THE JOURNAL OF CHEMISTRY Vol No. 4. by U.S.A. …changing the affinity of the NAM2 protein for this maturase, and that in the wild-type situation the NAM2 protein would interact with

THE JOURNAL OF BIOLOGICAL CHEMISTRY (c) 1991 by The American Society for Biochemistry and Molecular Biology, Inc

Vol . 266, No. 4. , Issue of February 5, pp. 2537-2541.1991 Printed in U.S.A.

Purification and Characterization of the Saccharomyces cerevisiae Mitochondrial Leucyl-tRNA Synthetase*

(Received for publication, August 6, 1990)

Wlodzimerz ZagorskiSil, Bertrand CastaingSII, Christopher J. Herbert$, Michel LabouesseS, Robert Martin**, and Piotr P. SlonirnskiS From the $Centre de Ginetique Moleculaire, Laboratoire Propre du Centre National de la Recherche Scientifique associi a 1’Uniuersiti Pierre et Marie Curie, Gif-sur-Yvette F-91198, France; the §Institute of Biochemistry and Biophysics, Polish Academy of Sciences, Rakowiecka 36 Str., Warsaw, Poland, and the **In.stitut de Biologie Moleculaire et Cellulaire du Centre National de la Recherche Scientifique, 15 Rue Decartes, Strasbourg F-67084, France

We have purified the product of the NAM2 gene, the mitochondrial leucyl-tRNA synthetase, from yeast mi- tochondria. The purified protein cross-reacts with an- tibodies raised against the product of a LacZJNAM2 gene fusion and antibodies raised against the purified Escherichia coli leucyl-tRNA synthetase. The mass as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis is about 100 kDa, consistent with the size predicted by the gene sequence (102 kDa). The N-terminal sequence of the protein has been deter- mined and shows that the first nine amino acids pre- dicted by the gene sequence have been removed, prob- ably during transport into the mitochondria.

Pre-mRNA splicing in yeast mitochondria is a complex process that requires proteins encoded by both the nuclear and the mitochondrial genome. The nuclear gene N A M 2 of Saccharomyces cerevisiae was originally identified and defined by dominant alleles able to suppress mutations that inactivate the maturase encoded by the fourth intron of the mitochon- drial cytochrome b gene (b14 maturase) (Dujardin et al., 1980). This maturase is required for the excision of two introns, its own intron b14, and the fourth intron (aI4) of the mitochon- drial gene encoding subunit I of cytochrome oxidase ( c o d ) (Dhawale et al., 1981; De La Salle et al., 1982; Labouesse and Slonimski, 1983; Labouesse et al., 1984). These two introns are similar, and both encode proteins. The a14 protein has an endonuclease activity (Delahodde et al., 1989; Wenzlau et al., 1989), but despite considerable homology with the b14 matu- rase it does not appear to act as an RNA maturase under normal conditions. However, a single substitution causing a glutamate to be replaced by a lysine activates a “latent” maturase activity in the a14 protein, which is then able to participate in the excision of both a14 and b14 (Dujardin et al., 1982). Furthermore, genetic studies on the N A M 2 sup- pressor alleles have indicated that the protein encoded by the a14 intron is essential for suppressor activity (Dujardin et al., 1983). This led to the hypothesis that the effect of the sup- pressor mutations is to enable the NAMB protein to activate

* This work was supported by the Centre National de la Recherche Scientifique and the Ligue Nationale Francaise Contre le Cancer. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

ll Recipient of a Poste rouge du Centre National de la Recherche Scientifique.

I( Recipient of a grant from the Ministre de Recherche et Technol- ogie.

the a14 maturase by a protein-protein interaction, possibly by changing the affinity of the NAM2 protein for this maturase, and that in the wild-type situation the NAM2 protein would interact with the b14 maturase (Herbert et al., 198813). This hypothesis is supported by the observation that when cloned on a multicopy plasmid the wild-type N A M 2 gene has a low but measurable suppressor activity (Labouesse et al., 1987).

The N A M 2 gene has been cloned and sequenced; analysis of the sequence shows that it could encode a protein of 894 amino acids with a molecular mass of 102 kDa. The deduced protein sequence shows considerable homology with the Esch- erichia coli leucyl-tRNA synthetase (LRS),’ 34% identical residues and 20% conservative replacements in 839 amino acids that can be unambiguously aligned, suggesting that the N A M 2 gene might encode the mitochondrial leucyl-tRNA synthetase (mLRS). This was confirmed by the observation that the levels of the mLRS in crude mitochondrial extracts varied with the copy number of the N A M 2 gene and that no activity was detected in extracts made from a strain in which the gene was deleted. Finally, antibodies raised to the C- terminal part of the NAMB protein inhibit mLRS activity in crude extracts (Herbert et al., 1988a). Independently, Tzago- loff et al. (1988) cloned the S. cerevisiae mLRS, and the coding sequence of the gene was identical to the sequence of the N A M 2 gene.

To date, two aminoacyl-tRNA synthetases have been shown to be involved in fungal pre-mRNA splicing, the tyrosyl-tRNA synthetase in Neurospora crmsa (Akins and Lambowitz, 1987) and the mLRS in S. cereuisiae (Herbert et al., 1988a). Thus, in addition to their aminoacyl-tRNA synthetase functions it is reasonable to assume that these proteins are able to bind to the pre-mRNA, and in the case of the mLRS, possibly form a complex with a pre-mRNA maturase. To further our studies on the role of the mLRS in mitochondrial pre-mRNA splicing and its interactions with other components of the splicing system we decided to purify the protein. Here we report a procedure for a rapid and simple purification of the mLRS, a preliminary characterization of the enzyme, and the deter- mination of the N-terminal sequence of the purified protein.

MATERIALS AND METHODS

Yeast Strains and Growth Medium-The rho+,mit+ wild-type strain CW04 has the nuclear genotype cu-his3-11,15,1eu2-3,112,ade2-l,ura3- l ,trpl-l,canl-lOO,NAM2+ (Banroques et al., 1986). HM200/1 is a rho+,mit+ strain carrying the plasmid YCpGMC068 and has the same nuclear genotype as CW04 except that it is nam2::LEUZ. Construc-

The abbreviations used are: LRS, leucyl tRNA synthetase; mLRS, mitochondrial LRS; mMRS, mitochondrial methionyl tRNA synthe- tase; SDS, sodium dodecyl sulfate.

2537

2538 Purification of S. cerevisiae Mitochondrial Leucyl-tRNA Synthetase tion of YCpGMC068 containing the S . cereuisiae NAM2 gene under the control of UASGALlO and the CYCl promoter has been described in detail elsewhere (Herbert et al., 1988a). In HM200/1, expression of the NAM2 gene is repressed by glucose and induced by galactose via the UASGALIO. The growth medium contained 3 g of yeast extract (Difco), 10 g of galactose, 0.8 g of (NH4)*S04, 0.7 g of MgSO,. 7H20, 0.5 g of NaCI, 1.0 g of KH2P04, and 8 pl of FeC13 in solution ( d = 1.27)/liter. After sterilization, the medium was supplemented with 20 mg/liter adenine, 10 mg/liter histidine, 2 ml/liter 1.8 M CaC12 and 2.5 mg/liter S-penicillin. Cultures were grown aerobically a t 28 "C for 36 h.

Chemicals-Heparin-Ultrogel was from IBF France, total E. coli tRNA was from Sigma (type XXI), and E. coli tRNA-Leu4 was a kind gift of Drs. N. Bilgium and C. G. Kurland. ~[U-"C]Leucine (342 mCi/mmol), ~-[~H]leucine (52 Ci/mmol), and ~-[~~S]methionine (>800 Ci/mmol) were from Amersham Corp. When necessary, specific activities were adjusted with appropriate cold L-amino acids. Glass fiber filters (GF/B) were from Whatman. Phenylmethylsulfonyl flu- oride was from Merck; ATP was from Sigma; Triton X-100, from Fluka. All other products were of analytical grade.

Aminoacyl-tRNA Synthetase Assay-Charging reactions were car- ried out at 20 "C in 25 pl (65 pl for initial velocity determinations) of a standard incubation mixture containing 100 mM Tris/HCl, pH 8.0; 1 mM dithiothreitol, 2.5 pg of bovine serum albumin, and 0.06 pCi of L-[U-"C]leucine (100 mCi/mmol) or 0.03 pCi of ~[~~Slmeth ionine (50 mCi/mol). Mg2"ATP concentrations were optimal, and tRNA substrates were in excess. With the tRNA-Leu4, the reaction was carried out a t 10 mM ATP, 4 mM MgC12; with total E. coli tRNA the reaction was carried out at 36 mM ATP, 25 mM MgC12 for leucine incorporation and at 10 mM ATP, 10 mM MgCL with methionine as a substrate. Initial velocity conditions were used for the determination of the kinetic constants and 10-min reactions for following the enzyme fractionation and optimizing the conditions. After incubation, 20 p1 (10 pl for initial velocity measurements) of the aminoacylation mixture was applied to a GF/B filter, and the cold trichloroacetic acid-insoluble radioactivity was counted in a liquid scintillation coun- ter.

Purification of the Mitochondrial Leueyl-tRNA Synthetase from HM2OO/l-Purification of the NAM2-encoded mLRS was followed by assaying the formation of the tRNA charged with radioactive leucine. Cultures (12 liters) were grown as specified previously. Mi- tochondria (about 440 mg of total protein, 31-32 mg of protein/ml) fully competent in translation and respiration were isolated and stored in liquid Nz as described by Kozlowski and Zagorski (1988). Frozen cell paste, mitochondria, and mitochondrial lysates main- tained an mLRS activity comparable to that detected in fresh prep- arations for several months (data not shown). All preparative steps were carried out at 0-4 "C.

Mitochondria (100-110 mg of protein) were lysed for 5 min on ice in 60 ml (final volume) of lysis buffer (10 mM Tris/HCl, pH 7.5, 5 mM 0-mercaptoethanol, 1 mM phenylmethylsulfonyl fluoride, 0.75% Triton X-100). The lysate (specific activity, 0.79 units/mg of protein, total activity 88.5 units) was centrifuged at 14,000 x g for 20 min, and saturated (NH4)2S04 solution (pH 7.0) was added to the super- natant. The fraction precipitating at 40% saturation was removed by centrifugation (14,000 X g, 20 min) and the supernatant collected and taken to 85% saturation. After an overnight incubation on ice, the enzyme-enriched precipitate was collected by centrifugation as above. The precipitate was dissolved in 5-8 ml of solubilization buffer (10 mM Tris-HC1, pH 7.5, 0.3% Triton X-10, 7 mM 0-mercaptoethanol, 20% glycerol, 1 mM phenylmethylsulfonyl fluoride) and dialyzed on ice three times for 45 min against 500-ml portions of buffer I (50 mM Tris/HCl, pH 8.0, 35 mM KCl, 15 mM MgC12, 8 mM P-mercaptoeth- anol, 0.5 mM EDTA-Na2, 1 mM phenylmethylsulfonyl fluoride, 10% glycerol). The dialyzed material (specific activity 1.5 units/mg of protein, total activity 87.2 units) was applied to a column of heparin- Ultrogel (25 ml, 1.6 X 12 cm) equilibrated previously with 150-200 ml of buffer I. After application of the enzyme, the column was eluted with 60-80 ml of buffer I followed by 100-150 ml of buffer I1 (buffer I with 150 mM KCl). The column was then eluted at a flow rate of 20 ml/h with a 400-ml linear gradient of KC1 (150-300 mM KCl) in buffer 11, and 1.5-ml fractions were collected. Active fractions (eluted between 0.225 and 0.245 mM KC1) were supplemented with Triton X-100 (to O.l%), concentrated separately 10-12 times with Centricon 30 (Amicon) ultrafilters, and then frozen in liquid N2 before storage at -90 "C. The specific activity of the homogeneous fraction was 350 units/mg of protein. The enzyme peak eluted from the heparin- Ultrogel column represented 57% of the total activity detected in the

lysate, and the homogeneous fractions represented 28% of the peak activity. One unit is defined as the amount of enzyme that catalyzes the leucylation of 1 nmol of tRNA/min at 20 "C.

Polyacrylamide Gel Electrophoresis and Zmmunoblotting"lO% Polyacrylamide-SDS gel electrophoresis was performed according to Laemmli (1970). When needed, appropriate portions from column fractions were concentrated before electrophoresis by precipitation with 80% acetone at -20 "C. Immunoblotting was performed with antiserum raised against purified E. coli LRS (kindly donated by Dr. Leberman, EMBL, Grenoble) or against a chimeric protein resulting from the fusion of the 5' region of Lac2 gene (280 N-terminal amino acids) and a 3' fragment of the NAMZ gene (363 C-terminal amino acids). Construction of fusion protein and preparation of antiserum have been described elsewhere (Herbert et al., 1988a).

After polyacrylamide-SDS gel electrophoresis immunoblotting was performed as described by Towbin et al. (1979). The membranes were incubated for 2 h at 37 "C with one of the antisera described above followed by incubation with a second antibody (anti-rabbit IgG) linked to alkaline phosphatase. Addition of nitro blue tetrazolium chloride and 5-bromo,4-chloro,3-indoyl phosphate p-toluidine salt caused the immunolabeled proteins to develop a purple colour.

N-terminal Microsequencing-Fractions 74 and 75 from the hepa- rin-Ultrogel column were pooled and trichloroacetic acid precipitated (final concentration, 5%) in the presence of 0.15% sodium deoxycho- late. The pellet was resuspended in Laemmli sample buffer (Laemmli 1970) and the pH adjusted to 7-8 with 1 M Tris base. Approximately 15 pg of mLRS was run on a 10% polyacrylamide minigel (Laemmli system, 1.5 mm thick). To avoid blocking the N terminus of the protein the preelectrophoresis (1 h at 10 mA) and the electrophoresis (2 h a t 10 mA) were carried out in.the presence of 1 mM sodium thioglycolate. After electrophoresis the gel was electroblotted onto Immobilon-P (3 h at 6.5 v/cm) using the Tris borate system of Bauw et al. (1989). The membrane was stained with amido blue-black, and the bands corresponding to the mLRS were combined for gas-liquid microsequencing. Sequencing was carried out using an Applied Bio- systems Sequencer (470A.ABI) as described by Le Caer and Rossier ( 1988).

Separation of the tRNA-Leu Zsoacceptors by CPC-5 Chromatogra- phy-A standard incubation mixture (60 pl) containing 16 pg/ml E. coli total tRNA (Sigma type XXI) and appropriate concentrations of ATP and MgC12 were incubated at 20" C for 10 min with either E. coli LRS or purified mitochondrial mLRS. Labeling with E. coli enzyme was done with [14C]leucine (8 mCi/mmol) and with mLRS with [3H]leucine (16.3 mCi/mmol). After incubation, aminoacylated mixtures were deproteinized by phenol-chloroform extraction; nucleic acids were precipitated with 78% cold ethanol and dried under vac- uum. The tRNA pellets from each assay were dissolved in water. Aliquots of the 3H- and "C-labeled products were mixed together and diluted with RPC-5 column buffer. These mixtures (500-600 pl) were then separated on an RPC-5 column as described by Roe et al. (1973).

RESULTS AND DISCUSSION

Aminoacyl-tRNA Synthetase Activities in Crude Extracts- Under normal conditions mitochondrial aminoacyl-tRNA synthetases are low abundance proteins. To facilitate our attempts to purify the S. cereuisiae mLRS we decided to use a vector that overexpresses the NAMZ gene. HM200/1 harbors a UASGALIO/NAM2 gene fusion cloned on a centromeric plasmid (Herbert et al., 1988a). In order to determine if this would be a suitable starting material for the purification we decided to assay the mLRS in whole cell and mitochondrial lysates of galactose-grown HM200/1 and the wild-type equiv- alent CW04. As an internal control the level of the mitochon- drial methionyl-tRNA synthetase (mMRS) was also meas- ured. Both enzymes are able to charge E. coli tRNA SO this was used as a readily available substrate (Schneller et at., 1976). The results in Table I show that the level of the mMRS is essentially the same in mitochondrial lysates of both HM200/1 and CW04. However, the level of the mLRS was approximately 45 times higher in mitochondrial lysates of HM200/1 compared with CW04. This indicated that galac- tose-grown HM200/1 would be a suitable starting material for attempts to purify the mLRS. It should be remembered that

Purification of S. cerevisiae Mitochondrial Leucyl-tRNA Synthetase 2539

as neither synthetase was measured with its cognate mito- chondrial tRNA the activities of the mLRS and mMRS measured in this experiment are not comparable.

Purification of the mLRS-Using antibodies raised against the product of a LacZINAM2 gene fusion, we were able to show that no NAM2 gene product was detectable in postmi- tochondrial supernatants of HM200/1 (data not shown). This indicated that essentially all of the overproduced mLRS is transported into the mitochondria. Thus, as mitochondrial proteins represent approximately 10% of total cellular pro- tein, the isolation of mitochondria gave a significant first purification step and removed the possible problem of con- taminating cytoplasmic LRS in the final preparation. The next step was an ammonium sulfate precipitation, most of the enzyme was found to precipitate from a cleared mitochondrial lysate at 40-85% saturation ammonium sulfate. This yielded a further 2-fold purification and concentrated the enzyme. After this the protein was redissolved and dialyzed to remove the ammonium sulfate and to change the buffer for the final stage of the purification. This was performed by column chromatography on heparin-Ultrogel using KC1 gradients. Upon chromatography the enzyme migrated as a single peak independent of the slope of the gradient. Steep gradients (35- 450 mM KC1) permitted the rapid isolation of a rather con- centrated fraction (with -50% purity); flat gradients (150- 300 mM KC1) led to the isolation of homogeneous enzyme with a specific activity of 350 units/mg of protein. The enzyme peak eluted from the column represented approximately 57% of the total activity detected in the lysate, the homogeneous fractions representing 28% of the peak activity. The activity measured with total E. coli tRNA co-migrated with the activ- ity charging E. coli tRNA-leu4 and the protein detected by the antibodies raised against the LacZ/NAM2 fusion protein (Fig. 1). Fractions eluted with -260 mM KC1 contained a single protein, whose molecular mass was assessed by SDS-poly- acrylamide gel electrophoresis and Coomassie staining to be 100 kDa (Fig. l ) , consistent with the size predicted from the NAM2 gene sequence (Labouesse et al., 1987). This protein cross-reacts with antibodies raised against the LacZ/NAM2 fusion protein and with antibodies raised against the purified E. coli LRS (Fig. 2), thus confirming the considerable simi- larity between the proteins predicted by a comparison of their primary sequences (Herbert et al., 1988a). This purification protocol is simple and rapid; the ability to construct NAM2 mutants in appropriate expression vectors and purify the corresponding proteins will facilitate the understanding of the role of the mLRS in mitochondrial pre-mRNA splicing.

N-terminal Sequence of the mLRS-The purified mLRS was prepared for sequencing, transferred to Immobilon-P, and the N-terminal sequence was determined as described by Le Caer and Rossier (1988). When the N-terminal sequence of 22 aminoacidsdeterminedfrom the protein LSTKRGPGPAV- KKLIAIGEKWK was compared with the sequence deduced from the gene it appeared that the first 9 amino acids had

been removed. Many proteins of the mitochondrial matrix have a short leader sequence which is necessary for targeting to the mitochondria and is removed during the transport process (reviewed by Hart1 et al., 1989). The matrix protease does not have a strict recognition sequence, but most sites fit the consensus sequence of basic, basic or hydrophobic/hydro- phobic. This is consistent with the site of proteolysis in the mLRS (MLSRPSSRF/LST), thus it is reasonable to infer that a 9 amino acid leader peptide is removed from the mLRS during or after transport into the mitochondria.

Characteristics of the Purified mLRS-The pH optimum of the purified mLRS is between pH 8.0 and 9.5. The activity is stimulated by KC1, with an optimum of 60-70 mM KC1 in the presence of 5 mM M e . When the M$' concentration is raised to 25 mM the KC1 optimum shifts to 75-175 mM (data not shown). The optimum temperature was found to be 37 "C; however, for practical reasons all incubations were carried out at 20 "C. At 4 mM M$+ and using E. coli tRNA-Leu4 as substrate, the optimal ATP concentration was around 10 mM. When purified mitochondrial tRNA-Leu was used at either 4 or 25 mM M e , the ATP optimum was 10-15 mM. However, when total E. coli tRNA was used as the substrate and the determinations were done at 4 and 25 mM Mg"', a higher ATP concentration was found to be optimal at 25 mM M e . This led us to examine more carefully the relationship be- tween the optimal ATP concentration and the Mg"' concen- tration. We observed that at 5 mM Mg2+ ATP influences the enzyme activity in a typical fashion with a well defined optimum at 7 mM ATP. At 25 mM M e , the enzyme activity reaches a plateau at low ATP concentrations (4-12 mM); higher concentrations of ATP result in additional charging with a broad optimum around 30-35 mM ATP (Fig. 3A). If the M e concentration is increased still further to 40 mM, the biphasic shape of the curve is maintained, but the ATP optimum is shifted to a higher value (data not shown). In contrast to the mLRS the purified E. coli LRS is inhibited by high ATP concentrations at both 5 and 25 mM Mg2+ (Fig. 3B) . At this point it should be remembered that in E. coli there are five tRNA-Leu isoacceptors (Holmes et al., 1977), but in yeast mitochondria there is only one (Berlani et al., 1980). In addition, there is a change in the genetic code between E. coli and yeast mitochondria; in E. coli there are six leucine codons, but in yeast mitochondria there are only two, the CUN family being threonine codons (Li and Tzago- loff, 1979). The E. coli isoacceptors that are equivalent to the yeast mitochondrial leucine tRNA are the ~ R N A - L ~ u ~ ' ~ (cor- responding to the codons UUA UUG). Thus, it is reasonable that the yeast mLRS would discriminate against the other E. coli isoacceptors and that the "extra" charging and second ATP optimum seen at high Mg?" concentration could corre- spond to the charging of the other E. coli isoacceptors. To investigate this total E. coli tRNA was charged by the purified mLRS at 5 and 25 mM M e . The charged tRNAs were separated on an RCP-5 column, and the results are shown in

TABLE I Methionyl- and leucyl-tRNA synthetase activities in whole cell and mitochondrial extracts of galactose-grown CW04

(wild-type NAM2) HM200/1 and (UASGALlOINAM2 fusion) Total E. coli tRNA was used to assay the mMRS, and E. coli tRNA-Leu4 was used to assay the mLRS.

Fraction Met-tRNA synthase Leu-tRNA synthase

unitslmg protein Mitochondrial lysate of HM200/1 0.218 0.860 Mitochondrial lysate of CW04 0.201 0.018 Total cell lysate of HM200/1 0.037 0.090 Total cell lysate of CW04 0.027 0.003

2540 Purification of S. cerevisiae Mitochondrial Leucyl-tRNA Synthetase

A B FIG. 1. Elution profile of the

mLRS from the heparin-Ultrogel 6 column. Panel A, enzyme activity was 4 followed by aminoacylation of total E. ;;; 0.015

- x

(A-A) as described under "Materials 2 o,o - 8000 2 coli tRNA (o"-o) or tRNA-Leu4 c 300 c

and Methods," using 3 PI of the column fractions. Panel B; polyacrylamide gel 250 E - 4000 2 electrophoresis of 1 pg of protein from 5 0.005 fractions 78-82. The position of the mo- 7 200 - L

lecular mass markers is shown on the : left. 0 40 80 120 150 - 0 "

- 12000 M r 81 LRS (Koa) 78 79 80 82 < &qc.-".-

E 92.5 - 01 .

= 0 N

"

66.2 - - u .

In c 3 0 I

45.0 - Fractlon number

A B C "- p;= 1. F. +- " k - " 92.5 kDa

, U C 6 6 . 2 kDa

i I I I

i ! I 1-. ~ (r + 45.0 kDa

FIG. 2. Western blot analysis of purified mLRS. A: Lane 1 , E. coli LRS (10 ng): Lane 2, S. cerevisiae mLRS (20 ng). The blot was visualized with antiserum raised against a chimeric P-galactosid- ase/NAM2 protein. B: Lane 1 , S. cerevisiae mLRS (20 ng): lane 2, E. coli LRS (10 ng). The blot was visualized with antiserum raised against the purified E. coli LRS. C: Lane I , E. coli LRS (1.5 pg); lane 2, molecular mass markers; lane 3, S. cerevisiae mLRS (0.5 pg). Polyacrylamide gel electrophoresis stained with Coomassie blue; the size of the molecular mass markers is indicated on the right of the figure.

5mMMg - A 15000] 25mM Mg M P

0 10 20 30 40

[ATPI (mM)

8 50000 5mMMg - U 0 a b 4 0 0 0 0 / p 30000

c - 20000

10000

0 10 20 30 40

[ATPI (mM)

FIG. 3. Effect of ATP concentration on the leucine charging of total E. coli tRNA at high and low Mg2+ concentrations. A, S . cerevisiae mLRS; B, E. coli LRS.

0 2 0 40 60 80

Fractlon number FIG. 4. Separation of aminoacylated tRNAs by RCP-5 chro-

matography. Total E. coli tRNA was aminoacylated using the E. coli LRS in the presence of ['Hlleucine (W), or the S. cerevisiae mLRS in the presence of ['"C]leucine (+-+). The products were then mixed and separated by chromatography on RCP-5. The prod- ucts of the aminoacylation reaction performed with the E. coli enzyme act as markers for the different tRNA-Leu isoacceptors. A, amino- acylation with both enzymes performed a t 5 mM Mg-" and 7 mM ATP. B, aminoacylation with the E. coli LRS performed a t 5 mM MB'+ and 7 mM ATP; aminoacylation with the S. cerevisiae mLRS performed a t 25 mM Mg-" and 32 mM ATP.

Fig. 4. From these results we can see that there is no substan- tial difference between the tRNAs charged a t low and at high M F concentrations. In particular, the same isoacceptors are charged under both conditions, notably the isoacceptor Leu' (which comprises -90% of the total tRNA-Leu) is not more charged at high M 2 + concentrations. Thus the biphasic curve obtained when ATP is varied a t high Mg2+ remains unex- plained. However, it should be remembered that it is only seen when the mitochondrial enzyme is used to charge total E. coli tRNA.

Kinetic Parameters-These data were calculated with total E. coli tRNA under two different conditions. At 5 mM M$+ and 7 mM ATP the K,,, values for leucine and tRNA were 7.1 and 1.02 pM, respectively. At 25 mM M e and 32 mM ATP

Purification of S. cerevisiae Mitochondrial Leucyl-tRNA Synthetase 2541

the K,,, values for leucine and tRNA were 7.6 and 2.0 pM, respectively.

Conclusions-We have presented here a simple and rapid purification scheme for the mLRS of S. cereuisiae which is encoded by the NAM2 gene. This protocol could be modified easily to allow the purification and characterization of mutant proteins originally described in genetic experiments. The pu- rification of the mitochondrial leucyl-tRNA synthetase is of particular interest for two reasons. First, the mitochondrial genetic code differs from the universal code, with the CUN family encoding threonine instead of leucine. Thus, this is the first reported purification of a synthetase that charges a nonuniversal set of tRNAs, and structural studies on the purified mLRS should give us considerable insight into the process of discrimination in tRNA selection. Second, the mLRS is involved in the excision of at least two mitochondrial introns, a14 and b14, as well as in tRNA aminoacylation. Genetic experiments have shown that the maturases encoded by these introns are also needed for splicing; however, it is not known if the mLRS interacts with the intron RNA, the maturases, or both. The availability of the purified enzyme for in vitro studies should help us to begin to answer these questions.

Acknowledgments-We would like to thank Drs. J.-P. Le Caer and J. Rossier for the N-terminal sequence determination, Dr. R. Buck- ingham for help with the RCP-5 chromatography, Dr. C. Zelwer for helpful discussions, and A. Harington for reading the manuscript. We are indebted to Dr. R. Leberman (EMBL, Grenoble) for generous gifts of purified E. coli LRS and antiserum raised against the enzyme. Dr. B. Jacrot is thanked for accepting W. Z. and B. C. at the EMBL laboratory in Grenoble where part of this work was done.

REFERENCES

Akins, R. A., and Lambowitz, A. M. (1987) Cell 50, 331-345 Banroques, J., Delahodde, A., and Jacq. C. (1986) Cell 46,837-844 Bauw, G., Van Damme, J., Puype, M., Vandekerckhove, J., Gesser,

B., Ratz, G. P., Laurissen, J. B., and Celis, J. E. (1989) Proc Natl. Acad. Sci. U. S. A. 86,7701-7705

Berlani, R. E., Bonitz, S. G., Coruzzi, G., Nobrega, M., and Tzagoloff, A. (1980) Nucleic Acids Res. 8, 5017-5030

Delahodde, A., Goguel, V., Becam, A. M., Creusot, F., Perea, J., Banroques, J., and Jacq, C. (1989) Cell 56,431-441

De La Salle, H., Jacq, C., and Slonimski, P. P. (1982) Cell 28, 721- 732

Dhawale, S., Hanson, D. K., Alexander, N. J., Perlman, P. S., and Mahler, H. R. (1981) Proc. Natl. Acad. Sci. U. S. A. 78,1778-1782

Dujardin, G., Pajot, P., Groudinsky, O., and Slonimski, P. P. (1980) Mol. Gev Genet. 179,469-482

Dujardin, G., Jacq, C., and Slonimiski, P. P. (1982) Nature 298,628- 632

Dujardin, G., Labouesse, M., Netter, P., and Slonimski, P. P. (1983) in Mitochondria (Schweyen, R.J., Wolf, R., and Kaudewitz, F., eds) pp. 233-250, Walter de Gruyter, Berlin

Hartl, F., Pfanner, N., Nicholson, D., and Neupert, W. (1989)

Herbert, C. J., Labouesse, M., Dujardin, G., and Slonimski, P. P.

Herbert, C. J., Dujardin, G., Labouesse, M., and Slonimski, P. P.

Holmes, W. M., Goldman, E., Miner, T. A., and Hatfield, G. .W

Kozlowski, M., and Zagorski, W. (1988) Anal. Biochem. 172, 332-

Labouesse, M., and Slonimski, P. P. (1983) EMBO J. 2, 269-276 Labouesse, M., Herbert, C. J., Dujardin, G., and Slonimski, P. P.

Laemmli, U. K. (1970) Nature 227,680-685 LeCaer, J.-P., and Rossier, J. (1988) Anal. Biochem. 169, 246-252 Li, M., and Tzagoloff, A. (1979) Cell 18,47-53 Roe, B. A., Marcu, K., and Dudock, B. (1973) Biochim. Biophys. Acta

Schneller, J. M., Schneller, C., Martin, R., and Stahl, A. J. C. (1976)

Towbin, H., Staehelin, T., and Gordon, J. (1979) Proc. Natl. Acad.

Tzagoloff, A., Akai, A., Kurkuos, M., and Repetto, B. (1988) J. Biol.

Wenzlau, J. M., Saldanha, R. J., Butow, R. A., and Perlman, P. S.

Biochim. B i ~ p h y ~ . Acta 988, 1-45

(1988a) EMBO J. 7,473-483

(1988b) Mol. Gen. Genet. 213, 297-309

(1977) Proc. Natl. Acad. Sci. U. S. A. 74, 1393-1397

391

(1987) EMBO J. 6, 713-721

319,25-36

Nucleic Acids Res. 3 , 1151-1165

Sci. U. S. A. 76,4340-4354

Chem. 263,850-856

(1989) Cell 56,421-430