Characterization of a yeast mitochondrial ribosomal protein structurally ...

THEJOURNAL OF BIOLOGICAL CHEMISTRY

Vol. 267, No. 8, Issue of March 15, pp. 5508-5514, 1992 Printed in U.S. A.

0 1992 by The American Society for Biochemistry and Molecular Biology, Inc.

Characterization of a Yeast Mitochondrial Ribosomal Protein Structurally Related to the Mammalian 68-kDa High Affinity Laminin Receptor* (Received for publication, September 30, 1991)

Stephen C. Davis, Alexander TzagoloffS, andSteven R. Ellis From the Department of Biochemistry, The University of Louisville, Louisville, Kentucky 40292 and the $Department of Biological Sciences, Columbia University, New York, New York10027

We have cloned the nuclear gene MRP4 coding for a mitochondrial ribosomal protein of the yeast,Saccharomyces cerevisiae.The genewas isolated by complementationofarespiratory-deficientmutant with a pleiotropic defect in mitochondrial gene expression. MRP4 revealed thatit has The nucleotide sequence of sequence similarity with Escherichia coli ribosomal protein 52 and related proteins of chloroplast ribosomes from different plants. Further characterization of the MRP4 protein revealed that it is a component of the 37 S subunit of mitochondrial ribosomes. Moreover, the phenotype of MRP4 is consistent cells carrying a disrupted copy of with the MRP4 protein being an essential component of the mitochondrial proteinsynthetic machinery. Finally, we note that theMRP4 protein and other members of the 52 family of ribosomal proteins have regions of sequence similarity with the mammalian68kDa high affinity laminin receptor.

Analysis of nuclear genes coding for mitochondrial ribosomal proteins of the yeast, Saccharomyces cereuisiue, has revealed some unusual aspects of the evolution of this class of ribosomes. Although the two rRNAs and a number of the ribosomal proteins are structurallyrelated to ribosomal components of Escherichia coli ( 1 4 ,indicating a common ancestry, the other mitochondrial ribosomal proteins have a more obscure origin. A number of mitochondrial ribosomal proteins have been identified that share no sequence similarity with proteins in current databases and therefore have either diverged from their eubacterial ancestors to such an extent as to preclude meaningful sequence comparisons or have origins unrelated to eubacterial ribosomal proteins (4-10). The mosaic nature of the mitochondrial ribosome is probably a reflection, on the one hand, of a core structure of homologous components responsible for carrying out central stepsof protein synthesis common to both eubacteria and mitochondria and, on the other, of a unique set of proteins with more specialized functions peculiar to mitochondrial protein synthesis. Here, we further define the composition of the eubacteria-

* This work was supported by Grant GM40632 from the National Institutes of Health (to S. R. E.) and Grant DBM 9004030 from the Nationat Science Foundation (to A. T.). 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 solelyto indicate this fact. The nucleotide sequence(s) reported in thispaperhas been submitted totheGenBankTM/EMBL Data Bankwith accession number($ M82841.

like core of mitochondrial ribosomes of S. cereuisiae by characterization of a gene coding for a mitochondrial ribosomal protein homologous to E. coli ribosomal protein S2 (11).The gene, designated MRP4, was cloned by complementation of a respiratory-deficient yeast strain with a pleiotropic defect in mitochondrial gene expression. The distribution of the MRP4 protein in mitochondrial extracts shows it to be a component of the small subunit of the mitochondrial ribosome. The phenotype of cells carrying a disrupted copy of the MRP4 gene indicates that theMRP4 protein hasan essential role in the function of the mitochondrial ribosome in vivo. We also report that the S2 family of ribosomal proteins have regions of sequence similarity with the mammalian 68kDa high affinity laminin receptor (12, 13). Among the S2like proteins, the MRP4 gene product has the highest degree of similarity with the laminin receptor, which initially drew our attention to thisrelationship and may explain why it has not been reported previously. The possible functional correlate for the structural relationship between the two families remains to be elucidated. MATERIALS AND METHODS

Yeast and Bacterial Strains-The yeast strains used in this study are listed in TableI. The respiratory-deficient mutant C75 (a,p+,mt6, mrp4-I) was derived from S. cerevisiae strain D273-10B/Al by mutagenesis with ethylmethanesulfonate (14). C75/U3 (a,p+,ura3l,mrp4-1) was obtained from a cross ofC75 with W303-1A (a,p+, ade2-l,his3-ll,15,leu2-3,112,ura3-I,trpI-l,canI-100). W3031AVMRP4(a,p~,ade2-l,his3-II,15,leu2-3,112,ura3-l,trpl-l,canl100,mrp4::HZS3) was created by the in situdisruption of MRP4, using the strategy outlined in Fig. 5A. Media used for the cultivation of yeast were YPD (1%yeast extract, 2% peptone, 2% glucose), YEPG (1%yeast extract, 2% ethanol, 2% peptone, 3% glycerol), and minimal (0.67% nitrogen base without amino acids, 2% glucose). Where appropriate, othergrowth supplements were added to minimal media in amounts described previously (15). The E. coli strain used was JM101. Antigen Preparation and Antibody Production-A PATH expression vector (16) was used to synthesize trpE-MRP4 fusion proteins for use in raising antibodies against the MRP4gene product. Briefly, a 1.5-kb’ PuuII-Hind111 fragment, coding for the 306 carboxyl-terminal residues of the MRP4 protein, was isolated from the plasmid pG54/ST1. The fragment was cloned into pATH2 digested with SmaI and HindIII. The resultant plasmid, pTE4-4, was transformed into JMlOl cells, and expression of the fusion protein was induced by tryptophan starvation. Cells were fractionated according to Koerner et al. (16), and the insoluble fraction was used as anenriched source of TE4 fusion protein. TE4 fusion protein was further purified by preparative SDS-polyacrylamide gel electrophoresis (17). The TE4 fusion protein was detected by staining with Coomassie Blue for 10 min, destained for 15 min in 5% acetic acid, and excised from the gel. Protein was electroeluted from gel fragments overnight in the presence of SDS (la). Amicon Centriprep 30 columns were used to

’ The abbreviations used are: kb, kilobase(s); bp, base pair($; SDS, sodium dodecyl sulfate.

5508

S2 Ribosomal Proteins and 68-kDa Laminin

Receptors are Related

5509

TABLEI Genotypes of S. cereuisk strains Strain

Genotype Ref. 14 R. Rothstein" R. Rothstein This study This study This study C75 X W303-1A Ref. 43

a,p+,met6 D273-10B/lA W303-1A a,p+,ade2-l,his3-ll,15,leu2-3,112,urd3-l,trpI-I,canl-l00 W303-1B a,p+,ade2-1,his3-ll,15,leu2-3,112,urd3-1,t~1-1,can1-100 W303-1AVMRP4 a,p~,ade2-I,his3-Il,I5,Ieu2-3,ll2,urd3-I,3 W303-1BVMRP4 a,p~,ade2-l,his3-ll,15,leu2-3,112,urd3-I, c75 a,p+,metG,mrp4-1 c75/u3 a,p+,ura3-1,mrp4-1 a,PO, ade, lys COP161po College of Physicians and Surgeons, Columbia University, New York.

6o

o i

FIG. 1. Localization of MRP4 within plasmid pG54/T1. The top left shows a partial restriction map and thelocations of two open reading frames in the nuclear insert of plasmid pG54/T1. The restriction map of the dashed region has notbeen determined. The locations of restriction enzyme cleavage sites are asfollows: S, SphI; B, BamHI; G, EglII; and K, KpnI. Ears below represent restriction fragments cloned into the yeast shuttle vector YEp352. The ability of each plasmid to complement the respiratory deficiency of strain C75/U3 is indicated by a plus if the plasmid restores growth on glycerol and a minus if it does not. The locations and orientations of the MRP4 reading frame and of the unidentified reading frame (ORF) as determined by DNA sequencing are represented by the open bones above the restriction map.

sucrose gradients in AMT-500 (10 mM Tris-HCI, pH 7.5, 10 mM MgC12, 500 mM NH4Cl, 6 mM 2-mercaptoethanol) and centrifuged for 16 h at 85,000 X g (21). The MRP4 protein was detected in immunoblots of mitochondrial extractseither by the 5-bromo-4chloro-3-indolyl phosphate/nitro blue tetrazolium colorometric assay using conditions described previously (20) or by the enhanced chemiluminescence method using a kit and protocols obtained from Amersham Corp. Disruption of MRP4-The MRP4 gene was disrupted by replacing amino acids 221-367 of the MRP4 open reading frame with the yeast HZS3 gene, as outlined in Fig. 5A. pG54/ST1 was cleaved with BglII, removing a 439-bp fragment that spans roughly the carboxyl-terminal half of the MRP4 reading frame. This fragment was replaced with a 1.7-kb BamHI fragment carrying HZS3. The resulting plasmid was digested with EcoRI, and a fragment of approximately 4 kb containing HZS3 flanked by 5' and 3' sequences derived from MRP4 was isolated and used to transform either W303-1A or W303-1B cells to histidine prototropy. Transformants were screened for respiratory growth on YEPG media and screened for a disrupted copy of MRP4 by Southern analysis (22). Miscellaneous Procedures-Standard methods were used for the analysis of DNA with restriction enzymes and other manipulations of DNA (23). All DNA sequences were obtained by the method of Maxam and Gilbert (24) using 5' end-labeled single-stranded restriction fragments. RESULTS

I

I Phenotypes of the pet Mutant, C75-A

MRP4

100 bp

FIG. 2. Restriction sites used to determine the sequence of MRPI. The restriction sites used for 5'-end labeling are indicated as follows: H, HoeIII; X,XhoI; K, KpnI; P , PuuII; T, TaqI;G, BglII; F , Hinfl; E , BamHI; E, EcoRI; R, EcoRII; and D, DdeI. The arrows indicate the direction and approximate extent of each sequence determination. The open box represents the MRP4 open reading frame. exchange the electroelution buffer with 0.9% NaCl and concentrate the TE4fusion protein. New Zealand White rabbits were immunized with TE4 protein (100 pg) emulsified with RIB1 adjuvant as suggested by Ribi ImmunoChem Research, Inc. Two weeks postimmunization, a n additional 100 pg of TE4 was emulsified with Freund's incomplete adjuvant and injected into the rabbits subcutaneously. Subsequent boosts of 100 pg of TE4 in Freund's incomplete adjuvant were a t 1month intervals, and blood was drawn 1 week after each boost. Sera were processed as described by Li et al. (19). Sera enriched for antibodies against the MRP4 protein were made by preabsorbing crude serawith an unrelated trpE fusion protein, followedby a second preabsorption with mitochondrial extracts prepared from W3031BVMRP4. Mitochondria Preparation and Fractionation-Yeast mitochondria were isolated as described by Dang et al. (20). Mitochondria were lysed with 1%deoxycholate and clarified by centrifugation at 14,000 X g for 10 min. Clarified extracts were loaded directly onto 10-30%

respiratory-deficient mutant of S. cereuisiae, C75(mrp4-1)is one of approximately 2000 independent pet strains selected for their inability to grow on nonfermentable carbon sources (25). C75 has been assigned to complementation group G54, which has one additional member. The respiration defect in C75 is complemented by a po tester strain, indicating that the mutation is in anuclear gene and is recessive (Table I). Complementation by the po tester also indicates the presence of wild-type mitochondrial DNA in C75. Absorption spectra of mitochondria from C75 show little or none of the cytochromes a, a3 or b, all of which are known to be products of the yeast mitochondrial translational machinery. The a! absorption bands of cytochromes c and cl, both of which are nuclearly encoded gene products, however, are present (data not shown). The above spectral phenotype is often an indication of a defect in mitochondrial protein synthesis. Decreased incorporation of [35S]methionineinto mitochondrially translated proteins in the mutant compared with wild-type is also consistent with a block in mitochondrial protein synthesis (data notshown). Cloning and Sequence Analysis of MRP4-A nuclear gene capable of complementing the respiratory defect of C75/U3 (ur&-l,mrp4-1) was isolated by transformation with a wildtype yeast genomic library and selection for growth on minimal glycerol medium. The genomic library used in the transformation consisted of partial Sau3A fragments of yeast nuclear DNA averaging 5-15 kb cloned into the URAS-containingshuttle vector YEp24 (26). Approximately 5,000 Ura+ clones were obtained by transforming 5 X 10' cells with 5 pg ofDNA. Of these, only one clone (C75/U3/T1) had also

S2 Ribosomal Proteins and 68-kDa Laminin Receptors are Related

5510

MRP4

M q R H V F A R N F R R L S L L R N P S L T K R F Q S S A S G A A N T P N N N D E V M L L q q K L L

-50

MRP4

Y O E I R S E L K S L S Q V P E D E I L P E L K K S L E ~ O K L S O K E Q Q L E A E L S D F F R N V-100

MRP4

A L L N K L F D S K T L D G Q S S T T T A A A T P T K P Y P N L I P S A N D K P Y S S Q E L F L R Q -150

MRP4 ECOSZ TOES2 HLAMR

FRifi !i![$$fl: l[Fj PDO&

R PL 4U R S S T Q S V I Y G EKVG I H I I D L N

HLAMR

L

"8"

4. primary sequence alignment of mitochondrial, eubacterial, andchloroplasts2withthegs-kDa high affinity laminin receptor.The yeast mitochondrial S2 homolog (MRP4), E. coli S2 (ECOSZ), Tobacco nicotiniu chloroplast S2 (TOBSZ), and the human laminin receptor ( H M M R ) werealignedwiththemotifrecogn~t~on and alignment program of Vingron and Argos (33), using two filtering rounds.

oFQM EQv

IY

K R K

soG

T L S V L K R A A K V V E G V S E S O G ~ I L F L-249

I Y I IN L K R T

u

E

K L L LA A R A ~~~~~~~

IV I A E

N

b " ::1

-70

I

sT

E I

sG I u

E K

G E Io

sN

-298 -120

1;;

Onlypositionswithidenticalorconservativesubstitutionsinthehumanlaminin receptor and in at least one of the three ribosomal proteins have been boxed. Conservative substitutions are defined as'Arg=Lys,Asp=G1u,Met=lle=

Leu=Val,Asn=Gln,Phe=Tyr=Trp, Ala = Gly, and Thr = Ser. MRP4 ECOSZ 70852 HLAM

GQOG

-383

-227 [rlR E G U S 0 -236 I C E G R S S V I R N P Y L ~ M R G T I S R E H P U E V M P D L Y F V R O P E E I E K E E A A A E K A V T K E E F Q G E U -231

-394 -245

S2 Ribosomal Proteins and 68-kDa Laminin

Receptors are Related 1

5511

2

80.0

’ 0.5

0.0

49.5

kDa

1.0kb

u

B

’

*

32.5

-231

-

27.5

9.4 6.8

4.4

18.5

kb

- 2.3

- 2.0 FIG. 5. Disruption of the MRP4 gene. A, disruption strategy. A 1.7-kilobase pair BamHI fragment carrying the yeast HZS3 gene was used to replace a 439-bp BglII fragment released from plasmid pG54/T1. Approximately 40% of the MRP4 open reading frame, including most of the region homologous to E. coli ribosomal protein S2 was deleted in this construction. The open box represents the MRP4 open reading frame, and the shaded box represents the HIS3 gene. Restriction enzyme sites are indicated as follows: S, SphI; E, EcoRI; B, BamHI; and G, BglII. A linear EcoRI fragment carrying the disrupted MRP4 gene was used to transform strains W303-1A/B t o histidine prototropy. The bar below the disruption strategy represents theprobe used in Fig. 5B. B, genomic DNA was prepared from strains W303-1A ( l a n e 1) or W303-1AVMRP4 ( l a n e 2). DNA was digested with EcoRI, and the resulting fragments were hybridized with a 1.65-kb EcoRI fragment derived from the wild-type MRP4 gene that spans the 5”flanking region and the bulk of the MRP4 reading frame.

FIG. 6. Antisera raisedagainst a trpE-MRP4 fusion protein recognizes the MRP4 gene product. The strategy used for the construction of a trpE-MRP4 hybrid gene and for the expression and purification of the fusion protein is described under “Materials and Methods.” Mitochondria were isolated from strains W303-1B ( l a n e 1) and W303-1BVMRP4 ( l a n e 2). Mitochondrial proteins were separated by SDS-polyacrylamide gel electrophoresis, transferred to nitrocellulose, and immunoblotted with sera raised to the trpE-MRP4 fusion protein.

0.3 d v)

N

4

0.1

TABLE I1 Geneticproperties of different MRP4 alleles Cells from strains 1 and 2 were mixed on YPDplates, grown overnight a t 30 “C, and replica-plated onto minimal medium. The following day, cells were replica-plated onto YEPG medium. A p l w indicates the diploids grew on glycerol and ethanol, and a minus indicates that the diploids were unable to grow onthese carbon sources. Strain 1

C75 (mrp4-I) W303-1AVMRP4 (mrp4::HZS3) W303-1BVMRP4 (mr&::HZS3)

Strain 2

W303-1Ap0 C75 (mrp4-1) COP161a0

Growth on YEPG

+-

acquired the ability to grow on glycerol. The Ura+ and Gly’ phenotypes of C75/U3/T1 cosegregated, confirming the presence of an autonomously replicating plasmid. The nuclear insert of the resident plasmid, pG54/T1, was determined by restriction mapping to be 3.4 kb in length (Fig. 1).To map the location of the complementing gene, different regions of pG54/T1 were transferred to the shuttlevector YEp352 (27) and tested for complementation of C75/U3. The results of these transformations allowed the gene to be localized to a 2.6-kb region, contained in pG54/ST1 and defined by the internal SphI site and the right-most edge of the insert in pG54/T1. Neither of the smaller subclones, pG54/ST2 and pG54/ST4, spanning this region complements the mutant (Fig. 1).The inability of the latter constructs to complement indicates that the gene must cross the unique KpnI and BarnHI sites. A region of approximately 2.6 kb representing

J

l

l

l

l

l

l

l

l

8

t

t

o

,

t

C

2 4 6 8 101214161820222426283032

Fraction Number

-.%.

F-T

MRP4b

FIG. 7. The MRP4 protein co-migrateswith the small subunit of mitochondrial ribosomes10-3070 on isokinetic sucrose gradients. Mitochondria were prepared from W303-1B p + cells and lysed in 1%deoxycholate, and theclarified extract was loaded directly onto 10-30% sucrose gradientsin AMT-500. Centrifugation was carried out a t 85,000 X g for 16 h. Gradients were fractionated, and the absorbance at 254 nm was monitored using an ISCO model 185 density gradient fractionator and a UA-5 absorbance detector. Fraction 1represents the topof the gradient. Fractions across the gradient were collected, and equal volumes from every other fraction were subjected to SDS-polyacrylamide gel electrophoresis. The immunoblot of the gradient fractions using the antibody raised against the trpE-MRP4 fusion protein is shown at thebottom of the figure.

almost the entire insert of pG54/ST1 was sequenced by the strategy outlined in Fig. 2. Two open reading frames were identified within this region. The smaller of the two putative genes codes for a protein of 15,216 Da, with no sequence similarity to any protein in the most recent release (release 68) of the translated GenBank. This gene is contained entirely in pG54/ST4, which fails to complement C75/U3, and on this basis, can be safely excluded from being responsible for the respiratory-deficient phenotype

5512


of G54 mutants. The longer of the two open reading frames present in pG54/ST1 has been named MRP4 and,for reasons detailed below, is proposed to be the gene responsible for complementation of C75/U3. The coding region of MRP4 starts from an ATG codonupstream of the KpnI site(nucleotide 189) in Fig. 3 and ends with an opal termination codon (nucleotide 1182) just past the BglII site of the pG54/ST4 insert. In addition to the KpnI site, the MRP4 reading frame spans the BamHI site that was inferred from the subcloning data to be essential for complementation of C75/U3. The reading frame identified as MRP4 encodes a primary translation product of 394 amino acids with a predicted molecular mass of 44,159 Da. We have compared the sequence of the MRP4 product with other sequences in the GenBank database, using the FASTA search program of Pearson and Lipman (28). The search revealed that portions of the MRP4 protein have sequences similar to E. coli ribosomal protein S2 (11) and to a related protein of chloroplast ribosomes from Spinacia oleracea (29) and several other plants (30-32). The alignment in Fig. 4 shows that the E. coli and S. oleracea proteins have sequence similarity over their entire lengths. The yeast mitochondrial protein shares the same region of similarity but, in addition, has an extra amino-terminal domain of some 170 residues that is absent in the other 52 proteins. Within the carboxyl-terminal half of the MRP4 protein from amino acid 176 there aretwo blocks of sequence conservation separated by a variable region. The first conserved block spans 111amino acids from residues 176 to 286 of the MRP4 protein. In thisregion, 34 and 27% of the amino acids areidentical with the S. oleracea and E. coli S2 proteins, respectively. The S. oleracea and E. coli proteins have 42% identity over the same region. The second conserved region spans 78 amino acids from residues 317 to 394 of the MRP4 protein. In this region, the MRP4 protein has 29 and 38% identity with the S. oleracea and E. coli proteins, respectively. Sequence identity between the S. oleracea and E. coli proteins in this region is 37%. These two blocks of conserved sequence are separated by a region that bears no sequence similarity to either of the other two proteins. Furthermore, a gap of 19 amino acids must be introduced into the MRP4 sequence in this region to align the second block of conserved sequence. In addition, our search also revealed a region of sequence similarity between the MRP4 protein and the human (12) and mouse (13) 68-kDa high affinity laminin receptors. The FASTA search program revealed 33% identical residues in a region spanning 76 amino acids from positions 318 through 394 in theMRP4 protein andamino acids 125 through 199 of the human laminin receptor. Monte Carlo analysis was used to demonstrate that thesimilarity in sequence over this region is not due to sequence bias (data not shown). This region of similarity corresponds to the second block of conserved sequence between the MRP4 protein and the eubacterial and chloroplast 52 proteins. When the multiple alignment program of Vingron and Argos (33) was used to align the human laminin receptor with the other three proteins, the region described above was aligned, as was a region upstream that roughly corresponds to the first conserved region between the MRP4 protein and the other 52 proteins (Fig. 4). Disruption of MRP4"Confirmation that the longer of the two open reading frames detected in the G54/ST1insert contains the wild-type allele of mrp4-1 was obtained by the following analysis. The chromosomal copy of MRP4 was disrupted by the one-step gene replacement procedure of Rothstein (34). The strategy for the disruption of MRP4 with the yeast HIS3 gene is illustrated schematically in Fig. 5A. A linear EcoRI fragment containing HIS3 flanked by sequences

derived from MRP4 was used to transform the yeast strain W303-1A to histidine prototropy. Histidine prototrophs (W303-1AVMRP4) were screened for growth on respiratory carbon sources. The substitution of the disrupted mrp4::HIS3 for the wild-type copy of the MRP4 gene was confirmed by Southern blot analysis of genomic DNA from several respiratory-defective, histidine-independent clones from the transformation. Representative data from this analysis are shown in Fig. 5B. Hybridization of EcoRI-digested genomic DNA isolated from W303-1A cells with a probe specific for MRP4 revealed a single fragment of about 1.6 kb, a size expected for the chromosomal copy of MRP4. Similar analysis of genomic DNA isolated from the respiratory-defective, histidine-independent clones revealed that the1.6-kb fragment was absent and replaced by a new fragment of about 4 kb. The presence of a 4-kb fragment is consistent with the size of the EcoRI fragment in a strain with the disrupted MRP4, based on the restriction map shown in Fig. 5A. The respiratory deficiency of W303-1AVMRP4 is not complemented by C75 (Table 11),indicating that themrp4::HIS3 and mrp4-1 mutations were likely to be allelic and that the restored respiratory capacity of the C75 strain upon transformation with plasmids containing MRP4 is due to true complementation rather than extragenic suppression. The failure of W303-1BVMRP4 to be complemented by a po mutant further suggests that theintroduction of the mrp4::HIS3 allele induces deletions or loss of mitochondrial DNA. Inactivation of genes essential for mitochondrial gene expression has been shown to cause cells to spontaneously lose portions of their mitochondrial DNA at a high frequency (35). The inability of the strain with a disruption of MRP4 to maintain wild-type mitochondrial DNA is consistent with its pleiotropic phenotype and impaired mitochondrial translation. C75 and other derivative strains contain intact mitochondrial genomes, as judged by complementation tests with po testers (Table 11). This is probably a consequence of the mrp4-1 allele, which may be sufficiently leaky to prevent a loss of wild-type mitochondrial DNA. Association of MRP4 Protein with the Small Subunit of Mitochondrial Ribosomes-The identification of MRP4 protein as the yeast mitochondrial S2 homolog is supported by immunochemical evidence. To determine if MRP4 codes for a mitochondrial ribosomal protein, we raised antibodies against the MRP4 protein and examined its distribution in mitochondrial extracts fractionated by differential centrifugation. Antibodies against the MRP4 protein were raised by immunizing rabbits with a trpE-MRP4 fusion protein expressed from the vector pATH2 in E. coli (16). Fig. 6 shows that antibody raised to the trpE-MRP4fusion protein recognizes the MRP4 gene product. Mitochondrial extracts prepared from yeast strains containing either wild-type or disrupted alleles of MRP4 were fractionated by SDS-polyacrylamide gel electrophoresis, transferred to nitrocellulose, and probed with antibody raised against the trpE-MRP4 fusion protein. Fig. 6, lune 1 shows that in wild-type mitochondrial extracts, the antibody recognizes a protein with an estimated molecular mass of 43 kDa, which is near the size predicted for the MRP4 gene product. The 43-kDa protein is not detectedin mitochondrial extracts from a strain carrying a disrupted copy of MRP4 (Fig. 6, lane 2), indicating that the 43-kDa protein recognized by the antibody raised to a trpEMRP4 fusion protein is the MRP4 gene product. The anti-MRP4 antibody was used to determine if the MRP4 protein is physically associated with mitochondrial ribosomes upon fractionation of mitochondrial extracts. Detergent-lysed mitochondrial extracts were applied directly to

S2 Ribosomal Proteins and 68-kDa Laminin Receptors are Related 10-30% sucrose gradients containing 500 mM NH4Cl, conditions known to dissociate mitochondrial ribosomal subunits (21). Fig. 7 shows an immunoblot of fractions recovered after centrifugation. The MRP4 protein is found in fractions that coincide with the absorbance peak for the 37 S ribosomal subunits. DISCUSSION

We have cloned a gene coding for the mitochondrial ribosomal protein MRP4of S. cereuisiae. The gene was cloned by complementation of a respiratory-deficient mutant, C75 ( r n r d - l ) , with a pleiotropic defect in mitochondrial gene expression. Sequence analysis revealed that the protein encoded by MRP4 has sequence similarity to E. coli ribosomal protein S2, indicating that theMRP4 protein was probably a component of the mitochondrial ribosome. This was confirmed by immunochemical evidence showing that theMRP4 protein cofractionates with 37 S ribosomal subunits during their separation by sucrose gradient centrifugation. Several mitochondrial ribosomal proteins have now been identified that arehomologous to ribosomal proteins of E. coli (2-5). These proteins presumably function in aspects of protein synthesis that are common to both eubacterial and mitochondrial ribosomes. In this regard, several studies have suggested that E. coli S2 contributes to ribosomal domains involved in tRNA binding. Addition of S2 was shown to stimulate tRNA binding to subunits deficient in this protein (36) or previously inactivated with the histidine-modifying agent Rose Bengal (37). Moreover, 30 S particles reconstituted withoutS2arepartially defective in tRNA binding (38). However, it has not been determined if S2 is essential for protein synthesis in uiuo, since mutants lacking this protein have not been isolated (39). The phenotypes of the mrp4-1 mutant, C75, and the disruption strain, W303-1BVMRP4, indicate that the MRP4 protein is required for a minimum level of protein synthesis necessary for maintenance of the mitochondrial genome. However, since the MRP4 protein is roughly 2 times the size of E. coli S2, we are unable to conclude if the phenotype is a consequence of a loss of a function attributable solely to the S2-like sequences present in the MRP4 protein. The MRP4 protein represents another member of a class of mitochondrial ribosomal proteins that have sequences similar to E. coli ribosomal proteins within the context of a much larger protein (2-4). Whether or not the sequences unrelated to E. coli ribosomal proteins contribute to the function or stability of this class of mitochondrial ribosomal proteins has not been determined but is currently under investigation. An intriguing finding that hasemerged from analysis of the primary structure of the MRP4 protein is its sequence similarity to mammalian 68-kDa high affinity laminin receptors. This is true not only of the mitochondrial but all the other bacterial and chloroplast S2 homologs as well, although the extent of sequence similarity is highest with the mitochondrial protein. The 68-kDa laminin receptor has been implicated in mediating the adhesion of certain cell types to the extracellular matrix (40), and its level of expression appears to correlate with the metastatic potential of cells (41). The significance of the sequence similarity between the S2 family of proteins and the 68-kDa laminin receptor is not immediately obvious. It has been suggested that the 68-kDa protein may have functions within the cell other than promoting cellular adhesion to the extracellular matrix (42). If this is correct, future elucidation of the role of the intracellularlaminin receptor may help in understanding the basis for the sequence similarity with the S2 proteins.

5513

Formosa' has identified a highly conserved homolog of the 68-kDa laminin receptor in S. cereuisiae. The MRP4 protein shows about the same degree of similarity with the yeast protein identified by Formosa as it does with the human and mouse 68-kDa laminin receptors, making it unlikely that the MRP4 proteinis a directdescendant of the protein identified by Formosa. Instead, it would appear that theMRP4 protein is of eubacterial descent, and if the S2 family of proteins and the mammalian 68-kDa high affinity laminin receptors did share a common ancestor, the gene duplication must have occurred prior to thedivergence of the eubacterial and eukaryotic lineages. Acknowledgments-We are grateful to Dr. Paul Joyce for valuable discussions, Cynthia Greaton for technical assistance, and Ellen Ford for assistance in the preparation of the manuscript. We are also grateful to Dr. Timothy Formosa, Department of Biochemistry, University of Utah for communicating unpublished results.

REFERENCES 1. Sor, F., and Fukuhara, H. (1980) C. R. Acud. Sci. Paris Ser. D 29 1,933-936 2. Fearon, K., and Mason, T. L. (1988) Mol. Cell. Biol. 8,3636-3646 3. Dang, H., and Ellis, S. R. (1990) Nucleic Acids Res. 1 8 , 68956901 4. Kitakawa, M., Grohmann, L., Graack, H.-R., and Isono, K. (1990) Nucleic Acids Res. 1 8 , 1521-1529 5. Myers, A. M., Crivellone, M. D., and Tzagoloff, A. (1987) J. Biol. Chem. 262,3388-3397 6. Partaledis, J. A,, and Mason, T. L. (1988) Mol. Cell. Biol. 8,36473660 7. Grohmann, L., Graack, H.-R., and Kitakawa, M. (1989) Eur. J. Biochem. 183,155-160 8. Matsushita, Y., Kitakawa,. M... and Isono, K. (1989) Mol. & Gen. Genet. 2 i 9 , i19-124 9. McMullin. T. W.. Haffter. P.. and Fox. T. D. (1990) . , Mol. Cell. Biol. 10,4590-4595 10. Kang, W., Matsushita, Y., and Isono, K. (1991) Mol. Gen. Genet. 225,474-482 11. Wittmann-Liebold, B., and Bosserhoff, A. (1981) FEBS Lett. 129,lO-16 12. Rao, C.N., Castronovo, V., Schmitt, M.C., Wewer, U.M., Claysmith, A. P., Liotta, L. A., and Sobel, M. E. (1989) Biochemistry 28,7476-7486 13. Hunt, L. T., and Barker, W.C. (1988) Nucleic Acids Res. 1 6 , 5195 14. Tzagoloff, A., Akai, A., and Needleman, R.B. (1975) J. Biol. Chem. 250,8228-8235 15. Ellis, S. R., Morales, M. J., Li, J.-M., Hopper, A. K., and Martin, N. C. (1986) J. Biol. Chem. 261,9703-9709 16. Koerner, T. J., Hill, J. E., Myers, A. M., and Tzagoloff, A. (1990) Methods Enzymol. 194,477-490 17. Laemmli, U. K. (1970) Nature 227,680-685 18. Garfin, D. E. (1990) Methods Enzymol. 1 8 2 , 425-441 19. Li, J.-M., Hopper, A. K., and Martin, N. C. (1989) J. Cell. Biol. 109,1411-1419 20. Dang, H., Franklin, G., Darlak, K., Spatola, A. F., and Ellis, S. R. (1990) J. Biol. Chem. 2 6 5 , 7449-7454 21. Grivell, L. A., Reijnders, L., and Borst, P. (1971) Biochim. Biophys. Acta 247,91-103 22. Southern, E. M. (1975) J. Mol. Biol. 98,503-517 23. Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982) Molecular Cloning: A LaboratoryManuul, pp. 98-185, 364-401, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY 24. Maxam, A.M., and Gilbert, W. (1977) Proc.Natl.Acud.Sci. U. S. A . 74,560-564 25. Dieckman, C. L., and Tzagoloff, A. (1985) J. Biol. Chem. 2 6 0 , 1513-1520 26. Botstein, D., and Davis, R. W. (1982) in Molecular Biology of the Yeast Saccharomyces cerevisiue Metabolism and Gene Expression (Strathern, J. N., Jones, E. W., and Broach, J. R., eds) pp. 607-636, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY I

,

* T. Formosa, personal communication.

5514


27. Hill, J. E.,Myers, A. M., Koerner, T. J., and Tzagoloff, A. (1986) Yeast 2,163-167 28. Pearson, W. R., and Lipman, D.'J. (1988) Proc.Natl.Acad.Sci. U.S. A. 05,2444-2448 29; Hudson, G. S., Mason, J. G., Holton, T. A., Koller, B., Cox, G. B., Whitfeld, P. R., and Bottomley, W. (1987) J. Mol. Biol. 196,283-298 30. Shinozaki, K.,Ohme, M., Tanaka, M., Wakasugi, T., Hayashida, N., Matsubayashi, T., Zaita, N., Chunwongse, J., Obokata, J., Yamaguchi-Shinozaki, K.,Ohto, C., Torazawa, K., Meng, B. Y., Sugita, M., Deno, H., Kamogashira, T., Yamada, K., Kusuda, J., Takaiwa, F., Kato, A., Tohdoh, N., Shimada, H., and Sugiura, M. (1986) EMBO J. 5,2043-2049 31. Hoglund, A.-S., and Gray, J. C. (1987) NucleicAcidsRes. 16, 10590 32. Ohyama, K., Fukuzawa, H., Kohchi, T., Shirai, H., Sano, T., Sano, S., Umesono, K., Shiki, Y., Takeuchi, M., Chang, Z., Aota, S.-I., Inokuchi, H., and Ozeki, H. (1986) Nature 3 2 2 , 572-573 33. Vingron, M., and Argos, P. (1991) J. Mol. Bwl. 218,33-43

34. Rothstein, R. (1983) Methods Enzymol. 101,202-210 35. Myers, A. M.,Pape, K., L. and Tzagoloff, A. (1985) EMBO J. 4 , 2087-2092 36. Randall-Hazelbauer, L. L., and Kurland, C. G. (1972) Mol. & Gen.Genet. 1 1 6 , 234-242 37. Thomas, G., Sweeney,R., Chang, C., and Noller, H. F. (1975) J. Mol. Biol. 95, 91-102 38. Nomura, M., Mizushima, S., Ozaki, M., Traub, P., and Lowry, C. V. (1969) Cold Spring Harbor Symp.Quunt. Biol. 34,49-61 39. Dabbs, E. R. (1986) in Structure,Function, and Genetics of Ribosomes (Hardesty B., and Kramer, G., eds) pp 733-748, Springer-Verlag, New York 40. Akiyama, S. K., Nagata, K., and Yamada, K. M. (1990) Biochem. Biophys.Acta 1031,91-110 41. Terranova, V. P., Rao, C. N., Kalebic, T., Margulies, I. M., and Liotta, L. A. (1983) Proc. Natl. Acad. Sci. U.S. A . 80,444-448 42. Rabacchi, S. A., Neve, R. L., and Drager, U. C. (1990) Deuelopment(Camb.) 109,521-531 M., M. Scott, R., Clements, L. S., and 43. Parikh, V. S., Morgan, Butow, R. A. (1987) Science 2 3 5 , 576-580