The Pichia pastoris PER6 gene product is a peroxisomal integral membrane protein essential for peroxisome biogenesis and has sequence similarity to the Zellweger syndrome protein PAF-1.

We report the cloning of PER6, a gene essential for peroxisome biogenesis in the methylotrophic yeast Pichia pastoris. The PER6 sequence predicts that its product Per6p is a 52-kDa polypeptide with the cysteine-rich C3HC4 motif. Per6p has significant overall sequence similarity with the human peroxisome assembly factor PAF-1, a protein that is defective in certain patients suffering from the peroxisomal disorder Zellweger syndrome, and with car1, a protein required for peroxisome biogenesis and caryogamy in the filamentous fungus Podospora anserina. In addition, the C3HC4 motif and two of the three membrane-spanning segments predicted for Per6p align with the C3HC4 motifs and the two membrane-spanning segments predicted for PAF-1 and car1. Like PAF-1, Per6p is a peroxisomal integral membrane protein. In methanol- or oleic acid-induced cells of per6 mutants, morphologically recognizable peroxisomes are absent. Instead, peroxisomal remnants are observed. In addition, peroxisomal matrix proteins are synthesized but located in the cytosol. The similarities between Per6p and PAF-1 in amino acid sequence and biochemical properties, and between mutants defective in their respective genes, suggest that Per6p is the putative yeast homolog of PAF-1.     

A stretch of positively charged amino acids at the N terminus of Hansenula polymorpha Pex3p is involved in incorporation of the protein into the peroxisomal membrane

Pex3p is a peroxisomal membrane protein that is essential for peroxisome biogenesis. Here, we show that a conserved stretch of positively charged amino acids (Arg(11)-X-Lys-Lys-Lys(15)) in the N terminus of Hansenula polymorpha Pex3p is involved in incorporation of the protein into its target membrane. Despite the strong conservation, this sequence shows a high degree of redundancy. Substitution of either Arg(11), Lys(13), Lys(14), or Lys(15) with uncharged or negatively charged amino acids did not interfere with Pex3p location and function. However, a mutant Pex3p, carrying negatively charged amino acids at position 13 and 15 (K13E/K15E), caused moderate but significant defects in peroxisome assembly and matrix protein import. Additional changes in the N terminus of Pex3p, e.g. replacing three or four of the positively charged amino acids with negatively charged ones, led to a typical pex3 phenotype, i.e. accumulation of peroxisomal matrix proteins in the cytosol and absence of peroxisomal remnants. Also, in these cases, the mutant Pex3p levels were reduced. Remarkably, mutant Pex3p proteins were mislocalized to mitochondria or the cytosol, depending on the nature of the mutation. Furthermore, in case of reduced amounts of Pex3p, the levels of other peroxisomal membrane proteins, e.g. Pex10p and Pex14p, were also diminished, suggesting that Pex3p maybe involved in the recruitment or stabilization of these proteins (in the membrane).     

Temperature-sensitive phenotype of Chinese hamster ovary cells defective in PEX5 gene

SK32 mutant cells, which were isolated as peroxisome-deficient Chinese hamster ovary (CHO) cells by an advantage of a visible peroxisome form of green fluorescent protein (GFP), were found to suffer from a functional loss of PEX5 gene encoding for PTS1R. The sequence analysis of cDNA indicated that PEX5 gene encoded for the two isoforms composed of 603 amino acids (PTS1RS) and 640 amino acids (PTS1RL). The mutation changed glycine to arginine at amino acid position 343 of PTS1RL (corresponding to the position 306 of PTS1RS) in SK32 cells. The mutant cells exhibited a temperature-sensitive (TS) phenotype on the peroxisomal localizations of the recombinant GFP and urate oxidase appending a genuine peroxisome targeting signal 1 (PTS1), a tripeptide of Ser-Lys-Leu (SKL) at the C-terminus, but did not on that of catalase harboring a divergent PTS1, Lys-Ala-Asn-Leu (KANL) sequence. 3-ketoacyl-CoA thiolase (hereafter referred to as thiolase), which harbors an extension sequence (PTS2) at the N-terminus, never appeared to be affected on the peroxisomal localization in the mutant cells. When thiolase was examined on the molecular size in the mutant cells, the enzyme existed as the larger precursor form in the peroxisomes at 37 degrees C and a considerable part (almost half) was converted to the mature size at 30 degrees C. These results indicate that the amino acid substitution, Gly306Arg in PTS1RS and/or Gly343Arg in PTSRL, gives rise to TS phenotype on the peroxisomal translocation of PTS1 proteins and the maturation of PTS2 protein.     

A novel aberrant splicing mutation of the PEX16 gene in two patients with Zellweger syndrome

Human Pex16p, a peroxisomal membrane protein composed of 336 amino acids, plays a central role in peroxisomal membrane biogenesis. A nonsense mutation (R176ter) in the PEX16 gene has been reported in the case of only one patient (D-01) belonging to complementation group D of the peroxisome biogenesis disorders. We have now identified two patients belonging to group D (D-02 and D-03) whose fibroblasts were found to contain no peroxisomal membrane structure ghosts. Molecular analysis of the PEX16 gene revealed aberrant cDNA species lacking 65 bp, corresponding to exon 10 skipping caused by a splice site mutation (IVS10 + 2T -->C). Both patients, although unrelated, were homozygous for this mutation. This mutation changes the amino acid sequence starting from codon 298 and introduces a termination codon at codon 336. As a consequence, the cell's ability to membrane synthesis and protein import is disrupted, which implies that the changed C terminus of the Pex16p in these patients likely affects its function.     

The peroxin pex3p initiates membrane assembly in peroxisome biogenesis

Rat cDNA encoding a 372-amino-acid peroxin was isolated, primarily by functional complementation screening, using a peroxisome-deficient Chinese hamster ovary cell mutant, ZPG208, of complementation group 17. The deduced primary sequence showed approximately 25% amino acid identity with the yeast Pex3p, thereby we termed this cDNA rat PEX3 (RnPEX3). Human and Chinese hamster Pex3p showed 96 and 94% identity to rat Pex3p and had 373 amino acids. Pex3p was characterized as an integral membrane protein of peroxisomes, exposing its N- and C-terminal parts to the cytosol. A homozygous, inactivating missense mutation, G to A at position413, in a codon (GGA) for Gly(138) and resulting in a codon (GAA) for Glu was the genetic cause of peroxisome deficiency of complementation group 17 ZPG208. The peroxisome-restoring activity apparently required the full length of Pex3p, whereas its N-terminal part from residues 1 to 40 was sufficient to target a fusion protein to peroxisomes. We also demonstrated that Pex3p binds the farnesylated peroxisomal membrane protein Pex19p. Moreover, upon expression of PEX3 in ZPG208, peroxisomal membrane vesicles were assembled before the import of soluble proteins such as PTS2-tagged green fluorescent protein. Thus, Pex3p assembles membrane vesicles before the matrix proteins are translocated.     

Human peroxisome assembly factor-2 (PAF-2): a gene responsible for group C peroxisome biogenesis disorder in humans.

Peroxisome-biogenesis disorders (PBD) are genetically heterogeneous and can be classified into at least ten complementation groups. We recently isolated the cDNA for rat peroxisome assembly factor-2 (PAF-2) by functional complementation using the peroxisome-deficient Chinese-hamster-ovary cell mutant, ZP92. To clarify the novel pathogenic gene of PBD, we cloned the full-length human PAF-2 cDNA that morphologically and biochemically restores peroxisomes of group C Zellweger fibroblasts (the same as group 4 in the Kennedy-Krieger Institute) and identified two pathogenic mutations in the PAF-2 gene in two patients with group C Zellweger syndrome. The 2,940-bp open reading frame of the human PAF-2 cDNA encodes a 980-amino-acid protein that shows 87.1% identity with rat PAF-2 and also restored the peroxisome assembly after gene transfer to fibroblasts of group C patients. Direct sequencing of the PAF-2 gene revealed a homozygous 1-bp insertion at nucleotide 511 (511 insT) in one patient with group C Zellweger syndrome (ZS), which introduces a premature termination codon in the PAF-2 gene, and, in the second patient, revealed a splice-site mutation in intron 3 (IVS3+1G-->A), which skipped exon 3, an event that leads to peroxisome deficiency. Chromosome mapping utilizing FISH indicates that PAF-2 is located on chromosome 6p21.1. These results confirm that human PAF-2 cDNA restores peroxisome of group C cells and that defects in the PAF-2 produce peroxisome deficiency of group C PBD.     

Peroxisome biogenesis and molecular defects in peroxisome assembly disorders

Peroxisome assembly in mammals requires more than 14 genes. So far, we have isolated seven complementation groups (CGs) of peroxisome biogenesis-defective Chinese hamster ovary (CHO) cell mutants, Z65, Z24/ZP107, ZP92, ZP105/ZP139, ZP109, ZP110, ZP114. Two peroxin cDNAs, PEX2 and PEX6, were first cloned by genetic phenotype-complementation assay using Z65 and ZP92, respectively, and were shown to be responsible for peroxisome biogenesis disorders (PBD) such as Zellweger syndrome, of CG-F (the same as CG-X in U.S.A.) and CG-C (the same as CG-IV), respectively. Pex2p is a RING zinc finger membrane protein of peroxisomes and Pex6p is a member of the AAA ATPase family. We likewise isolated PEX12 encoding a peroxisomal integral membrane protein in the RING family, by functional complementation of ZP109, demonstrating PEX12 to be responsible for CG-III PBD. We also cloned PEX1 by screening of human liver cDNA library, using ZP107. PEX1 mutation was delineated to be the genetic cause of PBD in the most highest incidence group, CG-E (the same a CG-I). Moreover, we recently found that Pex5p, using PEX5-defective ZP105 and ZP139. Thus, CHO cell mutants defective in peroxisome biogenesis are indeed shown to be very useful for the studies of peroxisome assembly and delineating pathogenic genes in PBD. Furthermore, we have isolated novel CGs of CHO mutants, ZP119 and ZP126.     

Accumulation of glycolipids in mutant Chinese hamster ovary cells (Z65) with defective peroxisomal assembly and comparison of the metabolic rate of glycosphingolipids between Z65 cells and wild-type CHO-K1 cells

The influence of peroxisomal dysfunction on glycosphingolipid metabolism was investigated using mutant Chinese hamster ovary (CHO) cells (Z65) with defective assembly of the peroxisomal membranes. In accordance with previous observations, the concentration of very long chain fatty acid (C24:0) was shown to be higher in Z65 cells than in control cells. We then compared the composition of glycolipids in Z65 cells with that in CHO-K1 cells, which are wild-type Chinese hamster ovary cells with intact peroxisomes, and found significantly increased concentrations of ceramide monohexoside (CMH) and ganglioside GM3 in Z65 cells. However, there were no differences in the concentrations of glycerophospholipids, triglycerides, free fatty acids and cholesterol between Z65 and CHO-K1 cells. Further, to investigate the metabolic rate of the major lipids, Z65 and CHO-K1 cells were pulse-labeled with [3-14C]serine. [3-14C]Serine was incorporated into phosphatidylserine, phosphatidylethanolamine and sphingomyelin more quickly in CHO-K1 than in Z65 cells. However, after 48 h, the radioactivity incorporated into those lipids, including CMH, was greater in Z65 cells than in CHO-K1 cells. Thus, the altered metabolism of glycosphingolipids, probably due to peroxisomal dysfunction, was thought to be responsible for the change in glycosphingolipid composition in Z65 cells.     

Peroxisomes are formed from complex membrane structures in PEX6-deficient CHO cells upon genetic complementation

Pex6p belongs to the AAA family of ATPases. Its CHO mutant, ZP92, lacks normal peroxisomes but contains peroxisomal membrane remnants, so called peroxisomal ghosts, which are detected with anti-70-kDa peroxisomal membrane protein (PMP70) antibody. No peroxisomal matrix proteins were detected inside the ghosts, but exogenously expressed green fluorescent protein (GFP) fused to peroxisome targeting signal-1 (PTS-1) accumulated in the areas adjacent to the ghosts. Electron microscopic examination revealed that PMP70-positive ghosts in ZP92 were complex membrane structures, rather than peroxisomes with reduced matrix protein import ability. In a typical case, a set of one central spherical body and two layers of double-membraned loops were observed, with endoplasmic reticulum present alongside the outer loop. In the early stage of complementation by PEX6 cDNA, catalase and acyl-CoA oxidase accumulated in the lumen of the double-membraned loops. Biochemical analysis revealed that almost all the peroxisomal ghosts were converted into peroxisomes upon complementation. Our results indicate that 1) Peroxisomal ghosts are complex membrane structures; and 2) The complex membrane structures become import competent and are converted into peroxisomes upon complementation with PEX6.     

A novel pex2 mutant: catalase-deficient but temperature-sensitive PTS1 and PTS2 import

We searched for Chinese hamster ovary (CHO) cell mutants defective in peroxisome biogenesis by using peroxisome targeting sequence (PTS) of Pex3p (amino acid residues 1-40)-fused enhanced green fluorescent protein (EGFP). From mutagenized wild-type CHO-K1 cells stably expressing rat Pex2p and Pex3p(1-40)-EGFP, cell colonies resistant to the 9-(1(')-pyrene)nonanol/ultraviolet treatment were examined for intracellular location of peroxisomal proteins, including EGFP chimera, catalase, and matrix proteins with PTS types 1 and 2. One clone, ZPEG309, showed a distinct phenotype: import defect of catalase, but normal transport of PTS1 and PTS2 proteins at 37 degrees C. PTS1 and PTS2 import was abrogated when ZPEG309 was cultured at 39 degrees C. Genetic defect of ZPEG309 was a nonsense point mutation in a codon for Arg50 in CHO PEX2 and a mutation resulting in a C-terminal truncation of the introduced rat Pex2p. Therefore, ZPEG309 is a novel pex2, catalase-deficient mutant with temperature-sensitive PTS1 and PTS2 import.     

Identification of the cDNA clone which encodes the 58-kDa protein containing the amino acid sequence of rat liver non-specific lipid-transfer protein (sterol-carrier protein 2). Homology with rat peroxisomal and mitochondrial 3-oxoacyl-CoA thiolases

The relationship between the rat liver non-specific lipid-transfer protein (nsLTP) and the 58-kDa protein cross-reactive with anti-nsLTP antibodies, was investigated by cDNA analysis. A 1945-bp cDNA clone was isolated which encodes a 58.7-kDa protein. This protein is identical to the 58-kDa immunoreactive protein determined by N-terminal sequence analysis of the purified 58-kDa protein. It consists of 546 amino acid residues, of which the 123 C-terminal residues are identical to the sequence of nsLTP. The N-terminal 400 amino acid residues of the 58.7-kDa protein were found to have 23.5% identity with the sequence of both mitochondrial and peroxisomal rat 3-oxoacyl-CoA thiolases, including a hypothetical substrate-binding site. The cDNA insert hybridizes with 1.1-kb, 1.7-kb, 2.4-kb and 3.0-kb mRNA species in RNA isolated from various rat tissues and from Chinese hamster ovary (CHO) cells. Southern blot analysis suggests that these mRNA species are generated from a single gene. Mutant CHO cells, deficient in peroxisomes, lack nsLTP. We have found that the mRNA encoding nsLTP is still present in these cells, which suggests that the absence of this protein is related to the lack of peroxisomes.     

Peroxisome targeting signal of rat liver acyl-coenzyme A oxidase resides at the carboxy terminus.

To identify the topogenic signal of peroxisomal acyl-coenzyme A oxidase (AOX) of rat liver, we carried out in vitro import experiments with mutant polypeptides of the enzyme. Full-length AOX and polypeptides that were truncated at the N-terminal region were efficiently imported into peroxisomes, as determined by resistance to externally added proteinase K. Polypeptides carrying internal deletions in the C-terminal region exhibited much lower import activities. Polypeptides that were truncated or mutated at the extreme C terminus were totally import negative. When the five amino acid residues at the extreme C terminus were attached to some of the import-negative polypeptides, the import activities were rescued. Moreover, the C-terminal 199 and 70 amino acid residues of AOX directed fusion proteins with two bacterial enzymes to peroxisomes. These results are interpreted to mean that the peroxisome targeting signal of AOX residues at the C terminus and the five or fewer residues at the extreme terminus have an obligatory function in targeting. The C-terminal internal region also has an important role for efficient import, possibly through a conformational effect.     

The primary structure of human Rieske iron-sulfur protein of mitochondrial cytochrome bc1 complex deduced from cDNA analysis

We isolated a cDNA encoding human Rieske Fe-S protein of mitochondrial cytochrome bc1 complex from a fibroblast cDNA library by colony hybridization. The cDNA contains the nucleotide sequence encoding all of the amino acids (274 residues) comprising the putative precursor to the protein. Based on the known amino acid sequence of bovine Rieske Fe-S protein, the N-terminal extension sequence is presumed to be composed of 78 amino acids with a molecular weight of 8053. The mature protein consists of the same number of amino acid residues as that of its rat and bovine counterparts, having a homology of about 92% with the latter.     

Peroxisome Formation Requires the Endoplasmic Reticulum Channel Protein Sec61

In peroxisome formation, models of near‐autonomous peroxisome biogenesis with membrane protein integration directly from the cytosol into the peroxisomal membrane are in direct conflict with models whereby peroxisomes bud from the endoplasmic reticulum and receive their membrane proteins through a branch of the secretory pathway. We therefore reinvestigated the role of the Sec 61 complex, the protein‐conducting channel of the endoplasmic reticulum (ER) in peroxisome formation. We found that depletion or partial inactivation of Sec 61 in yeast disables peroxisome formation. The ER entry of the early peroxisomal membrane protein Pex 3 engineered with a glycosylation tag is reduced in sec61 mutant cells. Moreover, we were able to reconstitute Pex 3 import into ER membranes in vitro, and we identified a variant of a signal anchor sequence for ER translocation at the Pex 3 N‐terminus. Our findings are consistent with a Sec 61 requirement for peroxisome formation and a fundamental role of the ER in peroxisome biogenesis.     

A missense mutation in the RING finger motif of PEX2 protein disturbs the import of peroxisome targeting signal 1 (PTS1)-containing protein but not the PTS2-containing protein

SK24 and PT54 mutant cells, which are peroxisome-deficient Chinese hamster ovary (CHO) cells isolated using peroxisomal forms of green fluorescent protein (GFP), were found to be defective in the PEX2 gene. The nucleotide sequences of PEX2 cDNA from the mutant cells were determined to identify mutation sites in the mutant cells. The mutation in SK24 cells changed cysteine to tyrosine at amino acid position 258, which is a component of the RING finger (C(3)HC(4)) motif in the carboxyl terminus of the protein. PT54 cells contained a nonsense mutation in the codon for glutamine at position 101, resulting in premature termination. The immunocytochemical analyses revealed distinct phenotypes between mutant cells defective in the PEX2 gene. Both mutant cells exhibited cytosolic mislocalizations on catalase and urate oxidase containing PTS1. On the other hand, on 3-ketoacyl-CoA thiolase containing PTS2, PT54 cells exhibited cytosolic mislocalization, but SK24 cells exhibited peroxisomal localization. When wild-type or mutant-type PEX2 cDNA was transfected into both mutant cells, the stable transformants restored the phenotype in accordance with the transfected cDNA. These observations indicate that an amino acid substitution, cysteine-258 to tyrosine, in the RING finger motif of PEX2 protein, whose function is required for peroxisomal localizations of both PTS1- and PTS2-containing proteins, results in a complete defect in the PTS1 pathway but not in the PTS2 pathway.     

The 70 kDa peroxisomal membrane protein: an ATP-binding cassette transporter protein involved in peroxisome biogenesis

The 70 kDa peroxisomal membrane protein (PMP70) is a major component of peroxisomal membranes. cDNAs for human and rat PMP70 have been cloned and sequenced and the gene mapped to the human chromosome 1p21-22. The predicted amino acid sequence showed homology to members of the ATP-binding cassette transporter family. In humans, mutations in the PMP70 gene have been found in a subset of patients with Zellweger syndrome, a lethal inborn error of peroxisome biogenesis. These results suggest that PMP70 functions in transporting molecules or possibility peptides across the peroxisomal membrane and has an important role in peroxisome assembly.     

Altered antigenic disposition of peroxisomal urate oxidase in PEX5-defective Chinese hamster ovary cells

Since Chinese hamster ovary (CHO) cells never express urate oxidase (UO), we tried to establish cell lines stably producing UO in order to elucidate the peroxisomal import process. The enzyme is a peroxisome targeting signal 1 (PTS1) protein harboring SKL motif at the carboxy-terminus [Biochem. Biophys. Res. Commun. 158 (1989) 991] and PEX5 protein (Pex5p) is supposed to be involved in the import process [Nat. Genet. 9 (1995) 115; J. Cell Biol. 130 (1995) 51]. We transfected a cDNA encoding rat UO into both wild type and PEX5-defective CHO cells to isolate each cell line stably producing the enzyme. While we examined the import process of UO in mutant cells, we noticed an interesting observation by using polyclonal antibody U1 or U2, which separately recognizes epitopes of UO. U1 antibody mainly interacts with epitopes in the amino-terminal region of UO. On the other hand, U2 antibody reacts with many epitopes distributed in the broad region of UO molecule. When UO produced in cultured cells was stained with U2 antibody, the enzyme was detected in peroxisomes of both wild type and PEX5-mutant cells. Whereas, U1 antibody stained the peroxisomal UO in wild type cells, but not in PEX5-mutant cells. These immunocytochemical observations suggest that the epitopes at the amino-terminal region of UO will be concealed in mutant cells. When the mutant cells were transfected with wild type PEX5 cDNA, U1 antibody came to react with UO in peroxisomes of mutant cells. The restoration indicates that the exposure of N-terminal epitopes of UO will depend upon the functional Pex5p. Immunoelectron microscopic observation showed that the peroxisomal import of UO was partially retarded in PEX5 mutant cells. The observation also supported the fact that UO was mainly localized in the peroxisomal matrix of wild type cells but in the membrane of mutant cells.     

Conversion of human interferon-β from a secreted to a phosphatidylinositol anchored protein by fusion of a 17 amino acid sequence to its carboxyl terminus

A number of cell-surface proteins are anchored in plasma membranes by a glycosylated phosphatidylinositol (PI) moiety that is covalently attached to the carboxyl-terminal amino acid of the mature protein. We have previously reported the construction of a cDNA clone of a truncated Platelet-derived growth factor (PDGF) receptor that consists of the extracellular domain without the transmembrane and cytoplasmic domains. In the construction of the vector, a sequence of 51 base pairs (bp) from the 3′-untranslated region of the receptor cDNA was linked in frame with the external domain coding sequence. The truncated receptor protein with the peptide VTSGHCHEERVDRHDGE fused to its carboxyl terminus was covalently attached to the membrane by a PI linkage and it was released by phosphatidylinositol specific-phospholipase C (PI-PLC). When the 51 bp sequence was deleted, the external domain receptor protein was secreted into the media. To determine whether the PI linkage of the protein was due to the 17 amino acids added, the peptide was fused to the carboxyl terminus of the secreted protein human Interferon-β (hu-IFN-β). Chinese hamster ovary (CHO) cells transfected with the hu-IFN-β cDNA secreted the protein to theconditioned media, whereas CHO cells transfected with the carboxyl terminus modified-hu-IFN-β cDNA did not secrete detectable levels of protein. CHO cells expressing the carboxyl terminus modified-hu-IFN-β were treated with PI-PLC, the media and cell lysates were analyzed by SDS-PAGE after immunoprecipitation with antibodies against hu-IFN-β. The modified protein is anchored to the plasma membrane by a PI linkage and it is specifically released by PI-PLC, whereas a control preparation of CHO cells expressing wild type hu-IFN-β does not show the same pattern. The 17 amino acid peptide fused to the carboxyl terminus of IFN-β directs attachment of a PI anchor and targets the fusion protein to the plasma membrane.