Peptidyl-tRNA hydrolase in Bacillus subtilis, encoded by spoVC, is essential to vegetative growth, whereas the homologous enzyme in Saccharomyces cerevisiae is dispensable



Peptidyl-tRNA hydrolase in Escherichia coli, encoded by pth, is essential for recycling tRNA molecules sequestered as peptidyl-tRNA as a result of pre-mature dissociation from the ribosome during translation. Genes homologous to pth are present in other bacteria, yeast and man, but not in archaea. The homologous gene in Bacillus subtilis, spoVC, was first identified as a gene involved in sporulation. A second copy of spoVC, under the control of the xyl promoter, was integrated into B. subtilis at the amy locus. In this background, interruption of the original gene was possible provided that expression of the copy at the amy locus was induced. When spoVC was interrupted, both vegetative growth and sporulation were dependent on xylose, showing that SpoVC is essential. The role of SpoVC in sporulation is discussed and appears to be consistent with previous hypotheses that a relaxation of translational accuracy may occur during sporulation. The homologous gene in Saccharomyces cerevisiae, yHR189W, has been interrupted in both haploid and diploid strains. The mutant haploid strains remain viable, as do the yHR189W mutant spores obtained by tetrad dis-section, with either glucose or glycerol as carbon source, showing that the yHR189W gene product is dispensable for cell growth and for mitochondrial respiration.


The translation of mRNA by ribosomes is not perfectly processive. Thus, a significant proportion of ribosomes that initiate translation terminate before reaching the stop signal, resulting in the dissociation of peptidyl-tRNA. This phenomenon, often called peptidyl-tRNA ‘drop-off’, is thought to occur at an average frequency of about 3 × 10−4 per codon (Manley, 1978; Jørgensen and Kurland, 1990). In Escherichia coli, the enzyme peptidyl-tRNA hydrolase (Pth) is required for recycling peptidyl-tRNA by cleaving the ester bound between peptide and tRNA (Cuzin et al., 1967; Atherly and Menninger, 1972). In Pth-deficient cells, protein synthesis is arrested, and cells finally die from starvation for essential species of tRNA (Menninger, 1978; 1979), in the first instance for tRNALys, which becomes sequestrated as peptidyl-tRNA and unavailable for protein synthesis (Heurgué-Hamard et al., 1996). Enzymes homologous to E. coli Pth, a protein of 194 amino acids, are widespread in microorganisms, and also appear to exist in higher organisms, including mice and man. Genome sequencing of archaebacteria, on the other hand, does not appear to reveal sequences homologous to the E. coli enzyme. Biochemical studies in Saccharomyces cerevisiae have demonstrated the presence of enzymatic activity that shares some characteristics of Pth in E. coli (Kossel and RajBhandary, 1968; Jost and Bock, 1969). More recently, sequencing of the S. cerevisiae genome has led to the identification of a homologous gene, yHR189W, encoding a protein of 190 amino acids (Ouzounis et al., 1995).

Attempts to purify Pth activity from rabbit reticulocyte lysates have suggested the presence of an enzyme that may fulfil the same role as the enzyme in E. coli but appears, curiously, to use a strikingly different mechanism. Whereas Pth in E. coli is an esterase, the activity isolated from rabbit reticulocytes is a phosphodiesterase that liberates peptidyl-AMP from peptidyl-tRNA (Gross et al., 1992a). On sucrose gradient centrifugation, the activity sediments with an apparent Mr of 65 000, considerably larger than the E. coli enzyme.

In Bacillus subtilis, a gene homologous to E. coli pth is present at 5° on the chromosome and encodes a protein of 188 residues (Ogasawara et al., 1994). The gene, named spoVC, was first identified in studies of B. subtilis mutants affecting sporulation. A mutation affecting spoVC renders B. subtilis temperature sensitive for sporulation; at non-permissive temperature, spo-285 (Ts) mutant cells cease sporulation at stage IV to stage V (Young, 1976), characterized by spore cortex development and coat formation. The behaviour of this mutant suggests that spoVC is expressed long before it assumes a particular role related to sporulation, consistent with the possibility that it might be an essential gene. Here, we show that SpoVC does indeed possess the Pth activity suggested by the homology of the protein to members of the Pth family, as it can complement a thermosensitive pth(ts) mutant in E. coli isolated by Atherly and Menninger (1972). We show rigorously that SpoVC is required for vegetative growth in B. subtilis and discuss the putative relation between Pth activity and sporulation. Finally, we show that the homologous gene in S. cerevisiae is not required for growth.


SpoVC complements an E. coli mutant deficient in Pth

The spoVC gene of B. subtilis is part of a large operon containing 14 open reading frames (ORFs), including several of unknown function. The gene is expressed from a σ37-controlled promoter located shortly upstream from the SpoVC coding sequence (Moran et al., 1981). The high level of amino acid sequence identity (31%) and similarity (49%) between SpoVC and Pth from E. coli strongly suggests that SpoVC should hydrolyse free peptidyl-tRNA and be able to complement the mutant in E. coli affecting pth and showing a thermosensitive growth phenotype (Atherly and Menninger, 1972). To test this hypothesis, the spoVC gene was cloned under the control of the B. subtilis xyl promoter. Expression of spoVC from this plasmid requires xylose in B. subtilis but is constitutive in E. coli. Complementation experiments of the thermosensitive pth mutation in E. coli were performed in two strain backgrounds with differing degrees of thermosensitivity, VH4 and VH805 (Heurgué-Hamard et al., 1996; Menez et al., 2000). In both cases, growth was restored independently of xylose, confirming the ability of SpoVC to complement the E. coli pth mutation.

Bacillus subtilis spoVC expression is required for both vegetative growth and sporulation

No equivalent of the E. coli pth(ts) mutant has been reported in B. subtilis. The spo-285 (Ts) mutant described by Young (1976) displays a block in sporulation but is unaffected during vegetative growth. To test whether spoVC is an essential gene, we deleted the spoVC locus while cloning a second spoVC gene at the amylase locus, under the control of the xylose-inducible promoter. Two plasmids were constructed for this purpose, according to the strategy of Guerout-Fleury et al. (1996). One plasmid (pVC9) was designed to exchange the spoVC locus for an MLS resistance gene, and a second (pVC7) to allow spoVC expression under the control of a xylose-inducible promoter, constructed for integration at the amylase locus (Fig. 1). In a second recombination, spoVC and the downstream gene yabK were replaced by the MLS resistance gene erm. Thus, transformants selected on MLS-xylose lose the entire spoVC gene at its normal locus, resulting in strain BS4 (Fig. 1), in which SpoVC synthesis should depend entirely on the activity of the PxylA promoter. After purification in the absence of xylose, strain BS4 was no longer able to grow, providing strong evidence that the spoVC gene is indeed essential for vegetative growth. In the presence of xylose, BS4 shows a wild-type sporulation phenotype.

Figure 1.

. A. Plasmids pVC7 and pVC9 used for spoVC interruption and integration of xylR , P xylA , spoVC at the amylase locus. MLS resistance in B. subtilis is encoded by erm and ampicillin resistance in E. coli by bla .

B. Strain BS4 derived from recombination of pVC7 and pVC9 at amyE and ctc/mfd loci. Cat represents the gene for chloramphenicol resistance. PxylA and xylR are cloned upstream of the spoVC open reading frame.

Interruption of the yHR189W gene encoding a bacterial-type Pth in the yeast S. cerevisiae

The sequencing of chromosome VIII in S. cerevisiae revealed an ORF (yHR189W) potentially encoding a close orthologue of bacterial Pth (Ouzounis et al., 1995). Compared with bacterial Pth, the putative yeast protein has no significant N-terminal extension, and analysis of the amino acid sequence by the neural network method of Nielsen et al. (1997) for a potential signal peptide that would suggest a mitochondrial location for the protein gave negative results. The ability of the yHR189W gene product to function as Pth was verified by cloning the gene into the plasmid vector pTrc99c under the control of the trc promoter inducible by IPTG and transforming the Pth thermosensitive strain VH4. At the non-permissive temperature of 37°C, growth of VH4 was restored by the recombinant vector and not by the control vector without insert, provided that IPTG was present in the medium.

To determine experimentally whether the gene, called yHR189W, was necessary for either mitochondrial function or normal aerobic cell growth, we attempted to interrupt the gene in order to delete the majority of the coding sequence. The interruption was made by transformation with linear DNA (Baudin et al., 1993) of the two haploid HIS3 strains yIB27 and yIB33, as well as the diploid strain constructed by crossing the two haploid strains. Transformation of the haploid strains as well as the diploid strain gave rise to His+ recombinants, suggesting that yHR189W was not required for viability. To confirm this conclusion, sporulation of the diploid His+ recombinant was induced, and tetrads were dissected. The four spores were viable, and the expected pattern of segregation of the nuclear markers among the spores was confirmed. The presence of inactivated yHR189W in spores T2D and T2A and the absence of the intact gene was demonstrated by Southern blot (Fig. 2). This shows that the normal EcoRV fragment containing yHR189W (see strain yIB27) is replaced by a larger fragment of the expected size in spores T2D and T2A, with both fragments being found in DNA from the diploid strain (2nHis+). The new larger fragment was seen to hybridize to probes specific for both yHR189W and HIS3. The presence of the interruption and the absence of the normal yHR189W gene in the recombinant strains were also verified by polymerase chain reaction (PCR). As peptidyl-tRNA drop-off in bacterial cells is a temperature-dependent process (Cruz-Vera et al., 2000), growth of normal and yHR189W-deleted spores was compared at different temperatures. However, over a temperature range of 28–40°C, no differences in growth rate were detected. Finally, we studied the growth of the four spores on media with glycerol as carbon source; normal growth under these conditions requires active mitochondrial oxidative phosphorylation. All four spores grew normally, indicating that the yHR189W gene is not required for normal mitochondrial function.

Figure 2.

. Southern blot analysis of intact and interrupted yHR189W in S. cerevisiae .

A. DNA from the haploid strain yIB27, the diploid strain after recombination as described in the text (His+/His3), and from the four spores of T2 was digested with EcoRV and analysed first with a probe specific for the region of the yHR189W gene, showing the replacement of the normal EcoRV fragment containing yHR189W by a larger fragment when interrupted by HIS3. Expected sizes are shown on the right.

B. The membrane was partially dehybridized and rehybridized with a probe specific for HIS3, showing hybridization only to the larger of the EcoRV fragments. Lane 1 in each case shows a size standard made by labelling a BstEII digest of phage λ DNA.


Earlier studies using a spoVC mutant showed the involvement of the gene product in the process of sporulation in B. subtilis (Young, 1976). Here, we show in a rigorous manner that SpoVC is essential for vegetative growth in B. subtilis and confirm that the protein possesses Pth activity by its capacity to complement a Pth mutant in E. coli. Previously, Pth had only been shown to be essential to cell growth in E. coli (Atherly and Menninger, 1972).

The generally accepted role of Pth is to hydrolyse peptidyl-tRNA that has dissociated from the ribosome as a result of some sort of translational accident. Thus, Pth is not able to hydrolyse peptidyl-tRNA on 70S E. coli ribosomes, although the possibility has not been excluded that dissociation of the ribosomal subunits might expose peptidyl-tRNA remaining associated with the large subunit to the action of Pth. Various proposals have been made regarding the type of event that may lead to peptidyl-tRNA drop-off from the ribosome. Kurland and Ehrenberg (1985) argued that drop-off is an unavoidable consequence of optimizing the opposed requirements for ac-curate aminoacyl-tRNA selection and high translational processivity. They imply that drop-off occurs largely from cognate peptidyl-tRNA–mRNA complexes. Menninger (1977) proposed that drop-off is a cellular strategy to correct translational errors – the ‘ribosomal editor hypothesis’. He supposed that drop-off occurs frequently as a result of erroneous incorporation of an amino acid into the growing polypeptide chain, after a proofreading failure in the ribosomal A-site. The rationale is that the weak interaction between a codon and the anticodon of a non-cognate tRNA fails to be maintained at some point between the peptidyl transfer to the non-cognate aminoacyl-tRNA and the subsequent peptidyl transfer, thereby leading to drop-off. More recently, it has been shown that a significant part of peptidyl-tRNA drop-off results from to a factor-catalysed process during the early stages of polypeptide chain synthesis (Heurgué-Hamard et al., 1998; 2000).

The necessity to recycle tRNA that becomes se-questered as peptidyl-tRNA may depend on the physiological state of the cell, such as the rate of cell division and tRNA synthesis in relation to the rate of peptidyl- tRNA drop-off. During exponential growth, the fraction of individual tRNA species present as free peptidyl-tRNA in the cell will stabilize at levels depending on the relative fluxes feeding and emptying the pools of peptidyl-tRNA, namely peptidyl-tRNA drop-off on the one hand, and Pth activity and tRNA synthesis on the other. Whether or not Pth activity is required for cell growth should depend on the magnitude of the respective rate parameters and the capacity of the cell to support some peptidyl-tRNA, as no proof exists that peptidyl-tRNA is toxic to cells other than by reducing the level of tRNA available for protein synthesis. In contrast, in non-dividing cells, in the absence of new tRNA synthesis, the recycling by Pth of tRNA sequestered as peptidyl-tRNA seems more likely to be essential to cell viability. This may be part of the explanation for the fact that the spo-285 (Ts) mutation in B. subtilis affects sporulation but not vegetative growth.

Recent in vitro studies of peptidyl-tRNA drop-off and of the substrate specificity of Pth tend to confirm that errors in aminoacyl-tRNA selection are a major cause of peptidyl-tRNA drop-off (Dinçbas et al., 1999; Heurgué-Hamard et al., 2000; V. Heurgué-Hamard, unpublished data), as first proposed by Menninger (1977). This suggests a further explanation for the important role of SpoVC in sporulation. Genetic studies in several lower eukaryotes strongly suggest that an increase in translational ambiguity may be a requirement at critical stages of cell differentiation (Picard-Bennoun, 1982; Silar and Picard, 1994). Increased ambiguity is very likely to be accompanied by increased peptidyl-tRNA drop-off and an increased requirement for SpoVC.

It should be pointed out that the thermosensitivity of the spo-285 mutation in sporulation may be a characteristic of peptidyl-tRNA drop-off rather than a property of the mutant enzyme. Thus, recent experiments suggest that Pth activity in the E. coli pth mutant that is thermosensitive for growth varies little with temperature, suggesting that the thermosensitivity must arise as a result of increased drop-off at non-permissive temperatures (Cruz-Vera et al., 2000).

No purification of Pth from S. cerevisiae has been described, but an enzyme has been isolated from the related yeast hybrid strain Saccharomyces fragilis/ Saccharomyces dobzhanskii that was able to hydrolyse N-substituted aminoacyl- or N-substituted peptidyl-tRNAs but not free aminoacyl-tRNAs (Jost and Bock, 1969). We do not know whether this activity is similar to that of the yHR189W gene product in S. cerevisiae. The demonstration that S. cerevisiae can dispense with yHR189W function without loss of viability raises the question as to whether the function of peptidyl-tRNA hydrolysis is essential in this organism, or whether another enzymatic system is present that may recycle tRNAs sequestered as peptidyl-tRNA, as has been suggested from studies in animal cells (Gross et al., 1992a,b). The apparent absence of the bacterial type of Pth in Archaea raises a similar question. The construction of a strain lacking the prokaryotic-like Pth may facilitate identification of such an alternative system in yeast, if it exists. In preliminary experiments, we have shown that cell extracts from a yHR189W-interrupted strain contain an enzymatic ac-tivity that can hydrolyse Ac2Lys-tRNALys and can be separated from ribonuclease activity by ion-exchange chromatography.

Experimental procedures

Bacterial strains

Strains of E. coli K-12, B. subtilis and S. cerevisiae are listed in Table 1. TG1 (Wain-Hobson et al., 1985) was used for shuttle vector cloning. MO1800 was used because it allows efficient xylose induction and facilitates recombination at the amylase locus. Competent B. subtilis cells were prepared and transformed according to the method of Anagnostopoulos and Spizizen, 1961), and sporulation was induced by exhaustion in Difco nutrient broth medium. PCR amplification for cloning was performed on MO963 as wild-type strain.

Table 1. .  E. coli K-12, B. subtilis and S. cerevisiae strains.
E. coli
 TG1 supE, thiD ( Δ lac-proAB) hsd5/F + , traD36, proAB + , lacIq Δ M15 Wain-Hobson et al. (1985 )
 VH4 supE44, thi1, thr1, leuB6, lacY1, tonA21, trp-, pth (Ts) Heurgué-Hamard et (1996)
 VH805 metb, glyV-55, pth (Ts) , recA ::Cm Menez et al. (2000 )
B. subtilis
 MO963 trpC2, pheA1 Frandsen et al. (1999 )
 MO1800 xylA, amyE :: erm F. Arigoni
 BS3 MO1800 , amyE:: pXyl- spoVC This work, BS3 × pVC7
 BS4BS3 spoVC::ermThis work, MO963 × pVC9
S. cerevisiae
 yIB27Mata ura3Δ851, leu2Δ1, his3Δ200, lys2Δ202Laboratory strain
 yIB33Matαtrp1Δ63, hisΔ200Laboratory strain
 yJMM116Mata ura3Δ851, leu2Δ1, yHR189::HIS3, lys2Δ202 
 yJMM117Matαtrp1Δ63, yHR189::HIS3 
 yJMM118Mata ura3Δ851, leu2Δ1, yHR189::HIS3 
 yJMM119/T2AHis+ ura trp lys leu+Spore A tetrad no. 2
 (or T2B, C, D) (or spore B, C, D)

Growth conditions

For E. coli, Luria–Bertani (LB) was used as rich medium. Antibiotics were added as necessary at the following concentrations: ampicillin 200 μg ml−1, tetracycline 12.5 μg ml−1 and chloramphenicol 15 μg ml−1. For B. subtilis, LB was also used as rich medium, supplemented as necessary with chloramphenicol (5 μg ml−1) or spectinomycin (100 μg ml−1), lincomycin (25 μg ml−1) and erythromycin (1 μg ml−1) for macrolide–lincosamide–streptogramin B (MLS) resistance. The Pspac plasmid promoter was induced when necessary by the addition of IPTG to a final concentration of 1 mM. The xylose promoter was induced by the addition of xylose to 50 mM in rich medium. Complementation studies in E. coli pth thermosensitive strains were performed on LB-ampicillin plates at non-permissive temperatures as stated in the text. For S. cerevisiae, the following media were used, YEP-glucose (2%) or YEP-glycerol (3%) as rich medium, complete synthetic medium with supplements as required by auxotrophies and minimum synthetic medium (Adams et al., 1997).

The spoVC locus under xylose control in a spoVC deletion strain

The construction of a strain containing a spoVC gene under xylose control used two plasmids (pVC9 and pVC7), which can recombine with the chromosome by double recombination events. The spoVC gene was amplified using primers spoVCBam (CGCGGATCCAGTTATGTGACTAAGGGAGGA) and spoVCXba (GCTCTAGATTGGTGTAGTATGACTTTGT). The PCR product was cleaved by BamHI and XbaI and cloned with two further fragments: the xylose-inducible promoter and its repressor carried on an XbaI–EcoRI fragment and excised from pDG1832 (Webb et al., 1997); and the amylase recombination sequences, the ‘amy back’ and ‘amy front’EcoRI–BamHI region from pDG1662 (Guerout-Fleury et al., 1996). Ligation of these three fragments creates pVC7, containing resistance genes for ampicillin, spectinomycin and chloramphenicol, from which spoVC expression depends on the xyl promoter (Fig. 1). A region of 1620 nucleotides around spoVC, from ctc to after yabK, was amplified by PCR using the following primers: OlctcBs (GATTCATCTGACCGGAGAAGC) and OlyabKBs (TTGAC GGGGTACAAAAGGACA). The PCR product was then cloned in pDG271 (Stragier et al., 1988) between the PstI and XbaI sites. A HindIII site between the cat and tet genes was eliminated by blunt ligation of StuI–EcoRV termini, resulting in pVC5. The erm resistance region of pDG939 was cloned in pUC19 to allow clonage in pVC5 between the remaining HindIII site and the SmaI site. The resulting plasmid (pVC6) contains the neighbouring upstream and downstream sequence of spoVC and an erm resistance gene present at the spoVC locus. Finally, as the HindIII deletion destroyed resistance to both chloramphenicol and tetracycline in pVC5, a new spectinomycin resistance cassette from pDG1726 (Guerout-Fleury et al., 1995) was cloned in pVC5 between EcoRI and BamHI (yielding pVC8). The spoVC::erm region was then recloned from pVC6 to pVC8 to give plasmid pVC9 (Fig. 1). MO1800 was first transformed with linearized pVC7, selecting for chloramphenicol resistance, screening for loss of MLS and spectinomycin resistance. The resulting strain (BS3) was then transformed with pVC9 linearized by PvuI digestion, selecting on LB-xylose-MLS plates. Loss of spectinomycin resistance was verified, and the resulting strain was BS4 (Fig. 1).

Deletion of yHR189W in S. cerevisiae

The following oligonucleotides were used to amplify the HIS3 gene in order to provide termini (underlined) homologous to the two ends of gene yHR189W. yHR189Wup: 5′-TGGAGAC TAGTGTCTGACCGGGATAGGCAATCCAGAGCCTCA GTACGCTGGAAAGGAAAGCGCGCCTC-3′; yHR189Wlow: 5′-GGCTATGAAATGTACTGAGTCAGAGCACGCCAGG CAGCAGGTTCACTCGCGCTAGGAGTCACTGC-3′.Recombination with deletion of most of the yHR189W gene was performed according to the method of Baudin et al. (1993).

Southern blot

After restriction enzyme digestion, DNA fragments were separated by gel electrophoresis in 0.7% agarose gels. Phage λ DNA digested with BstEII and labelled with [γ-32P]-ATP was used as molecular weight marker. Labelled probes for the yHR189W and HIS3 genes were prepared by PCR amplification of the complete genes followed by labelling with a Ready Prime II labelling kit (Pharmacia) according to the manufacturer's instructions.


This work was supported by the Centre National de la Recherche Scientifique (UPR 9073), Aventis Pharma France (convention de recherche PES0781098), l’Association pour la Recherche sur le Cancer and the Fondation pour la Recherche Médicale.