An auto-inducible phosphate-controlled expression system of Bacillus licheniformis

Background A promoter that drives high-level, long-term expression of the target gene under substrate limited growth conditions in the absence of an artificial inducer would facilitate the efficient production of heterologous proteins at low cost. A novel phosphate-regulated expression system was constructed using the promoter of the phytase encoding gene phyL from Bacillus licheniformis for the overexpression of proteins in this industrially relevant host. Results It is shown that the phyL promoter enables a strong overexpression of the heterologous genes amyE and xynA in B. licheniformis when cells were subjected to phosphate limitation. Whether B. licheniformis can use phytate as an alternative phosphate source and how this substrate influences the PphyL controlled gene expression under growth conditions with limited inorganic phosphate concentrations were also investigated. It is shown that B. licheniformis cells are able to use sodium phytate as alternative phosphate source. The addition of small amounts of sodium phytate (≤ 5 mM) to the growth medium resulted in a strong induction and overexpression of both model genes in B. licheniformis cells under phosphate limited growth conditions. Conclusions The PphyL controlled expression of the investigated heterologous genes in B. licheniformis is strongly auto-induced under phosphate limited conditions. The proposed PphyL expression system enables an overexpression of target genes in B. licheniformis under growth conditions, which can be easily performed in a fed-batch fermentation process. Electronic supplementary material The online version of this article (10.1186/s12896-018-0490-6) contains supplementary material, which is available to authorized users.


Background
Bacillus licheniformis is a saprophytic bacterium that is generally recognized as safe (GRAS) by the U.S. Food and Drug Administration [1]. The ability to produce and secrete high amounts of proteins into the extracellular medium (20-25 g/L) makes this bacterium to one of the most important industrial hosts for the large-scale production of industrial enzymes, such as amylases, proteases, phytases, and other specialty enzymes [2]. Another advantage of B. licheniformis is its ability to grow rapidly in simple media to high-cell-densities, which is favourable for an industrial-scale production.
The expression systems used in B. licheniformis were mostly developed for B. subtilis. At present, three types of expression systems that contain (i) constitutive promoters, (ii) inducer-specific promoters and (iii) auto-inducible promoters have been used for high-level production of heterologous proteins in Bacillus subtilis (e.g., [3][4][5]). Among them, expression systems containing inducer-specific promoters (e.g., Pspac and Pxyl) are the most widely used type. However, the requirement for specific inducers, such as IPTG or xylose, increases the cost of their large-scale application [6,7]. Constitutive expression systems allow for continuous transcription of their target gene, and thus, are not suitable for the production of potential toxic proteins. In contrast, auto-inducible expression systems that require no specific inducers are ideal for the industrial production of heterologous proteins at low cost. These expression systems are induced by a variety of environmental factors, which can be easily simulated in industrial fermentation processes [8]. For example, nutrient limitation, e.g.
glucose, is such a suitable signal for the induction of an auto-inducible promoter system [9]. Furthermore, an expression system using the phosphate starvation inducible pst promoter has been developed for B. subtilis [10]. However, a comparable auto-inducible promoter system has so far not been shown for B. licheniformis.
It has been recently demonstrated that phosphate starvation conditions induce a tightly regulated set of genes, which are involved in the mobilization of alternative phosphate sources by B. licheniformis cells. Among them, the phytase PhyL belongs to the most abundant extracellular protein under these conditions [11]. Phytase is an enzyme that catalyses the hydrolysis of phytate, the salt of phytic acid, to release a series of myo-inositol phosphate intermediates and inorganic phosphate (Pi) [12,13]. Phytate is the major storage form of phosphate in plant seeds such as cereal and oilseeds (1 to 5% by weight) [14]. The strong expression of the phyL gene indicated that B. licheniformis cells might use phytate as an alternative phosphate source when concentration of Pi becomes limiting [11,15]. It could be therefore concluded that the PphyL promoter would be a good candidate for the construction of a novel expression system that can be used for the production of heterologous proteins in B. licheniformis under phosphate limited growth conditions.
In this study, the suitability of the phyL promoter as a novel auto-inducible phosphate-regulated expression system for B. licheniformis was investigated by means of translational reporter gene fusions with the heterologous genes amyE and xynA from B. subtilis both at the transcriptional and translational level. Furthermore, the role of phytate as an alternative, natural phosphate source for the growth of B. licheniformis cells and as an inducer for the expression of the phyL promoter were studied.

Strains and cultivation
All bacterial strains used in this study are listed in Table 1. B. licheniformis MW3 (B. licheniformis DSM13 (ΔhsdR1, ΔhsdR2)), which ensures high transformation efficiencies due to the deletion of two type I restriction modification systems hsdR1 and hsdR2, was used as the host strain [16]. Belitzky minimal medium (BMM) was used in all growth experiments [17].

Construction of strains
The activity of the PphyL promoter was analyzed by means of translational reporter gene fusions. For this purpose, an approximately 300-bp fragment containing the PphyL promoter from B. licheniformis DSM13 (Additional file 1: Figure S2) was cloned in front of the α-amylase and xylanase reporter genes from B. subtilis 168 with the primer pairs 1/2 and 1/5, respectively (Additional file 1: Table S1). The amyE and xynA genes from B. subtilis 168 were amplified with the primer pairs 3/4 and 6/7, respectively (Additional file 1: Table S1). The PphyL′-′amyE and PphyL′-′xynA fusions were constructed by means of the precise gene fusion polymerase chain reaction strategy described by Yon and Fried [18] by using the primer pairs 1/4 and 1/7, respectively. The PCR-fusion fragments were then inserted into the XbaI and KpnI sites of the multi-copy plasmid pKUC (this shuttle vector is based on the pUC18 and pKTH290 plasmids [19]) resulting in vector pKUC3 and pKUC4, respectively. These vectors were used to transform the B. licheniformis strain MW3 by electroporation [20] resulting in the strains TH3 (pKUC3) and TH4 (pKUC4), respectively. These strains were then cultivated in phosphate-limited BMM as already described. Culture supernatants for enzyme assays were taken at different time points during the cultivation. The first sample was taken during the exponential growth phase (four hours after cultivation), the second sample during the transient phase and additional samples were taken 2, 4, 6, 8, 10, 12 and 14 h after onset of the stationary growth phase.

Enzyme assays
Activity of the amylase AmyE was determined with the Ceralpha kit (Megazyme International Ireland Ltd., Bray, Ireland). Amylase activity was calculated in "international units" (IU) by the equation: IU/mL = 4.6 x (ΔE 400 × 4.7 x Dilution) (ΔE 400 = Absorbance at 400 nm (reaction) -Absorbance at 400 nm (blank)). One international unit of activity is defined as the amount of enzyme required to release one micromole of glucose-reducing sugar equivalents per minute under defined conditions of temperature and pH (40°C, pH 6.5) [21]. Activity of the xylanase XynA was measured using the modified dinitrosalicylic acid (DNSA) method [22] with some modifications as described in details by Nguyen et al. [23]. One international unit (IU/mL) of xylanase activity was defined as the amount of enzyme that liberates one micromole of reducing sugar equivalent to xylose per minute under the assay conditions described.

Growth analysis
The growth of B. licheniformis cells on sodium phytate (Sigma-Aldrich Co, USA) as an alternative phosphate source was studied in a phosphate limited BMM by the addition of 0.5 mM sodium phytate 2 h after onset of the stationary growth phase. Growth experiments were done in 50 mL BMM in 250 mL Erlenmeyer flasks at 37°C under vigorous at 200 rpm. The growth of the B. licheniformis strains was determined by measuring the optical density (OD) of the cultures every two hours at a wavelength of 500 nm.
In order to elucidate whether phytate is an inducer of the PphyL promoter, B. licheniformis strains carrying the translational fusion of PphyL′-′amyE and PphyL′-′ xynA were cultivated in a phosphate limited BMM with sodium phytate, which was added to the growth medium at an OD (at 500 nm) of 1.0 with the final concentrations of either 0.5 mM or 5 mM. The cell samples were taken during the logarithmic growth phase at an OD of 1.0 and 1, 2, 3 and 4 h after onset of the stationary growth phase.

RNA isolation and northern blot analysis
Cell disruption was performed by using the RiboLyser Cell Disrupter (Thermo Electron Corporation, Germany) and total RNA was isolated and purified by using the KingFisher mL pipetting robot (Thermo Electron Corporation, Germany) by means of the MagNA Pure LC RNA isolation Kit I (Roche Diagnostics, Germany) as described in detail by Jürgen et al. [24]. The quality of the isolated total RNA was analyzed by means of the Bioanalyzer 2100 from Agilent (Germany).
The effect of phytate on the expression of the PphyL controlled expression of the heterologous amyE gene was determined by Northern blot analyses as described by Wetzstein et al. [25]. The specific hybridization reaction for the amyE mRNA was performed with appropriate digoxigenin-labeled RNA probes. The probes were synthesized with the T7 RNA polymerase from the T7 promoter-containing internal PCR products of the amyE gene using the primer pairs 8/9 (Additional file 1: Table S1).

Analysis of the phyL promoter sequence
A recent transcriptome analysis indicated a phosphate controlled expression of the phyL gene of B. licheniformis (Additional file 1: Figure S1) [11]. In addition, the study of Hoi et al. [11] also revealed that phytase is one of the most prominent proteins in the secretome of B. licheniformis cells under phosphate limited growth conditions. In order to elucidate the structure of the phyL promoter of B. licheniformis the predicted promoter region with the well-known promoter sequence of the phytase encoding gene phyC from Bacillus amyloliquefaciens FZB42, a close relative of B. licheniformis was aligned. It is shown that the phyL promoter region contains a σ A -like promoter sequence, which well resembles the reported − 10 and − 35 sequences of the phyC promoter (Fig. 1). In addition, the phyL promoter region contained another − 10-like consensus sequence, which is located between the putative − 10 and − 35 promoter sequence (Fig. 1). A pair of tandemly repeated PhoP TT(T/A/C)ACA binding boxes was found within and downstream of the − 35 consensus promoter sequence. Another putative PhoP binding box was located downstream of the − 10 consensus region of the phyL promoter. This suggests that the expression of the phyL gene in B. licheniformis is probably regulated by the PhoPR two component system as reported for the phyC gene in B. amyloliquefaciens FZB45 (Fig. 1).

PphyL controlled expression patterns of the amyE and xynA genes
The phosphate regulated promoter, PphyL, was cloned upstream of the reporter genes amyE and xynA, resulting in the plasmids pKUC3 (strain TH3) and pKUC4 (strain TH4) respectively. The expression pattern of the B. licheniformis strains containing one of those plasmids revealed that both promoter fusions started to express the amyE and xynA reporter genes when cells entered the transient phase (6 h after inoculation) in a phosphate limited BMM (Fig. 2). In order to verify that the synthetic constructs of the phyL promoter and the reporter genes are indeed mainly phosphate-dependent regulated during the onset of the stationary phase, the strain B. licheniformis TH3, carrying the PphyL-amyE fusion, was exemplarily cultivated with different phosphate concentrations (0.15, 0.3 and 0.6 mM) (Additional file 1: Figure S3). The data demonstrate that α-amylase activity was significantly increased under phosphate-limited conditions (0.15 mM). In contrast, only low α-amylase activities were detectable in cultivations with high phosphate concentrations. Furthermore, no amylase or xylanase activity was detected during the exponential phase without phosphate limitation.
The amylase activity in the PphyL′-′amyE fusion strain TH3 increased gradually throughout the stationary phase and reached a final maximal activity of about 2.5 IU/mL (Fig. 2a). The PphyL′-′xynA fusion strain TH4 showed a similar final xylanase activity throughout the stationary phase of about 1.5 IU/mL (Fig. 2b).
The SDS-PAGE analysis of the extracellular protein fraction of the strain TH3 revealed that the heterologous AmyE protein started to secrete and accumulate in the extracellular medium at the transient phase (6 h after cultivation). The AmyE level increased gradually throughout the stationary phase (Fig. 2c). The production pattern of the heterologous XynA protein in strain TH4 was similar to the production pattern observed for the AmyE protein (Fig. 2d).

Analysis of phytate as alternative substrate and inducer
B. licheniformis was then tested for its ability to use phytate as alternative phosphate source when cells were licheniformis cells was monitored in phosphate-limited BMM in the absence or presence of sodium phytate. In growth experiments without the addition of sodium phytate, B. licheniformis cells entered the transient phase after 6 h of cultivation at an OD 500 nm of about 0.8. The growth rate of the control and the phytate containing culture remained constant until an OD of around 1.0 was reached. However, in the culture with 0.5 mM sodium phytate the OD values increased gradually up to an OD of about 2. When higher amounts of sodium phytate (5 mM) were added to the growth medium, B. licheniformis cells grew more slowly to an OD value of 1.0 within 8 h, however, the biomass significantly increased up to an OD value of 2.7 at the end of the cultivation (Fig. 3).
In order to investigate whether phytate might serve as an inducer of the phyL promoter, the B. licheniformis strains TH3 and TH4 bearing the PphyL′-′amyE and PphyL′-′xynA fusions, were cultivated in phosphate limited BMM supplemented with different concentrations of sodium phytate. Under growth conditions without sodium phytate, the PphyL′-′amyE fusion strain TH3 showed a maximal amylase activity of about 2.6 IU/mL, while the PphyL′-′xynA fusion strain TH4 reached a maximal xylanase activity of about 1.5 IU/mL ( Fig. 3a  and b). When 0.5 mM sodium phytate was added to the growth medium, the phyL′-′amyE fusion strain TH3 reached a higher maximal amylase activity of 4.5 IU/mL (Fig. 3a). This observation was supported by Northern blot analyses of the strain TH3, which revealed a significant higher amyE transcript level under growth conditions with 0.5 mM sodium phytate compared to the control without phytate addition (Fig. 4). The addition of 0.5 mM phytate to the growth medium of the phyL′-′ xynA fusion strain TH4 resulted in higher maximal xylanase activity of 9.2 IU/mL (Fig. 3b). Such an increase in the expression pattern could also be observed for the XynA protein level in strain TH4 under growth conditions with 0.5 mM phytate (Fig. 3d). The application of a higher phytate concentration (5 mM) did not result in higher amylase or xylanase activities ( Fig. 3a and b). It is worth mentioning that the native phytase protein PhyL was overproduced and secreted when phytate was added to the phosphate limited BMM (Fig. 3d).

Discussion
A global transcriptome and proteome analysis revealed that more than 100 genes are significantly upregulated in B. licheniformis in response to phosphate limitation [11]. The phyL transcript belonged to the most strongly induced and abundant mRNAs during the transition phase from the exponential to stationary phase. A sequence analysis indicates that the phyL gene is similar to other phosphate-controlled genes regulated by the PhoPR two component system. Furthermore, the − 10 and − 35 regions of the phyL promoter reveal typical SigmaA-dependent sequences (Fig. 1 binding boxes were found within and downstream of the phyL promoter region. Such a specific PhoP control was recently shown for the highly similar phyC promoter of the closely related species B. amyloliquefaciens [26]. Data of this study reveal that the phyL promoter was only induced when cells are exposed to phosphate limitation. As suggested by Makarewicz et al. [26], both Eσ A RNAP holoenzyme and PhoP~P are necessary and sufficient to establish the transcriptional activation of the phyC promoter in B. amyloliquefaciens under such growth conditions. Thus, the expression of the phyL promoter in B. licheniformis might be similarly regulated as described for the phyC promoter in B. amyloliquefaciens. Data of this study indicate an efficient expression of two heterologous model proteins by using the tightly regulated and strongly inducible PphyL promoter in B. licheniformis under phosphate-limitation conditions. Neither protein bands in a 1D-SDS-PAGE nor activities of the two model proteins, the amylase or the xylanase, could be detected during the exponential growth phase. Furthermore, the growth behavior of the recombinant strains also indicates that perturbing effects of the proposed expression system on the growth of B. licheniformis cells during the exponential growth phase, can be excluded as long as sufficient inorganic phosphate is available. In addition, it is shown that sodium phytate is a suitable alternative phosphate source for the growth of B. licheniformis when cells were subjected to phosphate limited growth conditions. Experiments in this study suggest that moderate concentrations of sodium phytate (≤ 5 mM) would be more favorable to induce the activity of the phyL promoter in B. licheniformis. The addition of higher concentrations of phytate (e.g. 1% w/v or 15 mM) to the growth medium could hamper the activity of the promoter of the phytase gene in B. licheniformis (data not shown). This could be due to a critical increase of inorganic phosphate levels by the phytate hydrolysis of the induced phytase enzyme, which would down-regulate the activity of the phosphate responsive PhoPR two-component system and thus result in a lower PphyL activity.
For B. subtilis a similar phosphate controlled expression system based on the pst promoter was suggested [10]. However, a crucial starvation of an essential substrate, such as phosphate, during the protein over-production phase could diminish the protein synthesis capacity of the host. Therefore, to reach optimal yields, either the promoter of an appropriate expression system has to be switched on before the complete exhaustion of the critical substrate or the limited nutrient needs to be replaced by another suitable substrate, which does not lead to a down-regulation of the promoter system. Thus, the perfect alternative substrate should be metabolized and in parallel be an inducer of the system [9]. Data of this study indicated that the expression system using the phyL promoter is not only strong and tightly regulated by the level of inorganic phosphate but also easily inducible by the alternative phosphate source phytate. The suggested promoter system is comparable to the pst promoter system of B. subtilis [10] but exhibits an additional feature due to its expression stimulation by the alternative phosphate source phytate.