Simulating Intestinal Growth Conditions Enhances Toxin Production of Enteropathogenic Bacillus cereus

Abstract

Bacillus cereus is a ubiquitous bacterial pathogen increasingly reported to be the causative agent of foodborne infections and intoxications. Since the enterotoxins linked to the diarrheal form of food poising are foremost produced in the human intestine, the toxic potential of enteropathogenic B. cereus strains is difficult to predict from studies carried out under routine cultivation procedures. In this study, toxigenic properties of a panel of strains (n = 19) of diverse origin were compared using cell culture medium pre-incubated with CaCo-2 cells to mimic intestinal growth conditions. Shortly after contact of the bacteria with the simulated host environment, enterotoxin gene expression was activated and total protein secretion of all strains was accelerated. Although the signal stimulating enterotoxin production still needs to be elucidated, it could be shown that it originated from the CaCo-2 cells. Overall, our study demonstrates that the currently used methods in B. cereus diagnostics, based on standard culture medium, are not allowing a conclusive prediction of the potential health risk related to a certain strain. Thus, these methods should be complemented by cultivation procedures that are simulating intestinal host conditions.

Keywords: Bacillus cereus; CaCo-2 cells; enterotoxins; simulated intestinal conditions; toxic potential.

Figure 1

Growth of 19 B. cereus strains under simulated intestinal conditions (RPMI medium treated with CaCo-2 cells, 37°C, 7% CO₂). Strains are grouped consistent with their toxin profile. (A) Toxin profile A (nhe, hbl, cytK2). (B) Toxin profile C (nhe, hbl). (C) Toxin profile D (nhe, cytK2). (D) Toxin profile F (nhe).

Figure 2

Enterotoxin gene transcription of the B. cereus strains. nheB (A) and hblD (C) transcription was determined by qRT-PCR, normalized to 16S rrn levels of the same sample and relative to the transcript level of an external calibrator. hblD expression of reference strain F837/76 at 6 h growth in CGY medium (Jeßberger et al., 2015) was used as calibrator and set to 100% (log-2 = 0). With the 2^−ΔΔCT method, transcription of all other samples was compared to this calibrator. Transcription efficiencies of nheB (B) and hblD (D) were determined as relative transcript level/OD₆₀₀. Strains are grouped according to their enterotoxin gene profiles, which are separated by dotted lines.

Figure 3

Enterotoxin production of 19 B. cereus strains grown under simulated intestinal conditions. Reciprocal titers as well as the productivity (titer/OD₆₀₀) are shown. (A) NheB. (B) Hbl L2. (C) Hbl L1. (D) Hbl B. Strains are grouped according to their enterotoxin gene profiles, which are separated by dotted lines.

Figure 4

Total protein secretion of the B. cereus strains grown under simulated intestinal conditions. (A) Quantification of extracellular protein after 2, 4, 6, 8, and 10 h growth. (B) Efficiency of protein secretion determined by normalization of extracellular protein concentrations to the OD₆₀₀. Strains are grouped according to their enterotoxin gene profiles, which are separated by dotted lines.

Figure 5

Comparison of the B. cereus strain set grown in CGY (Jeßberger et al., 2015) and cRPMI (this study). (A) Transcription efficiency of nheB determined as % transcription per OD₆₀₀. (B) NheB productivity determined as reciprocal titer per OD₆₀₀. (C) Efficiency of protein secretion determined as extracellular protein concentration per OD₆₀₀.

Figure 6

Growth and enterotoxin production of 7 selected B. cereus strains comparing CaCo-2 treated cRPMI medium with untreated RPMI 1640 medium. Both media types were supplemented with 2% casein hydrolysate and 1% glucose. cRPMI medium had been pre-incubated for 22 h with differentiated CaCo-2 cells. Bacteria were grown in 80 cm² cell culture flasks at 37°C under 7% CO₂ atmosphere. OD₆₀₀ was recorded for 10 h and every 2 h samples of the supernatant were taken. Enterotoxin production was determined via sandwich EIA specific for NheB.

Figure 7

Toxicity tests of the 19 B. cereus strains on CaCo-2 cells. (A) Reciprocal titers for 50% lethal doses obtained in WST-1 bioassays are shown. The assay was performed only after 6 h of growth under simulated intestinal conditions. (B) Propidium iodide influx tests were performed after 2, 4, 6, 8, and 10 h of growth. The maximum linear slope of the fluorescence curves of each strain is shown. Strains are grouped according to their enterotoxin gene profiles, which are separated by dotted lines.