A New Bifidobacteria Expression SysTem (BEST) to Produce and Deliver Interleukin-10 in Bifidobacterium bifidum

Abstract

In the last years there has been a growing interest in the use of genetically modified bacteria to deliver molecules of therapeutic interest at mucosal surfaces. Due to the well-recognized probiotic properties of some strains, bifidobacteria represent excellent candidates for the development of live vehicles to produce and deliver heterologous proteins at mucosal surfaces. However, very few studies have considered this genus because of its complexity to be genetically manipulated. In this work, we report the development of a new Bifidobacteria Expression SysTem (BEST) allowing the production of heterologous proteins in Bifidobacterium bifidum. This system is based on: i) the broad host range plasmid pWV01, ii) a stress-inducible promoter, and iii) two different signal peptides (SPs) one issued from Lactococcus lactis (SP_Exp4) and issued from Bifidobacterium longum (SP_BL1181). The functionality of BEST system was validated by cloning murine interleukin-10 (IL-10) and establishing the resulting plasmids (i.e., pBEST_Exp4:IL-10 and pBEST_BL1181:IL-10) in the strain of B. bifidum BS42. We then demonstrated in vitro that recombinant B. bifidum BS42 harboring pBEST_BL1181:IL-10 plasmid efficiently secreted IL-10 and that this secretion was significantly higher (sevenfold) than its counterpart B. bifidum BS42 harboring pBEST_Exp4:IL-10 plasmid. Finally, we validated in vivo that recombinant B. bifidum strains producing IL-10 using BEST system efficiently delivered this cytokine at mucosal surfaces and exhibit beneficial effects in a murine model of low-grade intestinal inflammation.

Keywords: Bifidobacterium bifidum; IL-10; heterologous expression system; low-grade intestinal inflammation; microbiota; recombinant bacteria.

FIGURE 1

Schematic representation of plasmids construction.

FIGURE 2

Experimental protocol for the mouse chronic low-grade inflammation model.

FIGURE 3

IL-10 production of recombinant Bifidobacteria BS42-WT, BS42:BEST_Exp4:IL-10 and BS42:BEST_BL1181:IL-10 grown to OD_600nm = 0.6–0.8 in MRS-cysteine at 37°C under anaerobic atmosphere. Cultures were washed once with PBS and placed in fresh medium, for 30 min at 37°C under anaerobic atmosphere. IL-10 was measured by ELISA in both cell-pellet (C) and supernatant filtered samples (S) (n = 3). Comparison between cell-pellet, supernatant and strains involved the non-parametric Mann–Whitney test. ^∗p < 0.05.

FIGURE 4

Effect of heat-shock, pH and bile salts on IL-10 production. BS42:BEST_BL1181:IL-10 (A) and BS42:BEST_Exp4:IL-10 (B) were cultured in MRS-cysteine broth under anaerobic condition until OD_600nm = 0.6–0.8. Cultures were washed once with PBS and re-cultured in fresh medium for 30 min at 37°C (except for heat-shock). (C) Comparison of IL-10 production in supernatant samples of the two recombinant strains. NI, non-induced (pH6); heat-shock: 42°C; acid shocks: pH4 and pH8; salt-shocks: NaCl25 = 25 g/L of NaCl; SB100 = 100 mg/L of bile salts; SB500 500 mg/L of bile salts. IL-10 was measured with ELISA in Cell-pellet and filtered Supernatant. Data (means plus SEMs) are the result of four separate experiments. Comparison involved the non-parametric Mann–Whitney test. ^∗p < 0.05.

FIGURE 5

Impact of recombinant B. bifidum strains secreting murine IL-10 on a murine model of low grade inflammation induced by two rectal administrations of DNBS. Severity of the colitis reactivation was assessed by (A) change in body weight at day 33, (B) macroscopic scores and (C) fecal lipocalin-2 levels at day 34 and (D) histological score on colon and (E) Histological representation (H&E staining) (n = 2^∗8 = 16 mice/group). Statistically significant differences are calculated relative to the DNBS group using non-parametric Mann–Whitney test. ^∗p < 0.05, ^∗∗∗p < 0.0001.

FIGURE 6

Impact of recombinant B. bifidum strains secreting murine IL-10 on the systemic T-helper balance. Cytokines secreted by splenocytes from each mice group, stimulated for 48h with CD3/CD28, were measured using ELISA (n = 2^∗8 = 16 mice/group). Comparison was performed with the non-parametric Mann–Whitney test. Statistically significant differences were calculated relative to the DNBS group. ^∗p < 0.05, ^∗∗p < 0.001, ^∗∗∗p < 0.0001.