By rewiring the sporulation gene-regulatory network of Bacillus subtilis, we generated a novel expression system relying on derepression. The gene of interest is placed under the control of the abrB promoter, which is active only when Spo0A is absent, and Spo0A is controlled via an IPTG (isopropyl-β-d-thiogalactopyranoside)-inducible promoter.

Cells were grown overnight in TY medium (16) containing 100 μM IPTG and diluted 30-fold to an optical density at 600 nm (OD₆₀₀) of 0.05. At an OD₆₀₀ of ~1.0, cells were diluted 10-fold in minimal medium, and a 1-ml suspension (without glass beads) was treated with a Mini-Bead-Beater-8 (Biospec Products) for 1 min at maximal speed to separate cell chains into individual cells. Two hours later (end of log phase), another sample was measured. GFP fluorescence was measured using a Coulter Epics XL-MCL flow cytometer (Beckman Coulter). The average fluorescence of 20,000 cells was determined using WinMDI 2.8 (http://facs.scripps.edu/software.html) and plotted against IPTG concentrations (Fig. ​(Fig.2).2). As shown in Fig. ​Fig.2,2, GFP expression under the control of P_abrB is high without inducer but is strongly reduced upon increases in levels of Spo0A*. When the native spo0A gene is deleted, GFP expression is further increased. The maximum concentration of GFP in this strain was quantified by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) relative to bovine serum albumin standards. Dry-matter concentrations of biomass were calculated using a predetermined correlation factor of 0.33 g (dry weight) of cells per OD₆₀₀ unit (19). The concentration of GFP with an average fluorescence per cell of 550 arbitrary units was 16.7 mg GFP/g (dry weight). Interestingly, under inducing conditions (in the presence of Spo0A*), P_abrB is more tightly repressed in the spo0A (Δspo0A) mutant. In the wild-type strain, the abrB promoter is still leaky and shows optimal derepression of abrB when Spo0A* is induced with 50 μM IPTG.

Our results show that secreted β-toxin is more actively degraded in the presence of a functional spo0A gene. Many of the known extracellular proteases are under either the direct or indirect control of Spo0A (6). To examine whether the increased degradation of β-toxin is related to increased extracellular proteolytic activity, we assayed the growth medium for proteolytic activity. Total protease activity was measured using the Roche resorufin-labeled universal protease substrate as described by the manufacturer. Direct optical readout of the OD₅₇₄ was plotted. As depicted in Fig. ​Fig.3C,3C, only a minor proteolytic activity was measured in the Δspo0A strain. In the presence of spo0A, strong protease activity was observed starting after 150 min of resuspension in fresh medium, when cells enter stationary growth phase (data not shown).

The use of sporulation-deficient mutants (such as the Δspo0A strain) is common practice for large-scale high-density fermentation processes, mainly because it prevents the formation of spores, which are hard to remove from the growth system (8). Strain stability is not an issue for these mutants, since sporulation is purely an adaptive phenotype and not essential. The addition of IPTG in the case of our Spo0A* strain led to induction of Spo0A, and therefore undesirable sporulation could occur more rapidly. However, we did not observe sporulation under our culture conditions. The process of sporulation is tightly controlled, and when only spo0A is induced during logarithmic growth, not all essential components for sporulation are expressed, preventing premature sporulation of the culture (2).

In this paper we show how a naturally occurring gene-regulatory network can be adapted to serve as a tailored expression system. An additional benefit of using this endogenous pathway is that the system is based on self-cloning via homologous recombination, keeping the introduced foreign DNA to a minimum. Furthermore, upon activation of the abrB promoter, extracellular proteolytic activity is reduced.