Previous Article | Next Article ![]()
Journal of Bacteriology, October 2004, p. 6824-6829, Vol. 186, No. 20
0021-9193/04/$08.00+0 DOI: 10.1128/JB.186.20.6824-6829.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Curdin Ragaz,1,
Christian Baron,2 and Franz Narberhaus1,3*
Institut für Mikrobiologie, Eidgenössische Technische Hochschule, Zürich, Switzerland,1 Department of Biology, McMaster University, Hamilton, Ontario, Canada,2 Lehrstuhl für Biologie der Mikroorganismen, Fakultät für Biologie, Ruhr-Universität Bochum, Bochum, Germany3
Received 8 April 2004/ Accepted 14 July 2004
|
|
|---|
32), both copies on pAT were under the control of highly conserved ROSE (named for repression of heat shock gene expression) sequences in their 5' untranslated region. Secondary structure predictions of the corresponding mRNA strongly suggest that it represses translation at low temperatures by masking the Shine-Dalgarno sequence. The hspC gene was barely expressed (if at all) and not temperature responsive. |
|
|---|
A puzzling observation was recently made during studies on the temperature stability of the A. tumefaciens type IV secretion system, which translocates DNA and protein into plant cells. Concomitant with the accumulation of plant secondary metabolite-induced virulence proteins at low temperatures, small heat shock proteins (sHsps) were induced (see Discussion of reference 4). Synthesis of sHsps was also observed during symbiosis of Sinorhizobium meliloti with the legume Melilotus alba (30), and root exudates of the symbiotic host Alnus glutinosa induced sHsp synthesis in the actinomycete Frankia sp. strain ACN12a-tsr (12), suggesting a functional importance of sHsps during plant-microbe interaction. Generally, sHsps belong to a large and heterogeneous family of molecular chaperones, also called
-crystallin-type heat shock proteins (
-Hsps) (6, 26). As the name indicates, sHsps are small (around 20 kDa) and usually induced under heat stress conditions. The Escherichia coli Ibp proteins (inclusion body binding proteins IbpA and IbpB) were named Ibp, because they were found associated with inclusion bodies derived from massively overproduced recombinant protein (1). It has been established that sHsps exhibit chaperone-like activity (13, 14). They stably bind to unfolded proteins, maintain them in a folding-competent state, and thus prevent formation of large insoluble aggregates (7, 18). In contrast to most other bacteria, rhizobia produce multiple sHsps after heat shock (23). Two distinct classes could be distinguished on the basis of their primary amino acid sequences. Members of both classes have chaperone activity and assemble into large class-specific oligomers (42).
At least three different mechanisms are known to confer temperature regulation to sHsp-encoding genes in bacteria (26). In many organisms, alternative sigma factors, such as RpoH (
32), induce heat shock gene transcription under stress conditions (46). RheA, a repressor protein, inhibits transcription of Hsp18 in Streptomyces albus (40). Posttranscriptional control by ROSE (named for repression of heat shock gene expression), a conserved sequence in the 5' untranslated regions (5'-UTRs) of sHsp genes, in rhizobia has been reported (27, 32). The sequence is predicted to fold into a secondary structure that masks the Shine-Dalgarno (SD) sequence at low temperatures (31). Local melting of that structure at higher temperatures liberates the SD sequence and allows translation (5).
Regulation of the heat shock response in A. tumefaciens is only partially understood. The dnaKJ operon, clpB, and several other heat shock genes are under control of RpoH (19, 24, 38). The groESL operon was found to be under dual control of RpoH and the HrcA repressor (24, 39). Analysis of the heat shock proteome by two-dimensional protein gel electrophoresis confirmed these results but also raised additional questions, since 32 of 56 heat shock proteins were induced independently of RpoH and HrcA (34). The existence of at least one additional control mechanism was proposed. Our present study on regulation of the four A. tumefaciens sHsps revealed that ROSE-mediated control is one of these mechanisms.
|
|
|---|
Plasmid constructions. Recombinant DNA work was performed by standard protocols (35). Plasmids pAT5277, pAT5278, and pAT5279 (Table 1) were constructed as follows. PCR-generated fragments were digested with SmaI and PstI before they were ligated into pUC18 treated with the same enzymes. The correct nucleotide sequence was confirmed by automated sequencing. Subsequently, the fragments were transferred into pSUP482 to construct in-frame fusions to the lacZ gene. The entire hsp-lacZ fusions were then cloned via the SmaI and XbaI sites into pVS-BADNco. Reporter plasmids were transferred into E. coli via transformation and into A. tumefaciens via electroporation.
|
View this table: [in a new window] |
TABLE 1. Strains, plasmids and oligonucleotides used in this study
|
Immunoblot analyses and generation of antisera. Cells from liquid cultures or from plates were washed with 0.9% NaCl prior to use. Protein concentrations were determined with the Bradford assay (Bio-Rad). Equal amounts (20 µg of protein) of each extract were separated by sodium dodecyl sulfate-12% polyacrylamide gel electrophoresis and blotted onto Hybond-C extra nitrocellulose filters (Amersham). Antisera were used in 1/10,000 dilutions. Bound peroxidase-coupled secondary anti-rabbit immunoglobulin G was detected with the Pierce SuperSignal kit.
Rabbit serum against streptavidin-tagged HspL was generated as follows. The hspL gene was PCR amplified from A. tumefaciens C58 chromosomal DNA with oligonucleotides ibpA5 and ibpA3. The fragment was cleaved with Acc65I and PstI and ligated into similarly cleaved pT7-7StrepII. Transformed cells were plated on LB medium supplemented with AMP (50 µg/ml), and the desired clones were identified by restriction analysis and DNA sequencing. The T7 promoter-carrying plasmid was introduced into E. coli strain BL21(
DE3) for protein overproduction. Cells in 800-ml culture (4x 200 ml in 500-ml flasks) in LB medium containing AMP (100 µg/ml) were induced for 4 h at 37°C with 0.5 mM isopropyl-ß-D-thiogalactopyranoside (IPTG) for overproduction of the protein fused to the N-terminal affinity (StrepII) tag (sequence was WSHPQFEK). The overexpressed StrepII tag fusion protein was purified from cell lysates in one step by affinity chromatography on a streptactin Sepharose column (IBA, Göttingen, Germany) according to the manufacturer's instructions. The purified protein (700 µg) was lyophilized and injected into rabbits for the generation of a specific antiserum (BioGenes, Berlin, Germany).
RNA isolation and primer extension. Total RNA was prepared using hot acid phenol as described previously (2). The oligonucleotides used to map the transcription start sites of hspAT1, hspAT2, hspL, and hspC are listed in Table 1.
ß-Galactosidase assays. The ß-galactosidase activity in exponentially growing E. coli and A. tumefaciens cells was measured by standard protocols (20).
|
|
|---|
![]() View larger version (24K): [in a new window] |
FIG. 1. Overview of the four small heat shock genes encoded by A. tumefaciens. The genomic locations (pAT plasmid, linear chromosome, and circular chromosome) are indicated in the gene designations. Potential transcription start sites are represented by arrows. Presumed regulatory mechanisms are listed, and ROSE-type sequences are shown as grey boxes.
|
-crystallin domain and the IxI motif in the C-terminal extension (data not shown). Agrobacterial sHsps from different replicons exhibit around 45% identity, whereas both proteins derived from the pAT plasmid have 74% of their amino acids in common.
Temperature-dependent production of
-Hsps in A. tumefaciens was monitored by immunoblot analysis using antisera raised against HspL. sHsps were not detected during exponential growth at 20 or 30°C or after a shift from 20 to 30°C (Fig. 2). Continuous growth at 37°C and shifts from 20 to 37 or 42°C triggered sHsp synthesis. In line with the severity of the heat shock, an upshift from 30 to 37°C for 30 min induced much smaller amounts of sHsps than a shift from 30 to 42°C. The results demonstrate that A. tumefaciens precisely records the ambient temperature and adjusts the cellular level of sHsps accordingly.
![]() View larger version (23K): [in a new window] |
FIG. 2. Heat-induced production of sHsps in A. tumefaciens. Cells were grown in YEB medium at the temperatures indicated. For heat treatment, aliquots of the cells were shifted to higher temperatures for 30 min. Harvested cells were subjected to Western blot analysis as described in Materials and Methods using an anti-HspL serum. The detected band migrates at 18 kDa.
|
ROSE-controlled expression of pAT-encoded sHsps.
Both hspAT1 and hspAT2 are preceded by a sequence strongly resembling the ROSE motif of many other rhizobial small heat shock genes (31, 32). To assign the precise transcription start sites, primer extension experiments were performed with RNA from normally grown cells and heat-shocked cells. As in B. japonicum, Rhizobium sp. strain NGR234, and other species (29, 32), corresponding transcripts were detected only in RNA preparations from heat-treated A. tumefaciens cells (Fig. 3A and B). The deduced 10 and 35 regions of both hspAT genes are identical and conform to the consensus sequence of a
70-type promoter. Both approximately 80-nucleotide-long 5'-UTRs downstream of the transcription start sites are highly similar, in particular towards the 3' end where more than 30 residues including the SD sequences are completely identical (Fig. 3C). Computer-assisted secondary structure predictions of the 5'-UTRs suggest in each case an extended secondary structure resembling ROSE-like elements found in other rhizobia (shown for hspAT2 in Fig. 3D). The characteristic features include the SD sequence and AUG start codon being masked by base pairing as well as a strictly conserved bulged G residue opposite the SD sequence. The corresponding bulged residue in ROSE1 of B. japonicum (G83) was shown to be instrumental in translation initiation at elevated temperatures in vivo (31). Accordingly, thermal melting of an in vitro-synthesized transcript lacking that residue started to melt only at temperatures above 45°C (5).
![]() View larger version (39K): [in a new window] |
FIG. 3. Transcription start site mapping of pAT-encoded small heat shock genes. Primer extension experiments were performed with total RNA from A. tumefaciens grown in YEB medium at 30°C or heat shocked from 30 to 42°C for 30 min. (A) Reverse transcriptase reaction with primer Sig212, and (B) reverse transcriptase reaction with primer Sig214. The corresponding sequencing reactions (AGCT) with pUC18-derived plasmids carrying the hspAT1 and hspAT2 regions, respectively, are shown. The positions of reverse transcripts corresponding to +1 in panel C are indicated by black arrowheads to the left of the gels. (C) Alignment of the deduced promoter regions and ROSE sequences of hspAT1 (top strand) and hspAT2 (bottom strand). Transcription start sites (+1), 10 and 35 regions, Shine-Dalgarno sequences (SD), and translational start sites are indicated. (D) Secondary structure prediction of the hspAT2 ROSE region. The mfold program version 3.1 (http://www.bioinfo.rpi.edu/applications/mfold/old/rna/) was used (47). Nucleotides are numbered starting from the transcription start site (+1). The Shine-Dalgarno sequence and start codon are labeled. A conserved bulged G residue present in all known ROSE elements is indicated by a black arrow.
|
The hspL gene belongs to the RpoH regulon.
Primer extension experiments with oligonucleotides complementary to the 5' end of hspL revealed transcripts of hspL only under heat shock conditions (Fig. 4A). The deduced transcription start site is preceded by a 10 region and a 35 region with similarity to RpoH-dependent promoters from A. tumefaciens or E. coli (Fig. 4B). The presence of a RpoH-type promoter is consistent with proteome analyses of heat-shocked A. tumefaciens cells (34). Heat induction of a protein named HspD was lost in a
rpoH strain. The N-terminal sequence of HspD corresponds to the HspL sequence (Eliora Ron [Tel-Aviv University], personal communication). We suggest renaming HspD and using HspL for its chromosomal location. Interestingly, two different HspL spots were located on two-dimensional protein gels, suggesting covalent modification of the protein (34).
![]() View larger version (30K): [in a new window] |
FIG. 4. Transcription start site mapping of the hspL gene. Primer extension experiments were performed with total RNA from A. tumefaciens grown in YEB medium at 30°C or heat shocked from 30 to 42°C for 30 min. Primers Sig225 (A) and Sig218 (not shown) producing identical results were used. The corresponding sequencing reactions (AGCT) with a pUC18-derived plasmid carrying the hspL region, respectively, are shown. The position of the reverse transcript from which the transcription start site was deduced is indicated by an arrowhead to the right of the gel. (B) The deduced promoter region of hspL is listed and compared with consensus sequences of RpoH-dependent promoters from A. tumefaciens (Atu) and E. coli (Eco) (24).
|
In order to facilitate a comparative analysis of sHsp genes from different replicons, we constructed translational lacZ fusions to hspAT2, hspL, and hspC (Materials and Methods and Table 1). All constructs led to basal expression when A. tumefaciens was grown at 25°C (Fig. 5). Expression of hspAT2-lacZ was elevated during growth at 30°C and induced more than 10-fold at 37°C. The hspL-lacZ fusion was induced fivefold during growth at 37°C. Both results confirm that expression of the ROSE- and RpoH-controlled sHsp genes is temperature regulated. In contrast, the expression of hspC increased only slightly with temperature, indicating again that it is not significantly expressed and not temperature controlled.
![]() View larger version (18K): [in a new window] |
FIG. 5. Temperature-dependent expression of A. tumefaciens small heat shock genes. Cultures were grown to exponential phase in AB medium with glucose at the temperatures indicated before ß-galactosidase activity (in Miller units [MU]) was measured.
|
32 (Fig. 4B), expression of this gene was not induced by increasing the temperature in E. coli (Fig. 6). It has been observed previously that promoter sequences from
-proteobacteria were not recognized by E. coli RNA polymerase. E. coli
32-containing RNA polymerase was unable to transcribe the B. japonicum dnaK gene (22), and the
70-containing holoenzyme of E. coli did not recognize the housekeeping-type hspA-rpoH1 promoter of B. japonicum (31). The A. tumefaciens hspL gene appears to be another example of this phenomenon. On the other hand, the E. coli dnaK promoter was heat activated in A. tumefaciens (38), and the E. coli groESL promoter was transcribed by the B. japonicum RNA polymerase carrying sigma factor RpoH1 or RpoH2 (28). These data suggest that promoter recognition by RNA polymerases from
-proteobacteria is less strict than transcription initiation by the E. coli enzyme.
![]() View larger version (23K): [in a new window] |
FIG. 6. Expression of lacZ fusions to hspAT2 and hspL after a temperature upshift in A. tumefaciens and E. coli. A. tumefaciens cells grown in AB medium with glucose and E. coli grown in LB medium were shifted from 20 to 37°C before ß-galactosidase activity was measured at the time points indicated. The activities at timepoint zero were set to 1.
|
Our results identified the ROSE element in A. tumefaciens as one of the additional heat shock control systems proposed by Rosen et al. (34). However, only two genes, hspAT1 and hspAT2, are regulated by this translation mechanism. A search for additional ROSE-type sequences in the genome of A. tumefaciens demonstrated that no other genes bear such an RNA thermometer. Since 32 of 56 heat shock proteins were shown to be induced independently of RpoH and HrcA (34), induction of a substantial number of proteins still remains to be elucidated.
Work in Christian Baron's laboratory was supported by the Canadian Institutes of Health Research (grant MOP-64300). The work performed in Franz Narberhaus' laboratory was supported by the Swiss Federal Institute of Technology, Zürich, Switzerland.
S.B. and C.R. contributed equally to this work. ![]()
|
|
|---|
-crystallinsmall heat-shock protein superfamily. Int. J. Biol. Macromol. 22:151-162.[CrossRef][Medline]
-Crystallin can function as a molecular chaperone. Proc. Natl. Acad. Sci. USA 89:10449-10453.
32-like sigma factor. J. Bacteriol. 174:991-997.
32 dependent class of heat shock genes in Bradyrhizobium japonicum. Mol. Gen. Genet. 254:195-206.[CrossRef][Medline]
32 homolog (RpoH) and an HrcA repressor in the heat shock response of Agrobacterium tumefaciens. J. Bacteriol. 181:7509-7515.
32 homolog (RpoH) plays a major role in the heat shock response of Agrobacterium tumefaciens. J. Bacteriol. 183:5302-5310.
-Crystallin-type heat shock proteins: socializing minichaperones in the context of a multichaperone network. Microbiol. Mol. Biol. Rev. 66:64-93.
32-like protein is part of a unique heat shock gene cluster together with groESL1 and three small heat shock genes. J. Bacteriol. 178:5337-5346.
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»