Precursor of a mitochondrial enzyme accumulates in the cytoplasm of preen glands for a specialized function

Expression, purification, and characterization of human malonyl-CoA decarboxylase

The recombinant human malonyl-CoA decarboxylase (hMCD) was overexpressed in Escherichia coli with and without the first 39 N-terminal amino acids via a cleavable MBP-fusion construct. Proteolytic digestion using genenase I to remove the MBP-fusion tag was optimized for both the full length and truncated hMCD. The apo-hMCD enzymes were solubilized and purified to homogeneity. Steady-state kinetic characterization showed similar kinetic parameters for the MBP-fused and apo-hMCD enzymes with an apparent Km value of approximately 330-520 microM and a turnover rate kcat of 13-28s(-1). For the apo-hMCD enzymes, the N-terminal truncated hMCD was well tolerated over a broad pH range (pH 4-10); whereas the full-length hMCD appeared to be stable only at pH >/= 8.5. Our results showed that the N-terminal region of hMCD has no effect on the catalytic activity of the enzyme but plays a role in the folding process and conformation stability of hMCD.

Cytoplasmic accumulation of a normally mitochondrial malonyl-CoA decarboxylase by the use of an alternate transcription start site

Malonyl-CoA decarboxylase, a normally mitochondrial enzyme, accumulates in the cytoplasm of specialized glands to cause production of multiple methyl-branched fatty acids. Evidence was presented that a single copy of the decarboxylase gene present in the goose genome codes for both the mitochondrial form found in extremely low amounts in the liver and the cytosolic form found in large amounts in uropygial glands. To elucidate how a single gene encodes both forms, the malonyl-CoA decarboxylase gene and the cDNAs for both the mitochondrial (liver) and the cytoplasmic (gland) species were cloned and sequenced. The decarboxylase gene, found in a 21-kb segment of cloned genomic DNA, is composed of five exons of 0.521, 0.118, 0.156, 0.145, and 1.93 kb interrupted by 6.9, 1.5, 0.45, and 9.3-kb introns. Exon 1 revealed two ATGs in frame 150 bp apart. cDNA for the cytoplasmic form and mitochondrial form showed identical nucleotide sequence, except that the latter was longer than the former. The longest cDNA for the cytoplasmic form of the enzyme extended only 44 bp 5' to the second ATG and the position corresponded to the transcription initiation site of the cytoplasmic form revealed by primer extension and RNase protection. The cDNA for the mitochondrial form isolated from the library extended 19 bp further upstream. Primer extension and RNase protection indicated that transcripts for the mitochondrial form initiated upstream from the first ATG. The N-terminal segment of the open reading frame initiated at the first ATG showed an amphipathic signal sequence appropriate for mitochondrial import. A putative full length mRNA for the mitochondrial form of the enzyme when translated in vitro yielded a 55-kDa primary translation product which was processed by removal of about 5 kDa during uptake into goose liver mitochondria. These results strongly suggest that in most tissues transcription initiates 5'- to the first ATG, generating a transcript that would generate a protein with an N-terminal leader for transport into mitochondria. In the uropygial gland the use of an alternate promoter generates transcripts initiated between the two ATGs and the translation product accumulates in the cytoplasm since it lacks a mitochondrial targeting sequence.

Yeast adenylate kinase is active simultaneously in mitochondria and cytoplasm and is required for non-fermentative growth

Displacement of the single copy structural gene for yeast adenylate kinase (long version) by a disrupted nonfunctional allele is tolerated in haploid cells. Since adenylate kinase activity is a pre-requisite for cell viability, the survival of haploid disruption mutants is indicative of the presence of an adenylate kinase isozyme in yeast, capable of forming ADP from AMP and, thus, of complementing the disrupted allele. The phenotype of these disruption mutants is pet, showing that complementation occurs only under fermentative conditions. Even on glucose, growth of the disruption mutants is slow. Adenylate kinase activity is found both in mitochondria and cytoplasm of wild type yeast. The disruption completely destroys the activity in mitochondria, whereas in the cytoplasmic fraction about 10% is retained. An antibody raised against yeast mitochondrial adenylate kinase recognizes cross-reacting material both in mitochondria and cytoplasm of the wild type, but fails to do so in each of the respective mutant fractions. The data indicate that yeast adenylate kinase (long version, AKY2) simultaneously occurs and is active in mitochondria and cytoplasm of the wild type. Nevertheless, it lacks a cleavable pre-sequence for import into mitochondria. A second, minor isozyme, encoded by a separate gene, is present exclusively in the cytoplasm.

Developmental pattern of the expression of malonyl-CoA decarboxylase gene and the production of unique lipids in the goose uropygial glands

The abundant fatty acid synthase in the uropygial gland of goose generates multimethyl-branched fatty acids as the major product because of the unique presence of the cytoplasmic malonyl-CoA decarboxylase which assures that only methylmalonyl-CoA is available to the synthase. If this conclusion is valid, the developmental pattern of expression of the gene for this tissue-specific decarboxylase should correlate with the appearance of other lipogenic enzymes and the production of the unique lipids. To test this possibility the levels of the decarboxylase, acetyl-CoA carboxylase, and fatty acid synthase in the gland of the embryonic and neonatal goose were measured by immunodiffusion and immunoblot assays for the proteins as well as the enzyme assays for the catalytic activities. Malonyl-CoA decarboxylase appeared several days before hatching as did the other two lipogenic enzymes and reached half-maximal levels by hatching. The levels of expression of the malonyl-CoA decarboxylase gene and cytoplasmic actin gene, which is not expected to be developmentally regulated, were measured by dot-blot analysis using cloned cDNA for the two proteins. The decarboxylase transcripts appeared 4 days prior to hatching and reached maximal levels by hatching, whereas the levels of cytoplasmic actin gene transcripts showed very little change. The appearance of oil droplets in the glands was clearly seen soon after hatching. These results show that malonyl-CoA decarboxylase gene expression is developmentally regulated in a manner consistent with its proposed role in the synthesis of the unique lipids of the uropygial gland.

Amino-terminal extension generated from an upstream AUG codon is not required for mitochondrial import of yeast N2,N2-dimethylguanosine-specific tRNA methyltransferase

The TRM1 gene of Saccharomyces cerevisiae is necessary for the N2,N2-dimethylguanosine modification of both mitochondrial and cytoplasmic tRNAs. The DNA sequence of the TRM1 locus and the 5' ends of mRNAs expressed from this gene have been determined. The majority of the 5' ends map within a large open reading frame between two in-frame ATGs at positions +1 and +49. A small fraction of the 5' ends are located upstream of the first ATG. Both AUGs of the TRM1 mRNAs are used to initiate translation, and two forms of N2,N2-dimethylguanosine-specific tRNA methyltransferase, which differ by an amino-terminal extension of 16 amino acids, are made. Mitochondrial tRNAs are modified when the initiation of translation is limited to one or the other of the AUGs, suggesting that the amino-terminal extension is not necessary for import of the protein into mitochondria. Mitochondrial targeting information must, therefore, be located in a region of N2,N2-dimethylguanosine-specific tRNA methyltransferase that is found in both forms of the enzyme.

In vitro translation of mushroom tyrosinase

Mushroom tyrosinase was purified and antibodies prepared against the holo enzyme and a protein of 26,000 daltons. Both antibodies recognized the large subunit of the enzyme but only one recognized the 26,000 dalton protein. Poly A+ mRNA was isolated from mushrooms, translated in vitro, and a 41,000 dalton protein immunoprecipitated from the translation mix with either antibody. This 41,000 dalton protein presumably corresponds to the large subunit of the holoenzyme. Antibodies against the holoenzyme also immunoprecipitated another translation product with a molecular weight of 15,000 daltons corresponding to the small subunit of the holoenzyme. These results suggest that each subunit may be coded for by different genes and undergo posttranslational processing.

In vitro translation of the major capsid polypeptide from Ustilago maydis virus stain P1

Double-stranded RNA (dsRNA) from Ustilago maydis virus strain P1 was translated in vitro using a nuclease-treated rabbit reticulocyte lysate system. Following heat denaturation of the H2 double-stranded RNA segment in 90% dimethyl sulfoxide and incubation in the cell free extract, a primary translation product was observed which showed the same molecular weight and co-migrated with viral coat protein on 10% SDS-polyacrylamide gels. The in vitro product of the H2 dsRNA segment could also be immunoprecipitated with antibodies prepared against viral coat protein. Limited proteolysis of the in vitro product and authentic viral coat protein using Staphylococcus aureus V8 protease produced similar peptide patterns on SDS gels. In vitro translation products from other dsRNA segments that make up the P1 viral genome could not be precipitated by antibody to viral coat protein. These results complement the genetic data that indicated that information for coat formation and maintenance was contained within the H segments of dsRNA.

Transport of proteins into mitochondria

There is still much that is obscure concerning the transport of proteins into or through the mitochondrial membrane systems. In addition, as pointed out previously, it is unlikely that the details of the process are the same for proteins destined for different compartments of the organelle. A brief summary of the process for matrix proteins might be as follows: The proteins are synthesized on free polysomes as precursors of higher molecular weight than the native forms. These precursors are liberated into the cell cytosol and subsequently translocated into the mitochondria. This timing might be different in yeast under some circumstances, synthesis being completed in association with the mitochondria. The precursors interact with a receptor in the outer mitochondrial membrane interaction being mediated by the presequences of the precursors. The presequences therefore act as addressing signals as well as possibly playing a role in one or all of (a) solubilization of precursors, (b) prevention of premature assembly into multimeric structures, or (c) maintenance of nonnative configurations required for transport. Interaction occurs with a second receptor, this time in the inner membrane of the mitochondria, interaction being with multiple sites in the polypeptide chain. Transport across the inner membrane then occurs, this transport depending on a transmembrane electrochemical gradient of which the proton component is the essential part. Transport is accompanied or followed by proteolysis of the prepiece, and formation of the native structure. While steps 1 and 2 of this sequence can be considered well established, the remaining steps are still poorly understood or purely hypothetical. Nevertheless, this sequence of events is consistent with known facts about the process and provides a framework for future investigations.