*For correspondence. E-mail email@example.com; Tel. (+33) 1 45 68 88 42; Fax (+33) 1 45 68 89 38.
ClpP-dependent degradation of PopR allows tightly regulated expression of the clpP3 clpP4 operon in Streptomyces lividans
Article first published online: 7 MAY 2002
Volume 44, Issue 3, pages 633–643, May 2002
How to Cite
Viala, J. and Mazodier, P. (2002), ClpP-dependent degradation of PopR allows tightly regulated expression of the clpP3 clpP4 operon in Streptomyces lividans. Molecular Microbiology, 44: 633–643. doi: 10.1046/j.1365-2958.2002.02907.x
- Issue published online: 7 MAY 2002
- Article first published online: 7 MAY 2002
Five clpP genes have been identified in Streptomyces coelicolor. The clpP1 and clpP2 genes form one operon, the clpP3 and clpP4 genes form another, and clpP5 is monocistronic. Previous studies in Streptomyces lividans have shown that the first operon (clpP1 clpP2) is required for a normal cell cycle. Expression of the second operon (clpP3 clpP4) is activated by PopR if the first operon is nonfunctional. We show here that PopR degradation is primarily dependent on ClpP1 and ClpP2, but can also be achieved by ClpP3 and ClpP4. The carboxy-terminus of PopR plays an essential part in the degradation process. Indeed, replacement of the last two alanine residues by aspartate residues greatly increased PopR stability. These substitutions did not impair PopR activity and, as expected, accumulation of the mutant form of PopR led to very strong expression of the clpP3 clpP4 operon. Increased PopR levels led to delayed sporulation. The results obtained in this study support the notion of cross-processing between ClpP1 and ClpP2.
Energy-dependent proteases play a key role in cells, degrading non-functional proteins and specific short-lived regulators. Several ATP-dependent proteases have been characterized in Escherichia coli, including Lon, FtsH, HslUV and Clp (Gottesman, 1996). The proteolytic Clp complex consists of two types of subunit: the proteolytic ClpP subunit and the ATPase subunit, ClpA or ClpX. ClpP subunits are organized into two superimposed heptameric rings, which form a central chamber. Amino acid residues Ser-97, His-122 and Asp-171 of the 14 catalytic triads are located within the proteolytic chamber (Maurizi et al., 1990a; Wang et al., 1997). A hexamer of ATPase subunits binds to one or both ends of the tetradecamer (Grimaud et al., 1998), resulting in the recognition, unfolding and translocation to the proteolytic chamber of the substrate (Weber-Ban et al., 1999; Kim et al., 2000; Ishikawa et al., 2001). Clp proteases are not only involved in the degradation of misfolded proteins (Frees and Ingmer, 1999; Krüger et al., 2000), but also in the degradation of specific regulators. For example, the starvation sigma factor σS in E. coli (Schweder et al., 1996), the CtrA response regulator and the McpA chemoreceptor in Caulobacter crescentus (Jenal and Fuchs, 1998; Tsai and Alley, 2001) are degraded by ClpXP, whereas the ComK transcriptional regulator (Turgay et al., 1998) and the CtsR class three heat shock gene repressor in Bacillus subtilis are degraded by ClpP (Derréet al., 2000) associated with the ClpC HSP100 ATPase subunit (Krüger et al., 2001). Another set of Clp targets has been described in E. coli and B. subtilis: the SsrA-tagged proteins. SsrA (small stable RNA) functions as both a tRNA and an mRNA. The mRNA encodes a small peptide that is added co-translationally to truncated proteins. These tagged polypeptides are then targeted for degradation (for a review, see Karzai et al., 2000). ClpAP in E. coli and ClpXP in E. coli and B. subtilis are among the proteases that ensure the degradation of these SsrA-tagged proteins in the cytoplasm (Gottesman et al., 1998; Wiegert and Schumann, 2001). Several studies have highlighted the importance of the carboxy-terminal sequence of various substrates for recognition by ClpX. This is the case for the SsrA-tagged proteins, the MuA transposase in E. coli (Levchenko et al., 1997) or the McpA chemoreceptor in C. crescentus (Tsai and Alley, 2001). However, in the case of the bacteriophage λO replication protein, it is the N-terminus that carries the recognition signal (Gonciarz-Swiatek et al., 1999). ClpA has been shown to recognize the N-terminal extremity of RepA (Hoskins et al., 2000) and the N-terminus of the molecule subjected to the N-end rule (Tobias et al., 1991).
Streptomyces are Gram-positive soil bacteria with a high G+C content. They present a particularly complex growth cycle, during which numerous secondary metabolites are produced. Indeed, bacteria of the Streptomyces genus produce 70% of all commercially available antibiotics. The genome of this bacterium has now been entirely sequenced (http://www.sanger.ac.uk/Projects/S_coel icolor/). It is 8.7 Mb long, almost twice the length of the E. coli genome. Several multigene families have been identified. For example, five clpP-like genes have been identified in Streptomyces coelicolor, whereas only one clpP gene is present in most bacterial genomes. These clpP-like genes are organized into two apparent operons, one corresponding to clpP1 clpP2 and the other to clpP3 clpP4, and a monocistronic transcription unit corresponding to clpP5. ClpP4 has a modified catalytic triad, in which the His residue is out of alignment, and the Ser residue of ClpP5 is shifted out of the consensus alignment by one position. This raises questions concerning the possible role of ClpP4 and ClpP5 as bona fide ClpP proteases. We studied this gene family in Streptomyces lividans, a species very closely related to S. coelicolor. It has been shown that a mutation in clpP1 leads to growth cycle alteration. Both clpP1 and clpP2 are required to restore the wild-type phenotype, leading to the suggestion that the clpP1 mutation has a polar effect on clpP2 and that clpP1 and clpP2 form an operon (de Crecy-Lagard et al., 1999). As for clpP3 and clpP4, the putative translation initiation codon of clpP4 is 1 bp downstream from the clpP3 stop codon, which suggests that these two genes are organized as an operon. The study of this multigene family recently led to the characterization of interactive regulation between the two apparent operons. Indeed, the clpP3 gene is silent in the wild-type strain, but its expression was strongly induced in the clpP1 mutant strain. This induction has been shown to be mediated by an activator named PopR (Viala et al., 2000). Increased expression of the clpP3 gene in clpP1 mutants is not caused by induction of the activator because the level of popR transcription is low, but similar in both wild-type and clpP1 strains (Viala et al., 2000). We investigated clpP3 induction in the clpP1 mutant by analysing PopR stability in wild-type and clpP1 bacteria. We found that ClpP operon products have a key role in PopR degradation. We also showed that the carboxy-terminal residues of PopR play an essential part in the degradation signal.
PopR is stabilized in a clpP1 mutant
Previous results have shown that the clpP3 operon is induced if the clpP1 clpP2 operon is not functional. This induction is mediated by the PopR activator. However, the level of transcription of popR is similar in wild-type and clpP1 mutant strains (Viala et al., 2000), suggesting that PopR may be controlled at the post-transcriptional level. As the ClpP1 proteolytic subunit is involved in the induction of the clpP3 clpP4 operon, this suggested that PopR may be subject to proteolysis. We therefore investigated the stability of PopR. We carried out a Western blot with polyclonal antibodies directed against PopR, using crude extracts of various strains, S. lividans 1326 wild type (WT) and the clpP1 mutant, as well as derivatives of these strains in which popR was overexpressed. We were unable to detect PopR in the wild-type strain (Fig. 1, lane 1), but detected a small amount of this protein in the clpP1 mutant strain (Fig. 1, lane 2). Overexpression of popR, from a multicopy plasmid, under the control of its own promoter in strain WT (pJV100), allowed us to detect a weak signal corresponding to PopR (Fig. 1, lane 5). The introduction of the same construct into a clpP1 mutant strain resulted in a strong signal (Fig. 1, lane 6). As expected, crude extracts from strains carrying the control vector, pUWL219, did not display this higher level of PopR (Fig. 1, lanes 3 and 4). Thus, PopR was preferentially detected in a clpP1 mutant, indicating that PopR was stabilized in the absence of proteins encoded by the clpP1 operon. These results also show that the machinery that degrades PopR is probably saturated if multiple copies of popR are present because, under these conditions, PopR was detected in the wild-type strain.
Complementation of the clpP1 mutant by the complete clpP1 operon restores PopR degradation
We complemented the clpP1 mutant strain with one or both genes of the clpP1 clpP2 operon and measured PopR levels. To facilitate PopR detection, we used the clpP1 (pJV100) strain, which overexpresses popR, into which we introduced the integrative control vector, pHM11a, or one of the following constructs: pJV50, pJV51 or pJV52, carrying, respectively, the complete clpP1 operon, the clpP1 gene or the clpP2 gene, under the control of the strong constitutive promoter, Perm. As each of the two genes has several possible initiation codons (Fig. 2A), we generated constructs encoding proteins with N-termini chosen according to the codon usage rule, as described by de Crecy-Lagard et al. (1999). We therefore introduced a translation initiation codon downstream from Perm, such that the sequences of the ClpP1 molecules encoded by pJV50 and pJV51 began with MTNLMPS, whereas the sequence of the ClpP2 molecule encoded by pJV52 began with MNDFPG. Complementation with the complete clpP1 clpP2 operon restored PopR degradation (Fig. 2B, lane 2), but no degradation was observed if the mutant was complemented with either clpP1 (Fig. 2B, lane 3) or clpP2 (Fig. 2B, lane 4) alone. Thus, both ClpP1 and ClpP2 seem to be necessary for the degradation of PopR. This suggests that the functional proteolytic complex may be a heterologous complex consisting of both ClpP1 and ClpP2 or that ClpP1 and ClpP2 must interact in some way before acting separately.
Control of ClpP production and evidence for cross-processing of ClpP1 and ClpP2
The previously complemented clpP1 (pJV100) strains harbouring pHM11a, pJV50, pJV51 or pJV53 were analysed by Western blot experiments with polyclonal anti-ClpP1 antibodies and compared with the wild-type strain. In wild type (Fig. 3A, lanes 1 and 2) and in clpP1 (pJV100) carrying pJV50 (clpP1 and clpP2 genes; Fig. 3A, lane 3), two signals were detected at around 23 kDa. These two signals may correspond to ClpP1 and ClpP2 or to two forms of ClpP1, given that ClpP is autoprocessed in E. coli to release the first 14 N-terminal residues (Maurizi et al., 1990b). In clpP1 (pJV100) carrying pJV51 (clpP1 gene), only the signal corresponding to the larger protein was observed (Fig. 3A, lane 4) and, in clpP1 (pJV100) carrying pJV52 (clpP2 gene), no signal was detected (Fig. 3A, lane 5), suggesting that the antibodies used were highly specific for ClpP1 or that ClpP2 was not produced. We hypothesized that the antibodies were highly specific for ClpP1 and that the lower molecular weight signal corresponded to a processed form of ClpP1. We carried out immuno-precipitation with antibodies against ClpP1, using crude extract from clpP1 (pJV100) (pJV50), and the N-terminus of the smaller of the two proteins was sequenced. The N-terminal sequence was AGEPS. This corresponds to an internal sequence located eight amino acid residues downstream from the experimentally introduced translation initiation codon: MTNLMPSAAGEPS (Fig. 2A). This processed form of ClpP1 was detected only in the presence of the clpP2 gene, i.e. in the wild type and the clpP1 mutant carrying pJV50 (clpP1 and clpP2 genes). These data suggest that ClpP1 is not autoprocessed, but cross-processed by its counterpart, ClpP2. The processed form of ClpP1 accumulated in the wild-type strain (Fig. 3A, lanes 1 and 2).
The absence of a ClpP2-specific signal (Fig. 3A, lane 5) suggested that the polyclonal anti-ClpP1 antibodies did not recognize ClpP2. We investigated whether this was indeed the case and assessed ClpP2 production by enriching the anti-ClpP1 antiserum in antibodies directed against ClpP2 epitopes by immunopurification on a ClpP2 column. The previous Western blot was analysed with these enriched antibodies (Fig. 3B). The signals corresponding to the native and processed forms of ClpP1 were again detected (Fig. 3B, lanes 1–4). However, new signals were clearly visible on the Western blot. An additional band appeared in lanes 4–6, in which we expected to detect ClpP3 according to our results concerning PopR stability in these strains; this band was referred to as ClpP3. ClpP2 production from pJV52 (overexpressing clpP2) was clearly detected (higher molecular weight band in Fig. 3B, lane 5). The absence of this band from the wild type (Fig. 3B, lanes 1 and 2) and from the strain overexpressing clpP1 and clpP2 (Fig. 3B, lane 3) suggested that ClpP2 was processed. This is consistent with the characterization of a shortened ClpP2, with VIPRFV as its N-terminal sequence (de Crecy-Lagard et al., 1999). We suggest that this processed ClpP2 is the new signal, the higher molecular weight band appearing in Fig. 3B but not in Fig. 3A, lanes 1–3. Finally, the absence of a processed form if only ClpP1 or ClpP2 is produced (Fig. 3B, lanes 4 and 5) suggests that these two proteases are cross-processed.
The PopR degradation signal includes the two C-terminal alanines
The substrate degradation motif recognized by the Clp ATP-dependent protease has not been identified in all cases, but is often located in the N-terminal or C-terminal amino acid residues. The C-terminal SsrA tag has been well characterized. The SsrA system adds a peptide tag to the C-terminus extremity of incomplete polypeptides on stalled ribosomes. These tagged polypeptides are thus targeted for proteolysis (Karzai et al., 2000). This system is widespread among eubacteria, and the sequence tag is well conserved in the three last amino acid residues (http://www.indiana.edu/~tmrna/), Leu–Ala–Ala (Fig. 4A). The alanine residues are essential for the degradation signal (Gottesman et al., 1998).
The PopR peptide sequence ends in Leu–Ala–Ala, suggesting that this sequence may act as an SsrA-like tag (Fig. 4A). We investigated the possible involvement of these residues in PopR stability by replacing the two alanine residues with two aspartates. This new gene, ‘popR-DD’, was inserted into the multicopy vector pUWL219, under the control of its own promoter, to give pJV110. This construct was produced in the same way as pJV100, which contains the ‘popR-AA’ gene. Crude extracts of the wild-type strain harbouring pUWL219 (control vector), pJV100 (popR-AA) or pJV110 (popR-DD) were prepared, and PopR levels were determined by Western blotting (Fig. 4). An exposure time of 15 s was sufficient to detect a strong PopR-DD signal (Fig. 4B, lane 3), but too short to detect PopR-AA (Fig. 4B, lane 2). PopR-AA was detected after exposure for a longer period of time (data not shown). If a large amount of PopR was present in the crude extract, a weak band was also detected at 32 kDa (Fig. 4B, lanes 3 and 6), probably corresponding to a dimeric form of PopR. Cross-linking experiments using glutaraldehyde and purified PopR indicated that PopR was able to dimerize (data not shown).
Finally, Western blot experiments indicated that PopR-DD accumulated to very high levels in the wild type, so the two last alanine residues at the C-terminus are indeed essential to the degradation signal of PopR, as in the ssrA tagging system.
Other proteases recognize the PopR degradation signal
We compared the levels of PopR-AA and PopR-DD in a clpP1 genetic background, assuming that PopR-AA and PopR-DD levels should be similar if ClpP1 and ClpP2 were entirely responsible for PopR degradation. Crude extracts of the clpP1 mutant strain harbouring pUWL219 (control vector), pJV100 (popR-AA) or pJV110 (popR-DD) were prepared, and the amount of PopR was determined by Western blotting (Fig. 4). PopR-DD levels were similar in the wild type and in the clpP1 mutant (Fig. 4 lanes 3 and 6). However, for equivalent amounts of protein loaded, the PopR-DD signal (Fig. 4 lane 6) was considerably stronger than the PopR-AA signal (lane 5) in the clpP1 mutant.
This result suggests that other proteases also recognize the Ala–Ala motif on PopR for the degradation of this regulator.
ClpP3 and ClpP4 degrade PopR
As shown above, ClpP1 and ClpP2 are not the only proteases involved in controlling PopR levels within the cell. Likely candidates include ClpP3 and ClpP4. We therefore complemented the clpP1 (pJV100) strain, which over-expresses popR, with pJV41 carrying the clpP3 clpP4 operon downstream from the strong promoter Perm. Crude extracts were analysed by Western blot experiments with anti-PopR antibodies (Fig. 5). Efficient, but not total, PopR degradation was observed (Fig. 5, lane 3). Control experiments with the vector pHM11a (Fig. 5, lane 1) or with pJV50 carrying clpP1 and clpP2 (Fig. 5, lane 2) showed accumulation or total PopR degradation respectively.
These results indicate that ClpP3 and ClpP4 can also participate in PopR degradation, controlling their own synthesis through a negative feedback loop.
Phenotype linked to PopR accumulation
The modified protein, PopR-DD, was more stable than the wild-type protein. We therefore expected the level of clpP3 expression to be higher unless the amino acid substitution impaired PopR activity. We assessed clpP3 expression by primer extension experiments with mRNA extracted from the wild-type strain overexpressing popR-DD. Very strong expression was observed (data not shown), indicating that the PopR-DD form was active. The WT pJV110 strain expressing clpP1 clpP2 and overexpressing clpP3 clpP4 presented reduced red pigmentation (Fig. 6A) and slight growth retardation on solid R5 medium. Indeed, the aerial mycelium began to appear within 48 h in the wild-type strain harbouring the control vector (pUWL219) or multiple copies of popR-AA (pJV100) but not in the strain harbouring multiple copies of popR-DD (pJV110) (Fig. 6A). After 1 week, the wild type was sporulating, whereas the strain overexpressing popR-DD was forming aerial mycelium (Fig. 6B). The clpP1 mutant strain in which popR-DD was overexpressed also presented reduced red pigmentation with respect to the control strains. Interestingly, some of the bacteria appeared to undergo differentiation after 1 week (Fig. 6C). These differentiated structures, which did not complete sporulation, appeared preferentially in the mass of inoculum rather than at the periphery and did not spread throughout the population.
Which ATPase is involved in PopR recognition?
In Streptomyces, three ATPases, ClpX, ClpC1 and ClpC2, could potentially associate with ClpP proteins to form the proteolytic complex. As substrate recognition is controlled by the ATPase subunit, we considered it interesting to determine which ATPase worked with ClpP in PopR degradation. The construction of a clpX mutant in S. lividans (J. Viala and P. Mazodier, submitted) enabled us to carry out in vivo analysis of the role of ClpX in PopR degradation. Thus, PopR was detected by Western blotting, and clpP3 expression was analysed by primer extension in a clpX mutant. PopR did not accumulate in this genetic background, and clpP3 expression was not induced (data not shown), in contrast to what was observed in a clpP1 mutant. Thus, ClpX does not seem to be involved in the degradation of PopR, which suggests that one or both the ClpC ATPases may play a role in this process. Unfortunately, we could not test the involvement of ClpC1 in PopR degradation given the failure to obtain a mutation in this gene (de Crecy-Lagard et al., 1999).
Previous studies have shown that clpP3 expression is induced in the clpP1 genetic background and that the PopR activator mediates this induction. PopR binds specifically to the clpP3 promoter and does not auto-regulate its own synthesis (Viala et al., 2000). Here, we show that the clpP3 induction observed in a clpP1 mutant resulted from the stabilization of PopR in this genetic background. The degradation of PopR is ClpP1 ClpP2 dependent, but can also be carried out by ClpP3 and ClpP4. This observation points to the existence of an interactive network within the clpP multigenic family. PopR is the first target of a Clp protease to be described in Streptomyces.
The degradation of PopR requires a signal corresponding to the C-terminal amino acid residues Leu–Ala–Ala. Indeed, replacement of these two carboxy-terminal alanine residues by two aspartate residues greatly increases the stability of PopR. These features are reminiscent of those of the SsrA system. The most highly conserved part of the SsrA tag is the C-terminal Leu–Ala–Ala motif, and the two alanine residues are crucial for degradation by the Clp protease (Gottesman et al., 1998). Thus, PopR is a naturally tagged protein. Such dependence on the presence of two alanine residues at the C-terminus has been described for the proteolysis of CtrA, a general regulator in C. crescentus (Domian et al., 1997). Are two alanine residues at the C-terminus sufficient for degradation by ClpP1 and ClpP2 in Streptomyces? Apparently not, as the Lon protein, which ends in V–A–A at the C-terminus, does not accumulate in a clpP1 mutant (A. Bellier, unpublished results). Two alanine residues at the carboxy-terminus therefore seem to be necessary but not sufficient for degradation. This raises questions as to what else might be required for degradation: the Leu residue? Flynn et al. (2001) have studied recognition determinants within the 11-residue SsrA tag for degradation by ClpXP or ClpAP in E. coli. The tripeptide Leu-9–Ala-10–Ala-11 is very important for degradation by ClpXP. Furthermore, the tripeptide is sufficient to mark a protein as a substrate for degradation in some, but not all, cases. This suggests that other sequence features or structural characteristics could have an effect. The Leu-9 residue has also been shown to play an important role for degradation by ClpAP in E. coli, in addition to Ala-1, Ala-2, Ala-8 and Ala-10.
The clpP1 mutant was complemented with the clpP1 and/or the clpP2 genes, and the level of PopR was assessed in these strains. Both ClpP1 and ClpP2 were required for PopR degradation. There are several pos-sible explanations for this. First, it is possible that the only functional complex is made up of ClpP1 and ClpP2. Secondly, ClpP1 and ClpP2 may need each other to generate a mature protein and to form a complex together. Thirdly, ClpP1 and ClpP2 may need to interact to gen-erate a mature form before acting independently. We investigated whether this was the case by producing the mature form of ClpP1 directly from pJV54, and the potential mature form of ClpP2 described by de Crecy-Lagard et al. (1999) from pJV53. These plasmids were used in complementation experiments, and PopR stability was assessed. PopR was not degraded, but ClpP1 and ClpP2 were not produced from these constructs in large amounts (data not shown). Furthermore, the ClpP1 and ClpP2 proteins produced may be non-functional, as the proregion of proteases is commonly involved in the proper folding of the mature protease (Baker et al., 1993). To date, the role played by the ClpP propeptide is unclear. In E. coli, ClpP autoprocesses the amino-terminal 14-amino-acid propeptide. Cleavage requires an intact active site, and it has been shown that mutations modifying the active site result in the accumulation of an unprocessed form of ClpP. The unprocessed ClpP is able to self-associate, and in vitro experiments have shown that this molecule also interacts with ClpA. However, as the mutation affects the active site, the activity of this unprocessed form could not be tested (Maurizi et al., 1990a). The unprocessed form was analysed by electron microscopy, and the internal cavity was found to be filled or occluded by the propeptide (Kessel et al., 1995).
ClpP proteins are thought to act with an ATPase in the protein degradation process. No induction of clpP3 expression was observed in the clpX mutant, and PopR degradation persisted, indicating that ClpX is not essential for PopR recognition. This suggests that the ClpC ATPases may play a key role.
It was also shown that ClpP3 and ClpP4 could be involved in PopR degradation. Hence, ClpP3 and ClpP4 could exert a negative feedback control on their own synthesis. Interestingly, we observed complete degradation of PopR by ClpP1 and ClpP2, but slightly incomplete degradation of PopR by ClpP3 and ClpP4 (Fig. 5). This feature, if it is physiological, may correspond to tight post-translational control of PopR action by allowing total silencing of the clpP3 clpP4 genes in the presence of functional clpP1 clpP2 genes and adapted expression of clpP3 clpP4 under specific conditions.
These results led us to suggest a model of regulation for clpP3 operon expression (Fig. 7). In the wild-type strain, ClpP1 and ClpP2 proteins are produced and form a proteolytic complex with a Clp ATPase. This complex degrades the transcriptional activator PopR required for the expression of clpP3, preventing the expression of clpP3 and clpP4. In a clpP1 mutant, ClpP1 and ClpP2 are not produced, and PopR accumulates in the cell. PopR binds to the promoter region of the clpP3 operon and activates its transcription. The ClpP3 and ClpP4 proteins are produced, form a proteolytic complex and degrade PopR, leading to a steady-state equilibrium of ClpP3 and ClpP4 within the cell.
In E. coli, redundant activities have been demonstrated between ATP-dependent proteases. Thus, SsrA-tagged proteins are essentially degraded by ClpP proteases, but they can also be removed by FtsH (Herman et al., 1998); the σ32 sigma factor is degraded by FtsH but also by ClpYQ (Kanemori et al., 1997), and Lon and ClpYQ share a common substrate. Indeed, the cell division inhibitor SulA, normally degraded by Lon, is also a substrate of ClpYQ (Kanemori et al., 1999). Although ClpYQ ensures this activity, the lon mutant displays UV sensitivity resulting from the stabilization of SulA. This phenotype can be abolished by ClpYQ overproduction (Wu et al., 1999). Here, ClpP3 and ClpP4 also appear to be partially redundant with ClpP1 and ClpP2, as they are able to degrade PopR but do not reverse the bald phenotype resulting from the clpP1 mutation. Only strong overexpression of the clpP3 operon in the clpP1 mutant allows some bacteria to differentiate, as in the case of the clpP1 mutant accumulating PopR-DD. The overproduction of ClpP3 ClpP4 is likely to lead to the degradation of some ClpP1-specific targets involved in differentiation. As ClpP3 and ClpP4 cannot replace ClpP1 and ClpP2 at the differentia-tion level, the question remains as to why the organism has established interactive regulation between the clpP1 and clpP3 operons. The most simple explanation may be that ClpP3 and ClpP4 ensure the minimal activity required for cell survival, as the double mutation in clpP1 and clpP3 could not be obtained in previous work, suggesting that at least one copy of the clpP operons is essential for cell viability (Viala et al., 2000). The wild-type strain overexpressing popR-DD displays retarded growth. This may be because the large amounts of ClpP3 ClpP4 present in the wild-type strain overproducing these proteins compete with ClpP1 and ClpP2, thereby affecting growth.
Bacterial strains and media
Streptomyces lividans strain 1326 was obtained from the John Innes Culture Collection, and S. lividans 1326 clpP1::AmR (de Crecy-Lagard et al., 1999) and S. lividans 1326 clpX:vioR (J. Viala and P. Mazodier, submitted) were constructed in this laboratory. YEME medium was used for liquid cultures (Hopwood et al., 1985). Solidified NE medium (Murakami et al., 1989) and R5 medium (Hopwood et al., 1985) were used for Streptomyces cultures on plates. The antibiotics apramycin, viomycin, thiostrepton and hygromycin were added to final concentrations of 25, 30, 25 and 200 μg ml−1, respectively, to solid medium, and to final concentrations of 20, 10, 10 and 50 μg ml−1, respectively, to liquid medium.
Escherichia coli TG1 (Gibson, 1984) was used as the general cloning host. E. coli strains were grown in LB medium. Hygromycin and ampicillin were added to final concentrations of 200 and 100 μg ml−1 respectively.
Plasmids and plasmid construction
The E. coli/Streptomyces shuttle vectors used were pUWL219, which contains the replication functions of the Streptomyces multicopy plasmid pIJ101 (Wehmeier, 1995), and pHM11a, which allows a strong expression from the constitutive Perm promoter and contains an integration element directing its insertion into the Streptomyces genome at the mini-circle attachment site (Motamedi et al., 1995).
We constructed pJV41 by inserting the 1465 bp fragment corresponding to the S. lividans 1326 clpP3 clpP4 operon between the NdeI and BamHI sites of pHM11a. This fragment was obtained by polymerase chain reaction (PCR) amplification with oligonucleotides Ju 41 and Ju 43 (for oligonucleotides used, see Table 1). We constructed pJV50 by inserting the 1470 bp fragment corresponding to the S. lividans 1326 clpP1 clpP2 operon between the NdeI and HindIII sites of pHM11a. This fragment was obtained by PCR amplification with oligonucleotides Ju 62 and Ju 63. pJV51 was constructed by inserting the 620 bp fragment corresponding to the S. lividans 1326 clpP1 gene between the NdeI and HindIII sites of pHM11a. This fragment was obtained by PCR amplification with oligonucleotides Ju 62 and Ju 77. We constructed pJV52 by inserting the 810 bp fragment corresponding to the S. lividans 1326 clpP2 gene with the first clpP2 possible translation initiation codon between the NdeI and HindIII sites of pHM11a. This fragment was obtained by PCR amplification with oligonucleotides Ju 78 and Ju 63. We constructed pJV53 by inserting the 700 bp fragment encoding the putative mature form of the S. lividans 1326 clpP2 gene between the NdeI and HindIII sites of pHM11a. This fragment was obtained by PCR amplification with oligonucleotides Ju 79 and Ju 63. pJV54 was constructed by inserting the 594 bp fragment corresponding to the S. lividans 1326 clpP1 gene between the NdeI and HindIII sites of pHM11a. This fragment was obtained by PCR amplification with oligonucleotides Ju 82 and Ju 77.
pJV100 was constructed as described by Viala et al. (2000). pJV110 was constructed in the same way as pJV100, by inserting into the blunted BamHI site of pUWL219, the 780 bp fragment including the promoter region and the popR gene, modified to encode two aspartic acid residues before the stop codon instead of two alanines. This fragment was obtained by PCR amplification with oligonucleotides Ju 46 and Ju 69.
DNA manipulation, transformation and conjugation procedures
Plasmid DNA was extracted from E. coli with a Qiagen kit. DNA fragments were purified from agarose gels with Ultrafree-DA filters (Amicon-Millipore). Restriction enzymes were used as recommended by the manufacturers. DNA fragments were amplified by PCR (Mullis and Faloona, 1987; Saiki et al., 1988). Standard CaCl2 (Cohen et al., 1972) or electroporation procedures were used for E. coli transformation.
Streptomyces protoplasts were prepared and transformed as described by Hopwood et al. (1985).
RNA extraction and primer extension experiments
RNA extraction and primer extension experiments were carried out as described previously (Viala et al., 2000).
Protein extraction and Western blotting experiments
Cells were grown at 30°C. We collected 10 ml of culture and added 0.5 mm phenylmethylsulphonyl fluoride (PMSF) and 5 mm EDTA. Cells were pelleted, resuspended in 500 μl of 20 mm Tris, 5 mm EDTA, 1 mm dithiothreitol (DTT), 2× protease inhibitor cocktail (Roche Boehringer) and lysed by sonication. The resulting suspension was centrifuged for 30 min at 4°C, 12 000 r.p.m., and the protein concentration of the supernatant was determined by the Bradford (1976) method. Various quantities of protein extract (10–50 μg protein) were subjected to SDS–PAGE as described by Laemmli (1970). The proteins were transferred to a nitrocellulose membrane (Hybond C), which was then probed with rabbit poly-clonal anti-PopR (1:500–1:2000), anti-Streptomyces ClpP1 (1:10 000) or anti-Streptomyces ClpP1 enriched against ClpP2 epitope (1:1000) antibodies, which were detected with the Super Signal detection kit (Pierce).
Glutaraldehyde cross-linking of PopR
We incubated 3 μg of purified PopR (Viala et al., 2000) for 1 h at 37°C with 1 or 10 mm cross-linking reagent (glutaraldehyde) in a 10 μl reaction mixture consisting of 50 mm NaH2PO4, 50 mm NaCl and 10% glycerol. The reaction was stopped by adding SDS loading buffer; samples were boiled and subjected to SDS–PAGE. The gel was stained with Coomassie blue.
Immunoprecipitation of ClpP1
To immunoprecipitate ClpP1, anti-Streptomyces ClpP1 antibodies were covalently bound to a 1 ml Hi-Trap NHS-activated column (Pharmacia Biotech), as recommended by the manufacturer. The pellet of a 100 ml culture of S. lividans 1326 clpP1 pJV100 pJV50 was resuspended in 5 ml of 20 mm Tris, 5 mm EDTA, 1 mm DTT and subjected to sonication. The resulting suspension was centrifuged. The crude extract (the supernatant) was then loaded onto the prepared column and incubated overnight at 4°C. The column was washed with 20 column volumes of 20 mm Tris, pH 7.5, and 20 column volumes of 20 mm Tris, pH 7.5, 500 mm NaCl. ClpP1 proteins were eluted with 4× 1 ml of 100 mm glycine/HCl, pH 2.5. Eluates were collected in tubes containing 100 μl of 1 M Tris, pH 9. The column was washed with 20 column volumes of 20 mm Tris, pH 8.8. The remaining proteins were eluted in 4× 1 ml of 100 mm triethylamine, pH 11.5. Eluates were collected in tubes containing 100 μl of 2 M Tris, pH 6.8. The column was then washed with 20 column volumes of 20 mm Tris, pH 7.5. The first acid eluate was used for N-terminal sequencing.
Purification of antibodies
Purified ClpP1 or ClpP2 was covalently bound to a 1 ml Hi-Trap NHS-activated column (Pharmacia Biotech), as recommended by the manufacturer. We diluted 2 ml of rabbit serum obtained by immunization with ClpP1 1:5 in 20 mm Tris, pH 7.5, and filtered the resulting solution. This preparation was loaded onto the prepared columns and incubated overnight at 4°C. Washing and elution were performed as described in the previous paragraph.
We would like to thank T. Msadek for fruitful discussions and critical reading of the manuscript. We would also like to thank Edith Gouin and Roger Zenon for assistance with the production of polyclonal anti-ClpP1. We thank J. D’Alayer for N-terminal sequence determination. and A. Edelman and Associates for correcting this manuscript. This work was supported by research funds from the Institut Pasteur, Centre National de Recherche Scientifique and Université Paris 7. J.V. holds a fellowship from the Ministère de l’Education Nationale, de la Recherche et de la Technologie.
- 1993) The role of pro regions in protein folding. Curr Opin Cell Biol 5: 966–970. , , and (
- 1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248–254. (
- 1972) Nonchromosomal antibiotic resistance in bacteria: genetic transformation of Escherichia coli by R-factor DNA. Proc Natl Acad Sci USA 69: 2110–2114. , , and (
- 1999) Alteration of the synthesis of the Clp ATP-dependent protease affects morphological and physiological differentiation in Streptomyces. Mol Microbiol 32: 505–517. , , , , and (
- 2000) The CtsR regulator of stress response is active as a dimer and specifically degraded in vivo at 37 degrees C. Mol Microbiol 38: 335–347.DOI: 10.1046/j.1365-2958.2000.02124.x , , and (
- 1997) Cell type-specific phosphorylation and proteolysis of a transcriptional regulator controls the G1-to-S transition in a bacterial cell cycle. Cell 90: 415–424. , , and (
- 2001) Overlapping recognition determinants within the ssrA degradation tag allow modulation of proteolysis. Proc Natl Acad Sci USA 98: 10584–10589. , , , , , and (
- 1999) ClpP participates in the degradation of misfolded protein in Lactococcus lactis. Mol Microbiol 31: 79–87. , and (
- 1984) Studies on the Epstein–Barr Virus Genome. PhD Thesis, Cambridge University, UK. (
- 1999) Recognition, targeting, and hydrolysis of the lambda O replication protein by the ClpP/ClpX protease. J Biol Chem 274: 13999–14005. , , , , , et al. (
- 1996) Proteases and their targets in Escherichia coli. Annu Rev Genet 30: 465–506. (
- 1998) The ClpXP and ClpAP proteases degrade proteins with carboxy-terminal peptide tails added by the SsrA-tagging system. Genes Dev 12: 1338–1347. , , , and (
- 1998) Enzymatic and structural similarities between the Escherichia coli ATP- dependent proteases, ClpXP and ClpAP. J Biol Chem 273: 12476–12481. , , , , and (
- 1998) Degradation of carboxy-terminal-tagged cytoplasmic proteins by the Escherichia coli protease HflB (FtsH). Genes Dev 12: 1348–1355. , , , , and (
- 1985) Genetic Manipulation of Streptomyces. A Laboratory Manual. Norwich: The John Innes Foundation. , , , , , et al. (
- 2000) Substrate recognition by the ClpA chaperone component of ClpAP protease. J Biol Chem 275: 35361–35367. , , and (
- 2001) Translocation pathway of protein substrates in ClpAP protease. Proc Natl Acad Sci USA 98: 4328–4333. , , , , , and (
- 1998) An essential protease involved in bacterial cell-cycle control. EMBO J 17: 5658–5669. , and (
- 1997) Synergistic roles of HslVU and other ATP-dependent proteases in controlling in vivo turnover of sigma 32 and abnormal proteins in Escherichia coli. J Bacteriol 179: 7219–7225. , , , and (
- 1999) The ATP-dependent HslVU/ClpQY protease participates in turnover of cell division inhibitor SulA in Escherichia coli. J Bacteriol 181: 3674–3680. , , and (
- 2000) The SsrA-SmpB system for protein tagging, directed degradation and ribosome rescue. Nature Struct Biol 7: 449–455. , , and (
- 1995) Homology in structural organization between E. coli ClpAP protease and the eukaryotic 26S proteasome. J Mol Biol 250: 587–594. , , , , , , and (
- 2000) Dynamics of substrate denaturation and translocation by the ClpXP degradation machine. Mol Cell 5: 639–648. , , , , and (
- 2000) The Clp proteases of Bacillus subtilis are directly involved in degradation of misfolded proteins. J Bacteriol 182: 3259–3265. , , , , and (
- 2001) Clp-mediated proteolysis in Gram-positive bacteria is autoregulated by the stability of a repressor. EMBO J 20: 852–863. , , , , and (
- 1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680–685. (
- 1997) PDZ-like domains mediate binding specificity in the Clp/Hsp100 family of chaperones and protease regulatory subunits. Cell 91: 939–947. , , , , and (
- 1990a) ClpP represents a unique family of serine proteases. J Biol Chem 265: 12546–12552. , , , and (
- 1990b) Sequence and structure of Clp P, the proteolytic component of the ATP-dependent Clp protease of Escherichia coli. J Biol Chem 265: 12536–12545. , , , , , , and (
- 1995) Integrative vectors for heterologous gene expression in Streptomyces spp. Gene 160: 25–31.DOI: 10.1016/0378-1119(95)00191-8 , , and (
- 1987) Specific synthesis of DNA in vitro via a polymerase-catalyzed chain reaction. Methods Enzymol 155: 335–350. , and (
- 1989) Thiostrepton-induced gene expression in Streptomyces lividans. J Bacteriol 171: 1459–1466. , , and (
- 1988) Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239: 487–491. , , , , , et al. (
- 1996) Regulation of Escherichia coli starvation sigma factor (sigma s) by ClpXP protease. J Bacteriol 178: 470– 476. , , , and (
- 1991) The N-end rule in bacteria. Science 254: 1374– 1377. , , , and (
- 2001) Proteolysis of the Caulobacter Mcpa chemoreceptor is cell cycle regulated by a ClpX-dependent pathway. J Bacteriol 183: 5001– 5007. , and (
- 1998) Competence in Bacillus subtilis is controlled by regulated proteolysis of a transcription factor. EMBO J 17: 6730– 6738. , , , and (
- 2000) The clpP multigenic family in Streptomyces lividans: conditional expression of the clpP3 clpP4 operon is controlled by PopR, a novel transcriptional activator. Mol Microbiol 38: 602–612. , , and (
- 1997) The structure of ClpP at 2.3 Å resolution suggests a model for ATP-dependent proteolysis. Cell 91: 447–456. , , and (
- 1999) Global unfolding of a substrate protein by the Hsp100 chaperone ClpA. Nature 401: 90–93. , , , and (
- 1995) New multifunctional Escherichia coli-Streptomyces shuttle vectors allowing blue-white screening on XGal plates. Gene 165: 149–150. (
- 2001) SsrA-mediated tagging in Bacillus subtilis. J Bacteriol 183: 3885–3889. , and (
- 1999) Redundant in vivo proteolytic activities of Escherichia coli Lon and the ClpYQ (HslUV) protease. J Bacteriol 181: 3681–3687. , , and (