Rapid Detection of Colicin E9-Induced DNA Damage Using Escherichia coli Cells Carrying SOS Promoter-lux Fusions

Plasmids, bacterial strains, and media. E. coli JM83 (ara [Δlac-proAB] rpsL φ80lacZΔM15) was used as the host strain for cloning and mutagenesis. E. coli BL21(DE3) or ER2566 (Novagen) was used as the host strain for the expression vector pET21a (Novagen), which has a strong IPTG-inducible T7 polymerase promoter and a C-terminal polyhistidine tag (His tag) to facilitate the purification of overexpressed proteins. E. coli DPD1718 contains a fusion of the E. coli recA promoter region to the Photorhabdus luminescens luxCDABE reporter integrated into the lacZ locus of E. coli DPD1692 (8). E. coli DPD2377 is E. coli DM800(pDEW634) that contains a plasmid-borne yigN fusion to the P. luminescens luxCDABE reporter (27). All cultures were routinely grown in Luria-Bertani (LB) broth or on plates of LB agar, supplemented where required with ampicillin (100 μg ml⁻¹) or chloramphenicol (30 μg ml⁻¹). Plasmid pNP69 encodes the colicin E9 structural gene (ceaI) and the Im9 immunity gene (ceiI), followed by a C-terminal His tag, under the control of an inducible T7 promoter in pET21a. In experiments in which the effects of other DNase immunity proteins were investigated, the plasmids were pTrc99A (Pharmacia Biotech) constructs containing the respective immunity proteins as previously described (16). For the synchronization and protection experiments, we used a disulfide-locked colicin E9 construct, BH29, encoded by plasmid pBH29. This protein contains two cysteine mutations in the ColE9 receptor-binding domain that, in the reduced protein, do not affect its colicin activity. Formation of a disulfide bond under oxidizing conditions leads to loss of biological activity, as previously described (22). Plasmid pAG1 is derived from pML261 and contains a 2.4-kb EcoRI-HindIII fragment that encodes the complete btuB gene in the vector pUC8 (15).

Protein purification.pET vectors (Novagen) encoding ColE9 and its mutants, along with a polyhistidine-tagged Im9, were transformed into E. coli BL21(DE3) cells. ColE9/Im9 complexes were purified as previously described by metal chelate chromatography (10) with a phosphate-buffered saline elution buffer containing 0.5 M NaCl and 1 M imidazole, pH 7.4. Protein concentrations were determined by absorbance at 280 nm.

Where necessary, the Im9 protein was removed from the ColE9/Im9 complex by diluting the complex 1:2 in 6 M guanidine HCl, 2 mM dithiothreitol (DTT), followed by preparative Sephacryl S-100 size exclusion chromatography as previously described (30). Free colicin E9 was then refolded by extensive dialysis against phosphate-buffered saline, pH 7.4, and stored. The BH29/Im9 complex was treated in a similar way but was refolded by dialysis against phosphate-buffered saline, pH 7.4, containing 1 mM DTT and stored in the presence of 10 mM DTT.

Diamide oxidation and DTT reduction.Protein samples were dialyzed overnight against phosphate-buffered saline, pH 7.4, to remove the DTT and were then incubated with 1 mM diamide (N,N,N′,N′-tetramethylazodicarbamide) for 30 min in the dark at room temperature before extensive dialysis against phosphate-buffered saline, pH 7.4. For reduction, the protein samples were incubated with 10 mM DTT for 1 h at ambient temperature, followed by extensive dialysis against degassed phosphate-buffered saline, pH 7.4. To avoid the possibility of spontaneous oxidation during certain assays, reduced protein was alkylated using iodoacetamide (50 mM final concentration), followed by extensive dialysis.

Reporter assay conditions.All assays were done at 37°C using a microtiter plate luminometer (Lucy 1; Anthos Labtech, Salzburg, Austria), and the media, plates, and luminometer were prewarmed to this temperature to minimize the induction of a stress response due to cooling effects. Reporter strains were plated out fortnightly from frozen stocks, and single colonies were picked for overnight cultures and grown in the presence of appropriate antibiotics. Overnight cultures were diluted 1:50 and grown for 3 hours at 37°C (OD, ∼0.35 to 0.4), after which they were diluted 1:2 (100-μl total volume) into black 96-well plates with an optical bottom (Nunc) and with ColE9 or mutant protein added. Unless stated otherwise, colicin-immunity protein complexes were used for these experiments. Induction of luminescence was followed over a period of 1.5 to 3 hours, with readings taken every 300 or 600 s. Cell density was recorded via OD₄₉₂ values.

In the experiments in which the effects of immunity proteins were investigated in the presence of IPTG, the cells were grown to an OD₆₀₀ of 0.2, 1 mM IPTG was added, and the cells were incubated for another hour at 37°C, after which they were diluted 1:2 in 96-well plates as described above. For protease protection experiments, oxidized BH29 (free protein) was added to DPD1718 (OD₆₀₀, ∼0.3) in a 96-well plate at a final concentration of 0.4 nM. After 30 min, cell killing was induced by adding 1 mM DTT, and trypsin was added at regular time intervals between 0 min and 20 min at a final concentration of 0.05 mg/ml.

Data analysis.Luminescence values are mostly presented as relative luminescence units (RLU). The ratio of RLU over OD values was taken for the calculation of the “gamma,” values that were used to generate the dose-response curve and to represent the amount of protection offered by the immunity proteins or trypsin treatment. The gamma value is defined as the luminescence induced for any given sample concentration minus the luminescence of the control cells at the same time point divided by the luminescence of the control cells at that time point: (L_sample − L_control)/L_control (8). We mostly chose an arbitrary time point of 50 min for the calculation of the gamma value unless stated otherwise. All assays were performed at least twice with three to six replicates for each condition. Error bars, where shown, represent the standard error of the mean for at least two independent experiments.

Development of a lux-based DNA damage assay for ColE9.Transcriptional profiling of E. coli cells treated with colicin E9 for 10 min revealed the up-regulation of 28 genes of the LexA-regulated SOS response (28). A reporter system linked to transcription of any of these 28 genes should be specific for the DNA damage induced by ColE9. We investigated two previously described lux-based reporter strains that had been developed to assay genotoxicity. E. coli DPD1718 contains a fusion of the E. coli recA promoter region to the Photorhabdus luminescens luxCDABE reporter integrated into the lacZ locus of E. coli DPD1692 (8), while E. coli DPD2377 is E. coli DM800(pDEW634) that contains a plasmid-borne yigN fusion to the P. luminescencs luxCDABE reporter (27). For the initial characterization of the reporter strains, we used the known SOS response inducers mitomycin C (a DNA-alkylating agent) and nalidixic acid (a DNA gyrase inhibitor), as previously described (8). The chromosomal fusion was chosen for further studies because it gave a lower background and wider range of activity than the plasmid-borne reporter (data not shown).

The data in Fig. 1 show the change in cell density compared to the change in luminescence with time after the addition of 3 nM ColE9 to E. coli DPD1718 cells. An increase in luminescence over the background can be detected as early as 15 min after addition of colicin E9, with a maximum reached after about 1 hour. By reducing the sampling time, we detected an increase in luminescence after 11 min (data not shown). Previous reports with similar genotoxicity reporter systems had suggested that up to 60 min was required to see a response to DNA damage induction (8). The observed result is a significant improvement in the sensitivity of detection of DNA damage caused by ColE9 compared to the conventional cell growth assays in which OD₆₀₀ is routinely followed for 4 hours. Moreover, the traditional cell-killing assays monitor cell death, which is the final consequence of the DNA damage induced by ColE9 and which can be affected by DNA repair and other factors, which cannot always be controlled.

• Open in new tab
• Download powerpoint

FIG. 1.

Comparison of luminescence induction and growth inhibition by 3 nM ColE9. The RLU values for E. coli DPD1718 with no addition (▪) and treated with 3 nM ColE9 (▴) and the OD values for E. coli DPD1718 with no addition (□) and treated with 3 nM ColE9 (▵) were plotted with time.

The mechanism by which ColE9 induces the SOS response and thus results in bioluminescence in E. coli DPD1718 cells requires the generation of single-stranded DNA (ssDNA), which can be created by processing of DNA damage and stalled replication, and perhaps by other means (18). The ssDNA acts as a signal that activates a dormant coprotease activity of RecA to facilitate the proteolytic self-cleavage of the LexA repressor, thus inducing the LexA regulon. We have shown previously that the ColE9 DNase has a preference for cleaving DNA after Ts and that runs of Ts are readily cut by the enzyme but runs of As are not (23). The presence of runs of Ts in the E. coli genome may be the mechanism for generating the necessary ssDNA to activate the SOS response.

To characterize the detection limit and the linear range of our assay, we incubated E. coli DPD1718 with a range of ColE9/Im9 complex concentrations from 2 nM to 2 pM (final concentration) in serial twofold dilutions (Fig. 2, top). At lower concentrations, the bioluminescence increased with time, while at higher ColE9 concentrations, the maximum RLU values were much higher but then rapidly declined. We assume that this is due to an effect on the generation of bioluminescence associated with the overwhelming of DNA repair systems and finally the loss of viability, and probably even cell lysis, caused by high concentrations of ColE9. Gamma values calculated at 50 minutes were used to generate the dose-response curve shown in Fig. 2, bottom. The results showed that the assay is sensitive down to concentrations of 2 pM ColE9. In contrast, the spot test using purified ColE9/Im9 complex showed killing activity only to 1 nM (data not shown). Luminescence induction was linear over a concentration range from 0 to 0.5 nM (r² = 0.99) (Fig. 2, bottom). Higher colicin concentrations resulted in a response plateau, presumably because of the cell-killing action of ColE9. We have occasionally used colicin E9 concentrations outside the linear range in order to relate the results back to our conventional cell-killing assays (see below).

• Open in new tab
• Download powerpoint

FIG. 2.

Dose-response curve of luminescence induction by ColE9. (Top) Luminescences of cultures of E. coli DPD1718 with no addition (⧫) and treated with 2.0 nM (▪), 1.0 nM (▴), 0.50 nM (•), 0.25 nM (*), 0.13 nM (×), 0.063 nM (+), 0.031 nM (□), 0.016 nM (⋄), 0.0080 nM (▵), 0.0040 nM (○), and 0.0020 nM (-) ColE9. The graph on the bottom shows the linearity of the response expressed as a gamma value up to 0.50 nM of ColE9. The amount of luminescence induced (gamma) was calculated as described in Materials and Methods.

We investigated the specificity of luminescence induction by ColE9 by using two biologically inactive (devoid of cell-killing activity) mutants of ColE9. The H575A mutant is an active-site mutant of ColE9 that lacks DNase activity but still binds the outer membrane receptor BtuB and interacts with TolB (11, 29). The D35A TolB box mutant of ColE9 is unable to interact with TolB but still has BtuB-binding and DNase activities comparable to those of wild-type ColE9 (10). Both of these mutant proteins were unable to induce a bioluminescence response above the background in our reporter assay, even when used at a 10-fold-higher concentration (40 nM) than the wild-type ColE9 (4 nM). Together, these results indicate that the luminescence response is a consequence of ColE9 DNase activity in the cytoplasm of susceptible cells.

We isolated E. coli MV1710, a btuB mutant of the E. coli DPD1718 strain, by spotting 20-μl aliquots of 40 nM ColE9 onto an overlay of DPD1718 and picking colonies growing within the resulting zone of lysis. Transformation of this mutant with plasmid pAG1, which encodes BtuB, restored sensitivity to ColE9 (data not shown). Loss of a functional BtuB receptor in the outer membrane resulted in loss of the luminescence response in E. coli MV1710, even when 10-fold-higher concentrations of ColE9 were used (Fig. 3, bottom). When BtuB expression was restored via the pAG1 construct, lux induction was restored in the mutant reporter strain. The faster response of these cells may be related to the higher copy number of the pAG1-encoded btuB gene product in the outer membrane. It was shown previously that the amount of BtuB in the outer membrane of pAG1-bearing cells comprises about 20 to 40% of the outer membrane proteins (12). The lower overall luminescence induction may indicate an imbalance between the amounts of BtuB receptor and proteins encoded by the tol operon that are required for translocation of ColE9 and/or a faster cell-killing action by ColE9 under these conditions. The shift to the left in the timing of the peak of induction perhaps suggests that the latter explanation is more likely. A 12-fold-higher background luminescence was observed in the pAG1-containing cells than in E. coli DPD1718 and E. coli MV1710 cells without pAG1.

• Open in new tab
• Download powerpoint

FIG. 3.

Controls for the specificity of the luminescence induction of E. coli DPD1718. (Top) Luminescence induced by 4.0 nM ColE9 (⧫), 40 nM of the active-site ColE9 mutant H575A (▵), 40 nM of D35A TolB box ColE9 mutant (×), and no addition (⋄). For clarity, the data for the ColE9 D35A mutant protein and the no-addition cultures are not shown, as they were identical to that of the ColE9 H575A mutant. (Bottom) Absence of luminescence induction in E. coli MV1710 and restoration of the luminescence response when this strain was transformed with a plasmid encoding BtuB (pAG1). E. coli DPD1718 treated with 4.0 nM ColE9 (⧫), E. coli MV1710 treated with 4 nM (○) and 40 nM (*) ColE9, and E. coli MV1710 transformed with pAG1 with no addition (▵) and treated with 4 nM (▴) and 40 nM (×) ColE9.

Applications of the assay. (i) Protection by various immunity proteins—relation with K_d.A recent paper by Li et al. (16) compared the degrees of protection of bacteria against ColE9 activity by a set of immunity proteins (expressed intracellularly) whose affinities for the ColE9 DNase covered a range of 10 orders of magnitude. They used a spot test, a traditional cell-killing assay in which a series of decreasing concentrations of ColE9 are spotted onto a lawn of sensitive cells, and analyzed the numbers of surviving cells as a function of the affinities of the different immunity proteins for ColE9.

Since our assay is designed to investigate the DNA-damaging activities of DNase colicins in a more specific and sensitive way, we used it to look at a limited selection of these immunity proteins (K_d values ranging from 10⁻¹⁴ [Im9] to 10⁻⁶ [Im8] and 10⁻⁴ [Im7]) compared to the empty control vector pTrc99A. We used two concentrations of ColE9: 0.4 nM, within the linear range of our assay, and 4 nM, a lethal concentration used in the conventional assays. Aliquots of all immunity-expressing E. coli DPD1718 cells, before and after IPTG induction, were run on 16% sodium dodecyl sulfate-polyacrylamide gel electrophoresis to check for protein expression. Similar amounts of the immunity proteins were produced after 1 hour of IPTG treatment (data not shown). The luminescence results are presented in Fig. 4. The full protection offered by Im9 (K_d, 10⁻¹⁴) confirmed that the luminescence induction in our assay is a specific effect of the DNase activity of ColE9. Our data broadly support the analysis of Li et al. (16), which indicated that immunity proteins with a K_d of >10⁻⁶ offer no protection and that the K_d required for complete biological protection is <10⁻¹⁰ M (Fig. 4).

• Open in new tab
• Download powerpoint

FIG. 4.

Protection from DNA damage by intracellular expression of cognate (Im9) and noncognate (Im7 and Im8) immunity proteins. The luminescence induced by 0.4 nM and 4 nM ColE9 (expressed as a gamma value as described in Materials and Methods) in the presence of Im7, Im8, Im9, or no immunity proteins is shown. The solid bars are the mean gamma values of cultures in the absence of IPTG, and the open bars represent the mean effects of protein expression induced by 1 mM IPTG for 1 h. The error bars represent the standard errors of the mean for at least three independent experiments.

(ii) Synchronized cell death using a “disulfide-locked” colicin E9 protein.We recently presented data showing that a mutant ColE9 protein (BH29) with an engineered disulfide bond in its receptor-binding domain (via two strategically positioned cysteine mutations) was inactive in the oxidized form and that activity was restored when it was reduced through the addition of DTT (22). The current reporter assay, because of its faster response time and sensitivity, enabled us to probe the early events after DTT addition. The oxidized free protein (BH29) was incubated with cells for 45 min, at which time 1 mM DTT was added to certain wells. Reduced and alkylated free BH29 served as the positive control. Figure 5 shows resumption of colicin activity, as seen by a significant increase in luminescence above the background level of the oxidized protein a few minutes after addition of DTT that continued at a rate similar to that of the reduced and alkylated protein. The interval from the time of addition of reduced and alkylated BH29 until the 50% increase in the resulting luminescence was significantly longer than that seen after DTT addition to cells incubated with oxidized BH29, even though the latter must include the additional time taken for DTT to reduce BH29 before the cell-killing process can begin. The small increase in luminescence of the oxidized protein over the extended assay period is likely to be due to incomplete oxidation during preparation of the diamide-treated BH29 protein. Very similar findings were obtained when E. coli cells were preincubated with the oxidized protein for a fixed time, followed by removal of unbound protein by spinning and washing the cells in LB broth and addition of 1 mM DTT at the same time point as described above (data not shown). This suggests that DTT reduction of BH29 protein that is already bound to BtuB initiates the DNA damage and thus that some of the early events of receptor binding and even translocation have already occurred during incubation of E. coli cells with oxidized BH29.

• Open in new tab
• Download powerpoint

FIG. 5.

Effect of 1 mM DTT on luminescence induction by BH29. The luminescence induced in E. coli DPD1718 cells incubated with 0.4 nM reduced and alkylated BH29 (▪) or diamide-oxidized BH29 (▴) is shown with time after addition. One millimolar DTT was added after 46 minutes to some of the oxidized BH29-containing wells (×). The luminescence of E. coli DPD1718 cells with no addition (⧫) is shown.

(iii) Timing of cell death using protease protection.Limited digestion of ColE9 with trypsin under controlled conditions abolishes biological activity and results in three main cleavages, one at the N terminus of the translocation domain, one at the start of the receptor-binding domain, and one at the border of the receptor-binding and DNase domains (31). Trypsin was used to investigate how long after the addition of DTT E. coli cells incubated with oxidized BH29 could be rescued from colicin action (Fig. 6, top). Gamma values calculated at 60 min show that the amount of protection offered by trypsin activity decreased dramatically with increasing time lag between adding DTT (approximate time of initiation of colicin uptake) and trypsin (Fig. 6, bottom). The data show that after 3.5 min, only 50% of the cells can be protected by trypsin, and after 10 min, the cells can no longer be rescued from colicin action. The amount of trypsin (0.05 mg/ml) used for these experiments did not affect the growth of the reporter strain (data not shown).

• Open in new tab
• Download powerpoint

FIG. 6.

Trypsin protection of DNA damage induced by ColE9. (Top) E. coli DPD1718 cells were incubated with oxidized BH29 for 30 minutes, at which time 1 mM DTT was added to initiate cell killing. At the same time, 0.05 mg/ml trypsin was added at 2.5-min intervals (over a period of 20 minutes) to certain wells, and protection against cell killing was monitored by the reduction in luminescence induction. E. coli DPD1718 cells with no addition (⧫) and treated with 0.4 nM oxidized BH29 (▪); addition of 1 mM DTT to oxidized BH29 (▴); and addition of 1 mM DTT and 0.05 mg/ml trypsin at time zero (○), after 2.5 min (▵), after 5 min (⋄), after 7.5 min (□), after 10 min (×), after 15 min (*), and after 20 min (+). (Bottom) Decrease in protection offered by trypsin added at various times after the addition of DTT.