Use of Rgg quorum-sensing machinery to create an innovative recombinant protein expression system in Streptococcus thermophilus

Abstract

Streptococcus thermophilus holds promise as a chassis for producing and secreting heterologous proteins. Used for thousands of years to ferment milk, this species has generally recognized as safe (GRAS) status in the USA and qualified presumption of safety (QPS) status in Europe. In addition, it can be easily genetically modified thanks to its natural competence, and it secretes very few endogenous proteins, which means less downstream processing is needed to purify target proteins, reducing costs. Extracellular degradation of heterologous proteins can be eliminated by introducing mutations that inactivate the genes encoding the bacterium's three major surface proteases. Here, we constructed an inducible expression system that utilizes a peptide pheromone (SHP₁₃₅₈) and a transcriptional regulator (Rgg₁₃₅₈) involved in quorum-sensing regulation. We explored the functionality of a complete version of the system, in which the inducer is produced by the bacterium itself, by synthesizing a luciferase reporter protein. This complete version was assessed with bacteria grown in a chemically defined medium but also in vivo, in the faeces of germ-free mice. We also tested an incomplete version, in which the inducer had to be added to the culture medium, by synthesizing luciferase and a secreted form of elafin, a human protein with therapeutic properties. Our results show that, in our system, protein production can be modulated by employing different concentrations of the SHP₁₃₅₈ inducer or other SHPs with closed amino acid sequences. We also constructed a genetic background in which all system leakiness was eliminated. In conclusion, with this new inducible expression system, we have added to the set of tools currently used to produce secreted proteins in S. thermophilus, whose myriad applications include the delivery of therapeutic peptides or proteins.

Keywords: Rgg; Streptococcus thermophilus; heterologous protein production; inducible promoter; signalling peptide.

Fig. 1.. Schematic representation of the SHP/Rgg₁₃₅₈ quorum-sensing mechanism in S. thermophilus strain LMD-9 and of the genetic constructs used in the study. (a) The quorum-sensing signal is a pheromone encoded by the shp₁₃₅₈ gene. SHP₁₃₅₈ is a small hydrophobic peptide, whose mature form is produced by the cleavage of a precursor and then exported; the two processes are performed by the endopeptidase Eep and the transporter PptAB, respectively. At high cell densities, secreted SHP₁₃₅₈ is reimported into the cell by the Ami transporter and then interacts with the regulatory protein Rgg₁₃₅₈. When the resulting SHP/Rgg₁₃₅₈ complex binds to the promoter region of the two target genes, P_shp1358 and P_ster1357, it activates their transcription. (b) DNA constructs encoding the complete or truncated SHP/Rgg₁₃₅₈ system were fused to the luciferase-encoding gene (luxAB) or the elafin-encoding gene (names of corresponding strains on the right). The solid arrows represent genes (orientation and name underneath). The hatched arrows represent reporter or heterologous protein-encoding genes. The broken arrows represent promoters (name at head of arrow). Blue broken arrows represent promoters regulated by Rgg₁₃₅₈.

Fig. 2.. Levels of growth and luciferase luminescence for S. thermophilus strains grown under different conditions. (a) TIL1524 (blp::shp-rgg₁₃₅₈-P_ster1357-luxAB) in CDM (◯), (b) TIL1525 (blp::rgg₁₃₅₈-P_ster1357-luxAB) in CDM (◊) or in CDM to which synthetic SHP₁₃₅₈ had been added (final concentration of 1 µM) 2 h after the beginning of culture growth (♦), and (c) TIL1525 in CDM (◊) and TIL 1566 (blp::rgg₁₃₅₈-P_ster1357-luxAB ΔpptAB) in CDM (Δ). The growth curves (OD₆₀₀) are depicted using dotted lines, and the relative levels of luciferase luminescence (RLU/OD₆₀₀) are depicted using solid lines. Data shown are representative of three independent experiments.

Fig. 3.. Levels of growth and/or luciferase luminescence for the leakage-free strain, TIL1566 (blp::rgg₁₃₅₈-P_ster1357-luxAB ΔpptAB::erm), grown under different conditions. (a) Maximum relative levels of luciferase luminescence (RLU/OD₆₀₀) during strain growth in CDM across a range of SHP₁₃₅₈ (SHP3) concentrations (0–10 µM). Strain TIL1524 (blp::shp-rgg₁₃₅₈-P_ster1357-luxAB) was used as a control. The data shown are the means±sd of the results from four independent experiments. (b) Strain growth (OD₆₀₀) and luciferase luminescence (RLU/OD₆₀₀) in CDM to which SHP₁₃₅₈ was added at different time points; the coloured curves (•) show the results of SHP₁₃₅₈ being added at different time points, and the grey curves (×) show the results of the situations where no SHP₁₃₅₈ was added. The growth curves are depicted using dotted lines, and the relative levels of luciferase luminescence are depicted using solid lines. Data shown are representative of three independent experiments. (c) Maximum relative levels of luciferase luminescence (RLU/OD₆₀₀) during strain growth in CDM to which different SHPs had been added (1 µM). The data are the means±sd of the results from three independent experiments. To test for significant differences between the treatments and the control, one-way ANOVAs were performed, followed by Dunnett’s tests for multiple comparisons (**P<0.01; ****P<0.0001).

Fig. 4.. Bacteria abundance and luciferase luminescence in the faeces of germ-free mice that had been inoculated with S. thermophilus strain TIL1664 (blp::P32-luxAB-aphA3) (black open circles), TIL1524 (blp::shp-rgg₁₃₅₈-P_ster1357-luxAB-aphA3) (blue open circles) and TIL 1672 (blp::luxAB- aphA3) (red open circles). Faeces were sampled on days 2–3, 7, 10 and 14. (a) Counts (log₁₀ c.f.u. g⁻¹ faeces) were obtained by diluting and plating the faeces on M17lac. (b) Relative levels of luciferase luminescence (log₁₀ RLU×10000 c.f.u. g⁻¹ faeces) were measured using a 10⁻² dilution of the faeces immediately after their dilution in CDM. To test for significant differences among treatments, repeated-measures two-way ANOVAs were conducted, followed by Tukey’s tests for multiple comparisons (*P<0.05; ***P<0.001; and ****P<0.0001).

Fig. 5.. Presence of elafin in the supernatant of different S. thermophilus strains as detected by Western blot. (a) At OD₆₀₀=0.2 and 0.5 for strain TIL1536 (ΔhtrA::aphA3 ΔsepM::spec), TIL1551 (blp::shp-rgg₁₃₅₈-P_ster1357-elafin-P32cat ΔhtrA::aphA3 ΔywdF::spec) and TIL1552 (blp::rgg₁₃₅₈-P_ster1357-elafin-P32cat ΔhtrA::aphA3 ΔywdF::spec). (b) At OD₆₀₀=1 and 2 for strain TIL1536, TIL1551 and TIL1552. (c) At OD₆₀₀=0.2, 0.5 and 1 for strain TIL1567 (blp::rgg₁₃₅₈-P_ster1357-luxAB-aphA3 ΔhtrA::aphA3 ΔywdF::spec ΔpptAB::erm). Purified elafin was the positive control, and supernatant from strain TIL1536 was the negative control (i.e. the strain does not produce elafin). Elafin production was induced by adding synthetic SHP (+SHP) to the culture medium at OD₆₀₀=0.2.

Fig. 6.. Elastase activity inhibition of the supernatant of strain TIL1551(blp::shp-rgg₁₃₅₈-P_ster1357-elafin-P32cat ΔhtrA::aphA3 ΔywdF::spec) measured using an EnzChek elastase assay kit. Supernatants of strain TIL1536 (ΔhtrA::aphA3 ΔywdF::spec) and CDM were used as negative control. Ctrl+, positive control of the kit only containing the elastase enzyme and the fluorescent substrate, DQ elastin. Data shown are representative of three independent experiments.