A host-vector toolbox for improved secretory protein overproduction in Bacillus subtilis

Abstract

Target proteins in biotechnological applications are highly diverse. Therefore, versatile flexible expression systems for their functional overproduction are required. In order to find the right heterologous gene expression strategy, suitable host-vector systems, which combine different genetic circuits, are useful. In this study, we designed a novel Bacillus subtilis expression toolbox, which allows the overproduction and secretion of potentially toxic enzymes. This toolbox comprises a set of 60 expression vectors, which combine two promoter variants, four strong secretion signals, a translation-enhancing downstream box, and three plasmid backbones. This B. subtilis toolbox is based on a tailor-made, clean deletion mutant strain, which is protease and sporulation deficient and exhibits reduced autolysis and secondary metabolism. The appropriateness of this alternative expression platform was tested for the overproduction of two difficult-to-produce eukaryotic model proteins. These included the sulfhydryl oxidase Sox from Saccharomyces cerevisiae, which forms reactive hydrogen peroxide and undesired cross-linking of functional proteins, and the human interleukin-1β, a pro-inflammatory cytokine. For the best performing Sox and interleukin, overproducing and secreting variants of these new B. subtilis toolbox fermentation strategies were developed and tested. This study demonstrates the suitability of the prokaryotic B. subtilis host-vector system for the extracellular production of two eukaryotic proteins with biotechnological relevance. KEY POINTS: • Construction of a versatile Bacillus subtilis gene expression toolbox. • Verification of the toolbox by the secretory overproduction of two difficult-to-express proteins. • Fermentation strategy for an acetoin-controlled overproduction of heterologous proteins.

Keywords: Bacillus subtilis; Expression system; Fed batch; Human interleukin-1β; Overproduction; Protease deficiency; Protein secretion; Yeast sulfhydryl oxidase.

Fig. 1

Schematic overview (a) of the replaceable elements of the B. subtilis toolbox and (b) the array of the toolbox elements including restriction sites. The multiple cloning site of all toolbox vectors was adjusted as shown in Figure S1. GOI, gene of interest

Fig. 2

Sox amounts from SP screening. High amounts of Sox protein after 24 h (a) and 48 h (b) were calculated for the secretion signals of csn, nucB, yweA, and ylqB in comparison to significantly reduced amounts of Sox secreted by the amyE signal peptide. Dark blue bars represent Sox yields in μg/mL based on the quantification using a BSA standard (see Fig. S3b) while light blue bars illustrate the percentage distribution using the amyE-SP as reference (100%)

Fig. 3

Verification of the toolbox strains S1–S12 (see also Table 3) for the overexpression of Sox. a–c Growth curves of B. subtilis JK138 S1–12 in comparison to the negative control carrying the “empty vector” without expression cassette. Cultivations were performed in biological triplicates; standard deviations are presented as error bars. d–f Quantitative analyses of the secretory overproduction of Sox. Average protein amounts (normalized by OD) for the Sox strains S1–S12 in the culture supernatant 24 h and 48 h after the boost in EnpressoB medium. Protein concentrations are given as average values from three independent experiments. Standard deviations are presented as error bars. For comparison of protein yields, the target protein was quantified. “Empty vector” strains without expression cassette served as negative controls

Fig. 4

Activity of the heterologously produced sulfhydryl oxidase Sox. a Purification of Sox from the supernatant of B. subtilis WB600 pSox cultures; SN, supernatant; FT, flow through, W1–W2: wash fractions, E1–E4: elution fractions. b Temperature and pH profiles of purified Sox. The influence of temperature (light blue dots) and pH (dark blue dots) on Sox activity was determined using a miniaturized Sox assay in a semi-automated system applying a Tecan Freedom EVO® screening robot

Fig. 5

Development of a fermentation strategy for Sox overproduction. a Comparison of batch (light blue) and fed-batch (dark blue) fermentation strategies for Sox production in B. subtilis WB600 pSox. The cells were grown in a parallel fermenter system at 1.5 L scale at 37 °C either in batch or in fed-batch mode. Cell growth was monitored by the OD at 600 nm (triangles). Sox activity was determined in triplicates using purified enzyme (bars). b SDS-PAGE analysis of Sox accumulation in untreated culture supernatants of the batch and fed-batch fermentations

Fig. 6

a Comparison of growth and Sox production of the protease-deficient B. subtilis strains WB600 (light blue) and LS8P-D (dark blue). The cells were grown in batch fermentations at 37 °C (triangles). Sox activity was determined in triplicates using purified enzyme after 24 h (bars). b Verification of the best performing P_acoA-dependent S1 toolbox system in B. subtilis JK139 for the overexpression of the sulfhydryl oxidase Sox. The growth curve reflects three individual 1.5-L fed-batch fermentation experiments A–C. Shown are average protein amounts (normalized by OD) from the culture supernatant of the S1 expression strains from three independent experiments. Standard deviations are presented as error bars