Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Ahn, S. K. Articles by Nodwell, J. Search for Related Content PubMed PubMed Citation Articles by Ahn, S. K. Articles by Nodwell, J.

 Previous Article  |  Next Article 

Journal of Bacteriology, September 2007, p. 6655-6664, Vol. 189, No. 18
0021-9193/07/$08.00+0     doi:10.1128/JB.00717-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Investigation of Transcription Repression and Small-Molecule Responsiveness by TetR-Like Transcription Factors Using a Heterologous Escherichia coli-Based Assay

Sang Kyun Ahn,¹ Kapil Tahlan,¹ Zhou Yu,² and Justin Nodwell^1*

Department of Biochemistry and Biomedical Sciences, McMaster University, 1200 Main St. W., Hamilton, Ontario, Canada L8N 3Z5,¹ Department of Medical Genetics and Microbiology, University of Toronto, 1 King's College Circle, Toronto, Ontario, Canada M5S 1A8²

Received 7 May 2007/ Accepted 10 July 2007

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

 
The SCO7222 protein and ActR are two of 150 TetR-like transcription factors encoded in the Streptomyces coelicolor genome. Using bioluminescence as a readout, we have developed Escherichia coli-based biosensors that accurately report the regulatory activity of these proteins and used it to investigate their interactions with DNA and small-molecule ligands. We found that the SCO7222 protein and ActR repress the expression of their putative target genes, SCO7223 and actII-ORF2 (actA), respectively, by interacting with operator sequence in the promoters. The operators recognized by the two proteins are related such that O₇₂₂₃ (an operator for SCO7223) could be bound by both the SCO7222 protein and ActR with similar affinities. In contrast, O_act (an operator for actII-ORF2) was bound tightly by ActR and more weakly by the SCO7222 protein. We demonstrated ligand specificity of these proteins by showing that while TetR (but not ActR or the SCO7222 protein) interacts with tetracyclines, ActR (but not TetR or the SCO7222 protein) interacts with actinorhodin and related molecules. Through operator-targeted mutagenesis, we found that at least two nucleotide changes in O₇₂₂₃ were required to disrupt its interaction with SCO7222 protein, while ActR was more sensitive to changes on O_act. Most importantly, we found that the interaction of each protein with wild-type and mutant operator sequences in vivo and in vitro correlated perfectly. Our data suggest that E. coli-based biosensors of this type should be broadly applicable to TetR-like transcription factors.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

 
Three mechanisms commonly confer resistance to the antibiotic tetracycline: antibiotic degradation by TetX-like enzymes (31), ribosome protection by TetO-like proteins (7, 23), and, most frequently, antibiotic export by TetA-like efflux pumps (30). Many of the genes encoding these resistance determinants are under the direct control of repressor proteins referred to as the TetR-like transcription factors. Like most repressors, these proteins bind operator sequences in their target promoters preventing transcription initiation by RNA polymerase holoenzyme. A smaller number of TetR-like proteins have been implicated in transcriptional activation (6, 11).

TetR, the best-characterized member of this family, controls the expression of the efflux pump-encoding gene tetA in the transposon Tn10 (reviewed in reference 13). The tetR and tetA genes are divergently transcribed and separated by 80 bp of DNA that includes their promoters. The TetR protein binds tightly to two nearly identical 15-bp palindromic operator sequences (O_tet) to interfere with transcription of both genes. When tetracycline enters the cell, it binds the C-terminal ligand-binding domain in TetR as a complex with Mg²⁺, causing its release from O_tet (19, 25, 26). This relieves repression, permitting the expression of tetA and the export of the antibiotic out of the cell (16).

Genes encoding TetR-like transcription factors are common in bacterial genomes (reviewed in reference 20). For example, the filamentous antibiotic-producing bacterium Streptomyces coelicolor encodes at least 150 TetR-like transcription factors. It is noteworthy that many of these are closely linked (either as divergently transcribed genes or in operons) to genes that encode proteins similar to TetA-like efflux pumps and TetX-like monooxygenases. It is striking therefore, that relatively low concentrations of tetracycline (1 µg/ml) can impair this organism's growth (21). This constitutes considerable sensitivity given that 30 µg/ml is required to inhibit the growth of resistant Escherichia coli strains (1), and it suggests that the TetR-like transcription factors control many physiological processes unrelated to tetracycline. Only a few of the S. coelicolor TetR-like proteins have been linked to specific processes: ActR regulates export of the polyketide actinorhodin (3, 8, 24), Pip controls a putative multidrug resistance gene (10), PqrA controls a paraquat efflux pump (5), CprA and CprB may interact with -butyrolactones (17, 18), and ScbR controls production of the -butyrolactone SCB1 (27).

All of the TetR-like transcription factors that have been studied in molecular detail interact with small-molecule ligands that are chemically related or identical to the substrates of the proteins encoded by their target genes. TetR binds tetracycline (15), ActR interacts with actinorhodin and actinorhodin biosynthetic intermediates (24), and QacR binds various cationic lipophilic drugs (12). It is our view, therefore, that identification of the small-molecule ligands that interact with the C-terminal regulatory domains of TetR-like proteins of unknown function is a powerful means of deciphering their roles as well as those of the genes they control.

To facilitate this endeavor, we have developed a biosensor mechanism that we believe can be applied to many members of this family (24). The biosensors are based on synthetic promoters consisting of –10 and –35 promoter elements, separated by a putative binding site for a TetR-like transcription factor. These promoters are cloned upstream of the luxCDABE operon of Photorhabdus luminescens. LuxA and -B encode a luciferase enzyme, while LuxC, -D, and -E form a fatty aldehyde reductase complex that provides the luciferase substrate (29). The gene encoding the cognate TetR-like transcription factor is cloned into a second plasmid such that cells containing the lux plasmid are spontaneously bioluminescent while those containing both are not, due to repression of the synthetic promoter by the repressor. We showed previously with biosensors of this type based on TetR and ActR that bioluminescence could be induced by tetracycline and by actinorhodin and some of its biosynthetic intermediates, respectively (24).

In this study, we have explored the utility of this biosensor mechanism in greater depth, focusing on ActR, and the uncharacterized TetR-like SCO7222 protein. Based on the tetR/tetA paradigm, the target of the SCO7222 protein is predicted to be the divergently transcribed gene SCO7223, which encodes a probable TetX-like monooxygenase. We show that the SCO7222 protein interacts with palindromic sequences in the SCO7222/7223 intergenic sequence. These sequences are related to those bound by ActR, and indeed the two proteins exhibit considerable affinities for each other's operators, though we do not believe that these heterologous interactions are biologically relevant. SCO7222 protein does not respond to the ligands recognized by either ActR or TetR. Mutagenesis of the binding sites for the SCO7222 protein and ActR revealed that the two proteins recognize distinct nucleotides and, most importantly, that there is a perfect correlation between the in vivo and in vitro DNA-protein interactions. We suggest, therefore, that this biosensor mechanism is likely to be broadly applicable to the TetR-like proteins.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

 
Bacterial strains, plasmids, culture conditions, and bioluminescence measurements. Bacterial strains and plasmids used in this study are described in Table 1. E. coli cultures were grown using Luria broth (LB) or LB agar medium containing the appropriate antibiotics when required (22). Streptomyces cultures were grown at 30°C in YEME broth or maintained on solid MS agar medium (14). Isolated E. coli colonies were used to inoculate 1-ml amounts of reporter cultures, which were grown for 16 to 20 h before measurement of luminescence using a Lumat 9507 luminometer (Bertholt Technologies). In some cases, E. coli reporter cultures were supplemented with spent Streptomyces culture supernatant (prepared as described in reference 24) or with purified tetracyclines. The half-maximal concentrations of various tetracyclines required for induction were determined using a standard sigmoidal dose-response regression equation: y = bottom + (top – bottom)/(1 + 10^{(EC₅₀ – x) · hill slope}), where bottom (y_min) was set to 1, as data were normalized using signal/noise ratios, and where EC₅₀ represents the 50% effective concentration.

View this table: [in this window] [in a new window]

TABLE 1. Bacterial strains and plasmids used in this work

Procedures for DNA manipulation. Standard procedures were used for plasmid DNA isolation, manipulation, and analysis (22). Oligonucleotide primers used in this study were obtained from the Institute for Molecular Biology and Biotechnology (MOBIX) facility at McMaster University or from Sigma. PCRs were carried out using Vent DNA polymerase (New England Biolabs) with the following program: 95°C for 5 min followed by 30 cycles of 95°C and 55°C (30 s each), followed by 72°C (30 or 60 s, depending on the length of a template), with a final extension step at 72°C for 7 min. DNA sequencing was carried out by the MOBIX facility to select/isolate the appropriate PCR products.

Construction of lux-based reporter plasmids and expression vectors for the TetR, ActR, and SCO7222 protein biosensors. pO^tetlux and pO^actlux were constructed previously (24). Similarly, a DNA fragment containing a synthetic promoter—consisting of SCO7223 operator (O₇₂₂₃) flanked by the –35 and –10 regions from the Tn10 tetA promoter—was prepared by annealing T7223-1 and T7223-2 oligonucleotides (Fig. 1B and Table 2). These oligonucleotides were introduced into the BamHI-XhoI sites of pCS26-Pac to give reporter plasmid pO⁷²²³lux, in which the expression of the lux operon was under the control of the SCO7222 protein (Table 1).

View larger version (29K): [in this window] [in a new window]

FIG. 1. Putative binding sites for ActR and the SCO7222 protein. (A) Repeated elements in the actR/actA (modified from reference 24) and SCO7222/SCO7223 intergenic regions are shown. Between actR and actA, there are three weakly palindromic sequences (underlined) that exhibit a low degree of conservation. Between SCO7222 and SCO7223 there are three 15-bp repeats that are perfectly palindromic and highly conserved. Arrows indicate repeated sequence (actR/actA) and palindromes (SCO7222/SCO7223). (B) Sequence of the synthetic promoter used in the pO7223lux. The DNA fragment contained –35 and –10 regions from the Tn10 tetA promoter, flanking O7223. Arrows indicate the palindromic nucleotides in O7223 and Otet.

View this table: [in this window] [in a new window]

TABLE 2. Primers used in this study

Two expression vectors, pTetR and pActR (encoding TetR and ActR, respectively) were previously prepared (24). In parallel, the primer pair 7222-1/7222-2 (Table 2) along with S. coelicolor M145 chromosomal DNA (as template) was used to PCR amplify DNA fragments containing SCO7222. Gel-purified PCR product was then digested with BamHI and ligated to BamHI-EcoRV-treated pACYC184 to give pSCO7222, which served as the source of the SCO7222 protein in the biosensors.

Operator mutagenesis. Double-stranded DNA products, obtained by annealing the imperfectly matched primers TA3 and TA2 and T7223-3 and T7223-2, were inserted into the BamHI-EcoRV sites of pCS26-Pac to give pMAlux (TA3-TA2) and pMTlux (T7223-3-T7223-2), respectively. TA3 and T7223-3 are identical to TA1 and T7223-1, respectively, except that the operator sequences had been doped during synthesis as follows. At each operator position, while the concentration of the correct nucleotide was as usual during synthesis, the other three nucleotides were also present, as defined by x/3n, where x represents the concentration of principle nucleotide and n represents the length of the operator sequence. The result of this is that each mutagenic oligonucleotide would be expected to possess at least one nucleotide sequence change in the operator region, embedded in an otherwise "wild-type" synthetic promoter. In both cases, we mutagenized the top strands shown in Fig. 1B. The plasmids pMAlux and pMTlux were introduced into E. coli strains containing pActR or pSCO7222, respectively, to isolate colonies that produced luminescence, indicating that the respective repressors could not bind to the mutagenized operators. Plasmid DNA was isolated from these strains and reintroduced into E. coli to isolate kanamycin-resistant and chloramphenicol-sensitive colonies, which had lost pActR or pSCO7222 but still contained pMAlux or pMTlux. The DNA sequences of the operators cloned in pMAlux and pMTlux were determined for further analysis.

Expression and purification of His₆-ActR and His₆-SCO7222 protein. Previously prepared pET28a-ActR was used to express and purify N-terminal six-His-tagged ActR (His₆-ActR) in E. coli (24). The primers TOEV-1 and TOEV-2 (Table 2), along with pSCO7222 plasmid DNA (as template), were used to PCR amplify a 721-bp fragment containing SCO7222, which was introduced into the EcoRV site of pBlueScript SK II⁺ to give pBS7222. After sequencing SCO7222, the DNA fragment encoding SCO7222 protein was isolated as an NdeI-BamHI fragment from pBS7222 and was ligated to pET28a, giving pTO7222 (Table 1).

E. coli BL21(DE3) cultures containing pTO7222 were grown at 37°C to an optical density at 600 nm of 0.4 to 0.6 and were then induced with 1 mM isopropyl ß-D-thiogalactopyranoside for 20 h at 30°C. From this point, the same procedures were taken to purify His₆-ActR and His₆-SCO7222 (24).

EMSAs. The primers TET-EMSA-F and TET-EMSA-R (24), along with pO^actlux, pO⁷²²³lux, or the reporter plasmids with point-mutagenized operators (as templates), were used in PCRs to isolate double stranded DNA fragments containing operator regions, which served as probes for electrophoretic mobility shift assays (EMSAs). The DNA fragments were end labeled using [-³²P]ATP and T4 polynucleotide kinase (22).

Labeled probe (12.7 fmol), 1.5 to 3,200 fmol of purified protein and 90 ng of salmon sperm DNA were used in 15-µl reactions containing 1x reaction buffer (10 mM Tris-Cl [pH 7.8], 150 mM NaCl, 2 mM dithiothreitol, 10% glycerol). Reaction mixtures were incubated at 30°C for 10 min and were fractionated on 12% nondenaturing polyacrylamide gels containing 1.5% glycerol. The gels were exposed using a phosphor screen (Amersham), and bands were detected using a PhosphorImager (Molecular Dynamics).

Determination of K_D. ImageQuant software (Molecular Dynamics) was used to analyze EMSA results to determine the percentages of shifted and unshifted probes, which represent bound and unbound substrate, respectively. Saturation curves (% probe bound against [protein]) were drawn with SigmaPlot 2000 to determine the dissociation constant (K_D). Binding cooperativity was determined by Hill's equation: log (Y/1 – Y) = h log[S] – log K_D, where Y = protein-DNA complex/total DNA, S is the protein of interest, and K_D is the dissociation constant of protein binding to DNA (9). The slope of a straight line passing the point where 50% DNA binding occurs, is the Hill's coefficient (h) and indicates the binding cooperativity.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

 
Construction of biosensors using TetR, ActR, and the SCO7222 protein. Given the TetR paradigm, we predicted that the likely target of the SCO7222 protein would be SCO7223. Examination of the 110 bp that separates these genes revealed the sequences N1, N2, and N3 (Fig. 1A) corresponding to likely binding sites for the SCO7222 protein. All of these sequences exhibit significant similarity to the nearly perfect palindrome having the consensus sequence C/TTGGAACGNCGTTCCAG/C. This is similar to the intergenic DNA of actR and actA that includes the putative ActR-binding sequences P1, P2 (O_act), and P3. We carried out gel mobility shift assays using both intergenic fragments and purified His₆-SCO7222 and His₆-ActR. As shown in Fig. 2, both fragments formed three exceptionally tight complexes with their cognate repressors, consistent with the presence of three repressor binding sites in each. We determined approximate K_Ds for each of the SCO7222 protein complexes and found them to be 0.4 nM, 3.4 nM, and 17.3 nM for the L, M, and H complexes, respectively. A Hill plot of this data demonstrates a Hill's coefficient of 1.6, consistent with positive cooperative binding of the SCO7222 protein (Hill's coefficient of >1) to its three binding sites. The interaction of ActR with the actR/actA intergenic sequence was also characterized by strong binding (K_Ds of 0.1 nM, 0.3 nM, and 5.8 nM), but with a Hill's coefficient of 1 (data not shown), there was no evidence of cooperativity.

View larger version (50K): [in this window] [in a new window]

FIG. 2. Interactions of the SCO7222 protein and ActR with intergenic sequences. (A) A 32P-labeled probe containing the entire SCO7222/SCO7223 or actR/actA intergenic DNA (IR) sequences was incubated with the indicated concentrations of the SCO7222 protein or ActR at 30°C for 10 min and then separated in 12% nondenaturing polyacrylamide gels. Three distinguishable shifts, L (lower shift), M (middle shift), and H (higher shift), were observed in both cases. (B to D) The KD value of each SCO7222 protein shift was determined by plotting the percentage of bound DNA against the concentration of the SCO7222 protein. (E) Hill's plot was drawn to measure binding cooperativity by the SCO7222 protein.

As previously, we designed synthetic promoters in which each of the N1, N2, and N3 sequences was introduced between functional E. coli –10 and –35 promoter elements and introduced them upstream of luxCDABE operon in the vector pCS26-Pac (2). The resulting plasmids pN1lux, pN2lux, and pO⁷²²³lux were introduced into E. coli along with the previously constructed pO^tetlux and pO^actlux, and all conferred significant bioluminescence compared to promoterless reporter vector pCS26-Pac (Table 3). Similarly, we constructed an SCO7222 protein-producing plasmid by amplifying the gene with oligonucleotides that included a Shine-Dalgarno sequence upstream of the initiator codon and a BamHI site at the downstream end. This fragment was ligated downstream of a tetA promoter in the vector pACYC184 to create the plasmid pSCO7222. We introduced pTetR, pActR, pSCO7222, and the control pACYC184 into the E. coli strains bearing pO^tetlux, pO^actlux, pN1lux, pN2lux, and pO⁷²²³lux and tested each strain for bioluminescence. As previously reported, the pTetR and pActR eliminated pO^tetlux- and pO^actlux-dependent bioluminescence, respectively, while the control plasmid pACYC184 had no effect (24). Similarly, pSCO7222 eliminated the bioluminescence produced by pN1lux, pN2lux, and pO⁷²²³lux (Table 3), confirming direct binding as observed in vitro. The in vivo function of these three sites agrees with the mobility shift demonstrating three complexes between the SCO7222 protein and the SCO7222/SCO7223 intergenic sequence. This suggested that the SCO7222 protein is a repressor of SCO7223, and we focused the rest of our work on N3, which we refer to O₇₂₂₃. These results underscore the fact that in those cases (which are common) in which a tetR-like gene is paired with a putative target gene in a manner similar to the well-characterized tetR/tetA configuration, repeated palindromic sites in the intergenic sequences can be assumed to be candidate binding sites for the TetR-like protein.

View this table: [in this window] [in a new window]

TABLE 3. Effects of TetR/ActR/SCO7222 on various reporter plasmids

We noted that the putative operators recognized by ActR (ACGCGACCACCGTTCCAT) and the SCO7222 protein (CTGGAACGACGTTCCAG) were similar, particularly in their right half-sites. To determine whether either repressor could bind the other's operator, we combined pActR with pO⁷²²³lux biosensor and pSCO7222 with pO^actlux biosensor. We observed strong repression of pO⁷²²³lux-dependent bioluminescence by pActR (to background) and partial repression of pO^actlux-dependent bioluminescence (1.5-fold) by pSCO7222 (Table 3).

Effects of various ligands on TetR, ActR, and the SCO7222 protein. To determine whether tetracycline could relieve repression by SCO7222 protein, we introduced plasmid pUAO1, bearing the tetO gene from Campylobacter jejuni (28), into each biosensor strain to protect them against the antibiotic. The TetO gene product is a ribosomal protection protein that allows relatively high cytoplasmic concentrations of the antibiotic to be tolerated without substantial damage to the cells. The addition of pUAO1 had no major impact, although it slowed bacterial growth to 1 doubling per hour.

To determine whether the resulting strains could detect tetracycline, we cultured each of them in the presence of various concentrations of the drug (Fig. 3A). As discussed previously, we observed the dose-responsive induction of bioluminescence in the TetR biosensor that peaked when 4 µg/ml tetracycline was used (24). Higher concentrations of drug inhibited cell growth, in spite of the presence of the tetO gene, but even at levels where there was significant inhibition of growth, we could still detect bioluminescence in the TetR biosensor. Consistent with previous results, tetracycline did not induce bioluminescence in the ActR biosensor (24). The presence of tetracycline also did not relieve repression by SCO7222 protein. Similar analysis was conducted using various concentrations of demeclocycline, doxycycline, oxytetracycline, chlortetracycline, and methatetracycline, all antibiotics related in structure and mechanism to tetracycline. While the concentrations required for half-maximal induction varied (Table 4), all of the tetracycline derivatives analyzed were good inducers of the TetR biosensor (data not shown). Again, none of these molecules had any effect on the ActR-based (24) or the SCO7222 protein-based strain (data not shown) biosensors. We then applied culture supernatants from Streptomyces aureofaciens (a tetracycline producer) and S. coelicolor to all three biosensors to determine whether either streptomycete produced secondary metabolites that could interact with the SCO7222 protein. As demonstrated previously, the S. aureofaciens and S. coelicolor supernatants were able to activate bioluminescence in the TetR- and ActR-based biosensors, respectively. Neither, however, had any effect on the SCO7222 protein-based biosensor (Fig. 3B). Under the growth conditions we have employed in this work, therefore, neither strain produces an SCO7222 protein ligand, although it is possible that there was some inducing molecule in the S. coelicolor supernatant that could not cross the E. coli envelope.

View larger version (20K): [in this window] [in a new window]

FIG. 3. Effect of purified tetracycline (A) or S. coelicolor supernatant (B) on TetR-, ActR-, and SCO7222 protein-mediated repression. (A) One to 5 µg of pure tetracycline was added to E. coli biosensors harboring pUOA1 and either pOtetlux, pOactlux, or pO7223lux along with the vector expressing the cognate repressors, respectively. (B) Ten to 40% (vol/vol) S. coelicolor supernatant was added to the same biosensor strains described above. All values were measured in relative light units (RLU), and error bars indicate ±1 standard deviation of values obtained from three independent readings.

View this table: [in this window] [in a new window]

TABLE 4. Induction of TetR-controlled gene expression by various tetracyclinesa

Mutagenesis of operator sequences. Unlike pCS26-Pac, which is a low-copy-number plasmid, pACYC184 is propagated at a moderately high copy number in E. coli: 20 copies per cell (4). As a result, the TetR-like transcription factors in our biosensors are probably in excess of their targets, potentially impairing small-molecule induction and explaining the observed ActR-SCO7222 protein cross talk. To provide a basis for comparing the in vivo interactions of the repressors and operators with their biochemical affinities, we constructed mutants of O_act and O₇₂₂₃ using two strategies. In the first, we subjected each operator sequence to randomization and screened for nonfunctional operators that exhibited bioluminescence in the presence of their cognate repressors (see Materials and Methods). We screened 1,600 colonies (800 for each repressor) and isolated 16 O_act sequences impaired in their interaction with ActR and 18 O₇₂₂₃ sequences impaired in their interaction with the SCO7222 protein (Table 5).

View this table: [in this window] [in a new window]

TABLE 5. Sequences of O7223 and Oact mutants not recognized by the SCO7223 protein and ActRa

Interestingly, all but one (pMTlux10) of the repressor-resistant alleles exhibited multiple mutations, suggesting either that the high in vivo concentration of the repressor resulted in DNA binding that was relatively resistant to mutation or that the affinity of each repressor for its cognate operator was simply very high due to a large number of specific repressor-operator interactions. In O_act, many of the alleles were altered at position 8 or 11, while in O₇₂₂₃, many alleles had mutations at position 5, 14, or 15 (positions 14 and 15 were two nucleotides more frequently involved in O₇₂₂₃ double mutants [Table 5]). In our second strategy, therefore, we subjected these base pairs to site-directed mutagenesis. We generated the alleles O_actA8G, O_actG11C, O₇₂₂₃A5G, O₇₂₂₃C14G, and O₇₂₂₃C15A and assessed each for repression by their cognate repressors. Position 5 of O_act (O_actA5G) was also tested since it is one of the palindromic nucleotides in the operator. As shown in Table 6, while the point mutation at position 8 of O_act had no effect, those at positions 5 and 11 eliminated in vivo repression by ActR. None of the point mutations in O₇₂₂₃ altered repression of bioluminescence in vivo.

View this table: [in this window] [in a new window]

TABLE 6. Summary of in vivo and in vitro assay results with the SCO7222 protein

We also assessed the effects of each point mutation on cross-repression by ActR and the SCO7222 protein and found that the results were quite different. For example, O₇₂₂₃A5G and O₇₂₂₃C15A, which had no effect on repression by the SCO7222 protein, eliminated cross-repression by ActR. Clearly, therefore, while the operator sequences are similar, the manners in which they are recognized by the two proteins in the context of our biosensors are subtly different.

Correlation of binding strengths with in vivo repression. To assess the significance of these results, we measured the affinities of the SCO7222 protein and ActR for O₇₂₂₃ and O_act and for the relevant point mutants described above. Using purified His₆-SCO7222 and His₆-ActR, we carried out gel mobility shift experiments with radioactively labeled probes corresponding to each operator sequence. The interaction of both proteins with their cognate operators was exceptionally strong (Fig. 4). Figure 4A shows gel mobility shift results and saturation curves for the SCO7222 protein. Consistent with the fact that the SCO7222 protein caused complete inhibition of transcription on O₇₂₂₃ but not on O_act in vivo, K_Ds for these interactions were 2 nM for O₇₂₂₃ and 30 nM for O_act. Even though the affinity of the SCO7222 protein for O_act was much lower than that for O₇₂₂₃, binding at higher concentrations explains the consistent weak repression this protein brought about on the O_act-regulated promoter in vivo (Table 3). Figure 4B shows mobility shift results for ActR. ActR could bind to O₇₂₂₃ as efficiently as O_act, in agreement with the capacity of this protein to repress both O_act- and O₇₂₂₃-regulated transcription in vivo (Table 3). Unlike the SCO7222 protein, ActR produced two shifted bands with both O_act and O₇₂₂₃, raising the possibility that two ActR dimers interact with each site. Protein titration and Hill plot analysis did not indicate significant cooperativity in the assembly of these complexes (data not shown).

View larger version (57K): [in this window] [in a new window]

FIG. 4. Characterization of interactions between ActR or SCO7222 protein and Oact or O7223. A 12.7-fmol amount of 32P-labeled probes was incubated with 1.5 to 210 nM protein at 30°C for 10 min and then separated in 12% nondenaturing polyacrylamide gels. Interactions shown on the gels are between the SCO7222 protein and O7223/Oact (A), ActR and O7223/Oact (B), and SCO7222 protein/ActR and O7223A5G (C). CP, protein-DNA complex. KDs were determined by drawing the saturation curves and obtaining [SCO7222 protein] at half-maximal saturation.

We went on to assess the effect of each point mutation in O_act and O₇₂₂₃ on the affinities of these DNAs for ActR and the SCO7222 protein and found a nearly perfect correlation between in vitro binding and in vivo repression (Fig. 4C and Table 6). For example, in E. coli, the SCO7222 protein repressed bioluminescence from the O₇₂₂₃A5G operator, but ActR did not. Consistent with this, the SCO7222 protein bound to O₇₂₂₃A5G with a similar affinity to O₇₂₂₃, while much higher concentrations of ActR were required to form a complex with O₇₂₂₃A5G compared to either O_act or O₇₂₂₃ (Fig. 4C). In summary, the K_Ds of the SCO7222 protein for the three mutant operators it was able to repress in vivo (O₇₂₂₃A5G, O₇₂₂₃C14G, and O₇₂₂₃C15A) were all in the 2 to 3 nM range, whereas its K_Ds for those operators it could not repress in vivo were at least five times higher, and in one case, O_actG11C, we observed no interaction at all.

This work raises interesting questions about possible cross talk between these two repressors. In our view, it is unlikely that the interaction of these proteins with heterologous operators is biologically meaningful. SCO7222 and -7223 are located far outside the act gene cluster, which contains most, if not all, of the genes necessary for actinorhodin biosynthesis and self-resistance. How might cross talk be avoided between these two and perhaps other members of this very large gene family in vivo? One possibility is that coupled transcription and translation, which is a characteristic of all bacteria, results in the direct delivery of a TetR protein from the ribosome on which it is synthesized to its cognate operator. This would be consistent with the fact that the repressor-encoding genes tend to be closely linked to their target genes. If, as is the case for TetR, ActR and the SCO7222 protein regulate their own production as well as that of their target resistance genes, they would limit their own intracellular accumulation and hence minimize or eliminate cross talk. Temporal and spatial regulations might also be responsible for restricting the cross talk.

More importantly for our immediate purposes, the high degree of correlation between in vitro binding and in vivo repression and the fact that we were able to specifically detect several tetracyclines as pure molecules and from a natural source suggest that this biosensor mechanism is a valuable tool for investigating TetR-like transcription factors.

 
We thank Michael Surette and Gerard Wright for supplying pCS26-Pac and for sharing various tetracyclines, respectively. We also thank Eric Brown, John Capone, and Gerard Wright for use of their laboratory equipment.

K.T. was supported by a postdoctoral fellowship from the Natural Science and Engineering Research Council. This work was funded by grant MOP-57684 from the Canadian Institutes for Health Research.

 
* Corresponding author. Mailing address: Department of Biochemistry and Biomedical Sciences, HSC 4H21, McMaster University, 1200 Main Street West, Hamilton, Ontario, Canada L8N 3Z5. Phone: (905) 525-9140, ext. 27335. Fax: (905) 522-9033. E-mail: nodwellj{at}mcmaster.ca

Published ahead of print on 20 July 2007.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

 

1
1. Bannam, T. L., and J. I. Rood. 1999. Identification of structural and functional domains of the tetracycline efflux protein TetA(P) from Clostridium perfringens. Microbiology 145:2947-2955.[Abstract/Free Full Text]
2
2. Bjarnason, J., C. M. Southward, and M. G. Surette. 2003. Genomic profiling of iron-responsive genes in Salmonella enterica serovar Typhimurium by high-throughput screening of a random promoter library. J. Bacteriol. 185:4973-4982.[Abstract/Free Full Text]
3
3. Caballero, J. L., F. Malpartida, and D. A. Hopwood. 1991. Transcriptional organization and regulation of an antibiotic export complex in the producing Streptomyces culture. Mol. Gen. Genet. 228:372-380.[Medline]
4
4. Chang, A. C. Y., and S. N. Cohen. 1978. Construction and characterization of amplifiable multicopy DNA cloning vehicles derived from the P15A cryptic miniplasmid. J. Bacteriol. 134:1141-1156.[Abstract/Free Full Text]
5
5. Cho, Y.-H., E.-J. Kim, H.-J. Chung, J.-H. Choi, K. F. Chater, B.-E. Ahn, J.-H. Shin, and J.-H. Roe. 2003. The pqrAB operon is responsible for paraquat resistance in Streptomyces coelicolor. J. Bacteriol. 185:6756-6763.[Abstract/Free Full Text]
6
6. Christen, S., A. Srinivas, P. Bahler, A. Zeller, D. Pridmore, C. Bieniossek, U. Baumann, and B. Erni. 2006. Regulation of the Dha operon of Lactococcus lactis: a deviation from the rule followed by the TetR family of transcription regulators. J. Biol. Chem. 281:23129-23137.[Abstract/Free Full Text]
7
7. Connell, S. R., C. A. Trieber, G. P. Dinos, E. Einfeldt, D. E. Taylor, and K. H. Nierhaus. 2003. Mechanism of Tet(O)-mediated tetracycline resistance. EMBO J. 22:945-953.[CrossRef][Medline]
8
8. Fernandez-Moreno, M. A., J. L. Caballero, D. A. Hopwood, and F. Malpartida. 1991. The act cluster contains regulatory and antibiotic export genes, direct targets for translational control by the bldA tRNA gene of Streptomyces. Cell 66:769-780.[CrossRef][Medline]
9
9. Fersht, A. 1985. Enzyme structure and mechanism, 2nd ed. W. H. Freeman and Company, New York, NY.
10
10. Folcher, M., R. P. Morris, G. Dale, K. Salah-Bey-Hocini, P. H. Viollier, and C. J. Thompson. 2001. A transcriptional regulator of a pristinamycin resistance gene in Streptomyces coelicolor. J. Biol. Chem. 276:1479-1485.[Abstract/Free Full Text]
11
11. Girard, G., S. Barends, S. Rigali, E. Tjeerd van Rij, B. J. J. Lugtenberg, and G. V. Bloemberg. 2006. Pip, a novel activator of phenazine biosynthesis in Pseudomonas chlororaphis PCL1391. J. Bacteriol. 188:8283-8293.[Abstract/Free Full Text]
12
12. Grkovic, S., M. H. Brown, N. J. Roberts, I. T. Paulsen, and R. A. Skurray. 1998. QacR is a repressor protein that regulates expression of the Staphylococcus aureus multidrug efflux pump QacA. J. Biol. Chem. 273:18665-18673.[Abstract/Free Full Text]
13
13. Hillen, W., and C. Berens. 1994. Mechanisms underlying expression of Tn10 encoded tetracycline resistance. Annu. Rev. Microbiol. 48:345-369.[Medline]
14
14. Kieser, T., M. J. Bibb, M. J. Buttner, K. F. Chater, and D. A. Hopwood. 2000. Practical Streptomyces genetics. The John Innes Foundation, Norwich, England.
15
15. Kisker, C., W. Hinrichs, K. Tovar, W. Hillen, and W. Saenger. 1995. The complex formed between Tet repressor and tetracycline-Mg²⁺ reveals mechanism of antibiotic resistance. J. Mol. Biol. 247:240-280.
16
16. McMurry, L., R. E. Petrucci, Jr., and S. B. Levy. 1980. Active efflux of tetracycline encoded by four genetically different tetracycline resistance determinants in Escherichia coli. Proc. Natl. Acad. Sci. USA 77:3974-3977.[Abstract/Free Full Text]
17
17. Natsume, R., Y. Ohnishi, T. Senda, and S. Horinouchi. 2004. Crystal structure of a gamma-butyrolactone autoregulator receptor protein in Streptomyces coelicolor A3(2). J. Mol. Biol. 336:409-419.[CrossRef][Medline]
18
18. Onaka, H., T. Nakagawa, and S. Horinouchi. 1998. Involvement of two A-factor receptor homologues in Streptomyces coelicolor A3(2) in the regulation of secondary metabolism and morphogenesis. Mol. Microbiol. 28:743-753.[CrossRef][Medline]
19
19. Orth, P., W. Saenger, and W. Hinrichs. 1999. Tetracycline-chelated Mg²⁺ ion initiates helix unwinding in Tet repressor induction. Biochemistry 38:191-198.[CrossRef][Medline]
20
20. Ramos, J. L., M. Martinez-Bueno, A. J. Molina-Henares, W. Terán, K. Watanabe, X. Zhang, M. T. Gallegos, R. Brennan, and R. Tobes. 2005. The TetR family of transcriptional repressors. Microbiol. Mol. Biol. Rev. 69:326-356.[Abstract/Free Full Text]
21
21. Rodriguez-Garcia, A., P. Combes, R. Perez-Redondo, M. C. Smith, and M. C. Smith. 2005. Natural and synthetic tetracycline-inducible promoters for use in the antibiotic-producing bacteria Streptomyces. Nucleic Acids Res. 33:e87.[Abstract/Free Full Text]
22
22. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
23
23. Sougakoff, W., B. Papadopoulou, P. Nordmann, and P. Courvalin. 1987. Nucleotide sequence and distribution of the gene tetO encoding tetracycline resistance in Campylobacter coli. FEMS Microbiol. Lett. 44:153-159.[CrossRef]
24
24. Tahlan, K., S. K. Ahn, A. Sing, T. D. Bodnaruk, A. R. Willems, A. R. Davidson, and J. R. Nodwell. 2007. Initiation of actinorhodin export in Streptomyces coelicolor. Mol. Microbiol. 63:951-961.[CrossRef][Medline]
25
25. Takahashi, M., L. Altschmied, and W. Hillen. 1986. Kinetic and equilibrium characterization of the Tet repressor-tetracycline complex by fluorescence measurements. Evidence for divalent metal ion requirement and energy transfer. J. Mol. Biol. 187:341-348.[CrossRef][Medline]
26
26. Takahashi, M., J. Degenkolb, and W. Hillen. 1991. Determination of the equilibrium association constant between Tet repressor and tetracycline at limiting Mg²⁺ concentrations: a generally applicable method for effector-dependent high-affinity complexes. Anal. Biochem. 199:197-202.[CrossRef][Medline]
27
27. Takano, E., R. Chakraburtty, T. Nihira, Y. Yamada, and M. J. Bibb. 2001. A complex role for the gamma-butyrolactone SCB1 in regulating antibiotic production in Streptomyces coelicolor A3(2). Mol. Microbiol. 41:1015-1028.[CrossRef][Medline]
28
28. Taylor, D. E., K. Hiratsuka, H. Ray, and E. K. Manavathu. 1987. Characterization and expression of a cloned tetracycline resistance determinant from Campylobacter jejuni plasmid pUA466. J. Bacteriol. 169:2984-2989.[Abstract/Free Full Text]
29
29. Winson, M. K., S. Swift, P. J. Hill, C. M. Sims, G. Griesmayr, B. W. Bycroft, P. Williams, and G. S. Stewart. 1998. Engineering the luxCDABE genes from Photorhabdus luminescens to provide a bioluminescent reporter for constitutive and promoter probe plasmids and mini-Tn5 constructs. FEMS Microbiol. Lett. 163:193-202.[CrossRef][Medline]
30
30. Yamaguchi, A., T. Udagawa, and T. Sawai. 1990. Transport of divalent cations with tetracycline as mediated by the transposon Tn10-encoded tetracycline resistance protein. J. Biol. Chem. 265:4809-4813.[Abstract/Free Full Text]
31
31. Yang, W., I. F. Moore, K. P. Koteva, D. C. Bareich, D. W. Hughes, and G. D. Wright. 2004. TetX is a flavin-dependent monooxygenase conferring resistance to tetracycline antibiotics. J. Biol. Chem. 279:52346-52352.[Abstract/Free Full Text]

---

Journal of Bacteriology, September 2007, p. 6655-6664, Vol. 189, No. 18
0021-9193/07/$08.00+0     doi:10.1128/JB.00717-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Ahn, S. K. Articles by Nodwell, J. Search for Related Content PubMed PubMed Citation Articles by Ahn, S. K. Articles by Nodwell, J.