Lys-Leu, essential for the peroxisomal import. Although the Lon1 isoform has been annotated as plant mitochondrial Lon protease (
      • Sarria R.
      • Lyznik A.
      • Vallejos C.E.
      • Mackenzie S.A.
      ), we can conclude that the isoform is the peroxisome-specific isoform in plants. Interestingly, the Lon3 in plants is considered to be a chloroplast isoform (
      • Sarria R.
      • Lyznik A.
      • Vallejos C.E.
      • Mackenzie S.A.
      ). If this is the case, plants have three distinct Lon proteases in three organelles; mitochondria, chloroplast, and peroxisome. In peroxisomes, two other proteins containing the AAA domain are known (Pex1p and Pex6p). As their names suggest, they are involved in peroxisomal biogenesis and the import of peroxisomal matrix proteins. Lon protease in yeast is located in the mitochondrial matrix and is involved in maintenance of respiration and intramitochondrial proteolysis (
      • Suzuki C.K.
      • Suda K.
      • Wang N.
      • Schatz G.
      ). Although physiological functions of Lon protease in peroxisome has yet to be elucidated, it is possible that the protease is involved either in the folding process of the matrix proteins imported across the peroxisomal membranes as molecular chaperone, or in the degradation process of the misfolded proteins.
      Another unknown protein of interest is a bi-functional enzyme containing an ACDH domain (Fig. 3B). This protein has been annotated as similar to AI987948. The C-terminal sequence of the enzyme is Ala-Arg-Met, which can function as a PTS-1 (
      • Mohan K.V.K.
      • Som I.
      • Atreya C.D.
      ). It is, therefore, highly likely that the protein having an ACDH domain is an intrinsic peroxisomal protein. ACDHs in mitochondria are involved in the first step of mitochondrial β-oxidation of fatty acids. Contrary to the mitochondrial ACDHs, the peroxisomal enzyme has a phosphotransferase domain (APH) in its N-terminal half. It is of interest how this bi-functional enzyme is involved in the peroxisomal fatty acid metabolism.
      The third protein of possible functional importance is the putative protein that is similar to a bovine endozepine-related protein precursor with a theoretical molecular mass of 57 kDa. It has an acyl-CoA-binding protein (ACBP) domain in the N-terminal. ACBP, a 10-kDa lipid-binding protein that may regulate the availability of acyl-CoA esters, is formed from the precursor protein and directly regulated by peroxisome proliferator-activated receptor-γ and -α through the intronic peroxisome proliferator response element (
      • Helledie T.
      • Gr⊘ntved L.
      • Jensen S.S.
      • Kiilerich P.
      • Rietveld L.
      • Albrektsen T.
      • Boysen M.S.
      • N⊘hr J.
      • Larsen L.K.
      • Fleckner J.
      • Stunnenberg H.G.
      • Kristiansen K.
      • Mandrup S.
      ). It is intriguing to assume a physiological role of this C-terminal domain produced from a larger precursor in the transport/metabolism of lipid metabolites.
      In summary, we isolated a peroxisome fraction that is practically free from contamination of other cellular organelles, and we identified many known peroxisomal proteins. Semi-quantitative analysis produced a rough image of relative content of each protein; a proteome of rat liver peroxisome. A few novel proteins were added to the list, one of which is the newly identified peroxisome-specific isozyme of Lon protease. Judging from the relative content of the protein, the Lon protease is an abundant peroxisomal matrix protein. In rat liver, this protein is mainly associated with the electron-dense core structure in the peroxisomal matrix. Since the Lon proteases function as chaperones, the peroxisome-specific Lon isozyme is probably involved directly in peroxisome biogenesis. The peroxisomal isozyme distinct from the mitochondrial form is found not only in mammals but also in plants. An intriguing hypothesis is that there is a distinct isoform for each of the three plant organelles: mitochondria, chloroplast, and peroxisome. This hypothesis has yet to be substantiated, although the roles played by Lon proteases in organelle biosynthesis are clearly important.

      References

      1. Deleted in proof

        • O'Donovan C.
        • Bairoch A.
        • Apweiler R.
        Trends Biotechnol. 2001; 19: 178-181
        • Shevchenko A.
        • Jensen O.N.
        • Podtelejnikov A.V.
        • Sagliocco F.
        • Wilm M.
        • Vorm O.
        • Mortensen P.
        • Shevchenko A.
        • Boucherie H.
        • Mann M.
        Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 14440-14445
        • Pandey A.
        • Mann M.
        Nature. 2000; 405: 837-846
        • Washburn M.P.
        • Wolters D.
        • Yates J.R.
        Nat. Biotechnol. 2001; 19: 242-247
        • Wu C.C.
        • Yates J.R.
        Nat. Biotechnol. 2003; 21: 262-267
        • Jung E.
        • Heller M.
        • Sanchez J.C.
        • Hochstrasser D.F.
        Electrophoresis. 2000; 21: 3369-3377
        • Bell A.W.
        • Ward M.A.
        • Blackstock W.P.
        • Freeman H.N.M.
        • Choudhary J.S.
        • Lewis A.P.
        • Chotai D.
        • Fazel A.
        • Gushue J.N.
        • Paiement J.
        • Palcy S.
        • Chevet E.
        • Lafrenière-Roula M.
        • Solari R.
        • Thomas D.Y.
        • Rowley A.
        • Bergeron J.J.M.
        J. Biol. Chem. 2001; 276: 5152-5165
        • Andersen J.S.
        • Lyon C.E.
        • Fox A.H.
        • Leung A.K.L.
        • Lam Y.W.
        • Steen H.
        • Mann M.
        • Lamond A.I.
        Cur. Biol. 2002; 12: 1-11
        • Hanson B.J.
        • Schulenberg B.
        • Patton W.F.
        • Capaldi R.A.
        Electrophoresis. 2001; 22: 950-959
        • Gygi S.P.
        • Rist B.
        • Gerber S.A.
        • Turecek F.
        • Gelb M.H.
        • Aebersold R.
        Nat. Biotechnol. 1999; 17: 994-999
        • Imanaka T.
        • Shiina Y.
        • Takano T.
        • Hashimoto T.
        • Osumi T.
        J. Biol. Chem. 1996; 271: 3706-3713
        • Imanaka T.
        • Lazarow P.B.
        • Takano T.
        Biochim. Biophys. Acta. 1991; 1062: 264-270
        • Shevchenko A.
        • Wilm M.
        • Vorm O.
        • Mann M.
        Anal. Chem. 1996; 68: 850-858
        • Taniguchi H.
        • Suzuki M.
        • Manenti S.
        • Titani K.
        J. Biol. Chem. 1994; 269: 22481-22484
        • Konishi H.
        • Yamauchi E.
        • Taniguchi H.
        • Yamamoto T.
        • Matsuzaki H.
        • Takemura Y.
        • Ohmae K.
        • Kikkawa U.
        • Nishizuka Y.
        Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 6587-6592
        • Yoshihara T.
        • Hamamoto T.
        • Munakata R.
        • Tajiri R.
        • Ohsumi M.
        • Yokota S.
        J. Histochem. Cytochem. 2001; 49: 1123-1131
        • Shimozawa N.
        • Tsukamoto T.
        • Suzuki Y.
        • Orii T.
        • Shirayoshi Y.
        • Mori T.
        • Fujiki Y.
        Science. 1992; 255: 1132-1134
        • Suzuki Y.
        • Jiang L.L.
        • Souri M.
        • Miyazawa S.
        • Fukuda S.
        • Zhang Z.
        • Une M.
        • Shimozawa N.
        • Kondo N.
        • Orii T.
        • Hashimoto T.
        Am. J. Hum. Genet. 1997; 61: 1153-1162
        • Shimozawa N.
        • Nagase T.
        • Takemoto Y.
        • Ohura T.
        • Suzuki Y.
        • Kondo N.
        Am. J. Med. Genet. 2003; 120A: 40-43
        • Sarria R.
        • Lyznik A.
        • Vallejos C.E.
        • Mackenzie S.A.
        Plant Cell. 1998; 10: 1217-1228
        • Adam Z.
        • Adamska I.
        • Nakabayashi K.
        • Ostersetzer O.
        • Haussuhl K.
        • Manuell A.
        • Zheng B.
        • Vallon O.
        • Rodermel S.R.
        • Shinozaki K.
        • Clarke A.K.
        Plant Physiol. 2001; 125: 1912-1918
        • Barakat S.
        • Pearce D.A.
        • Sherman F.
        • Rapp W.D.
        Plant Mol. Biol. 1998; 37: 141-154
        • Mohan K.V.K.
        • Som I.
        • Atreya C.D.
        J. Virol. 2002; 76: 2543-2547
        • Lazarow P.B.
        • de Duve C.
        Proc. Natl. Acad. Sci. U. S. A. 1976; 73: 2043-2046
        • Mozdy A.D.
        • McCaffery J.M.
        • Shaw J.M.
        J. Cell Biol. 2000; 151: 367-379
        • Alfonzo J.D.
        • Thiemann O.H.
        • Simpson L.
        J. Biol. Chem. 1998; 273: 30003-30011
        • Miyauchi K.
        • Masaki R.
        • Taketani S.
        • Yamamoto A.
        • Akayama M.
        • Tashiro Y.
        J. Biol. Chem. 1991; 266: 19536-19542
        • Fowler S.
        • Remacle J.
        • Trouet A.
        • Beaufay H.
        • Berthet J.
        • Wibo M.
        • Hauser P.
        J. Cell Biol. 1976; 71: 535-550
        • Goldberg A.L.
        Eur. J. Biochem. 1992; 203: 9-23
        • Goldberg A.L.
        • Moerschell R.P.
        • Chung C.H.
        • Maurizi M.R.
        Methods Enzymol. 1994; 244: 350-375
        • Van Dyck L.
        • Pearce D.A.
        • Sherman F.
        J. Biol. Chem. 1994; 269: 238-242
        • Wang N.
        • Gottesman S.
        • Willingham M.C.
        • Gottesman M.M.
        • Maurizi M.R.
        Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 11247-11251
        • Suzuki C.K.
        • Suda K.
        • Wang N.
        • Schatz G.
        Science. 1994; 264: 273-276
        • Helledie T.
        • Gr⊘ntved L.
        • Jensen S.S.
        • Kiilerich P.
        • Rietveld L.
        • Albrektsen T.
        • Boysen M.S.
        • N⊘hr J.
        • Larsen L.K.
        • Fleckner J.
        • Stunnenberg H.G.
        • Kristiansen K.
        • Mandrup S.
        J. Biol. Chem. 2002; 277: 26821-26830