Previous Article | Next Article ![]()
Journal of Bacteriology, September 2005, p. 6333-6340, Vol. 187, No. 18
0021-9193/05/$08.00+0 doi:10.1128/JB.187.18.6333-6340.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Kinki University, Department of Agricultural Chemistry, Nakamachi, Nara 631-8505, Japan,1 Nippon Institute for Biological Science, Division of Molecular Biology, Ome, Tokyo 198-0024, Japan,2 Hosei University, Faculty of Engineering and Research Center for Micro-Nanotechnology, Koganei, Tokyo 184-8584, Japan,3 National Institute of Genetics, Department of Molecular Genetics, Mishima, Shizuoka 411-8540, Japan4
Received 20 February 2005/ Accepted 21 June 2005
|
|
|---|
|
|
|---|
Escherichia coli contains at least four zinc transport pumps, two zinc transporters (importers), ZnuABC (25) and ZupT (12), and two zinc exporters, ZntA (1, 28) and ZitB (11). The levels of these zinc transport proteins are considered to be regulated at the transcriptional level in a zinc-dependent manner. Two species of cytosolic transcription factor, ZntR and Zur, are involved in transcription regulation of the zinc response genes (3, 24). Transcription of the divergently transcribed znuA and znuBC genes is repressed by Zur when external zinc ions are high, leading to shutoff of zinc uptake. On the other hand, ZntR is an activator of zntA transcription for export of excess zinc. E. coli cells respond to external zinc ions at millimolar concentrations, but both Zur and ZntR respond in vitro to free zinc ions at femtomolar concentrations (24; K. Yamamoto and A. Ishihama, in preparation), suggesting that zinc ions within E. coli cells are present as complexes with proteins or zinc-chelating compounds but free zinc ions are maintained only at low concentrations.
In order to get insights into the response of E. coli to external zinc ions, we performed, in this study, microarray analysis of whole transcripts in the presence and absence of external zinc. In addition to the known set of zinc-responsive genes for import and export of zinc through cell membranes, two groups of genes were found to be highly induced, a set of genes for the synthesis of cysteine and a set of genes under the control of the RpoE sigma factor. The microarray data were confirmed by S1 nuclease mapping of individual mRNAs.
|
|
|---|
M15 galK2 galK22] (17). Cells were grown at 37°C in Luria-Bertani (LB) medium with reciprocal shaking (120 rpm). The cell density was monitored by measuring the turbidity at 540 nm with a Klett-Summerson photometer. At the middle of exponential phase (50 to 60 Klett units), the culture was divided into two portions. Various concentrations of ZnCl2 were added to one aliquot. Culture with or without ZnCl2 addition was continued under the same conditions. After 5 min, cells were harvested and total RNAs were prepared with hot-phenol method, precipitated with ethanol, and dissolved in RNase-free H2O. After digestion with RNase-free DNase I (Takara), RNA was reextracted, precipitated with ethanol, dissolved in RNase-free water, and stored at 80°C until use. Preparation of the fluorescence-labeled cDNA, hybridization on DNA chips, scanning microarrays, and data analysis were performed according to the published procedure (23, 30) with slight modification (O. Ozoline, K. Yamamoto, and A. Ishihama, in preparation). The E. coli CHIPs (Takara) used were Takara Ver. 1 products. All fluorescent intensity data were statistically analyzed as described (23, 30; O. Ozoline, K. Yamamoto, and A. Ishihama, in preparation).
S1 nuclease extension assay of mRNA.
The S1 nuclease protection assay of individual mRNA was carried out as described previously (30-32). Primers used are described in Table 1. Radiolabeling of primers was performed in vitro using 10 µCi [
-32P]ATP (5,000 Ci/mmol) and T4 polynucleotide kinase (Toyobo), and labeled primers were purified by polyacrylamide gel electrophoresis. Total RNAs were purified from E. coli with or without prior treatment by zinc. Mixtures of one 32P-end-labeled probe and total RNAs were incubated for 10 min at 75°C, followed by incubation at 37°C overnight for hybridization. The mixtures were digested with S1 nuclease (Takara) at 37°C for 10 min, and the remaining undigested products were extracted with phenol, precipitated with ethanol, and analyzed by electrophoresis on gels containing 6 M urea.
|
View this table: [in a new window] |
TABLE 1. Oligonucleotides used in this study
|
|
|
|---|
![]() View larger version (44K): [in a new window] |
FIG. 1. Response of wild-type E. coli W3110 to external zinc. [A] E. coli W3110 culture grown overnight at 37°C In LB medium was 100-fold diluted with fresh LB medium with 0 mM (), 0.5 mM ( ), 1.0 mM ( ), and 3.0 mM ( ) of ZnCl2, and incubated at 37°C with reciprocal shaking. Cell growth was monitored by measuring turbidity with a Klett-Summerson photometer. [B] E. coli W3110 was grown in the absence of ZnCl2 until exponential phase, and then ZnCl2 was added at a final concentration of 0.5 mM. Before (lanes 2 and 7) and at 5 min (lanes 3 and 8), 10 min (lanes 4 and 9), and 30 min (lanes 5 and 10) after addition of ZnCl2, cells were harvested and total RNA was isolated with the hot-phenol method (30, 32). The S1 nuclease protection assay was carried out with the zntA (lanes 1 to 5) or znuC (lanes 6 to 10) promoter probe as described in Materials and Methods. 32P-labeled probes were prepared by PCR. Lanes 1 and 6 indicate Maxam-Gilbert AG sequencing reactions. Arrows on the left were the bands protected against S1 nuclease.
|
Differential response of the zinc-responsive genes to external zinc. To confirm the differential response of the zinc importer and exporter genes to external zinc, we next determined the induction of zntA transcription and the repression of znuC transcription in the continuous presence of various concentrations of ZnCl2. When the E. coli culture was exposed to ZnCl2 for a short period (as in Fig. 1B), both zntA induction (Fig. 2-A1) and znuC repression (Fig. 2-A2) were observed at ZnCl2 concentrations above 10 µM (Fig. 2-A3). On the other hand, upon exposure to zinc for a prolonged time (ZnCl2 was added from the beginning of cell culture), the induction of ZntA was observed essentially at the same zinc concentration, above 10 µM (Fig. 2-B1), but the maximum repression of znuC was detected at ZnCl2 concentrations as low as 0.01 µM (Fig. 2-B2 and 2-B3). This finding indicates that in the continuous presence of zinc, repression of synthesis of the zinc importer (ZnuC) takes place at a lower concentration of external zinc than that needed for induction of the zinc exporter (ZntA).
![]() View larger version (52K): [in a new window] |
FIG. 2. Zinc-induced activation of zntA transcription and repression of znuC transcription. [A] E. coli W3110 was grown in the absence of ZnCl2. At the exponential phase of cell growth, various concentrations of ZnCl2 were added and after 5 min, total RNAs were prepared and subjected to S1 nuclease mapping using 32P-labeled zntA [A1] or znuC [A2] probes. [A3] The amounts of zntA ( ) or znuC ( ) RNA relative to those in the absence of ZnCl2 addition were measured. [B] ZnCl2 was added from the beginning of cell culture. The levels of zntA [B1] and znuC [B2] RNA and the relative levels of these RNAs relative to those in the absence of zinc addition [B3] were determined as in A.
|
Profiling of genome transcription in response to external zinc. Based on a combination of proteome analysis and radioactive zinc blotting assays, we estimated that the total number of zinc-binding protein species in E. coli is as much as 3% of the total proteins expressed in the exponential growth phase (15). Some of these zinc-binding proteins were considered to increase after zinc shock and play roles in trapping excess free zinc. To identify the entire set of zinc-responsive genes, we then performed the DNA microarray assay using the E. coli DNA chip (Takara, Japan).
E. coli W3110 was grown to the exponential phase, and then treated for a short period with 0.5 mM ZnCl2 (the minimum concentration required for maximum induction of zntA and maximum repression of znuC). Since the microarray assay should detect the genes which are affected, directly and indirectly, by the addition of zinc, samples were prepared at an early time (less than 5 min) after addition of ZnCl2 for detection of the genes directly induced by the addition of zinc. Total RNAs were prepared from both zinc-treated and untreated cultures, and subjected to the microarray assay under the conditions described in Materials and Methods. In brief, cDNAs were synthesized using Cy3- or Cy5-labeled nucleotide substrates and avian myeloblastosis virus reverse transcriptase. One cDNA sample was labeled with Cy3 and the other with Cy5. Two cDNA samples were mixed and hybridized with the whole set of E. coli genes spotted on a single slide glass.
Here we analyzed two independent cultures. In the first experiment, we labeled untreated sample with Cy3 and the zinc-treated sample with Cy5. The labels were reversed in the second experiment. The fluorescent intensity of each gene spot was measured and statistically analyzed for normalization. Table 2 shows the list of genes which are activated or repressed more than twofold after zinc addition. A total of 26 genes were up-regulated and three genes were down-regulated. As expected, the zntA transcript was included in the up-regulated gene group, while the znuC transcript was a member of the down-regulated genes. E. coli contains the second zinc importer ZupT and the second exporter ZitB, which were identified by genetic analyses using znuABC and zntA mutants, respectively (11, 12). Our microarray data, however, showed that the expression of zupT and zitB was not significantly affected under the conditions employed (5 min after addition of 0.5 mM ZnCl2 at the exponential growth phase) (data not shown).
|
View this table: [in a new window] |
TABLE 2. Zinc-responsive genes in Escherichia coli W3110a
|
![]() View larger version (73K): [in a new window] |
FIG. 3. Zinc-induced activation of the genes for cysteine biosynthesis. [A] E. coli W3110 was grown in LB in the absence of ZnCl2 until exponential phase and then divided into two equal aliquots. ZnCl2 was added to one aliquot at a final concentration of 0.5 mM. After 5 min incubation with (lane 2) and without (lane 3) ZnCl2, cells were harvested and total RNA was isolated with the hot-phenol method as described for Fig. 1. The S1 nuclease protection assay was carried out with the cysS (a), cysK (b), cysP (c), cysD (d), cysJ (e), nirB (f), cysE (g), and metB (h) promoter probes as described in Materials and Methods. Lane 1, Maxam-Gilbert AG sequencing reactions. Arrows on the left were the bands protected against S1 nuclease. [B] A total of 70 genes for cysteine biosynthesis are organized into nine transcriptional units. The first gene of each operon is shown, together with the location of the gene on the E. coli genome on the right. The bold arrows indicate zinc-inducible promoters. The dashed arrow indicates an unidentified promoter for the cysZ operon. [C] The genes activated after addition of external zinc include a number of the genes involved in the synthesis of cysteine. The activated genes, shown by stars, are indicated along the pathway of cysteine biosynthesis.
|
Induction of the RpoE regulon. The microarray data showed more than fourfold induction by zinc of the htrA gene encoding a periplasmic protease, which is involved in degradation of damaged proteins (Table 2). The htrA gene is transcribed by RNA polymerase containing the RpoE sigma factor (reviewed in reference 21). Including htrA, a total of 26 genes are known to be transcribed by RNA polymerase holoenzyme containing RpoE sigma (7). In addition to htrA, four genes of the RpoE regulon, fkpA, rseA, rseC, and yggN, were 1.5- to 2.0-fold up-regulated by zinc (see Table 3). We then carried out detailed analysis of transcriptional response to zinc of some RpoE regulon genes, including not only htrA, fkpA, and yggN but also two well-characterized sigma genes, rpoE and rpoH. S1 nuclease experiments showed two bands for fkpA, htrA, and yggN; three bands for rpoE; and five bands for rpoH (Fig. 4A). The location of two fkpA promoters, upstream p1 and downstream p2, agreed with those previously identified by Danese and Silhavy (6). fkpAp1 was induced markedly by zinc [Fig. 4A(a)]. This p1 promoter was found to contain the conserved promoter 10 and 35 sequences for RpoE sigma (Fig. 4B) as noted by Dartigalongue et al. (7) (note that the promoter naming is reversed in this report).
|
View this table: [in a new window] |
TABLE 3. RpoE-dependent zinc response genesa
|
![]() View larger version (56K): [in a new window] |
FIG. 4. Zinc-induced activation of the RpoE-dependent genes. [A] E. coli W3110 was grown in the absence of ZnCl2 until exponential phase, and then treated with 0.5 mM ZnCl2 as in Fig. 3A. After 5 min culture with (lane 2) and without (lane 3) ZnCl2, cells were harvested and total RNA was isolated with the hot-phenol method as described for Fig. 1. The S1 nuclease protection assay was carried out with the fkpA (a), htrA (b), rpoE (c), rpoH (d), and yggN (e) promoter probes as described in Materials and Methods. Lane 1, Maxam-Gilbert AG sequencing reactions. Arrows on the left were promoters which were detected as the bands protected against S1 nuclease. [B] RpoE-dependent and zinc-inducible promoter sequences. The underlines show the conserved promoter 10 and 35 sequences recognized by RpoE (7).
|
Both the rpoH and rpoE genes carry multiple promoters, of which at least one promoter for each gene is known to be under the control of RpoE. S1 mapping indicates that rpoEp1 and rpoEp2 were 21 nucleotides downstream and 4 nucleotides upstream, respectively, from the sites previously identified (27). Besides these two promoters, a new upstream promoter was detected for the rpoE gene [p0 in Fig. 4A(c)]. However, zinc activated only the rpoEp2 promoter, which is known to be under the control of RpoE sigma. The gene for RpoH sigma has been reported to be transcribed from six promoters (27), of which rpoHp2 is transcribed by RNA polymerase containing RpoE. The locations of three promoters, rpoHp1, rpoHp3, and rpoHp4, were the same as identified previously (9, 10), whereas the initiation sites of rpoHp5 and rpoHp6 transcription were 2 nucleotides upstream and 14 nucleotides downstream, respectively, from the reported sites (22). Previously one promoter, rpoHp3, was identified to be under the direct control of RpoE. Our S1 assay, however, indicates significant induction of two promoters, rpoHp3 and rpoHp6, after exposure to zinc [Fig. 4A(d)]. Accordingly the consensus 10 and 35 sequences were identified not only for the rpoHp3 but also the rpoHp6 promoter (Fig. 4B). These observations suggest that zinc provides a signal for the induction of the RpoE regulon.
|
|
|---|
Transcription stimulation by zinc was also observed for many members organized in the RpoE regulon (see Table 3 and Fig. 4). Zinc homeostasis is maintained inside E. coli cells, but proteins integrated into the cell wall and surface membrane are unavoidably exposed to external metals. Interaction of cell surface proteins with excess metals may lead to denaturation. The abnormal forms of proteins are recognized by refolding systems for repair or protease systems for degradation. The induction of the rpoE regulon by zinc agrees well with the fact that the genes encoding the proteins for repair or degradation of abnormal proteins are under the control of RpoE sigma (7). Under normal growth conditions, unused RpoE sigma is stored as a membrane-bound form with anti-RpoE sigma (8, 20). Thus, the anchoring target of RpoE sigma on cell membranes plays a role in sensing zinc level in environment. The CpxRS two-component system is also involved in induction of several enzymes for protein folding (29). Exposure of E. coli to excess zinc, however, does not induce the CpxRS regulon genes (data not shown), implying that CpxRS sense a different signal(s) or condition(s) for induction of a protein-folding system.
Brocklehurst and Morby (4) performed microarray assay for E. coli strain TG1, which adapted to survive in the continuous presence of external zinc, for identification of the genes involved in iron tolerance. The sets of increased and decreased genes are, however, different from the sets of up- and down-regulated genes in E. coli W3110 which was exposed to zinc for a short period (Tables 2 and 3).
This work was supported by Grants-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology, and the CREST fund from the Japan Science Corporation.
|
|
|---|
E and the Cpx signal transduction systems control the synthesis of periplasmic protein-folding enzymes in Escherichia coli. Genes Dev. 11:1183-1193.
E regulon. J. Biol. Chem. 276:20866-20875.
E-mediated response to extracytoplasmic stress in Escherichia coli is transduced by RseA and RseB, two negative regulators of
E. Mol. Microbiol. 24:373-385.[CrossRef][Medline]
E (RpoE) heat-shock transcription-factor activity by the RseA, RseB and RseC proteins. Mol. Microbiol. 24:355-371.[CrossRef][Medline]
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»