Previous Article | Next Article ![]()
Journal of Bacteriology, April 2009, p. 2894-2898, Vol. 191, No. 8
0021-9193/09/$08.00+0 doi:10.1128/JB.01715-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Laboratory for Molecular Biology, Department of Biological Sciences, University of Illinois at Chicago, Chicago, Illinois 60607
Received 8 December 2008/ Accepted 5 February 2009
|
|
|---|
|
|
|---|
In Escherichia coli, five proteases (Tsp, FtsH, Lon, ClpXP, and ClpAP) recognize the SsrA sequence and can participate to some extent in degrading polypeptide chains attached to it (3, 6, 9, 24), but ClpXP is thought to be the major protease responsible for the degradation of SsrA-tagged proteins in vivo under normal conditions in two gram-negative bacterial species (E. coli and Caulobacter crescentus) and in the gram-positive Bacillus subtilis (2, 6, 27). An adaptor protein, SspB, in E. coli and its ortholog (SspB
) in C. crescentus promote degradation of SsrA-tagged proteins by ClpXP, while no similar adaptors are known in gram-positive bacteria (2, 7, 14, 15). For most other bacterial species, SsrA-tagged proteins are speculated to be substrates of ClpXP (10). However, ClpX is not conserved in all bacteria, while FtsH is, and for bacteria lacking ClpX, FtsH or Lon is considered to be the likely protease for SsrA-tagged proteins (8, 10). The mycoplasmas, for example, which lack ClpX and ClpP orthologs, possess a Lon protease with exceptional activity on SsrA-tagged substrates (8).
An SsrA tagging system is active in Streptococcus pneumoniae (pneumococcus), but the specific proteases responsible for the recognition and degradation of its SsrA-tagged proteins are unknown (18). The pneumococcal SsrA sequence (AKNNTSYALAA) resembles the one in E. coli (AANDENYALAA) in length and also in the sequence recognized directly by ClpX (underlined) (5). There are fewer cellular proteases in S. pneumoniae than in E. coli or in B. subtilis. As it has just five putative clp genes (clpC, clpE, clpL, clpP, and clpX) and carries ftsH, encoding a well-characterized membrane-bound protease, S. pneumoniae lacks genes for proteases of the Lon, Tsp, ClpA, and ClpYQ classes (20). Thus, ClpXP and FtsH are both good candidates for SsrA-specific proteases; however, it should be noted that while degradation by ClpXP in vivo requires the presence of the adaptor protein SspB in E. coli (4), no apparent pneumococcal homologs of SspB are known.
To identify pneumococcal cytoplasmic proteases that are involved in degrading SsrA-tagged proteins in vivo, we constructed synthetic genes for model SsrA-tagged proteins and determined the effects of individual protease mutations on the level of the tagged proteins. The results showed that proteins tagged with SsrA at their C terminus are strongly stabilized by mutations of clpX and clpP but not by mutations affecting any other Clp protease or FtsH and suggest that ClpXP recognition of SsrA tags may not require an adaptor protein.
GFP-SsrA is unstable in S. pneumoniae.
To determine the fate of GFP-SsrA in vivo, we constructed strains in which genes encoding GFP-SsrA or green fluorescent protein (GFP) were inserted in the chromosome of S. pneumoniae downstream of the native site of the gene aga, creating a transcriptional fusion dependent on the raffinose-inducible aga promoter (Table 1; Fig. 1). The resulting expression cassette, aga-gfp-ssrA-aphA, also includes a constitutive promoter for the kanamycin resistance marker, aphA. To ensure maximal expression of GFP-SsrA, a synthetic ribosome binding site, 5'-AGGAGGTA, was positioned 6 bases upstream of the start codon of GFP. This sequence is found at pneumococcal ribosomal protein genes and is predicted to be a strong ribosome binding site (23). The constructs were confirmed by PCR analysis using primers binding outside and within the insertions and by sequencing each entire insert (Table 2). As aga encodes the enzyme
-galactosidase, which hydrolyzes p-nitrophenyl-D-galactopyranoside to the yellow product p-nitrophenol, the expression of the synthetic operon can be readily assayed spectrophotometrically (22).
|
View this table: [in a new window] |
TABLE 1. Bacterial strains used in this study
|
![]() View larger version (36K): [in a new window] |
FIG. 1. Organization of the aga locus and construction of cassettes for regulated synthesis of GFP- and SrA-tagged proteins. Fragments containing a truncated aga gene ('aga) or the aphA (kanamycin resistance) gene and a truncated rafE gene (rafE') were amplified with primer pairs E-L and F-G, respectively, from the CP1372 DNA template. gfp-mut2 was first cloned into pMAL-c2X after amplification from plasmid pKL147, kindly provided by Adam Driks (Loyola University), using primers A and B. ssrA was amplified from CP1250 with primers C and D and inserted downstream of gfp-mut2. A heterologous fragment with gfp-ssrA was amplified in two steps: one to add a synthetic ribosome binding site (primers H and J) using plasmid pMAL-c2X (gfp-ssrA) as a template and the second to add a restriction site to the first product (primers I and J). After digestion at restriction sites incorporated by the PCR primers, the second product was ligated with the targeting aga and aphArafE fragments as indicated. After transformation into CP1250 with Kan selection, the structure of the new synthetic locus in CP1903 was verified by sequencing. Filled arrows, genes manipulated during the construction; open arrows, conserved genes of the raffinose locus and its flanking genes. Promoters: Praf, aga promoter; Pc, artificial constitutive promoter derived from amiA. CP1911, CP1957, and CP1954 were constructed similarly but by using primer pairs H-K/I-K, E-Z/X-Y/B1-C1/A1-G, and E-S/T-U/V-G.
|
|
View this table: [in a new window] |
TABLE 2. Oligonucleotides used in this study
|
-galactosidase activity. Lysates of the wild type and of strains carrying GFP-SsrA or GFP constructs all displayed 10-fold induction of aga expression compared to controls without raffinose, indicating comparable expression levels for both gfp constructs (data not shown). Strains CP1903 and CP1911, expressing the GFP-SsrA or GFP genes, respectively, were examined for accumulation of GFP by a Western blotting assay using anti-GFP antibody. GFP-SsrA was not detectable, whereas untagged GFP was readily detected as a band at a position corresponding to the size of purified GFP (Fig. 2). We conclude that GFP-SsrA is degraded in vivo in the wild-type background but the untagged version of GFP is more stable, implying that the SsrA tag renders GFP a substrate of cytoplasmic proteases.
![]() View larger version (30K): [in a new window] |
FIG. 2. Stabilization of GFP-SsrA in clpX and clpP mutants. (A) Strains expressing gfp-ssrA in clpP (CP1904), clpC (CP1906), clpE (CP1907), clpL (CP1908), ftsH (CP1910), or clpX (CP1905) mutant backgrounds were harvested, lysed, and loaded in lanes 1 to 5 and 8 of a sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel. CP1903, expressing gfp-ssrA in the wild-type background (wt*), and the wild-type with no gfp insertion at the aga locus (wt) were loaded in lanes 6 and 7, respectively. (B) Strains expressing untagged gfp in clpP (CP1912), clpX (CP1913), clpC (CP1914), clpE (CP1915), or clpL (CP1916) mutant backgrounds or in the wild-type background (wt*; CP1911) were harvested, lysed, and loaded in lanes 1 to 6. Strains expressing gfp-ssrA or untagged gfp in protease-proficient backgrounds are represented as wt*. Cultures were grown to an optical density at 550 nm of 0.25 in a casein hydrolysate yeast extract medium (13) supplemented with raffinose (1 g/liter) for maximal induction of the aga locus. Cell pellets were resuspended and heated in a 1/100 volume of lysis buffer (100 mM Tris-HCl [pH 6.8], 4% sodium dodecyl sulfate, 0.2% bromophenol blue, 20% glycerol, and 200 mM dithiothreitol) for determination of GFP or GFP-SsrA by Western blotting with anti-GFP antibody (Roche). After sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transfer to polyvinylidene difluoride membranes, the membranes were probed with a mouse anti-GFP primary antibody (1:1,000; Roche) and with an anti-mouse immunoglobulin G secondary antibody conjugated to horseradish peroxidase (1:10,000; GE). Detection was performed with an enhanced chemiluminescence substrate (ECL Plus; GE) and Hyblot CL film (Denville Scientific) with exposures between 10 and 600 s.
|
The level of untagged GFP was not elevated in any of these protease-deficient backgrounds, suggesting that GFP itself is not a substrate of any of the corresponding proteases (Fig. 2). GFP-SsrA was detected in clpX and clpP mutants but not in the wild-type background or in the other mutant strains (Fig. 2). The parallel effects of clpP and clpX mutations strongly suggest that the observed stabilizing effect of these mutations on GFP-SsrA arises from disruption of ClpXP protease activity and not from any polar effects of the two mutations. We conclude that ClpXP is the major pneumococcal Clp protease that degrades SsrA-tagged GFP. However, the possibility that FtsH or another protease has some minor role in degradation of SsrA-tagged proteins cannot be ruled out, as ClpX is present in the other protease mutants and could mask minor additional activities.
GFP tagged with E. coli SsrA is degraded in S. pneumoniae. The sequence of the E. coli SsrA tag (SsrAEc), AANDENYALAA, is similar to that of S. pneumoniae (SsrASp), AKNNTSYALAA. The first 7 amino acids, AANDENY of the E. coli SsrA tag contain the binding site for its adaptor protein, SspB, while the terminal 3 amino acids (LAA) interact with ClpX (5). In vitro, SspB enhances the degradation rate of proteins tagged with SsrA by at least 10-fold and is required for their degradation by ClpXP in vivo (4). The three terminal residues of the pneumococcal SsrA tag are identical to those of the E. coli tag; the tags only differ in the portion specific for the adaptor protein. We reasoned that if S. pneumoniae requires an adaptor protein for ClpXP specific for AKNNTSY, then proteins tagged with SsrAEc would not be degraded, as the AANDENY portion of the tag recognized by the E. coli adaptor protein is considerably different in sequence and overall charge. To seek an indication of a pneumococcal adaptor with a different specificity for SsrA, we constructed a strain containing gfp fused to the ssrAEc tag and determined the levels of GFP-SsrAEc and GFP in vivo by Western blotting. Surprisingly, GFP-SsrAEc was completely absent from wild-type cells, while untagged GFP was readily detected in parallel controls, as expected. Suspecting that ClpXP may be responsible for the absence of GFP-SsrAEc, as it is for GFP-SsrASp, we transformed the SsrAEc construct into clpX and clpP mutants. GFP-SsrAEc was readily detectable in the clpX and clpP mutants at levels comparable to GFP-SsrASp in the same backgrounds (Fig. 3), strongly suggesting that ClpXP degrades GFP tagged with SsrAEc, just as it degrades GFP-SsrASp. This result implies that ClpXP likely does not require an adaptor to degrade SsrA-tagged protein in S. pneumoniae.
![]() View larger version (31K): [in a new window] |
FIG. 3. Targeting of E. coli SsrA by ClpXP in S. pneumoniae. (A) Strains expressing gfp-ssrAEc in clpP (CP1958) and clpX (CP1959) null mutant backgrounds were harvested, lysed, and loaded in lanes 1 and 2 and 7 and 8, respectively. The wild type with no gfp insertion at the aga locus (CP1250) and a strain expressing gfp-ssrAEc in the wild-type background (wt*; CP1957) were loaded in lanes 3 and 4 and lanes 5 and 6, respectively. (B) Strains expressing gfp in clpP (CP1912), clpX (CP1913), or wt (wt*; CP1911) backgrounds were loaded in lanes 1 and 2, 3 and 4, or 5 and 6, respectively. Cultures were induced with raffinose and prepared for Western analysis as described for Fig. 2, but all the samples were loaded in duplicate.
|
|
View this table: [in a new window] |
TABLE 3. Stabilization of CAT-SsrA by clpP or clpX mutations
|
In summary, ClpXP appears to be the primary protease of S. pneumoniae that degrades proteins tagged with SsrA signals of S. pneumoniae or E. coli and it appears to act without the participation of an adaptor protein.
Nucleotide sequence accession numbers. Nucleotide sequences of the new cassettes reported here have been deposited as GenBank accession numbers FJ495555 to FJ495558.
This material is based upon work supported in part by the National Science Foundation under grant no. MCB 0543187.
Published ahead of print on 13 February 2009. ![]()
|
|
|---|
-proteobacteria. J. Bacteriol. 189:272-275.This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»