The yeast ARG7 gene product is autoproteolyzed to two subunit peptides, yielding active ornithine acetyltransferase.

Yeast ornithine acetyltransferase has been purified from total yeast extracts as a heterodimer of two subpeptides (Liu, Y., Van Heeswijck, R., Hoj, P., and Hoogenraad, N. (1995) Eur. J. Biochem. 228, 291-296), confirmed to derive from a single ARG7-encoded precursor (Crabeel, M., Abadjieva, A., Hilven, P., Desimpelaere, J., and Soetens, O. (1997) Eur. J. Biochem. 250, 232-241). By Western immunoblotting, we show that Arg7p is also present as two subpeptides in isolated mitochondria, but that processing occurs before targeting to the mitochondria: deletion of the N-terminal leader peptide results in cytosolic accumulation of N-Arg7p, whereas C-Arg7p partially reaches the organelle by itself. When artificially co-expressed from separate genes, the two subpeptides can complement an arg7 mutation; ornithine acetyltransferase activity is measurable. Maturation of Arg7p occurs at threonine 215 (N-side), in the region most conserved among the 17 ornithine acetyltransferases characterized. Changing this conserved residue to alanine completely abolishes maturation. Furthermore, Arg7p is both processed and active in Escherichia coli, a heterologous background, and is also cleaved in vitro when produced by coupled transcription/translation in a reticulocyte lysate. Together, these data suggest classic autoproteolysis initiated by threonine 215. Most importantly, maturation is required for the enzyme to be functional, since the T215A substitution mutant is catalytically inactive and incapable of genetic complementation, despite its correct targeting to the mitochondria.

Yeast ornithine acetyltransferase has been purified from total yeast extracts as a heterodimer of two subpeptides (Liu, Y., Van Heeswijck, R., Hoj, P., and Hoogenraad, N. (1995) Eur. J. Biochem. 228, 291-296), confirmed to derive from a single ARG7-encoded precursor (Crabeel, M., Abadjieva, A., Hilven, P., Desimpelaere, J., and Soetens, O. (1997) Eur. J. Biochem. 250, 232-241). By Western immunoblotting, we show that Arg7p is also present as two subpeptides in isolated mitochondria, but that processing occurs before targeting to the mitochondria: deletion of the N-terminal leader peptide results in cytosolic accumulation of N-Arg7p, whereas C-Arg7p partially reaches the organelle by itself. When artificially co-expressed from separate genes, the two subpeptides can complement an arg7 mutation; ornithine acetyltransferase activity is measurable. Maturation of Arg7p occurs at threonine 215 (N-side), in the region most conserved among the 17 ornithine acetyltransferases characterized. Changing this conserved residue to alanine completely abolishes maturation. Furthermore, Arg7p is both processed and active in Escherichia coli, a heterologous background, and is also cleaved in vitro when produced by coupled transcription/translation in a reticulocyte lysate. Together, these data suggest classic autoproteolysis initiated by threonine 215. Most importantly, maturation is required for the enzyme to be functional, since the T215A substitution mutant is catalytically inactive and incapable of genetic complementation, despite its correct targeting to the mitochondria.
Ornithine is an important intermediate in the arginine biosynthetic pathway. Its synthesis in five steps starts with acetylation of the ␣-amino group of glutamate by acetylglutamate synthase (preventing cyclization of glutamate-␥-semialdehyde as occurs in proline biosynthesis) and ends with deacetylation of acetylornithine. In Escherichia coli and a few other bacteria using the linear pathway of arginine biosynthesis, the latter step is catalyzed by acetylornithinase. Most often, however, a more economic cyclic pathway is used in which ornithine acetyltransferase transfers the acetyl group of acetylornithine to glutamate, thereby regenerating acetylglutamate. In this case, the first enzyme of the pathway, acetylglutamate synthase, plays merely an anaplerotic role. Ornithine acetyltrans-ferase (N-acetyl-L-ornithine:L-glutamate N-acetyltransferase, EC 2.3.1.35) is encoded by the bacterial argJ genes and, in the yeast Saccharomyces cerevisae, by ARG7 (for reviews, see Refs. 1 and 2). In yeast, the acetylated derivatives cycle occurs in the mitochondria (3); ornithine produced in the matrix requires the mitochondrial ornithine carrier Arg11p for its export to the cytosol, where it is further processed to arginine (4).
Ornithine acetyltransferase was purified to homogeneity in the laboratory of Hoogenraad (5). On non-dissociating polyacrylamide gels, purified Arg7p migrated as a single band corresponding to an apparent molecular mass of 57 kDa. The protein in this band, however, turned out to be a heterodimer, since it resolved, when electrophoresed on denaturing polyacrylamide gels, to two subunit peptides with respective apparent molecular masses of 26.3 and 30.7 kDa. The evidence further suggested that the two peptides were derived from a single precursor since (i) a single in vitro translation product of approximately 57 kDa was immunoprecipitated by an antibody raised against the purified small subunit; (ii) the N-terminal sequences of the purified small and large subunits were respectively 40% and 45% identical to the N-terminal and central amino acid sequences of the homologous protein Arg J of Neisseria gonorrhoeae (5).
Our characterization of the ARG7 gene of S. cerevisiae (6) provided confirmation that ornithine acetyltransferase is indeed encoded by a single nuclear gene. It further revealed, by comparison with the N-terminal amino acid sequence determined by Hoogenraad, the presence of a short (8-residue), cleavable mitochondrial targeting leader peptide at the N-terminal end of the pre-enzyme. The calculated molecular mass of the encoded Arg7 protein is 47.8 kDa, lower than the apparent molecular mass determined by gel electrophoresis in Hoogenraad's laboratory. From the N-terminal sequence of the 30.7-kDa C-Arg7p subpeptide determined by Hoogenraad (XLLG-FIVTD . . . ), it emerges that the proteolysis site lies between alanine 214 and threonine 215 of the DNA-derived amino acid sequence.
The Arg7p acetyltransferase also displays some acetylglutamate synthase activity, enabling it, when the ARG7 gene is overexpressed in either yeast or E. coli, to complement mutations leading to the absence of acetylglutamate synthase activity (6).
Here we have used various HA epitope-tagged derivatives of Arg7p to re-examine the protein's structural properties. We confirm that the enzyme is detected as two subunit peptides even in extracts of isolated mitochondria. We show that processing (i) precedes targeting to the mitochondria, (ii) occurs in a heterologous E. coli background, (iii) occurs in vitro, when Arg7p is produced in a coupled transcription/translation system, and (iv) depends on threonine 215. Together, the data strongly suggest that maturation occurs by autoproteolysis.
* 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.
‡ Recipient of a doctoral fellowship granted by Vlaams Instituut voor de bevordering van het Wetenschappelijk-Technologisch Onderzoek in de Industrie. HIS3), two analogous vectors designed to make N-terminal translational fusions with the HA epitope (7). The full-length ARG7 ORF followed by its stop codon was inserted downstream from the HA coding sequence in pHP1, using a NotI/EcoRI PCR-fragment amplified with primers HP3/AA4. The inserts in pAA3 and pHP3, lacking, respectively, residues 2/3/4 and 2-16, were amplified by PCR using primers AA3/AA4 and HP5/AA4, respectively.
Plasmid pHP4 was derived from vector pYX223 (2-HIS3) (from R&D Systems) and obtained after insertion of an EcoRI/MluI restriction fragment obtained from a PCR-fragment amplified with oligonucleotide primers HP8/HP9. The forward primer HP8 bears an EcoRI restriction site followed by an AAA ATG (an initiator in an efficient translation context), and it primes amplification of the ARG7 ORF from codon 17 onward. The reverse primer HP9 bears a MluI site followed by a stop codon, the nine codons encoding the HA epitope, and a sequence complementary to the 3Ј end of the ARG7 ORF. In this vector expression is driven by the GAL1 promoter. The plasmids described in this section are schematized in Fig. 1.

Construction of Plasmids
Expressing N-Arg7p and C-Arg7p as Separate Peptides-Plasmid pHP6 was constructed by cloning a PCR fragment encoding amino acids 1-214 of Arg7p under the control of the GAL1 promoter in vector pYX213 (2, URA3) (from R&D Systems). PCR amplification was carried out on plasmid pYeA7-1 as DNA template using oligonucleotides AA37 (bearing an EcoRI restriction site) and AA38 (bearing a BamHI restriction site, followed by a stop codon).
Plasmid pHP7 was constructed similarly but using vector pYX223 (2-HIS3) from R&D Systems and the PCR fragment encoding amino acids 215-441 of Arg7p followed by its stop codon. This fragment was obtained using primers AA39 (bearing an EcoRI restriction site) followed by AGCATG (providing a properly contexted initiation codon) and AA40 (bearing a BamHI restriction site).
The PCR inserts of both pHP6 and pHP7 and the novel junctions were completely DNA sequenced to check for the absence of PCRinduced errors.
The control plasmid pHP9 was constructed by inserting into vector pYX213 the EcoRI-BamHI fragment obtained from a PCR fragment amplified on plasmid pYeA7-1 using primers AA37 and AA40.
Construction of the T215A Mutant of Arg7p-Plasmid pHP12 (schematized in Fig. 1) was generated by recombinant PCR. Two PCR fragments were amplified, consisting of overlapping portions of the ARG7 ORF (encoding, respectively, amino acids 1-220 and 211-441) followed by the sequence encoding the nine residues of the HA epitope. The fragments were amplified on plasmid pAA2 DNA using the primer pairs AA1/HP57 and HP58/AA2. HP57 and HP58 were complementary oligomers bearing on the sense strand the sequence GCC instead of ACC, so that threonine 215 of the wild type would be replaced by an alanine. The two overlapping PCR fragments were purified on Wizard columns (Promega) and used in a self-priming PCR amplification system, together with the external oligomers AA1 and AA2. Thus, exponentially increasing amounts of mutant DNA (encoding T215A Arg7p) were produced, together with a small amount of wild-type ARG7 DNA linearly amplified from the pAA2 template. After purification, this DNA was cut with BamHI and NotI (5Ј add-on restriction sites present on AA1 and AA2), purified, and ligated with similarly restricted and purified pYeF2 vector. Plasmid pHP12 was obtained from an E. coli transformant and shown by DNA sequencing of the full ARG7 ORF to contain no other modification than the intended G to A mutation causing threonine 215 to be replaced by an alanine.
Construction of Plasmids pAA31 and pAA32-Wild-type ARG7 and mutant arg7 encoding T215A Arg7p were PCR-amplified on plasmids pYeA7-1 and pHP12, respectively, using oligonucleotide AA69 as forward primer and AA70 as reverse primer. They contain, respectively, an NdeI site followed by an ATG and a BamHI site followed by a stop codon. The two fragments were inserted into the pET-19b vector from Novagen, under the transcriptional control of the IPTG-inducible T7-RNA polymerase and in translational frame with an N-terminal, enterokinase-removable 10ϫHis tag. The resulting plasmids were called pAA31 and pAA32, respectively.
Yeast and E. coli Total Protein Extracts for Western Blots-Yeast cells were grown to OD 1 (or less) in 100 ml of minimal ammonium minimal medium (10) containing 2% galactose as the carbon source (to induce GAL-promoter-driven gene expression) and the required amino acids (50 mg/liter each). After centrifugation at 4°C and washing by resuspending in water, the cell pellet was resuspended in 200 l (for a harvesting OD of 1, or proportionately less, otherwise) of ice-cold lysis buffer (5 mM Tris-HCl, pH 8, 0.3 M NaCl, 0.5% SDS, 12 mM EDTA, 0.5% ␤-mercaptoethanol, and "complete" antiprotease mixture from Roche Molecular Biochemicals). The suspension was transferred to a microcentrifuge tube containing the same volume of glass beads (BDH). Resuspended cells were vortexed for 1 min, then cooled on ice for 1 min. This was repeated five times. The cell debris were then centrifuged for 10 min at 12,000 rpm in a refrigerated microcentrifuge. A 10-l aliquot of supernatant was diluted in 90 l of water for determination of the protein concentration by the Folin procedure. The rest of the supernatant was immediately mixed with an equal volume of 2ϫ loading buffer (100 mM Tris-HCl, pH 6.8, 10% ␤-mercaptoethanol (freshly added), 5% SDS, 0.2% bromophenol blue, 20% glycerol), incubated for 5 min in a boiling water bath, and either used directly for SDS-PAGE or stored at Ϫ70°C (in which case, it was reheated in a boiling bath before loading). Such extracts usually contained about 25 mg of protein/ml. E. coli cells were grown in 10 ml of rich medium (2ϫ YT) containing, as required, either 50 mg/liter ampicillin or ampicillin and 30 mg/liter chloramphenicol. At OD 0.4, 2 mM IPTG was added. After induction for 3-5 h, the cells were collected by centrifugation, resuspended in 2.5 ml of buffer (50 mM Tris-HCl, pH 7.5), sonicated for 3 min (Branson Sonifier, model 250), and microcentrifuged (10 min, 12,000 rpm, 4°C). Aliquots (10 and 20 l) of these extracts were used for SDS-PAGE.
Preparation of Mitochondrial and Cytoplasmic Cell Fractions-The method of Daum et al. as optimized by Yaffe (11) was used to isolate mitochondria, except that "complete" antiprotease mixture (Roche Molecular Biochemicals) was added to the lysis buffer. We usually started with 2 liters of yeast cells grown on galactose medium to OD 2.
SDS-PAGE and Western Immunoblotting-Standard protocols were used for Western immunoblotting (as in Ref. 6). SDS-PAGE was performed on 10%, 12%, or 15% SDS-polyacrylamide gels. Rainbow molecular size standards from Amersham Pharmacia Biotech were used as markers. Proteins were transferred to Millipore Immobilon-P membranes by electroblotting with a Bio-Rad cell. Mouse monoclonal anti-HA (12CA5) from Roche Molecular Biochemicals was used to recognize HA epitope-tagged proteins; this was followed by detection of chemiluminescence with Roche's Western blotting kit based on peroxidase conjugates of anti-mouse IgG.
Coupled in Vitro Transcription and Translation-The Promega TNT quick-coupled transcription/translation system was used according to the manufacturer's instructions. For radioactive labeling, we used Redivue L-[ 35 S]methionine from Amersham Pharmacia Biotech (1000 Ci/ mmol at 10 mCi/ml).
Ornithine Acetyltransferase Activity Assay-Yeast cell cultures, cell extracts, and enzymatic reaction conditions were as described previously (6). A modified step was separation of the acetylglutamate formed in the reaction mixture from the [ 14 C]glutamate added to the reaction mixture, by means of prefilled Dowex AG 50-W (-X8 resin; 200 -400mesh) chromatography columns from Bio-Rad.
Spot Test Complementation Assay-The different yeast strains were grown overnight on a selective minimal medium. The cells were diluted in fresh medium the next morning and allowed to grow to exactly OD 0.5. Then 1 ml of each culture was centrifuged, washed with sterile water, recentrifuged, and resuspended in 1 ml of sterile water. This suspension was serially diluted down to 10 Ϫ5 , and 10-l aliquots of each dilution were spotted in a row on a Petri dish containing the relevant medium. Plates were incubated for 2 days at 30°C.

Arg7p Is Obtained as Two Subpeptides in Yeast Total Protein Extracts Prepared under Conditions That Minimize Trivial
Proteolytic Digestion-In Hoogenraad's laboratory, ornithine acetyltransferase was isolated as a heterodimer from a commercial preparation of compressed yeast. We wanted to test whether two subpeptides would also be obtained from ornithine acetyltransferase purified from our ⌺1278b-derived laboratory strains and from mutant strain BJ5459 devoid of vacuolar proteinases A and B. The extracts were prepared in the presence of a mixture of protease inhibitors.
To monitor the ARG7 gene product(s) by Western immunoblotting, we constructed a series of gene fusions expected to produce HA epitope-tagged derivatives of the enzyme. pAA2 yields Arg7p bearing the epitope at its C terminus, pHP1 yields Arg7p tagged at the intact N terminus; pAA3 and pHP3 also encode N-tagged Arg7p, but with partial (pAA3) or total (pHP3) truncation of the mitochondrial leader peptide (Fig. 1).
Western immunoblots showed only subunit peptides. When the harbored plasmid was pAA2, the only peptide to appear was C-Arg7p (apparent molecular mass: 31 kDa), the subunit constituting the C-terminal portion of Arg7p; when the plasmid present was pAA3 or pHP3, the detected polypeptide was N-Arg7p (apparent molecular mass: 26 kDa), i.e. the smaller, N-terminal portion of Arg7p (data not shown, but see further below for illustrated Western immunoblotting experiments).
No product was detected when the tag was on the intact N terminus of Arg7p (use of plasmid pHP1). This was as expected, since the N-terminal mitochondrial targeting peptide is cleaved off during targeting to the mitochondria.
Thus, Arg7p is processed to two subunit peptides in various yeast genetic backgrounds. A very small quantity of uncleaved precursor, with an estimated apparent molecular mass of 48 kDa, was sporadically observed in overloaded lanes. In some experiments, C-Arg7p appeared as a double band.
Arg7p Is Processed Independently of Its Targeting to the Mitochondrial Matrix-To analyze the mechanism of the proteolytic cleavage of Arg7p, we examined the subcellular localization of N-Arg7p and C-Arg7p produced from wild-type Arg7p and from versions lacking the mitochondrial leader peptide. We expected to observe cytosolic accumulation of the full-length Arg7p precursor if its maturation were catalyzed by a mitochondrial processing peptidase. We used C-terminally HA epitope-tagged derivatives of Arg3p (ornithine carbamoyltransferase, 39 kDa) and Arg8p (acetylornithine transaminase, 47 kDa) as markers of the cytosolic and mitochondrial fractions, respectively. Fig. 2A shows what we actually observed when the plasmid insert (pHP3) encoded an N-terminally tagged Arg7p lacking the leader peptide. As expected, the detected polypeptide did appear to accumulate exclusively in the cytosol, but it was not the full-length precursor. Instead, what we detected was the 26-kDa N-Arg7p subpeptide. Thus, deletion of the first 16 Nterminal amino acids does, as anticipated, prevent targeting to the mitochondrial matrix, but it does not impede processing of Arg7p to its two subunit peptides.
It was checked, moreover, that the C-terminally tagged 31-kDa C-Arg7p produced from pAA2 was mitochondrially located, as expected (not shown here, but see control lanes 3 and 4 of Fig. 5B).

When the Encoded Product Is a C-terminally HA Epitopetagged Arg7p Lacking the N-terminal Leader Peptide, the Large Subunit Appears to Localize Principally to the Mitochondria-If
Arg7p is processed to two subpeptides prior to its import into the mitochondrion, the C-Arg7p peptide might be able to target itself to the mitochondrial matrix. To test this possibility, we monitored the fate of C-terminally tagged Arg7p lacking the entire leader peptide (pHP4). As shown in Fig. 2B, only the 31-kDa C-Arg7p was detected, and a considerable fraction of it displayed a mitochondrial location (about 60% was found in the matrix). This confirms that the leader peptide is not required for processing and further shows that C-Arg7p can indeed target itself, albeit partially, to the mitochondrion. This moderately efficient mitochondrial targeting might be mediated by the stretch comprising amino acids 228 -240 (i.e. amino acids 14 -26 of the predicted N terminus of C-Arg7p, considering that C-Arg7p begins with threonine 215). This sequence does display features of a potential non-cleavable mitochondrial targeting sequence, namely a predicted amphipathic ␣-helical structure with a large hydrophobic moment and a positively charged opposite side with several hydroxylated amino acids (Fig. 3). It is reasonable to suppose that in a wild-type strain C-Arg7p is completely targeted to its correct mitochondrial site thanks to its association with N-Arg7p, which possesses a bona fide 8-residue cleavable N-terminal targeting peptide (6).
Co-expression of Two Artificial Genes, One Expressing N-Arg7p and the Other, C-Arg7p: the Reconstituted Enzyme Can Complement an arg7 Mutation and Is Active in Vitro-If Arg7p is active in vivo as a heterodimer consisting of an N-Arg7p and a C-Arg7p subpeptide, a functional enzyme might possibly be reconstituted from the two peptides produced separately in the same cells through co-expression of two distinct artificial genes. To test this possibility, we constructed pHP6, expressing a "gene" encoding N-Arg7p under the control of the GAl1 promoter of plasmid pYX213 (2, HIS3), and pHP7, expressing a "gene" encoding C-Arg7p under the control of the Gal1 promoter of plasmid pYX223 (2,URA3). We then transformed strain14S40c (ura3 Ϫ , his3 Ϫ , arg7 Ϫ ) with all combinations of empty vectors and of the plasmids just mentioned. Table I shows that the arg7 Ϫ strain's bradytrophy can be almost fully remedied when both Arg7p subunits are produced in the same cells. Neither subunit produced separately has any complementing effect. Furthermore, ornithine acetyltransferase activity was detectable only in cases of complementation. The measured enzyme activity amounted to 2.5% of the activity displayed by cells expressing a full-length ARG7 gene likewise driven by the GAl1 promoter (pHP9), and to about 10% of the activity measured in cells expressing ARG7 from its own promoter.
These data show that the two subunits that make up ornithine acetyltransferase can, to some extent at least, fold nor-   Roche's Rainbow molecular mass markers were used as molecular size standards (not shown here, but as in Fig. 6A). mally as separate domains with sufficient affinity for each other to allow their proper association and correct targeting to the mitochondria. The fact that complementation is only partial and enzyme activity not fully restored might have several causes, among which the presence of a terminal methionine in the C-Arg7 peptide seems to us the most probable. No activity was detected when we tried to reconstitute the enzyme by mixing two separate extracts, one presumably containing N-Arg7p and the other C-Arg7p. This suggests that at least one of the subunits might be unstable when produced singly.
The Fully Conserved Threonine 215 at the N Terminus of C-Arg7p Is Absolutely Required for Processing-A large number of ornithine acetyltransferases have been characterized to date and their alignment (Fig. 4) shows that the most conserved region in this family overlaps with the maturation site predicted from the data by Hoogenraad's team. Threonine 215 in particular, the predicted N-terminal amino acid of C-Arg7p, is conserved in all members of the family. To evaluate the role of this residue in ornithine acetyltransferase maturation, we constructed a T215A mutant protein, i.e. one in which threonine 215 is changed to an alanine. The mutant enzyme, produced from plasmid pHP12 in strain BJ5459, was shown in total protein extracts to accumulate exclusively as an uncleaved precursor (Fig. 5A).However, it was shown by subcel-  mitochondria (panel B). Panels A and B are two distinct 12% SDS-polyacrylamide gel electrophoreses. Panel A, total protein extracts of yeast strains bearing plasmid pAA2 expressing wild-type Arg7p HA-tagged at its C terminus (lane 1) or plasmid pHP12 expressing similarly tagged T215A Arg7p (lane 2) or plasmids pHP3 and pOS13 expressing, respectively, an arg7p deleted of its mitochondrial targeting peptide and HA-tagged at its N terminus and Arg8p HA-tagged at its C terminus. Panel B, mitochondrial (Mt) and cytosolic (Cy) extracts of strains bearing plasmid pHP12 (lanes 1 and 2) or plasmid pAA2 (lanes  3 and 4). Roche's Rainbow molecular mass markers were used as molecular size standards (not shown here, but as in Fig. 6A). lular fractionation to be correctly targeted to the mitochondrion (Fig. 5B).
The fact that threonine 215 is absolutely required for Arg7p processing suggests that an autoproteolytic mechanism could be responsible for maturation. Such self-processing events mediated by the -OH group of a threonine or serine, or by the -SH group of a cysteine, have been shown to be quite frequent (12,13). If maturation of Arg7p does rely on autoproteolysis, it should not be restricted to the yeast background.
Arg7p, Which Is Functional When Produced in E. coli, Is Processed in This Heterologous Background as Well-If Arg7p maturation is autocatalytic rather than catalyzed by a distinct endoprotease, the enzyme could be correctly processed in a heterologous context as well. We have shown previously by complementation and enzymatic assays that Arg7p is functional in E. coli when its ORF is expressed from a prokaryotic promoter (6). To see whether Arg7p maturation occurs in this heterologous background, we produced E. coli transformants harboring either the empty pTrc99A vector or the derived pAA7 plasmid expressing functional Arg7p. The strains were induced with IPTG, total protein extracts prepared, and SDS-PAGE was used to separate the extracted proteins. Coomassie Blue staining revealed two extra protein bands appearing only in the presence of pAA7 (estimated apparent molecular mass: 31 and 26 kDa). This shows that Arg7p is cleaved in E. coli as well (data not shown). We repeated the same experiment with E. coli strains containing either the vector pET19b or a derived plasmid expressing the gene for either wild-type Arg7p (pAA31) or the T215A variant, under the control of an IPTGinducible T7-RNA polymerase. As shown in Fig. 6A, the wildtype enzyme is processed to the usual 31-and 26-kDa subpeptides, while the mutant enzyme accumulates as an unprocessed precursor with an apparent molecular mass of 47 kDa, corresponding to the molecular mass of 47.8 kDa calculated for Arg7p on the basis of the DNA sequence.
Arg7p Is Cleaved in an in Vitro System-Although Liu et al. (5) obtained results showing that ornithine acetyltransferase produced from yeast mRNA in a reticulocyte lysate was uncleaved, showing up as a 57-kDa precursor, we decided to reassess the structure of the enzyme produced in vitro.
Transcription/translation reactions were carried out in a rabbit reticulocyte system using pET19b, pAA31, or pAA32 plasmid DNA as the template in the presence of 35 S-labeled methionine. As shown in Fig. 6B, only a 47-kDa precursor (not a 57-kDa protein) was obtained from the mutant template, while the wild-type template gave rise to 31-and 26-kDa subunits, with only a faint band at 47 kDa. The 31-kDa band appeared fainter than the 26-kDa band. This is explained in part by the presence of only 4 methionines in C-Arg7p versus 8 in N-Arg7p. Furthermore, C-Arg7p appeared as a double band, a fact not yet understood.
The T215A Mutant Displays No Detectable Ornithine Acetyltransferase Activity and Does Not Complement an arg7 Mutation-In enzyme assays, we measured the ornithine acetyltransferase activity of the yeast mutant strain JD1(arg7⌬) harboring either plasmid pAA2, encoding wild-type Arg7p, or plasmid pHP12, encoding the T215A Arg7p mutant. After growth on galactose medium to induce GAL1-promoter-driven expression of the enzyme, we detected no activity in the extract expected to contain the mutant Arg7p. The background level was the same as for the same strain carrying the empty pYeF2 cloning vector. This is in sharp contrast with the activity measured (300 nmol of acetylglutamate formed/min/mg of protein) when extracts of pAA2-harboring cells were used (Table II).
We have shown previously that a strain deleted of the ARG7 ORF has an arginine-leaky phenotype (6). In keeping with the results of the enzyme assays, we detected no improvement of  7. The mutant protein T215A Arg7p is unable to complement a strain deleted of its chromosomal ARG7 gene. Yeast strain JD1 transformed with the empty vector pYeF2 (row 1), with pAA2 expressing wild-type Arg7p (row 2), or with pHP12 expressing the mutant protein T215A Arg7p was grown on inducing galactose medium and spotted after serial dilution on solid galactose medium without (left) or with (right) added arginine. Arg7p Activation by Autoproteolysis