Characterization of the signal peptide at the amino terminus of the rat peroxisomal 3-ketoacyl-CoA thiolase precursor

The amino-terminal presequences of rat peroxisomal 3-ketoacyl-CoA thiolase precursors (types A and B) were reported to be cleavable signal peptides for peroxisomal protein translocation. In the present study, this was proven by immunoelectron microscopy of the cultured Chinese hamster ovary cells stably expressing fusion proteins of the amino-terminal sequences of the thiolase precursor and Escherichia coli dihydrofolate reductase. The fusion proteins were processed into mature forms of the apparently correct sizes. Site-directed mutagenesis studies of the charged residues in the B-type presequence (26 amino acid residues) revealed that arginine at position -24 and histidine at position -17 were both indispensable. Even replacement of these residues with other basic amino acids abolished the import activity. Both Arg-24 and His-17 were also required in a longer presequence (36 amino acid residues) of the thiolase A, thereby suggesting that the signal can function in an internal position. When glutamic acid at position -11 was changed to amino acids other than aspartic acid, the signal peptide became apparently effective in both peroxisomal and mitochondrial targeting. All of these data indicate that the thiolase signal peptide is a newly defined type of peroxisomal targeting signal recognized by a mechanism presumably different from that for a known peroxisomal signal, the carboxy-terminal Ser-Lys-Leu-COOH motif.

[PMP70, the 70-kDa peroxisomal membrane protein: a member of the ATP-binding cassette transporters]

We have isolated the cDNA of the 70-kDa peroxisomal membrane protein (PMP70) from rat and human liver cDNA libraries. The nucleotide sequence of the cDNA of PMP70 contains an open reading frame of 1977 bp which encodes an amino acid sequence of 659 residues. Possible two domains were identified by hydropathy analysis. One is a hydrophobic region, which presumably contains six transmembrane segments. The other is a hydrophilic domain, which shows striking similarity to the sequences of the ATP-binding cassette (ABC) transporter proteins, including bacterial periplasmic transport proteins, the human multidrug resistance P-glycoprotein (MDR1), cystic fibrosis transmembrane conductance regulator (CFTR), and the putative adrenoleukodystrophy gene product (ALDP). Based on its transmembrane structure and the homology to ABC proteins, PMP70 may be involved in ATP-dependent transport through peroxisomal membrane.

Chicken ovalbumin upstream promoter-transcription factor (COUP-TF) represses transcription from the promoter of the gene for ornithine transcarbamylase in a manner antagonistic to hepatocyte nuclear factor-4 (HNF-4)

Chicken ovalbumin upstream promoter-transcription factor (COUP-TF) and hepatocyte nuclear factor-4 (HNF-4) are orphan members of the steroid/thyroid receptor superfamily and exhibit ubiquitous and liver-enriched tissue distribution, respectively. The gene for rat ornithine transcarbamylase (OTC), an ornithine cycle enzyme, is mainly expressed in the liver and is under the control of the promoter and the 11-kilobase upstream enhancer, both of which are liver-selective. Two sites of the promoter region and two sites of the enhancer region of the OTC gene, as well as the ovalbumin promoter site, were recognized by both HNF-4 and COUP-TF, showing that these two factors have closely related binding specificities. Since HNF-4 activated expression from the OTC promoter in cotransfection analysis, this factor appears to participate in liver-selective activation of the OTC gene. On the other hand, COUP-TF repressed the expression from the OTC promoter, whereas it activated expression from several other promoters. Therefore, COUP-TF plays a dual regulatory role depending on the promoter context. Repression of a tissue-specific promoter by a ubiquitous transactivator and derepression by a related tissue-enriched transactivator is potentially an important mechanism for tissue-specific activation of a gene.

Structure and expression of the genes encoding peroxisomal beta-oxidation enzymes

To investigate the mechanism of induction of the peroxisomal beta-oxidation enzymes by peroxisome proliferators, genes of these enzymes were cloned from rat liver and their structures analyzed. The acyl-CoA oxidase gene was found to produce two forms of the enzyme differing in their amino acid sequences in a limited region, through alternative splicing of the two copies of the third exon. The amino acid sequence of the bifunctional enzyme suggests, compared with those of its mitochondrial counterparts, that the enoyl-CoA hydratase and 3-hydroxyacyl-CoA dehydrogenase activities are located on the amino- and the carboxyl sides, respectively. Two copies of the 3-ketoacyl-CoA thiolase gene were identified per haploid genome. One gene named A is constitutive and encodes a thiolase precursor carrying 36 amino acid residues of amino-terminal presequence, whereas the other, termed B, is remarkably induced by peroxisome proliferators and specifies a precursor having a 26-residue presequence. Functional analysis of the upstream sequence of the acyl-CoA oxidase gene revealed three functionally different regions, one of which had the character of ciprofibrate-responsive enhancer. In this region, two sequences were identified as binding-sites of rat liver nuclear proteins. A gene transfection study indicated that these sequence elements (termed A and B) play important roles in the induction of the gene, the former acting positively whereas the latter probably is acting negatively.

Signal peptide for peroxisomal targeting: replacement of an essential histidine residue by certain amino acids converts the amino-terminal presequence of peroxisomal 3-ketoacyl-CoA thiolase to a mitochondrial signal peptide

The role of the histidine residue at position -17 of the amino-terminal signal peptide of rat peroxisomal 3-ketoacyl-CoA thiolase was studied in vivo, employing site-directed mutagenesis. Among the nine amino acids tested, only glutamine could partially substitute for the histidine. Mutants carrying basic amino acids, arginine and lysine, and hydrophobic residues, leucine and valine, in place of histidine were all translocated to mitochondria, but not to peroxisomes. These results indicate that the signal peptide of the thiolase is recognized by a mechanism totally different from that for the SKL motif, a known peroxisomal targeting signal. Relationship of the thiolase signal peptide to those of mitochondrial proteins is discussed.

Carboxyl-terminal consensus Ser-Lys-Leu-related tripeptide of peroxisomal proteins functions in vitro as a minimal peroxisome-targeting signal

The minimal sequence requirement for a peroxisome-targeting signal was investigated using an in vitro import system. Carboxyl-terminal sequences Ser-Lys-Leu (SKL) and Leu-Gln-Ser-Lys-Leu (LQSKL) of acyl-CoA oxidase (AOX) directed to peroxisomes the fused proteins with import-incompetent forms of AOX and catalase that had been truncated, implying that the SKL tripeptide functions as a targeting signal. Elimination of the entire SKL sequence or deletion of any 1 or 2 amino acids in the sequence abolished the import activity of AOX. Substitution of alanine for serine did not affect the import activity. Topogenic activity was retained when lysine was mutated to either arginine or histidine, whereas mutation to glutamic acid completely abolished the activity. A synthetic peptide comprising the carboxyl-terminal 10 amino acid residues of AOX inhibited the import of the authentic AOX polypeptide, whereas other peptides in which SKL was mutated, deleted, or internally located were not effective. The uptake of AOX was little affected by the peptide with an amidated alpha-carboxyl group. These results strongly suggest that the carboxyl-terminal SKL motif sequence (Ser/Ala)-(Lys/Arg/His)-Leu functions as a topogenic signal in translocation of proteins into peroxisomes, requiring the whole tripeptide sequence with a free alpha-COOH group at the carboxyl terminus.

Molecular basis of 3-ketothiolase deficiency: identification of an AG to AC substitution at the splice acceptor site of intron 10 causing exon 11 skipping

3-Ketothiolase deficiency (3KTD) is the result of a deficiency in mitochondrial acetoacetyl-CoA thiolase (T2). The molecular basis of 3KTD was analyzed in a patient (GK10) and his family at the protein, cDNA and gene levels. Protein analyses showed that GK10's T2 protein was undetectable in fibroblasts even with the pulse-protein labeling method and that his parents were carriers of 3KTD. Complementary DNA analyses with PCR showed that T2 cDNA in the patient lacked the normal exon 11 sequence and that his parents were obligatory carriers of the DNA sequence which canceled exon 11. When the PCR-amplified genomic fragments around exon 11 were sequenced, an AG to AC mutation at the 3' splice site of intron 10 was detected. This mutation is presumed to be responsible for exon 11 skipping.

A novel isoform of rat hepatocyte nuclear factor 4 (HNF-4)

Based on the published nucleotide sequence for rat hepatocyte nuclear factor 4 (HNF-4; Sladek, F.M., Zhong, W., Lai, E. and Darnell, J.E., Jr. (1990) Genes Dev. 4, 2353-2365), we have cloned a cDNA by means of polymerase chain reaction amplification of reverse-transcribed RNA (RT-PCR). Our clone contained an extra segment of 30 bp, which was not found in the previously reported clone, in the coding region near the C-terminus. Further RT-PCR analysis demonstrated that both isoforms of HNF-4 mRNA, i.e., with or without the 30 nucleotide segment, occur in rat liver and kidney, presumably by differential splicing.

Analysis of carrot genes for proliferating cell nuclear antigen homologs with the aid of the polymerase chain reaction

We designed a polymerase chain reaction-based strategy to obtain information about the origin and distribution of a newly discovered proliferating cell nuclear antigen (PCNA) homolog. Carrot genomic segments were amplified using degenerate primers for two conserved regions of known PCNA homologs. The genes encoding the larger PCNA as well as typical PCNA contained introns. Thus, unlike processed PCNA pseudogenes in mammals, the larger homolog is not generated through reverse transcription of a typical PCNA mRNA. Moreover, introns of the larger PCNA homolog were positioned at the characteristic sites in plant PCNA genes. Attempts to amplify cDNA for an additional PCNA homolog from mammalian cells have been unsuccessful. These results suggest that the larger PCNA homolog was generated, presumably through gene duplication, after the divergence of the Planta and Animalia.

Identification of three mutant alleles of the gene for mitochondrial acetoacetyl-coenzyme A thiolase. A complete analysis of two generations of a family with 3-ketothiolase deficiency

3-Ketothiolase deficiency (3KTD) stems from a deficiency of mitochondrial acetoacetyl-coenzyme A thiolase (T2). We analyzed the molecular basis of 3KTD in two generations of a family. A boy (patient 2, GK04), his father (patient 1, GK05), his mother, and his brother were studied; three mutant alleles of T2 gene were identified. Patient 1 is a compound heterozygote: one allele has a point mutation of G to A at position 547 on his T2 cDNA, causing Gly150 to Arg substitution of the mature T2 subunit, and the other allele has GT to TT transition at the 5' splice site of intron 8, causing exon 8's skipping of the T2 cDNA. Patient 2 is also a compound heterozygote: one allele inherited from his mother has AG to CG transition at the 3' splice site of intron 10, causing exon 11's skipping of the T2 cDNA, and the other allele derived from patient 1 has the G to A mutation (Gly to Arg). The brother of patient 2 is an obligatory carrier with the mutant allele causing the exon 8 skipping. This report seems to be the first complete molecular definition of 3KTD at the gene level.

Complementation study of peroxisome-deficient disorders by immunofluorescence staining and characterization of fused cells

Genetic heterogeneity in peroxisome-deficient disorders, including Zellweger's cerebrohepatorenal syndrome, neonatal adrenoleukodystrophy and infantile Refsum disease, was investigated. Fibroblasts from 17 patients were fused using polyethylene glycol, cultivated on cover slips, and the formation of peroxisomes in the fused cells was visualized by immunofluorescence staining, using anti-human catalase IgG. Two distinct staining patterns were observed: (1) peroxisomes appeared in the majority of multinucleated cells, and (2) practically no peroxisomes were identified. Single step 12-(1'-pyrene) dodecanoic acid/ultraviolet (P12/UV)-selection confirmed that the former groups were resistant to this selection, most of the surviving cells contained abundant peroxisomes, and the latter cells died. In the complementary matching, [1-14C]lignoceric acid oxidation and the biosynthesis of peroxisomal proteins were also normalized. Five complementation groups were identified. Group A: Zellweger syndrome and infantile Refsum disease; Groups B, C and D: Zellweger syndrome; Group E: Zellweger syndrome, neonatal adrenoleukodystrophy and infantile Refsum disease. We compared these groupings with those of Roscher and identified eight complementation groups. There was no obvious relation between complementation groups and clinical phenotypes. These results indicate that the transport, intracellular processing and function of peroxisomal proteins were normalized in the complementary matching and that at least eight different genes are involved in the formation of normal peroxisomes and in the transport of peroxisomal enzymes.

Nucleotide sequence of the human 70 kDa peroxisomal membrane protein: a member of ATP-binding cassette transporters

The cDNA sequence of human liver 70 kDa peroxisomal membrane protein (hPMP70) was determined. The nucleotide sequence contains an open reading frame of 1977 base pairs and encodes an amino acid sequence of 659 residues which exhibits 95.0% identity with that of rat liver PMP70. hPMP70 shares close similarity to the members of a superfamily of ATP-binding transport proteins.

Different intracellular localization of peroxisomal proteins in fibroblasts from patients with aberrant peroxisome assembly

We investigated intracellular localization of peroxisomal proteins in fibroblasts from patients with Zellweger syndrome and neonatal adrenoleukodystrophy in whom peroxisomes were morphologically deficient or severely decreased. Indirect immunofluorescence staining revealed that catalase was mainly detected in the cytosol of fibroblasts from these patients, but a small amount of catalase was detected in granular pattern in a small percentage of cells. Double immunofluorescence staining revealed that catalase-containing particles in these patients also contained acyl-CoA oxidase and nonspecific lipid transfer protein. However, a 70 kD integral membrane protein and 3-ketoacyl-CoA thiolase were detected in all cells in granular pattern. Subcellular fractionation using digitonin after cell labeling revealed that a small amount of acyl-CoA oxidase and about half of thiolase in the precursor form were detected in the particulate fraction. These data suggest that the mechanisms of the transport and processing of catalase, acyl-CoA oxidase and nonspecific lipid transfer protein are different from those of the 70 kD integral membrane protein and 3-ketoacyl-CoA thiolase.

Amino-terminal presequence of the precursor of peroxisomal 3-ketoacyl-CoA thiolase is a cleavable signal peptide for peroxisomal targeting

To examine the function of the amino-terminal presequence of rat peroxisomal 3-ketoacyl-CoA thiolase precursor, fusion proteins of various amino-terminal regions of the precursor with non-peroxisomal enzymes were expressed in cultured mammalian cells. On immunofluorescence microscopy, all constructs carrying the presequence part exhibited punctate patterns of distribution, identical with that of catalase, a peroxisomal marker. Proteins lacking all or a part of the prepiece were found in the cytosol. These results indicate that the presequence of the thiolase has sufficient information for peroxisomal targeting.

Structure and expression of the human mitochondrial acetoacetyl-CoA thiolase-encoding gene

To examine 3-ketothiolase deficiency at the gene level, we analyzed the structure of the human mitochondrial acetoacetyl-CoA thiolase (MAT; EC 2.3.1.9)-encoding gene (MAT). From the genomic library of a normal subject in lambda EMBL3, we isolated seven overlapping clones covering the entire length of MAT and the structural organization was determined. The gene spans approx. 27 kb and contains twelve exons interrupted by eleven introns. The 5'-flanking region of the gene lacks a conventional TATA box, but is G + C-rich and contains two CAAT boxes. Included are a putative binding site for the transcription factor, Sp1, and sequences resembling the binding sites of several other transcription factors, all features characteristic of housekeeping genes. A CAT assay revealed that a 101-bp DNA fragment immediately upstream from the cap site has promoter activity, and suggested that a DNA fragment from bp -888 to -102 probably contains a negative regulatory element(s).

Evidence for a structural mutation (347Ala to Thr) in a German family with 3-ketothiolase deficiency

The molecular basis of 3-ketothiolase deficiency (3KTD) was examined in a 3KTD family. Immunochemical analyses showed that mitochondrial acetoacetyl-CoA thiolase (T2) biosynthesized in the patient's fibroblasts (GK06) was unstable and that the parents and brother were obligatory carriers of 3KTD. When sequencing the PCR-amplified patient's T2 cDNA, we noted a G to A replacement which caused 347Ala to Thr substitution of the mature T2 subunit. Transfection analysis revealed that this substitution resulted in an instability of the T2 protein. Analyses of the T2 cDNA and gene of the family indicated that the patient was a compound heterozygote; the allele that derived from the mother had a point mutation (347Ala to Thr) and the other allele from the father has a mutation which would abolish the T2 gene expression. This report is apparently the first definition of a mutant allele for 3KTD, at the gene level.

Two cis-acting regulatory sequences in the peroxisome proliferator-responsive enhancer region of rat acyl-CoA oxidase gene

To clarify the mechanism of transcriptional induction of the rat liver acyl-CoA oxidase gene by hypolipidemic agents, we searched for cis-acting regulatory sequences in the 5' upstream region of the gene by transfection studies. We found that the sequence between -639 and -472 acts as a peroxisome proliferator-responsive, tissue-specific enhancer. Footprint analysis revealed two protein binding sites in this region. One of these sequences exhibited a positive, whereas the other a negative, regulatory activity in transcription assays.

Molecular cloning and sequence of the complementary DNA encoding human mitochondrial acetoacetyl-coenzyme A thiolase and study of the variant enzymes in cultured fibroblasts from patients with 3-ketothiolase deficiency

Complementary DNAs encoding the precursor of human hepatic mitochondrial acetoacetyl-CoA thiolase (T2) (EC 2.3.1.9) were cloned and sequenced. The cDNA inserts in these clones were 1,518 bases in length when overlapped, and encoded the 427-amino acid precursor of this enzyme (45,199 mol wt). This amino acid sequence included a 33-residue leader peptide moiety and a 394-amino acid subunit of the mature enzyme (41,385 mol wt). The T2 gene expression in fibroblasts from four patients with 3-ketothiolase deficiency was analyzed by Northern blotting. The T2 mRNA in all four cell lines had the same 1.7 kb as that of the control. However, the amounts of T2 mRNA differed: the content was reduced in two cell lines (cases 1 and 3), whereas it was within a normal range in others (cases 2 and 4). Pulse labeling followed by subcellular fractionation revealed that the T2 proteins in the fibroblasts from these patients are present in the mitochondria. These results suggest that different mechanisms are involved in the enzyme defects in the four patients.

Molecular cloning, sequencing, and expression of mouse ferrochelatase

The cDNA encoding mouse ferrochelatase (protoheme ferrolyase, EC 4.99.1.1) was isolated from a mouse erythroleukemia (MEL) cell cDNA library in lambda gt11 expression vector, by immunoscreening with a polyclonal antibody. Two full-length clones containing cDNA inserts of 2.2 and 2.90 kilobases were obtained. These clones have the same entire enzyme coding region, but alternative putative polyadenylation sites in the 3'-noncoding regions. From the deduced primary structure, a putative leader sequence of 53 amino acid residues resulted in a precursor protein of 420 amino acid residues (Mr 47,130) and a mature protein of 367 residues (Mr 41,692). The cDNA allows for the expression of active ferrochelatase by transfected culture cells. RNA blot analysis showed two species of ferrochelatase mRNA consistent with findings of two polyadenylation sites. Both the mRNAs increased by treatment of the MEL cells with dimethyl sulfoxide. The band pattern of the RNA of the mouse liver was the same as that of the MEL cells. Based on these results, we deduce that ferrochelatase in erythroid and hepatic cells can be only of one type.