Constructing of Bacillus subtilis-Based Lux-Biosensors with the Use of Stress-Inducible Promoters

Here, we present a new lux-biosensor based on Bacillus subtilis for detecting of DNA-tropic and oxidative stress-causing agents. Hybrid plasmids pNK-DinC, pNK-AlkA, and pNK-MrgA have been constructed, in which the Photorhabdus luminescens reporter genes luxABCDE are transcribed from the stress-inducible promoters of B. subtilis: the SOS promoter PdinC, the methylation-specific response promoter PalkA, and the oxidative stress promoter PmrgA. The luminescence of B. subtilis-based biosensors specifically increases in response to the appearance in the environment of such common toxicants as mitomycin C, methyl methanesulfonate, and H2O2. Comparison with Escherichia coli-based lux-biosensors, where the promoters PdinI, PalkA, and Pdps were used, showed generally similar characteristics. However, for B. subtilis PdinC, a higher response amplitude was observed, and for B. subtilis PalkA, on the contrary, both the amplitude and the range of detectable toxicant concentrations were decreased. B. subtilis PdinC and B. subtilis PmrgA showed increased sensitivity to the genotoxic effects of the 2,2′-bis(bicyclo [2.2.1] heptane) compound, which is a promising propellant, compared to E. coli-based lux-biosensors. The obtained biosensors are applicable for detection of toxicants introduced into soil. Such bacillary biosensors can be used to study the differences in the mechanisms of toxicity against Gram-positive and Gram-negative bacteria.


Introduction
For testing chemical impurities (toxins) and other biologically active substances, primarily medications, luminescent biosensor cells (lux-biosensors) are currently used in two ways: (1) based on bioluminescence quenching [1][2][3], and (2) based on bioluminescence induction [4][5][6][7][8][9][10][11][12][13]. The existing lux-biosensors based on genetically engineered Escherichia coli cells and a number of naturally luminescent enterobacteria are of little use for working with soil pollution [14]. The design of lux-biosensors based on Gram-positive Bacillus subtilis bacteria could contribute to the analysis of the mechanisms of toxicity of medicinal products that specifically interact with Gram-positive bacteria, or the ecological monitoring of soil contamination.
The construction of a lux operon that is efficiently expressed in Gram-positive bacteria is complicated due to the need to replace the Shine-Dalgarno sequence upstream of each reading frame. Previously, such works were carried out and luminescent bacteria of Streptococcus, Staphylococcus, Bacillus, and a number of other genera were obtained, which were used mainly for medical applications [2,15,16]. The luminescence of these bacteria was constitutive, and it was used to determine the integral effect of biologically active substances on bacterial cells.
Promoters with the highest response amplitude are considered more promising for the development of lux-biosensors. According to works [17,18] in the SOS-regulon of B. subtilis, the dinC gene promoter has the highest response amplitude-36-175 folds. The activation of DNA glycosylase promoters in response to DNA alkylation in B. subtilis cells is not so high and for PalkA it is only 2.5 times, which was determined by the changes in the released 7-methylguanine after treatment with nitrosoguanidine [19]. E. coli has two regulons specifically activated by the superoxide anion radical (SoxR/S) and hydrogen peroxide (OxyR/S) [20][21][22]. There is not such a clear division of the response under oxidative stress in B. subtilis; genes that respond to oxidative stress can be activated by both hydrogen peroxide and paraquat [23]. The greatest increase in transcription is observed for promoters of katA and mrgA-the genes encoding catalase and ferritin-like DNA-binding protein, respectively [23][24][25][26].
In this work, we present whole-cell lux-biosensors based on B. subtilis cells, constructed using stress-inducible promoters: PdinC, PalkA, and PmrgA.

Characterization of the B. subtilis 168 pNK-DinC Lux-Biosensor for SOS-Response Detection
The characteristics of B. subtilis 168 pNK-DinC were investigated using the antibiotic mitomycin C (MitC), a well-known drug inducing the SOS-response in E. coli cells [27]. The cells were preliminarily grown to OD = 0.4 at 37 • C and divided into 200 µL portions. After MitC was added to final concentrations of 10 µM, 1 µM, 100 nM, and 10 nM, cells were incubated at room temperature for 3 h with periodic luminescence measurements. The kinetic curves on Figure 1 are the results of individual experiments, results of which are representative and consistent with the other experiments.
The data shown in Figure 1 indicate that the designed B. subtilis 168 pNK-DinC luxbiosensor is induced by MitC and, therefore, is sensitive to DNA damage. The threshold concentration of MitC is about 10 nM, which generally corresponds to the sensitivity to MitC of the best E. coli-based lux-biosensors for determining the SOS response, in which the PcolD promoter of the cda gene from the conjugative plasmid pColD-CA23 [22] or the PdinI promoter [28] are fused with luxCDABE. The time of the onset of biosensor activation is about 1 h with a maximum activation after 3 h. It is approximately two times slower than that for E. coli-based biosensor cells. The maximum response amplitude exceeds one order of magnitude and is 40-fold for 10 µM MitC.

Characterization of the B. subtilis 168 pNK-MrgA Lux-Biosensor for Oxidative Stress Detection
The functionality of B. subtilis 168 pNK-MrgA was tested by adding hydrogen peroxide to the biosensor cell culture to final concentrations of 10 mM, 1 mM, and 100 µM (Figure 2A). In E. coli cells, the homologue of the mrgA gene is dps [29]; therefore, we compared the characteristics of these biosensors. We used the E. coli MG1655 pDps biosensor with the same H 2 O 2 concentrations ( Figure 2B). The B. subtilis and E. coli cell cultures were preliminarily grown at 37 • C to OD = 0.4 and 0.1, respectively, and divided into portions, which were supplemented with H 2 O 2 in different concentrations.  From the data in Figure 2A, it can be seen that the PmrgA promoter in B. subtilis cells is inducible by H 2 O 2 . Biosensor induction begins 1 h after H 2 O 2 addition and reaches its maximum amplitude in 2.5 h. The threshold concentration of H 2 O 2 for this biosensor is 100 µM. The maximum response amplitude exceeds an order of magnitude and is 14 times for an H 2 O 2 concentration of 10 mM. The biosensor E. coli MG1655 pDps is characterized by high sensitivity to H 2 O 2 and the rather pronounced toxic effect of it at high concentrations (10 mM and 1 mM of H 2 O 2 ); immediately after the addition of the toxicant ("0" point at time axis), the biosensor's luminescence drops by two orders of magnitude and three times, respectively ( Figure 2B).
B. subtilis 168 pNK-MrgA shows about an order of magnitude reduced sensitivity to H 2 O 2 compared with E. coli MG1655 pDps ( Figure 2B). However, the width of working concentrations range for both biosensors is the same and is more than two orders of magnitude: from 100 µM to 10 mM for B. subtilis 168 pNK-MrgA and from 10 µM to 1 mM for E. coli MG1655 pDps. The response time of the E. coli-based biosensor is shorter-its induction begins in approximately 15 min and reaches the maximum in around 2.5 h. Thus, the optimal measurement time for both biosensors is about 2.5-3 h. The amplitude of the response for biosensors B. subtilis 168 pNK-MrgA and E. coli MG1655 pDps is approximately the same-10-15 times.

Characterization of the B. subtilis 168 pNK-AlkA Lux-Biosensor for DNA Alkylation Detection
The characteristics of the B. subtilis 168 pNK-AlkA biosensor were studied using the known alkylating agent methyl methanesulfonate (MMS) (Figure 3), which activates the alkA gene promoter in E. coli [30]. MMS was used at final concentrations of 1 mM, 100 µM, and 10 µM-the measurement time was prolonged to 4.5 h to obtain a higher response amplitude. As can be seen from the data in Figure 3, B. subtilis 168 pNK-AlkA is inducible by MMS. The threshold concentration is about 100 µM. The response time of the biosensor is about 2 h, with a maximum activation after 4 h. The maximum response amplitude does not exceed an order of magnitude and is 6 times for the MMS concentration of 1 mM. The B. subtilis 168 pNK-AlkA has the same sensitivity as the E. coli-based biosensor with PalkA transcriptionally fused with the β-galactosidase gene according to data from [30]. However, lux-biosensors that use luciferase as a reporter are usually more sensitive. This could be illustrated by comparing the sensitivity of biosensors from [11,31] and [30] to MNNG (N-methyl-N'-nitro-N-nitrosoguanidine)-threshold concentrations of MNNG were 10 nM for the lux-biosensor and 100 nM for the β-galactosidase-based biosensor.

Main Characteristics of the Obtained Biosensors
Data on the main characteristics of the obtained biosensors, namely, sensitivity, induction amplitude, dynamic range (working concentrations range) and response time are represented in Table 1.
One can see from Table 1  For biosensors E. coli pDinI and B. subtilis pNK-DinC, the cross-induction by non-specific toxicants was observed. These biosensors are also inducible by H 2 O 2 and MMS at high concentrations. This is quite expected because large-scale oxidative damage can stop the replication fork and induce the SOS response [31]. Multiple alkylation of nucleotide bases with drugs such as MMS can also lead to the appearance of extended single-stranded regions during repair and induce the SOS response [9]. According to the data obtained, the threshold concentration of BBH for B. subtilis 168 pNK-DinC is 0.1 g/L. For the E. coli MG1655 pDinI biosensor-1 g/L, the same as that for E. coli MG1655 pColD-lux biosensor, according the work [32]. The maximum response amplitude for B. subtilis 168 pNK-DinC was 50-fold, while for E. coli pDinI, it was only 4 ( Figure 4C). The time of the beginning of biosensors induction by BBH is approximately the same for E. coli and B. subtilis-about 1.5 h.
A pair of biosensors, B. subtilis 168 pNK-MrgA and E. coli pDps, was used to determine the ability of BBH to induce oxidative stress ( Figure 5).
The threshold BBH concentration for B. subtilis 168 pNK-MrgA is between 0.1 and 1 g/L ( Figure 5A). This is a higher sensitivity compared to E. coli pDps (1 g/L) ( Figure 5B) and the E. coli pOxyR-lux biosensors (specific to H 2 O 2 , according to [32], its sensitivity to BBH is 1 g/L), and it is almost equal to the sensitivity of the E. coli pSoxS-lux biosensor, which is specific to the superoxide anion radical. The maximum response amplitude for both B. subtilis 168 pNK-MrgA and E. coli pDps was about 10 times. The initiation time of biosensor induction is approximately the same-about 1 h.    subtilis 168 pNK-DinC biosensor cell cultures after adding MitC directly to the medium, soil, and soil supplemented with MitC. "Ctrl"-biosensor cells without the addition of toxicant; "MitC"-100 nM MitC added; "sand"-biosensor cells with sandy soil added; "peat"-peat soil; "ps"-sandy-peat soil; "s + Mit"-biosensor cells with soil, which was preliminary supplemented with MitC (final concentration of MitC in medium with biosensor cells and soil-mix was 100 nM); "p + Mit"-peat with MitC; and "ps + Mit"-peat-sand soil with MitC.
As can be seen from the data shown in Figure 6, the studied soils themselves do not possess genotoxic properties. The addition of sandy soil leads to an approximately 2-fold decrease in the response amplitude of the biosensor to MitC. Soils containing peat decrease the response amplitude by about 10 times. Such a decrease in the response can be explained by the partial shielding of luminescence by dark soil particles and partial binding of the toxicant. Soil acidity also plays a significant role, while peat bogs in Karelia are characterized by low pH [33], which could significantly decrease the characteristics of E. coli-based biosensors [14].

Discussion
Whole-cell biosensors are widely used for assessing compounds' toxicity and ecological control [34,35]. Gram-positive and Gram-negative bacteria can have different sensitivities to the same toxicant. Here, we present a new lux-biosensor based on B. subtilis for detecting DNA-damaging and oxidative-stress-causing agents. Characteristics of the obtained biosensor strains B. subtilis 168 pNK-DinC, B. subtilis 168 pNK-AlkA, and B. subtilis 168 pNK-MrgA were determined in experiments with the addition of standard toxicants, causing the SOS response, DNA alkylation, and oxidative stress, respectively (Figures 1-3). B. subtilis-based biosensors showed comparable characteristics with biosensors based on E. coli cells. New Gram-positive biosensors complement the bacterial test-systems, which consist of well-proven E. coli-based lux-biosensors [4,11,22,36].
It is known from the works [32,37] that strained hydrocarbon compounds are capable of causing DNA damage and oxidative stress, but not DNA alkylation in living cells. In E. coli cells, there are no homologues to the dinC gene of B. subtilis; therefore, the luxbiosensor E. coli pDinI, related to the SOS regulon and having a relatively high induction amplitude [28], was taken for comparison. Studies of the toxic properties of BBH have shown (Figures 4 and 5) that the biosensors B. subtilis 168 pNK-DinC and B. subtilis 168 pNK-MrgA have a higher sensitivity and response amplitude compared to analogous biosensors based on E. coli cells. This can be explained by the difference in permeability of the cell wall of Gram-positive and Gram-negative cells by compounds of the BBH type. It should be noted that there is no homologue of the E. coli SoxS/SoxR type regulation system in B. subtilis, which specifically reacts to the appearance of the superoxide anion radical [23]. This may increase the toxicity of BBH to B. subtilis compared to E. coli and serve as another explanation for the effect obtained. Consequently, the sensitivity of the mrgA gene promoter appears to be higher than that of Pdps ( Figure 5).
Previously, it has been shown that E. coli-based biosensors are not suitable for testing acidic peaty soils due to a catastrophic decrease in the ability of cells to luminescence and activation of stress promoters [14]. In contrast, the B. subtilis 168 pNK-DinC biosensor strain can be used for direct studies of the genotoxicity of soil samples adjusted to light shielding by soil particles (Figure 6).

Strains and Plasmids
Bacterial strains and plasmids used in the current work are presented in Table 2. E. coli MC1061 cells were used for constructing biosensor plasmids for B. subtilis cells. E. coli MG1655 cells were used for transformation by pDps and pDinI plasmids and obtaining E. coli-based biosensors. Table 2. Plasmids and bacterial strains used in the study.

Name Description Source
Bacterial strains The E. coli Pdps promoter was cloned into pDEW201 [5] vector and transcriptionally fused to the reporter genes luxCDABE P. luminescens. Ap r [39] pDinI The E. coli PdinI promoter was cloned into pDEW201 [5] vector and transcriptionally fused to the reporter genes luxCDABE P. luminescens. Ap r [28]

Enzymes and DNA Manipulation
Plasmid DNA was isolated by the GeneJET Plasmid Miniprep Kit (ThermoFisher Scientific, Waltham, MA, USA). The E. coli cell transformation with hybrid plasmids, agarose gel electrophoresis, and isolation of plasmid and total DNA was performed according to [40]. Restriction was carried out using the SacI restriction enzyme (Promega, Madison, WI, USA). Ligation was conducted with the use of Gibson Assembly, prepared according to [41]. B. subtilis cells were transformed according to [42,43].

Constructing of Biosensor Plasmids
The vector plasmid pPL_ABCDExen (see the plasmid map in Figure S1) was linearized by the SacI restriction enzyme. PmrgA, PalkA, and PdinC promoter regions were amplified by PCR on a matrix of B. subtilis 168 genome DNA. Primers, which were used for amplification of promoters and sequencing of insertion in the pPL_ABCDExen vector, are given in Table S1 (Supplementary Materials). Ligation of promoter-containing DNA fragments with linearized vector was conducted with the use of Gibson Assembly. Resulting plasmids, containing promoters of B. subtilis mrgA, alkA, and dinC genes, were named pNK-MrgA, pNK-AlkA, and pNK-DinC, correspondently. For initial transformation and plasmid DNA isolation, E. coli MC1061 was used. For sequencing of inserts in the resulting biosensor plasmids, promrev and promdir primers were used (see Table S1). Sequences of promoter regions inserted to obtain pNK-MrgA, pNK-AlkA, and pNK-DinC are given in the supplementary file. Isolated from E. coli cells, pNK-MrgA, pNK-AlkA, and pNK-DinC plasmids were used for transformation of B. subtilis 168 cells for obtaining biosensor strains.

Culture Medium and Growth Conditions
E. coli cell cultures were grown in Lysogeny Broth (LB); B. subtilis cells were grown at 37 • C in Brain Heart Infusion (BHI) supplemented with 50 mg/L tryptophan. When measuring luminescence, B. subtilis cells were grown with an additional 1% glycerol in the medium. The LB medium was composed of 1% tryptone, 0.5% yeast extract, and 1% NaCl. For obtaining solid media, agar was added to a final concentration of 1.5%. The media were supplemented with 100 µg/mL ampicillin or 10 µg/mL chloramphenicol. For culturing of B. subtilis-based biosensor strains, preferably, chloramphenicol was used.
For experiments on determining the sensitivity of biosensors, overnight cultures were used to inoculate liquid LB or BHI to an optical density (OD) of 0.01; the resulting cultures were grown with continuous agitation at 37 • C to the final OD of approximately 0.1 for E. coli and 0.4 for B. subtilis. The OD of cell suspensions was measured with a KFK-3 photometer (ZOMP, Moscow, Russia). Bioluminescence intensity of 200 µL portions of cell culture was measured in the 96well plates using the SynergyHT (Biotek Instruments, Winooski, VT, USA) equipment, or in capless microtubes using Biotox-7BM (BioPhysTech, Moscow, Russia), which is 100 times more sensitive. Luminescence values were expressed in relative light units (RLU), specific to each luminometer. Cell culture portions were incubated at room temperature for 3 to 5 h with periodical bioluminescence measurements. Longevity of incubation with measurements was determined by the average response time for every lux-biosensor individually.

Data Processing
All experiments were conducted in 5 replications. Kinetic curves on graphs were the results of individual experiments, results of which are consistent with the others in a row. For calculations of the main characteristics of biosensors' maximum induction coefficients, induction start time, etc., average values and standard deviations were calculated for five independent replications. The biosensor induction factor was calculated by dividing the luminescence of induced cells by the luminescence of control cells.

Conclusions
In general, results of the investigations demonstrate the operability of the constructed B. subtilis-based biosensors (Figure 7). Tests with standard toxicants causing the SOS response, DNA alkylation and oxidative stress showed comparable characteristics of biosensors based on B. subtilis cells with those based on E. coli cells. An important advantage of B. subtilis-based biosensors is their applicability for direct measurements of the toxicity of polluted soils. Lux-biosensors constructed in this work can complement the E. coli-based test system created earlier. The combination of E. coli-based biosensors with newly constructed B. subtilis ones can be used to study the mechanisms of toxicity of medicines and pollutants that could have different effects on Gram-positive and Gram-negative bacteria.

Conflicts of Interest:
The authors declare no conflict of interest.