Expression of degQ gene and its effect on lipopeptide production as well as formation of secretory proteases in Bacillus subtilis strains

Abstract Bacillus subtilis is described as a promising production strain for lipopeptides. In the case of B. subtilis strains JABs24 and DSM10T, surfactin and plipastatin are produced. Lipopeptide formation is controlled, among others, by the DegU response regulator. The activating phospho‐transfer by the DegS sensor kinase is stimulated by the pleiotropic regulator DegQ, resulting in enhanced DegU activation. In B. subtilis 168, a point mutation in the degQ promoter region leads to a reduction in gene expression. Corresponding reporter strains showed a 14‐fold reduced expression. This effect on degQ expression and the associated impact on lipopeptide formation was examined for B. subtilis JABs24, a lipopeptide‐producing derivative of strain 168, and B. subtilis wild‐type strain DSM10T, which has a native degQ expression. Based on the stimulatory effects of the DegU regulator on secretory protease formation, the impact of degQ expression on extracellular protease activity was additionally investigated. To follow the impact of degQ, a deletion mutant was constructed for DSM10T, while a natively expressed degQ version was integrated into strain JABs24. This allowed strain‐specific quantification of the stimulatory effect of degQ expression on plipastatin and the negative effect on surfactin production in strains JABs24 and DSM10T. While an unaffected degQ expression reduced surfactin production in JABs24 by about 25%, a sixfold increase in plipastatin was observed. In contrast, degQ deletion in DSM10T increased surfactin titer by threefold but decreased plipastatin production by fivefold. In addition, although significant differences in extracellular protease activity were detected, no decrease in plipastatin and surfactin produced during cultivation was observed.


| INTRODUC TI ON
B. subtilis is one of the best characterized gram-positive bacteria and serves as a model organism for fundamental and applied research.
The knowledge about the physiology of B. subtilis made this strain an important microbial host in biotechnology (Stein, 2005). In this context, B. subtilis is used as a super-secreting cell factory due to benefits such as excellent fermentation properties, high product yields in gram per liter range, and the lack of toxic by-products (van Dijl & Hecker, 2013). In addition to the production of industrially relevant enzymes and vitamins (Cui et al., 2018), B. subtilis natively forms a variety of secondary metabolites. Among these compounds, three lipopeptide families, in particular, namely surfactin, iturin, and fengycin, are reported to have broad bioactivity based on a common amphipathic structure comprising a fatty acid linked to a peptide moiety Marvasi et al., 2010). Different amino acid sequences in the circular peptide and variable fatty acid chain lengths give each lipopeptide unique properties (Zhao et al., 2017).
In the genome of Bacillus spp., bacteria encoding for fengycin biosynthesis also show the ability to produce surfactin (Kim et al., 2010;Yaseen et al., 2018). In this context, regulatory crosstalk of nonribosomal peptide synthetases (NRPSs) is conceivable (Vahidinasab et al., 2020;Yaseen et al., 2018). Surfactin is described as one of the most powerful microbially produced biosurfactants and has great potential to be used in many industrial sectors such as cosmetics, pharmaceuticals, as well as food . The benefits of surfactin are not limited to emulsifying activity, as some studies reported antimicrobial and anticancer properties (Béven & Wroblewski, 1997;Kameda et al., 1974). Fengycins, including plipastatin as a member of this group, have been shown to have several antagonistic effects for soil-borne fungal phytopathogens and may act as elicitors for systemic plant resistance (Cawoy et al., 2015). Moreover, fengycin has been described to have antiviral, antibacterial, and anticancer properties (Huang et al., 2006;Ongena et al., 2007;Raaijmakers et al., 2010;Yin et al., 2013). Due to these characteristics, fengycin has great potential for future agricultural applications.
Different physiological adaptations are associated with the DegU regulon, including the formation of extracellular enzymes, genetic competence, and biofilm formation (Dahl et al., 1992;Kobayashi, 2007;Mäder et al., 2002;Msadek et al., 1990;Shimotsu & Henner, 1986). Moreover, also surfactin and plipastatin production are affected by DegU regulation (Miras & Dubnau, 2016;Tsuge et al., 1999). As a response regulator, DegU is part of the two-component DegS-DegU system. After activating phospho-transfer from histidine kinase DegS to the response regulator DegU, the phosphorylated DegU version (DegU-P) can regulate the expression of various genes (Murray et al., 2009). In addition to this process, DegQ, a small protein of 46 amino acids, stimulates the autophosphorylation of DegS and is important for the complete activation of DegU by phosphorylation (Do et al., 2011;Yang et al., 1986). In the case of the domesticated laboratory model strain B. subtilis 168, a single base mutation in the −10 box silences degQ gene expression (Stanley & Lazazzera, 2005). As a result, phospho-transfer for DegU activation is reduced.
In this study, the lipopeptide-producing B. subtilis strain JABs24, an sfp + derivative of B. subtilis 168, and the wild-type strain DSM10 T , which exhibits a native degQ expression, were used to analyze the effect of degQ expression on lipopeptide production and formation of secretory proteases.

| Chemicals, materials, and standard procedures
All chemicals were purchased from Carl Roth GmbH & Co. KG, if not otherwise mentioned. Standard molecular techniques were performed as described by Sambrook and Russell (2006). PCRs were carried out on a PCR thermal cycler (peqSTAR 96X VWR GmbH) using DNA polymerase (Phusion High-Fidelity #M0530S, New England BioLabs). PCR reactions were purified after agarose-based gel electrophoresis using QIAquick PCR & Gel Cleanup Kit (Qiagen). Plasmid DNA was extracted with innuPREP Plasmid Mini Kit (Analytik Jena AG), and chromosomal DNA was purified using the ready-to-use in-nuPREP Bacteria DNA Kit (Analytik Jena AG) according to the manufacture's instruction.

| Strain construction, plasmids, and transformation method
All strains and plasmids used in this study are summarized in Table 1.
The oligonucleotides used to construct the strains and plasmids are listed in Table 2. Escherichia coli JM109 was used for plasmid propagation and cloning. Transformation of naturally competent B. subtilis strains was performed according to the "Paris method" (Harwood & Cutting, 1990). Depending on the selection marker, transformants were selected on Lysogeny Broth agar supplemented with ampicillin (100 µg/ml), spectinomycin (100 µg/ml), or erythromycin (10 µg/ml for E. coli and 5 µg/ml for B. subtilis). All plates were incubated at 37℃.
For the construction of BCKN1, the degQ gene of B. subtilis DSM10 T including native promoter region (+ 510 bp) and terminator structure was integrated into the amyE locus of B. subtilis JABs24 using plasmid pMAV5 (Vahidinasab et al., 2020) BKE31720 carrying the deletion of the degQ gene (ΔdegQ::erm) (Koo et al., 2017). The plasmids for the construction of the P degQ reporter strains were cloned using Gibson Assembly (New England BioLabs).
Therefore, the pKAM446 plasmid was digested with NheI and NdeI before integrating amplified degQ promoter regions from JABs24 and DSM10 T , respectively. The correctness of all mutant strains was ensured by sequencing (Eurofins Genomics Germany GmbH).

| Cultivation and growth conditions
The composition of the mineral salt medium used in this study was , and (endo)-protease activity. Surfactin and plipastatin concentration were measured as previously described by Geissler et al., (2017). Specifically, 2 ml of cell-free supernatant was extracted three times with chloroform/methanol (2:1). The pooled solvent layers were dried using a rotary evaporator at 10 mbar and 40℃. Dried samples were resolved in 2 ml methanol and applied in 6 mm bands on a silica HPTLC plate. As a mobile phase, chloroform/methanol/ water (65:25:4) was used with a migration distance over 60 mm.

Surfactin standard from Sigma Aldrich and plipastatin standard from
Lipofabrik were used for quantification.

| Data analysis
For the conversion of OD 600 into cell dry weight (CDW), the correlation factor of 3.762 was determined in a preliminary experiment described by Willenbacher et al. (2014). The product yield of biomass

| Assay for extracellular protease activity
The total activity of the degrading proteins in the cultivation medium was analyzed by azocasein assay. The measurement method was adapted from Charney and Tomarell (1947) (2) and (3). (1) (2)

TTTCCACACTCCTTT
The calculation of the absorption difference (ΔA) is described in Equation (2). The following volumetric peptidase activity is defined in Equation (3)  The corresponding β-galactosidase activity showed that the P degQ promoter region of DSM10 T exhibited a significantly higher expression level compared with that of JABs24 ( Figure 2). In the transition between the exponential and stationary phase, approx. 14-fold higher Miller Units were detected for the lacZ fusion with degQ promoter from DSM10 T .
In addition to the confirmation that degQ expression is nearly silenced in strain JABs24, the results have shown that both degQ promoter versions were more active in the DSM10 T background. Thus, threefold increased promoter activity was detected for P degQ DSM10T and sixfold higher Miller Units were calculated for P degQ JABs24 in the transition phase.

| Effect of degQ expression on the formation of lipopeptides and secretory proteases
It is already known that DegQ acts as a stimulator for autophos- Since degQ expression is also directly associated with secretory protease production, JABs24 and DSM10 T wild-type strains as well as their degQ mutants were examined for their ability to form extracellular proteases. Therefore, endopeptidase activity was measured in cell-free supernatants using an azocasein assay. In detail, JABs24 showed a basal activity for extracellular proteases during cultivation with a comparatively small increase to 8.7 ΔA/(h·mL) during the late exponential phase (Figure 3a). In contrast, strain DSM10 T showed the highest protease activity of up to 42.8 ΔA/ (h·mL), which reached a plateau after 12 h of cultivation (Figure 3b).
In respect of the stability of lipopeptides, no reduction of surfactin and plipastatin concentration was detected for both JABs24 and DSM10 T suggesting that secretion of proteases has an inferior impact on lipopeptide production.
In comparison, integration of a natively expressed degQ version from the DSM10 T strain into JABs24 increased secretory protease activity 2 times (17.6 ΔA/(h·mL) after 24 h) compared to JABs24 ( Figure 3c). A comparably great effect was observed for BCKN2, resulting in a continuous basal level of up to 6.2 ΔA/(h·mL) at the end of cultivation (Figure 3d). In this way, deletion of the degQ gene in DSM10 T reduced extracellular protease activity sevenfold. Table 3 gives an overall summary of the effect of degQ gene expression on the lipopeptide productivity and secretory protease formation of the wild-type strains JABs24 and DSM10 T and their degQ mutant strains.

| DISCUSS ION
Due to the point mutation within the degQ promoter region, B. subtilis JABs24, the lipopeptide-forming derivative of B. subtilis 168, shows a drastically reduced degQ gene expression. This circumstance was already described by Stanley and Lazazzera (2005) and confirmed by using lacZ reporter strains for a time-resolved comparison of the expression of the two degQ promoter versions until the transient growth phase (Figure 2). In this process, the wild-type strain DSM10 T showed much higher P degQ promoter activity compared to JABs24. Since DegQ is directly involved in the activation of the DegU response regulator, it is reasonable to assume that

| CON CLUS IONS
The degQ loci of the lipopeptide-producing strains DSM10 T and JABs24 differ by a single point mutation that leads to a drastic reduction of degQ gene expression in JABs24. Based on opposing regulatory mechanisms related to the DegU regulator, the presented strains show beneficial yields in surfactin or plipastatin production, which was confirmed by constructed degQ mutant strains. An additional negative effect of silenced degQ expression in JABs24 was furthermore quantitatively examined on the formation of extracellular proteases. Although a lipopeptide degradation cannot be excluded, different signal strengths of the protease activities measured during the cultivation processes did not lead to a decrease in lipopeptide concentration.

E TH I C S S TATEM ENT
None required.

ACK N OWLED G M ENTS
We thank Jens Pfannstiel and Philipp Hubel for their critical reading of the manuscript and fruitful discussions. The study was finan-

CO N FLI C T O F I NTE R E S T S
None declared.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data generated or analyzed during this study are included in this published article. An overview of the collected data is available in Zenodo at https://doi.org/10.5281/zenodo.5511929