Les personnes âgées sont souvent victimes de dénutrition pouvant conduire à de multiples dégradations de la santé. La prévention de cet état de dénutrition est une des principales préoccupations des autorités de santé. C’est dans cet objectif que le projet OPTIFEL a conçu de nouveaux produits alimentaires optimisés, afin d’améliorer l’alimentation des personnes âgées, en tenant compte de leurs besoins et souhaits, tout en préservant leur sécurité sanitaire. Bacillus cereus est une bactérie pathogène sporulée capable de se développer dans des conditions variables comme le froid, l’absence d’oxygène, des bas pH ainsi que des matrices alimentaires diverses. Elle est l’une des principales causes de Toxi-Infection Alimentaire Collective (TIAC) surtout en Institut Médico-Sociaux (IMS). Ce travail de thèse avait donc pour objectif de déterminer le comportement de B. cereus à différentes étapes de la vie d’un produit alimentaire. La conservation au froid peut permettre la multiplication de souches psychrotolérantes de B. cereus qui peuvent alors représenter un risque d’infection pour le consommateur. Les conditions anaérobies réduisent la capacité d’adaptation de B. cereus au froid et aux bas pH. La composition de l’aliment et sa préparation influence également le comportement de B. cereus, comme nous l’avons constaté avec un bouillon de carotte dans lequel la présence d’oxygène crée un effet létal pour la bactérie. Certaines souches psychrotolérantes de B. cereus sont capables de produire de la toxine émétique appelée céréulide. Nous avons montré que cette production peut se faire à partir de 8°C et atteindre des concentrations pouvant causer des intoxications à partir de 10°C. S’il est produit dans l’aliment, le cereulide ne serait pas détruit lors de la préparation de l’aliment. Par contre les cellules végétatives de B. cereus sont très thermosensibles et celles qui auraient pu se développer pourront être détruites par un réchauffage du produit avant sa consommation. Ce travail a permis de définir les paramètres importants pour limiter le risque d’infection alimentaire par B. cereus. / Elderlies are often victims of undernutrition which can lead to multiple health damages. Prevention of this state of undernutrition is one of the main concerns of health authorities. With this objective, the OPTIFEL project designed new optimized foodstuffs, to improve foods for elderlies, by taking into account their needs and wishes, while preserving their safety. Bacillus cereus is a spore-forming bacterial pathogen able to grow in various conditions as low temperature, absence of oxygen, low pH and in different types of food. It is one of the main causes of foodborne outbreaks, particularly in socio-medical institutions. The aim of this work was to determine the behavior of B. cereus at different steps of the life of the food product. Cold storage can allow multiplication of psychrotolerant B. cereus strains that can represent a risk of infection for consumers. Anaerobic conditions reduce the adaptation capacity of B. cereus at low temperature and low pH. The composition and the preparation of the food product also influence the behavior of B. cereus, as we showed with carrot broth in which oxygen created a lethal effect for the bacterium. Certain psychrotolerant B. cereus strains are able to produce an emetic toxin named cereulide. We showed that this production starts at 8 °C and can reach at 10 °C concentrations sufficient to cause intoxications. If this toxin is produced in food, it would not be destroyed during the food preparation. In contrast, vegetative cells of B. cereus are very heat-sensitive and those who would have been able to develop could be destroyed by a reheating of the food product before its consumption. This work allowed the identification of important parameters to limit the risk of foodborne disease by B. cereus.

<p>Silicon-based electric chip arrays were developed for characterization of Bacillus</p><p>cereus with respect to the capacity to produce toxins involved in food poisoning and foodborne infections. Bacteria of the B. cereus group contain different sets of four toxins encoded by eight genes. The purpose of this work was to develop a fast method for determination of the presence of these genes in colonies from primary enrichment cultures. The specific DNA detection was based on immobilization of DNA capture probes, which hybridize to specific sites on the target genes. Biotin-labeled detection probes were designed to hybridize with the target DNA adjacent to the capture probes. An extravidin - alkaline phosphatase complex was subsequently bound to the hybridized detection probes. Finally, p-aminophenyl phosphate was added as substrate for the enzyme, and the product p-aminophenol was brought in contact with the interdigitated gold electrode on the silicon chips surface. The p-aminophenol was oxidized at the anode to quinoneimine, which was then reduced back to paminophenol at the cathode. This redox recycling generates a current that was used as the DNA-chip response to the target DNA. Two versions of the assay were used. In the first version the capture probes were immobilized on magnetic beads and all</p><p>chemical reactions until and including the enzymatic reaction took place in an</p><p>eppendorf tube while the redox recycling was used to measure the amount of paminophenol produced after transfer from the tube to the silicon chip surface. In the second version a silicon chip array was used with 16 parallel electrode positions, each activated by immobilization of one type of capture probes on the gold electrodes. With this system all chemical reactions took place at the chip surface. The kinetics of cell disruption and DNA fragmentation from B. cereus by ultrasonication was determined. Maximum cell disruption was achieved within 5 min and the chip response increased in proportion to the ultrasonic time. Further ultrasonication up to 10 min resulted in further increasing current although no further cell disruption was observed. If the sonication time was extended above 10 min the signal declined. Based on analysis of the DNA size distribution by early end-point PCR and gel electrophoresis, it is suggested that the first 5 min ultrasonication increased the signal by increasing the release of target DNA molecules. Thereafter the signal was increased by fragmentation of target DNA which increases the diffusion rate and also the accessibility of the hybridization site. Finally, the DNA fragment sizes approached that of the hybridization site (51-bp) which may reduce the signal because of cleavage of the target DNA in the hybridization region. These studies were performed with the bead-based hybridization assay. The assay was highly specific to the target gene (hblC) of both B. cereus and B. thuringiensis with no response from negative control</p><p>cells of B. subtilis. The 16 positions of the silicon chip array were activated by</p><p>immobilization of all known toxin-coding genes of B. cereus and also included both a positive control and a negative control electrode positions. When these chips were exposed to ultrasonicated B. cereus, the gold electrodes were fouled by some component in DNA cell lysates. To circumvent this, the released large DNA was first extracted and then ultrasonicated again, since the extract mainly contains large molecular weight DNA. This DNA extract was applied to characterize one “diarrheal” and one “emetic” strain of B. cereus with the DNA chip arrays. The results agreed with PCR control analysis which means that these electric DNA chip arrays can be used to characterize bacterial colonies with respect to the genes coding of all known toxins of B. cereus: haemolysin (hblA, hblC, hblD), non-haemolytic enterotoxin (nheA, nheB, nheC), cytotoxin K-2 (cytK-2), and cereulide (ces). The chip assay required about 30 min after application of DNA samples. Due to the generic properties of the chips, this technique should also be applicable for characterization of the pathogenicity potential of many other organisms. Keywords: Bacillus cereus, haemolysin, non-haemolytic enterotoxin, cytotoxin K-2, cereulide, toxin-coding genes, bacterial colony, electric DNA chip, ultrasonication, DNA fragmentation.</p>

Silicon-based electric chip arrays were developed for characterization of Bacillus cereus with respect to the capacity to produce toxins involved in food poisoning and foodborne infections. Bacteria of the B. cereus group contain different sets of four toxins encoded by eight genes. The purpose of this work was to develop a fast method for determination of the presence of these genes in colonies from primary enrichment cultures. The specific DNA detection was based on immobilization of DNA capture probes, which hybridize to specific sites on the target genes. Biotin-labeled detection probes were designed to hybridize with the target DNA adjacent to the capture probes. An extravidin - alkaline phosphatase complex was subsequently bound to the hybridized detection probes. Finally, p-aminophenyl phosphate was added as substrate for the enzyme, and the product p-aminophenol was brought in contact with the interdigitated gold electrode on the silicon chips surface. The p-aminophenol was oxidized at the anode to quinoneimine, which was then reduced back to paminophenol at the cathode. This redox recycling generates a current that was used as the DNA-chip response to the target DNA. Two versions of the assay were used. In the first version the capture probes were immobilized on magnetic beads and all chemical reactions until and including the enzymatic reaction took place in an eppendorf tube while the redox recycling was used to measure the amount of paminophenol produced after transfer from the tube to the silicon chip surface. In the second version a silicon chip array was used with 16 parallel electrode positions, each activated by immobilization of one type of capture probes on the gold electrodes. With this system all chemical reactions took place at the chip surface. The kinetics of cell disruption and DNA fragmentation from B. cereus by ultrasonication was determined. Maximum cell disruption was achieved within 5 min and the chip response increased in proportion to the ultrasonic time. Further ultrasonication up to 10 min resulted in further increasing current although no further cell disruption was observed. If the sonication time was extended above 10 min the signal declined. Based on analysis of the DNA size distribution by early end-point PCR and gel electrophoresis, it is suggested that the first 5 min ultrasonication increased the signal by increasing the release of target DNA molecules. Thereafter the signal was increased by fragmentation of target DNA which increases the diffusion rate and also the accessibility of the hybridization site. Finally, the DNA fragment sizes approached that of the hybridization site (51-bp) which may reduce the signal because of cleavage of the target DNA in the hybridization region. These studies were performed with the bead-based hybridization assay. The assay was highly specific to the target gene (hblC) of both B. cereus and B. thuringiensis with no response from negative control cells of B. subtilis. The 16 positions of the silicon chip array were activated by immobilization of all known toxin-coding genes of B. cereus and also included both a positive control and a negative control electrode positions. When these chips were exposed to ultrasonicated B. cereus, the gold electrodes were fouled by some component in DNA cell lysates. To circumvent this, the released large DNA was first extracted and then ultrasonicated again, since the extract mainly contains large molecular weight DNA. This DNA extract was applied to characterize one “diarrheal” and one “emetic” strain of B. cereus with the DNA chip arrays. The results agreed with PCR control analysis which means that these electric DNA chip arrays can be used to characterize bacterial colonies with respect to the genes coding of all known toxins of B. cereus: haemolysin (hblA, hblC, hblD), non-haemolytic enterotoxin (nheA, nheB, nheC), cytotoxin K-2 (cytK-2), and cereulide (ces). The chip assay required about 30 min after application of DNA samples. Due to the generic properties of the chips, this technique should also be applicable for characterization of the pathogenicity potential of many other organisms. Keywords: Bacillus cereus, haemolysin, non-haemolytic enterotoxin, cytotoxin K-2, cereulide, toxin-coding genes, bacterial colony, electric DNA chip, ultrasonication, DNA fragmentation. / QC 20101111