The fate of extracellular proteins tagged by the SsrA system of Bacillus subtilis

Marc A. B. Kolkman1 and Eugenio Ferrari2

1 Genencor International BV, Archimedesweg 30, 2333 CN Leiden, The Netherlands
2 Genencor International, Inc., 925 Page Mill Road, Palo Alto, CA 94304, USA

Correspondence
Marc A. B. Kolkman
mkolkman{at}genencor.com


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In bacteria, SsrA, a highly conserved RNA molecule, functions in a mechanism meant to rescue stalled ribosomes. In this process, a peptide tag encoded by SsrA is cotranslationally added to truncated polypeptides, thereby targeting these molecules for proteolytic degradation, at least when they stay inside the cell. This study examined the fate of two extracellular proteins that were tagged by the SsrA system of Bacillus subtilis. Gene constructs encoding human interleukin-3 (hIL-3) fused to a signal peptide and B. subtilis {alpha}-amylase, both lacking an in-frame stop codon, were used as models to achieve ribosome stalling and activation of the SsrA system. Introduction of these gene constructs into B. subtilis led to tagging of the gene products by SsrA RNA. The tagged protein products bound to antibodies that were raised against the proteolysis tag encoded by B. subtilis SsrA [(A)GKTNSFNQNVALAA]. The apolar C-terminal SsrA-tag does not function as a specific signal for proteolytic degradation of SsrA-tagged amylase directly after trans-translation or during the secretion process. Also, SsrA-tagged amylase appeared to be very stable once outside the cell. In contrast, hIL-3 molecules tagged with the native, apolar SsrA-tag were considerably less stable than hIL-3 molecules that received a negatively charged control tag. Not one specific protease, but several non-specific proteases seem to play a role in the rapid degradation of SsrA-tagged hIL-3. The polarity of the C-terminus of heterologous hIL-3 protein proved to be an important determinant for protein stability when produced by B. subtilis. As observed previously in Escherichia coli and B. subtilis, SsrA tagging also occurs frequently in normally growing Gram-positive bacilli and it appears that intracellular proteins are the predominant natural substrates of SsrA.


Abbreviations: hIL-3, human interleukin-3; mIL-6, mouse interleukin-6


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Proteolysis in bacteria serves to rid the cell of abnormal and misfolded proteins. A unique mechanism for the destruction of abnormal proteins resulting from abortive termination of translation is provided by the SsrA-mediated tagging and degradation system (for a review, see Karzai et al., 2000). SsrA, also called 10Sa RNA or tmRNA, is a highly conserved RNA molecule in eubacteria. It is a unique molecule that can act as both a tRNA and an mRNA in a process referred to as trans-translation (Atkins & Gesteland, 1996; Jentsch, 1996; Keiler et al., 1996). This mechanism provides the cell with a way to release ribosomes that are stalled on untranslatable mRNAs, e.g. mRNAs lacking in-frame stop codons. In the model for SsrA action, SsrA charged with alanine enters the A site of a stalled ribosome, mimicking a tRNA. The alanine is added to the incomplete polypeptide chain; and then, serving as an mRNA, SsrA provides a short reading frame followed by a stop codon as a template to add a short peptide to the nascent polypeptide before translation terminates and a tagged protein is released. The peptide tag (encoded by SsrA) functions as a proteolytic degradation signal, and in Escherichia coli four proteases have been identified that degrade proteins tagged by SsrA. ClpXP, ClpAP and FtsH (HflB) degrade SsrA-tagged proteins in the cytoplasm (Gottesman et al., 1998; Herman et al., 1998), while SsrA-tagged proteins with signal peptides that are exported to the periplasm of E. coli are degraded by Tsp (Prc) protease (Keiler et al., 1996).

Several determinants have been discovered that lead to activation of the SsrA tagging system. SsrA tagging occurs when ribosomes stall at clusters of rare codons in an mRNA when the cognate tRNA is scarce (Roche & Sauer, 1999; Hayes et al., 2002b), or at poorly efficient stop codons (Collier et al., 2002). Also, the C-terminal amino acid sequence of nascent chains has been found to be a major determinant for SsrA tagging (Hayes et al., 2002a; Sunohara et al., 2002). The whole story leading to the elucidation of SsrA function started with the observation by Tu et al. (1995) that a fraction of mouse interleukin-6 (mIL-6) overexpressed in E. coli is truncated and contains the SsrA-tag. It is not clear why some of the mIL-6 molecules were tagged by SsrA. Perhaps mIL-6 mRNA is relatively unstable in E. coli, leading to transcripts that are trimmed at the 3' end by nucleases, thereby losing their stop codon. Alternatively, rare low tRNA codons in mIL-6 mRNA and/or overexpression of mIL-6 itself may lead to translational problems, thereby activating the SsrA tagging system. Whatever the reason, contamination of recombinant proteins with molecules that are truncated and/or tagged by the SsrA system restricts their usefulness in pharmaceutical and industrial applications. Therefore, we studied the SsrA-mediated peptide-tagging system in Bacillus subtilis, an industrially important species used for the commercial production of various proteins. We asked ourselves what would happen to a protein that is tagged by the SsrA system of B. subtilis and targeted outside the cell. Are these molecules degraded directly after the trans-translation event, during the secretion process, and/or by extracellular proteases in the medium? Or are these SsrA-tagged molecules able to escape proteolysis?

B. subtilis SsrA was isolated and sequenced several years ago (Ushida et al., 1994) and the sequence of the proteolysis tag encoded by B. subtilis SsrA [(A)GKTNSFNQNVALAA] has been predicted (Williams, 2000, 2002). This sequence was recently confirmed by mass spectroscopy (Ito et al., 2002). Wiegert & Schumann (2001) showed that the ClpXP protease is responsible for the degradation of intracellular SsrA-tagged proteins in B. subtilis. In our study, we focused on SsrA tagging of two different extracellular proteins. Here we show that a native extracellular protein, AmyE, tagged by the SsrA system of B. subtilis is not subjected to proteolytic degradation. Heterologous human interleukin-3 (hIL-3) protein that received the native, apolar SsrA-tag proved to be more susceptible to proteolytic degradation than the same molecule tagged with a polar control tag. We also provide data showing that polarity of the extreme C-terminus of a heterologous protein produced by B. subtilis is an important determinant for protein stability.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Plasmids, bacterial strains and media.
Table 1 lists the plasmids and bacterial strains used in this study. E. coli strains were grown in or on 2x YT medium (Bacto Tryptone, 16 g l-1; yeast extract, 10 g l-1; NaCl, 5 g l-1). B. subtilis strains were grown in TSB (Tryptone Soya Broth from Oxoid, 30 g l-1) or 2x SMM (Spizizen's minimal medium; Harwood & Cutting, 1990), or on SMA (Spizizen's minimal agar; Harwood & Cutting, 1990) or HI agar (Heart Infusion agar from Difco, 40 g l-1). When appropriate, media were supplemented with ampicillin, 100 µg ml-1; chloramphenicol, 5 µg ml-1; neomycin, 10 µg ml-1; spectinomycin, 100 µg ml-1; and/or tetracycline, 10 µg ml-1.


View this table:
[in this window]
[in a new window]
 
Table 1. Plasmids and strains

 
DNA techniques.
Plasmid DNA was isolated with the QIAprep spin miniprep kit (Qiagen) according to the instructions, except that B. subtilis cells were incubated with lysozyme (5 mg ml-1 in buffer P1) for 10 min at 37 °C prior to addition of lysis buffer (buffer P2). Chromosomal DNA was isolated as described by Harwood & Cutting (1990). Procedures for DNA restriction, ligation, agarose gel electrophoresis, and transformation of E. coli were carried out as described in Sambrook et al. (1989). Transformation of competent cells was used to transfer DNA (plasmids, linear DNA) into B. subtilis (Harwood & Cutting, 1990). PCR was carried out with High Fidelity Platinum Taq DNA Polymerase and if required PCR fragments were purified with the Qiaquick PCR purification kit (Qiagen). DNA primers (Table 2) and enzymes were purchased from Invitrogen and DNA sequencing was performed by BaseClear (Leiden, The Netherlands). To construct a B. subtilis 168 {Delta}ssrA mutant, ssrA and its flanking regions (approx. 2·2 kb) were amplified by PCR with primers pSsrAFW and pSsrARV and cloned into pCR2.1-TOPO, resulting in plasmid pTPSsrA. Plasmid pSsrASp was obtained by inserting a pDG1726-derived Sp resistance marker (Guérout-Fluery et al., 1995) into the unique SacI site in the ssrA gene of pTPSsrA. Finally, strain 168 {Delta}ssrA was obtained by a double-crossover recombination event between the disrupted ssrA gene of pSsrASp and the chromosomal ssrA gene in B. subtilis 168. The SsrADD-expressing B. subtilis strain was made as follows. A fragment consisting of the 5'-terminal part of ssrA, including the ssrA promoter region, was amplified with the primers pSsrAHindIIIfw and pSsrADDintRV (contains the alteration of the two alanine codons in the SsrA-tag sequence into codons for aspartic acid residues). In addition, an overlapping the 3'-terminal part of ssrA was amplified with the primers pSSrADDintFW (also containing the alteration of the two alanine codons into codons for two aspartic acid residues) and pSsrASphIRV. Both fragments were assembled in a fusion PCR with primers pSsrAHindIIIfw and pSsrASphIRV, and cloned in pCR2.1-TOPO, resulting in plasmid pSsrADD. The correct sequence of the fusion product in pSsrADD was confirmed by DNA sequencing. Next, a selective marker (the Tc resistance cassette derived from pDG1515; Guérout-Fluery et al., 1995) that functions in B. subtilis, was cloned into the EcoRV site of pSsrADD, resulting in plasmid pSsrADDTc. Finally, B. subtilis 168 IssrADD was obtained by a Campbell-type integration (single crossover) of pSsrADDTc into one of the disrupted ssrA regions on the chromosome of B. subtilis 168 {Delta}ssrA. This strain contains an active copy of the ssrADD gene on the chromosome (under control of the native ssrA promoter) and a disrupted copy of wild-type ssrA (insertion of the Sp resistance marker), as confirmed by PCR. Plasmid pX-AT was created by cloning a PCR fragment, amplified with primers pAmyE(SpeI)fw and pAmyETT(BamHI)rv, containing amyE with the folC transcription terminator inserted in front of its stop codon, into pX. Linearized pX-AT was transformed into B. subtilis 168, 168 {Delta}ssrA, 168 IssrADD and the triple protease-negative strain BSE23. Transformants that contain the amyE-terminator (amyE-TERM) construct behind the xylA promoter in the chromosomal amyE locus were selected by PCR and these strains were designated 168-XAT, 168 {Delta}ssrA-XAT, 168 IssrADD-XAT and BSE23-XAT, respectively. Plasmid pX-hIT was created by cloning a PCR fragment, amplified with the primers hIL3(SpeI)fw and hIL3T(BamHI)rv on pLATIL3 as template, containing the amyL-hIL3 gene with the folC transcription terminator inserted in front of its stop codon, into pX. Linearized pX-hIT was transformed into B. subtilis 168, 168 {Delta}ssrA, 168 IssrADD, 168 {Delta}yvjB, 168 {Delta}ctpA, OS14 and BSE23, resulting in the strains 168-XhIT, 168 {Delta}ssrA-XhiT, 168 IssrADD-XhIT, 168 {Delta}yvjB-XhIT, 168 {Delta}ctpA-XhIT, OS14-XhIT and BSE23-XhIT, respectively. Plasmid pLATILBStag was obtained by PCR using pLATIL3 as template and primers pLATIL3T2FW and pIL3BStagRV. The resulting PCR fragment was purified, digested with XhoI, self-ligated, and transformed into B. subtilis. Clones were checked by DNA sequencing and one correct clone was selected and named pLATIL3BStag. Plasmid pLATIL3DDtag and pLATIL3ECtag were made in the same way, but instead of primer pIL3BstagRV, primer pIL3DDtagRV and primer pIL3EctagRV were used, respectively. To construct B. subtilis 168 {Delta}ctpA, B. subtilis 168 was transformed with chromosomal DNA of BSE23. In BSE23, the ctpA gene was replaced by a Sp resistance cassette (Edwin Lee, Genencor International, Palo Alto, unpublished data). B. subtilis 168 {Delta}yvjB was obtained as follows. yvjB and its flanking regions (approx. 3·5 kb) were amplified by PCR with the primers pYvjBFW and pYvjBRV, and cloned in pCR2.1-TOPO, resulting in plasmid pTPYvjB. Plasmid pTPYvjBTc was obtained by replacing an internal SmaI–AccI fragment of the yvjB gene in pTPYvjB with a pDG1515-derived Tc resistance marker (Guérout-Fluery et al., 1995). Finally, B. subtilis 168 {Delta}yvjB was obtained by a double-crossover recombination event between the disrupted yvjB gene of pTPYvjBTc and the chromosomal yvjB gene. To construct B. subtilis 168 IclpP, the 5' end region of the clpP gene was amplified by PCR with the primers pClpPEcoFW containing a EcoRI site and pClpPBamRV containing a BamHI site. The amplified fragment was cleaved with EcoRI and BamHI, and cloned into EcoRI/BamHI-digested pMutin2 (Vagner et al., 1998), resulting in plasmid pMutClpP. B. subtilis 168 IclpP was obtained by a Campbell-type integration (single crossover) of pMutClpP into the clpP region on the chromosome. Cells of this strain are depleted for ClpP by growing them in medium without IPTG (Vagner et al., 1998).


View this table:
[in this window]
[in a new window]
 
Table 2. Oligonucleotides

 
Western blot analysis.
To obtain anti-BsSsrAtag antibodies (antibodies that recognize proteins with a C-terminal B. subtilis SsrA-tag), synthetic peptide AGKTNSFNQNVALAA [coupled via an N-terminal cysteine residue to KLH (keyhole limpet haemocyanin) carrier] was injected into rabbits (Eurogentec). Serum of the final bleed of one of the rabbits was selected for affinity purification, and this purified serum was used in the Western blot procedures. Evaluation of these antibodies revealed that they also recognize proteins with AGKTNSFNQNVALDD as C-terminal tag (DD-tag). Antibodies against hIL-3 were mouse monoclonals (Van Leen et al., 1991). Antibodies against AmyE were kindly provided by Marie-Françoise Petit-Glatron. Immunoblotting and detection was performed with alkaline-phosphatase-labelled conjugate and the BM Chromogenic Western Blotting kit (Roche Diagnostics) according to the instructions of the manufacturer.

Protein labelling, SDS-PAGE and fluorography.
Pulse–chase labelling of B. subtilis and SDS-PAGE was essentially as described by Van Dijl et al. (1991). However, samples collected after chase times of 0, 5, 10, 30 and 60 min were centrifuged for 10 s, and only the extracellular proteins (in the culture supernatant) were precipitated with trichloroacetic acid and eventually subjected to SDS-PAGE. Fluorography was performed with Amplify fluorographic reagent (Amersham-Pharmacia Biotech). Protein bands were quantified using the Storm PhosphorImager system (Molecular Dynamics).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Expression of the amyE gene lacking an in-frame stop codon leads to SsrA-tagged but stable amylase molecules
To analyse SsrA-mediated tagging of {alpha}-amylase (AmyE), a native, extracellular B. subtilis protein encoded by the amyE gene, we first created two ssrA mutants (see Methods). The first mutant, 168 {Delta}ssrA, is an SsrA-deficient strain as confirmed by Northern analysis (data not shown). The second mutant, 168 IssrADD, expresses an SsrA variant (SsrADD) in which the final two codons of the peptide reading frame are changed to encode aspartic acid residues instead of alanines. In E. coli, it was shown that an SsrADD variant mediates the addition of a peptide tag (DD-tag) that does not lead to rapid degradation (Abo et al., 2000; Keiler et al., 1996). As a model to study the SsrA system in B. subtilis, a gene construct (amyE-TERM) was created in which a transcription terminator was inserted into the amyE gene, just in front of its stop codon. The amyE-TERM gene construct under control of the xylose-inducible (and glucose-repressible) xylA promoter (XAT) was introduced into the B. subtilis chromosome by the pX system (Kim et al., 1996). Expression of amyE-TERM will result in transcripts lacking an in-frame stop codon. According to the tmRNA model for SsrA-mediated tagging of proteins (Keiler et al., 1996), translation of these transcripts will result in ribosome stalling, and subsequently recruitment of SsrA and peptide tagging. Expression of amyE-TERM was studied in a wild-type background in B. subtilis 168-XAT, in an ssrA-deficient background (168 {Delta}ssrA-XAT), in an ssrADD strain (168 IssrADD-XAT), and in the triple protease-negative strain BSE23-XAT. Cells were grown to the end of exponential phase and then induced with xylose. After 2 h of induction, glucose was added to block transcription of amyE-TERM and at four time-points SsrA-tagged AmyE in the medium was analysed with anti-BsSsrAtag and anti-AmyE antibodies (Fig. 1). A culture of B. subtilis 168-XAT that was not induced with xylose was included as a control. SsrA-tagged amylase was detected with the anti-SsrAtag antibodies in the medium of wild-type B. subtilis expressing amyE-TERM, but only when cells were induced with xylose and not when SsrA was absent. The protein bands that were detected with the anti-SsrAtag antibodies also reacted with the anti-AmyE antibodies, confirming that these bands are indeed SsrA-tagged amylase. Even when produced in a wild-type background, SsrA-tagged AmyE appeared to be very stable in the medium (still present after 2 h). Deletion of two major extracellular proteases of B. subtilis (AprE, NprE) and membrane-associated protease CtpA, a homologue of E. coli Tsp (Marasco et al., 1996), did not affect the amount of SsrA-tagged amylase in the medium. Furthermore, the level of tagged amylase detected in the medium of 168 IssrADD was similar to that of cells expressing wild-type SsrA, indicating that SsrA-tagged amylase is also not subjected to tag-specific degradation directly after trans-translation, during the secretion process, and/or while passing through the B. subtilis cell wall.



View larger version (29K):
[in this window]
[in a new window]
 
Fig. 1. SsrA-tagged amylase is not rapidly degraded. Cultures of strains 168-XAT, 168 {Delta}ssrA-XAT, 168 IssrADD-XAT and BSE23-XAT were grown to the end of the exponential phase and then induced with 0·4 % xylose (except where indicated: ‘uninduced’). After 2 h, glucose was added (0·5 %) to block transcription of amyE-TERM and samples were taken at this point (t=0 min) and after 30, 60 and 120 min. After centrifugation, the proteins in the culture supernatants were analysed by SDS-PAGE and Western blotting with anti-BsSsrAtag or anti-AmyE antibodies. The tagged AmyE bands are arrowed.

 
Degradation of hIL-3 molecules tagged by the SsrA system of B. subtilis
As a model for SsrA-mediated peptide tagging of a heterologous protein produced in B. subtilis, a gene construct was made that contained the hIL-3 gene (encoding human interleukin-3) preceded by a signal peptide sequence of the Bacillus licheniformis amyL gene, and in which the folC transcription terminator was inserted just in front of the stop codon of hIL-3. As described above for the amyE-TERM construct, the pX system (Kim et al., 1996) was used to introduce this new gene construct (hIL3-TERM) under control of the xylA promoter (XhIT), into the amyE locus of B. subtilis. Expression of hIL3-TERM was studied in a wild-type background in B. subtilis 168-XhIT, in an ssrA mutant strain (168 {Delta}ssrA-XhIT), an SsrADD-expressing strain (168 IssrADD-XhIT), and in several protease-negative strains. Cells were grown to the end of exponential phase and induced with xylose for 2 h. Then glucose was added to block transcription of hIL3-TERM and SsrA-tagged hIL-3 in the medium was analysed at four time-points using anti-SsrAtag antibodies (Fig. 2). As expected, no SsrA-tagged protein was present in the SsrA-deficient strain. SsrA-tagged hIL-3 was also not detected in the medium of the wild-type strain, indicating that these molecules are rapidly degraded somewhere. In the SsrADD strain, in which hIL-3 molecules received a DD-tag, a protein band was detected that reacted with the anti-SsrA-tag antibodies. We confirmed with anti-hIL-3 antibodies that this band is SsrADD-tagged hIL-3 and these molecules were only detected when cells were induced with xylose (data not shown). We conclude that hIL-3 molecules with the apolar SsrA-tag at the C-terminus are more susceptible to proteolytic degradation than molecules with the DD-tag at the C-terminus. Nevertheless, even with the DD-tag, hIL-3 molecules are still relatively unstable (completely degraded within 60 min). This, however, relates to the instability of hIL-3 itself and presumably also applies to many other heterologous proteins expressed in B. subtilis. To examine if ClpP, the proteolytic compound of ClpXP protease that is responsible for degradation of intracellular SsrA-tagged proteins in B. subtilis (Wiegert & Schumann, 2001), two major extracellular proteases (AprE, NprE), or two homologues of E. coli Tsp in B. subtilis (CtpA, YvjB) play a role in the degradation of tagged hIL-3, we also expressed hIL3-TERM in several protease-negative strains. The clpP mutation did not increase the amount of tagged hIL-3 in the medium, indicating that ClpXP protease is not responsible for intracellular degradation of SsrA-tagged hIL-3. Minor amounts of SsrA-tagged hIL-3 were just detectable when hIL3-TERM was expressed in the ctpA mutant or in the aprE nprE double mutant, but not in the medium of the yvjB mutant. The triple aprE nprE ctpA mutant showed levels of SsrA-tagged hIL-3 in the medium close to that observed in the SsrADD strain. These observations suggest that SsrA-tagged hIL-3 is not degraded by one specific protease, but rather by several (non-specific) proteases.



View larger version (32K):
[in this window]
[in a new window]
 
Fig. 2. SsrA-tagged hIL-3 is rapidly degraded and several proteases seem to play a role in this degradation. Cultures of strains 168-XhIT, 168 {Delta}ssrA-XhIT, 168 IclpP-XhIT, 168 {Delta}yvjB-XhIT, 168 {Delta}ctpA-XhIT, OS14-XhIT, BSE23-XhIT and 168 IssrADD-XhIT were grown to the end of the exponential phase and then induced with 0·4 % xylose. After 2 h, glucose was added (0·5 %) to block transcription of amyL-hIL3-TERM and samples were taken at this point (t=0 min) and after 30, 60 and 120 min. After centrifugation, the proteins in the culture supernatants were analysed by SDS-PAGE and Western blotting using anti-BsSsrAtag antibodies. The tagged hIL-3 bands are arrowed.

 
Stability of hIL-3 variants with different C-terminal tags produced by B. subtilis
The observation that hIL-3 molecules with a polar, charged tag at the C-terminus are less susceptible to proteolytic degradation than molecules with an apolar C-terminus prompted us to further investigate the stability of tagged hIL-3 molecules produced by B. subtilis. For this purpose, we used plasmid pLATIL3, which is a derivative of pGB/IL322 (Van Leen et al., 1991), containing an expression cassette for the production of hIL-3. Also in this construct, the B. licheniformis {alpha}-amylase (AmyL) signal peptide is used to direct secretion of mature hIL-3 into the medium. Three variants of pLATIL3 were created. Plasmid pLATIL3BStag contains a gene variant encoding hIL-3 fused at the C-terminus to the B. subtilis SsrA peptide tag (AGKTNSFNQNVALAA); plasmid pLATIL3ECtag contains a gene variant encoding h-IL3 fused at the C-terminus to the E. coli SsrA-tag (AANDENYALAA). The third plasmid, pLATIL3DDtag, contains a gene encoding h-IL-3 fused at the C-terminus to the sequence encoding a DD-tag (AGKTNSFNQNVALDD). This tag is the same as the B. subtilis SsrA-tag (AA-tag), except that instead of two alanines at the extreme C-terminus it contains two aspartic acid residues. Equal amounts of total extracellular protein (checked by Coomassie staining; data not shown) produced by cells of B. subtilis 168 containing pLATIL3, pLATIL3BStag, pLATIL3DDtag or pLATIL3ECtag were analysed by Western blotting (Fig. 3a). The amount of the hIL-3-DDtag present in the medium was found to be roughly five times higher than that of wild-type hIL-3, hIL-3-AAtag or hIL-3-ECtag. As stated above, hIL-3 molecules produced by wild-type B. subtilis are relatively unstable due to proteolytic degradation, and the results represented in Fig. 3(a) suggest that addition of a C-terminal SsrA-tag does not lead to increased degradation of hIL-3 molecules. It is important to note, however, that in E. coli, proteins tagged cotranslationally by the SsrA system are degraded more rapidly than proteins with essentially the same sequence in which the SsrA-tag is DNA encoded (Gottesman et al., 1998). Strikingly, addition of the DD-tag (with two charged, polar residues at the extreme C-terminus) leads to a higher level of extracellular hIL-3, indicating that DD-tagged hIL-3 is less susceptible to proteolytic degradation. To explore this further, a pulse–chase assay was performed with the B. subtilis strains 168(pLATIL3BStag) and 168(pLATIL3DDtag) (Fig. 3b, c). The initial level (chase time 0 min) of hIL-3-DDtag in the medium is approximately four times higher than that of hIL-3-AAtag. In addition, the hIL-3-AAtag variant was degraded with a half-life of <2 min (Fig. 3c), whereas the half-life of hIL-3-DDtag was somewhat increased (approx. 5 min; Fig. 3c). The latter observation supports the notion that DD-tagged hIL-3 is less susceptible to extracellular proteases compared to hIL-3 with an AA-tag. However, the observation that the initial level of hIL3-DDtag in the medium is considerably higher than that of hIL3-AAtag suggests that hIL3-AAtag is also subject to proteolytic degradation before the molecules reach the medium, e.g. during passage through the cell wall of B. subtilis.



View larger version (45K):
[in this window]
[in a new window]
 
Fig. 3. Stability of hIL-3 variants with different C-terminal tags. (a) Western blot analysis of hIL-3 protein variants produced by B. subtilis 168 transformed with plasmid pLATIL3 (lane 1), pLATIL3BStag [expression of hIL-3 with a C-terminal B. subtilis SsrA-tag (AA-tag): hIL-3-AGKTNSFNQNVALAA; lane 2)], pLATIL3DDtag (expression of hIL-3 with a DD-tag: hIL-3-AGKTNSFNQNVALDD; lane 3), and pLATIL3ECtag [expression of hIL-3 with a C-terminal E. coli SsrA-tag (EC-tag): hIL-3-AANDENYALAA; lane 4]. Culture supernatants of cells entering the stationary phase were collected and in each lane equal amounts of extracellular protein were analysed by SDS-PAGE and Western blotting with anti-hIL-3 antibody. (b) Pulse–chase assays. Cells of B. subtilis 168 (pLATIL3BStag) and 168 (pLATIL3DDtag) were labelled with [35S]methionine for 1 min prior to chase with excess non-radioactive methionine. Samples were withdrawn at the times indicated, centrifuged, and the culture supernatants analysed by SDS-PAGE and fluorography. (c) The amounts of hIL-3-AAtag and hIL3-DDtag in (b) were quantified by determination of the radioactivity in the dried gel using a PhosphorImager (Molecular Dynamics) and plotted.

 
SsrA tagging of native B. subtilis proteins
B. subtilis 168 IssrADD expressing the variant SsrADD RNA, containing the protease-resistant DD-tag sequence, was analysed by Western blotting using the anti-BsSsrAtag antibodies to detect native proteins of B. subtilis that are tagged through the SsrA system. As controls, cells of B. subtilis 168 (expressing wild-type SsrA) and 168 {Delta}ssrA were used. Intracellular and extracellular proteins, from cells taken either during exponential growth or in the stationary phase, were analysed separately. A large number of intracellular proteins were detected by anti-BsSsrAtag antibody when cells expressed SsrADD (Fig. 4, lanes 2 and 8), while almost all of these bands were absent in cells expressing either wild-type SsrA (lanes 1 and 7) or no SsrA (lanes 3 and 9). As observed earlier in E. coli (Abo et al., 2000) and more recently in B. subtilis (Fujihara et al., 2002), these results indicate that many endogenous cellular proteins were tagged by the SsrA system, resulting in chimeric proteins. The proteins with the wild-type SsrA-tag (AA-tag) are subsequently degraded by proteases, while proteins with the DD-tag escape proteolysis. Whereas in the exponential growth phase the majority of the reacting bands were of relatively low molecular mass (lane 2), in the stationary phase a shift was observed towards proteins with a higher molecular mass (lane 8). In the exponential growth phase, no SsrADD-tagged proteins could be detected in the medium (lane 5), and in the stationary phase only a vague smear was observed (lane 11). This is most likely due to cell lysis, by which some intracellular SsrADD-tagged proteins end up in the medium. It appears that SsrA-mediated trans-translation occurs quite frequently in normally growing bacilli, and most natural substrates of SsrA seem to be intracellular proteins.



View larger version (71K):
[in this window]
[in a new window]
 
Fig. 4. Tagging of native B. subtilis proteins. Total intracellular or extracellular proteins produced by cells in the exponential or stationary phase of B. subtilis 168 expressing wild-type SsrA (AA), 168 IssrADD expressing SsrADD (DD), or 168 {Delta}ssrA containing no SsrA RNA (–) were analysed by Western blotting using anti-BsSsrAtag antibody. Protein sample loaded in each lane was normalized by measuring the optical density of the cell cultures. Position of molecular standards are indicated on the left.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
To study SsrA-mediated tagging of extracellular proteins in B. subtilis, the {alpha}-amylase gene and the hIL-3 gene containing a signal peptide sequence, both lacking an in-frame stop codon, were used as a model to achieve ribosome stalling and activation of the SsrA system. Introduction of the amyE gene construct, lacking an in-frame stop codon, into B. subtilis leads to tagging of the gene product by SsrA RNA. The tagged protein product bound to antibodies that were raised against the predicted proteolysis tag encoded by B. subtilis SsrA: (A)GKTNSFNQNVALAA. Once in the medium, SsrA-tagged {alpha}-amylase is very stable and not subjected to proteolytic degradation. The level of tagged {alpha}-amylase produced in an SsrADD strain is similar to that produced in a wild-type ssrA strain, indicating that the SsrA-tag also does not function as a specific signal for proteolysis directly after the trans-translation process, during the secretion process and/or while passing the B. subtilis cell wall. In contrast, SsrA-tagged hIL-3 molecules are rapidly degraded. Part of this degradation can be attributed to the instability of hIL-3 protein itself when produced by B. subtilis. However, hIL-3 tagged by SsrADD RNA, receiving a tag with two aspartic acid residues at the extreme C-terminal end, is more stable than hIL-3 tagged by wild-type SsrA (same tag but with two alanines at the extreme C-terminus). Not one specific protease of the SsrA system, but several (non-specific) proteases seem to play a role in the degradation of tagged hIL-3. Among these proteases are AprE and/or NprE, the two major extracellular proteases of B. subtilis, and CtpA. ClpXP is the protease that is primarily responsible for the intracellular degradation of SsrA-tagged proteins in B. subtilis (Wiegert & Schumann, 2001). A clpP mutation, however, does not increase the amount of SsrA-tagged hIL-3, indicating that tagged hIL-3 is not subjected to tag-specific degradation directly after the trans-translation process in the cytoplasm.

One of the functions of the SsrA system is to degrade incomplete protein fragments which otherwise might have inappropriate cellular activities. But is it necessary for cells to degrade tagged proteins that are targeted outside the cell? By using a B. subtilis mutant that expresses SsrADD encoding a protease-resistant tag, we showed that SsrA tags many intracellular proteins during growth, a finding similar to the observations of Fujihara et al. (2002). In future studies, we intend to investigate which proteases (besides ClpXP) play a role in the degradation of all these intracellular, SsrA-tagged proteins by studying the stability of these proteins in specific, protease-negative B. subtilis strains. No distinct bands were detected with the anti-SsrA-tag antibodies when the extracellular proteins produced by the SsrADD-expressing strain were analysed. Although there may be a low level of stalling-induced tagging of extracellular proteins that could not be detected by Western blot analysis, it seems that tagging of native, extracellular proteins does not occur in B. subtilis or occurs only to a very limited extent. Thus it appears that B. subtilis does not really need quality-control proteases that specifically degrade extracellular, SsrA-tagged proteins, e.g. in order to facilitate the reuse of degradation products (small peptides and amino acids) of tagged proteins to minimize the loss of energy and valuable compounds. The fact that SsrA-tagged {alpha}-amylase escapes proteolytic degradation confirms that B. subtilis lacks quality-control proteases specific for SsrA-tagged proteins that are targeted outside the cell. However, tagged molecules that are translocated may also end up in the bacterial membrane or cell wall, there possibly causing harmful effects to the cell. To deal with this, B. subtilis would need a membrane- and/or cell-wall-associated protease that specifically degrades SsrA-tagged proteins that (partly) end up in the membrane–cell wall interface after the secretion. Several observations indicate that CtpA may fulfil such a quality-control function in B. subtilis. In the first place, it is predicted that CtpA is localized into the bacterial membrane (Tjalsma et al., 2000). Moreover, CtpA is a homologue of E. coli Tsp (Marasco et al., 1996), which is a periplasmic protease that degrades SsrA-tagged proteins that are exported to the periplasmic compartment of E. coli (Keiler et al., 1996; Karzai et al., 2000). We also showed that CtpA plays a (minor) role in the degradation of SsrA-tagged hIL-3, supposedly in one of the stages of the secretion process. To confirm that CtpA indeed has this function, it would be interesting to study tagging and degradation of a model protein that contains a cell wall retention signal and/or a lipoprotein.

The apolar character of the SsrA tag makes this tag a degradation signal (Gottesman et al., 1998; Herman et al., 1998; Keiler et al., 1996). This suggests that not only SsrA-tagged proteins, but also heterologous proteins with natural, apolar C-terminal tails, will form good substrates for proteases of the SsrA quality-control system. It should be possible to make foreign proteins with apolar C-terminal tails less susceptible to degradation by these proteases by altering their C-termini, e.g. by adding a couple of charged residues at the extreme end or by replacing some apolar residues by charged amino acids. In this study, it was shown that addition of a DD-tag to the C-terminus of hIL-3 (a molecule with a rather apolar C-terminus) leads to increased stability of the molecules and a higher yield, when produced by B. subtilis. In this particular case, it seems that the charged residues at the C-terminus not only protect hIL-3 protein against a tail-specific protease like CtpA, but also against proteases with wider substrate specificity such as AprE and NprE. This result confirms that making apolar C-terminal tails of foreign secretory proteins more polar may in some cases diminish degradation of these proteins, ultimately resulting in an increased production level.

Although tagging of native extracellular protein thus seems to be a relatively rare event in B. subtilis, there may be (artificial) conditions that lead to tagging of extracellular protein. This is illustrated by the observation that overexpression of a heterologous protein (mIL-6) in E. coli leads to a subpopulation of molecules that are tagged by the SsrA system (Tu et al., 1995). This may also occur when certain recombinant proteins are overproduced in B. subtilis. The results of this study suggest that depending on the type of protein, these tagged protein products may or may not be degraded. Even when these tagged protein products are degraded in wild-type bacilli one should realize that, for industrial production of heterologous proteins, generally multiple protease-negative strains are used to prevent degradation of the protein products. If these proteases also fulfil a function in the degradation of SsrA-tagged proteins, the final protein product may still be contaminated with a variable amount of (nonfunctional) tagged molecules. This restricts its use when high quality of the protein product is required, e.g. when proteins are used for pharmaceutical purposes. Knowledge of determinants that lead to SsrA tagging and of the quality-control proteases involved in this system should prove useful in overcoming this potential problem.


   ACKNOWLEDGEMENTS
 
We thank Edwin Lee for providing strain BSE23, and Roopa Ghirnikar, Bob Caldwell and Walter Weyler for critical reading of the manuscript. This work was supported in part by a grant (QLK3-CT-1999-00413) of the European Commission.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Abo, T., Inada, T., Ogawa, K. & Aiba, H. (2000). SsrA-mediated tagging and proteolysis of LacI and its role in the regulation of lac operon. EMBO J 19, 3762–3769.[Abstract/Free Full Text]

Atkins, J. F. & Gesteland, R. F. (1996). A case for trans translation. Nature 379, 769–771.[CrossRef][Medline]

Collier, J., Binet, E. & Bouloc, P. (2002). Competition between SsrA tagging and translational termination at weak stop codons in Escherichia coli. Mol Microbiol 45, 745–754.[CrossRef][Medline]

Fujihara, A., Tomatsu, H., Inagaki, S., Tadaki, T., Ushida, C., Himeno, H. & Muto, A. (2002). Detection of tmRNA-mediated trans-translation products in Bacillus subtilis. Genes Cells 7, 343–350.[Abstract/Free Full Text]

Gottesman, S., Roche, E., Zhou, Y. N. & Sauer, R. T. (1998). The ClpXP and ClpAP proteases degrade proteins with carboxy-terminal peptide tails added by the SsrA-tagging system. Genes Dev 12, 1338–1347.[Abstract/Free Full Text]

Guérout-Fluery, A.-M., Shazand, K., Frandsen, N. & Stragier, P. (1995). Antibiotic-resistance cassettes for Bacillus subtilis. Gene 167, 335–336.[CrossRef][Medline]

Harwood, C. R. & Cutting, S. M. (1990). Molecular Biological Methods for Bacillus. Chichester, UK: Wiley.

Hayes, C. S., Bose, B. & Sauer, R. T. (2002a). Proline residues at the C terminus of nascent chains induce SsrA tagging during translation termination. J Biol Chem 277, 33825–33832.[Abstract/Free Full Text]

Hayes, C. S., Bose, B. & Sauer, R. T. (2002b). Stop codons preceded by rare arginine codons are efficient determinants of SsrA tagging in Escherichia coli. Proc Natl Acad Sci U S A 99, 3440–3445.[Abstract/Free Full Text]

Herman, C., Thévenet, D., Bouloc, P., Walker, G. C. & D'Ari, R. (1998). Degradation of carboxy-terminal-tagged cytoplasmic proteins by the Escherichia coli protease HflB (FtsH). Genes Dev 12, 1348–1355.[Abstract/Free Full Text]

Ito, K., Tadaki, T., Lee, S., Takada, K., Muto, A. & Himeno, H. (2002). Trans-translation mediated by Bacillus subtilis tmRNA. FEBS Lett 516, 245–252.[CrossRef][Medline]

Jentsch, S. (1996). When proteins receive deadly messages at birth. Science 271, 955–956.[Medline]

Karzai, A. W., Susskind, M. M. & Sauer, R. T. (1999). SmpB, a unique RNA-binding protein essential for the peptide-tagging activity of SsrA (tmRNA). EMBO J 18, 3793–3799.[Abstract/Free Full Text]

Karzai, A. W., Roche, E. D. & Sauer, R. T. (2000). The SsrA-SmpB system for protein tagging, directed degradation and ribosome rescue. Nat Struct Biol 7, 449–455.[CrossRef][Medline]

Keiler, K. C., Waller, P. R. H. & Sauer, R. T. (1996). Role of a peptide tagging system in degradation of proteins synthesized from damaged messenger RNA. Science 271, 990–993.[Abstract]

Kim, L., Mogk, A. & Schumann, W. (1996). A xylose-inducible B. subtilis integration vector and its application. Gene 181, 71–76.[CrossRef][Medline]

Kunst, F., Ogasawara, N., Moszer, I. & 148 other authors (1997). The complete genome sequence of the Gram-positive bacterium Bacillus subtilis. Nature 390, 249–256.[CrossRef][Medline]

Marasco, R., Varcamonti, M., Ricca, E. & Sacco, M. (1996). A new Bacillus subtilis gene with homology to Escherichia coli prc. Gene 183, 149–152.[CrossRef][Medline]

Roche, E. D. & Sauer, R. T. (1999). SsrA-mediated peptide tagging caused by rare codons and tRNA scarcity. EMBO J 18, 4579–4589.[Abstract/Free Full Text]

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Sunohara, T., Abo, T., Inada, T. & Aiba, H. (2002). The C-terminal amino acid sequence of nascent peptide is a major determinant of SsrA tagging at all three stop codons. RNA 8, 1416–1427.[Abstract/Free Full Text]

Tjalsma, H., Bolhuis, A., Jongbloed, J. D. H., Bron, S. & Van Dijl, J. M. (2000). Signal peptide-dependent protein transport in Bacillus subtilis: a genome-based survey of the secretome. Microbiol Mol Biol Rev 64, 515–547.[Abstract/Free Full Text]

Tu, G.-F., Reid, G. E., Zhang, J.-G., Moritz, R. L. & Simpson, R. J. (1995). C-terminal extension of truncated recombinant proteins in Escherichia coli with a 10Sa RNA decapeptide. J Biol Chem 270, 9322–9326.[Abstract/Free Full Text]

Ushida, C., Himeno, H., Watanabe, T. & Muto, A. (1994). tRNA-like structures in 10Sa RNAs of Mycoplasma capricolum and Bacillus subtilis. Nucleic Acids Res 22, 3392–3396.[Abstract]

Vagner, V., Dervyn, E. & Ehrlich, S. D. (1998). A vector for systematic gene inactivation in Bacillus subtilis. Microbiology 144, 3097–3104.[Abstract]

Van Dijl, J. M., de Jong, A., Smith, H., Bron, S. & Venema, G. (1991). Non-functional expression of Escherichia coli signal peptidase I in Bacillus subtilis. J Gen Microbiol 137, 2073–2083.[Medline]

Van Leen, R. W., Bakhuis, J. G., van Beckhoven, R. F. W. C. & 7 other authors (1991). Production of human interleukin-3 using industrial microorganisms. Biotechnology 9, 47–52.[Medline]

Wiegert, T. & Schumann, W. (2001). SsrA-mediated tagging in Bacillus subtilis. J Bacteriol 183, 3885–3889.[Abstract/Free Full Text]

Williams, K. P. (2000). The tmRNA website. Nucleic Acids Res 27, 165–166.[CrossRef]

Williams, K. P. (2002). The tmRNA website: invasion by an intron. Nucleic Acids Res 30, 179–182.[Abstract/Free Full Text]

Received 2 April 2003; revised 3 October 2003; accepted 13 October 2003.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Kolkman, M. A. B.
Articles by Ferrari, E.
Articles citing this Article
PubMed
PubMed Citation
Articles by Kolkman, M. A. B.
Articles by Ferrari, E.
Agricola
Articles by Kolkman, M. A. B.
Articles by Ferrari, E.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2004 Society for General Microbiology.