Etablierung eines Frühwarnsystems zur Erkennung lebensmittelbedingter Risiken in Bayern – risikoorientierte Lebensmittelüberwachung weiter gefasst

Recent and past food scandals highlight the urgent need for food safety authorities to anticipate future risks in order to enable improved and proactive response mechanisms. With this goal in mind, the Bavarian Health and Food Safety Authority (LGL) has established an early warning system aimed at early identification of potential health risks and fraudulent practices in the food sector. The early warning approach allows intervention at different stages of risk development. On the one hand, the scanning of a broad area of the food production chain (horizon scanning) may enable prospective risk identification. To this end, relevant factors of influence (drivers) need to be identified, observed and analyzed. On the other hand, the retrospective analysis of case studies may unravel novel cause-effect associations that could be transferred to other products and production technologies, possibly leading to the identification of new drivers (root cause analysis). Additionally, various datasets emanating from food safety authorities, for example data from the European Rapid Alert System for Food and Feed (RASFF) and from Bavarian-wide laboratory analyses and food business inspections, will be closely observed and analyzed. The multidisciplinary character of the LGL has proven to be essential for the realization of the early warning approach, the analysis of inflowing data and implementation of consequences such as food business inspections and sampling for analysis. This article details the objectives and methodical approaches of the project. The acquisition of systematically compiled information as well as its analysis and assessment are comprehensively described. The early warning system presented here has already contributed significantly to the risk-oriented food inspection practiced to date, by integrating novel risk-relevant aspects.     

Effect of Oxime Ether Incorporation in Acyl Indole Derivatives on PPAR Subtype Selectivity

Compounds that simultaneously activate peroxisome proliferator‐activated receptor (PPAR) subtypes α and γ have the potential to effectively treat dyslipidemia and type 2 diabetes (T2D) in a single pharmaceutically active molecule. The frequently observed side effects of selective PPARγ agonists, such as edema and weight gain, were expected to be overcome by using additive PPARα activity, leading to dual PPARα/γ agonists with balanced activity for both subtypes. Herein we report the discovery, synthesis, and optimization of a new series of α‐ethoxyphenylpropionic acid bearing 5‐ or 6‐substituted indoles. The incorporation of oxime ethers on the carbonyl portion of the benzoyl group can bring the PPARα/γ potency ratio equal to or slightly greater than one, as is the case for compounds 20 c and 21 a . Compound 20 c shows high efficacy in an ob/ob mouse model of T2D and dyslipidemia, similar to that of rosiglitazone and tesaglitazar, but with a significant increase in body weight gain. In contrast, compound 21 a , less potent as a dual PPARα/γ activator than 20 c , showed an interesting pharmacological profile, as it elicits a decrease in body weight relative to reference compounds.     

Quantitative proteomic analysis of bacterial enzymes released in cheese during ripening

Due to increasingly available bacterial genomes in databases, proteomic tools have recently been used to screen proteins expressed by micro-organisms in food in order to better understand their metabolism in situ. While the main objective is the systematic identification of proteins, the next step will be to bridge the gap between identification and quantification of these proteins. For that purpose, a new mass spectrometry-based approach was applied, using isobaric tagging reagent for quantitative proteomic analysis (iTRAQ), which are amine specific and yield labelled peptides identical in mass. Experimental Swiss-type cheeses were manufactured from microfiltered milk using Streptococcus thermophilus ITG ST20 and Lactobacillus helveticus ITG LH1 as lactic acid starters. At three ripening times (7, 20 and 69 days), cheese aqueous phases were extracted and enriched in bacterial proteins by fractionation. Each sample, standardised in protein amount prior to proteomic analyses, was: i) analysed by 2D-electrophoresis for qualitative analysis and ii) submitted to trypsinolysis, and labelled with specific iTRAQ tag, one per ripening time. The three labelled samples were mixed together and analysed by nano-LC coupled on-line with ESI-QTOF mass spectrometer. Thirty proteins, both from bacterial or bovine origin, were identified and efficiently quantified. The free bacterial proteins detected were enzymes from the central carbon metabolism as well as stress proteins. Depending on the protein considered, the quantity of these proteins in the cheese aqueous extract increased from 2.5 to 20 fold in concentration from day 7 to day 69 of ripening.     

Implications of Lactobacillus collinoides and Brettanomyces/Dekkera anomala in phenolic off-flavour defects of ciders

Different Lactobacillus collinoides and Brettanomyces/Dekkera anomala cider strains were studied for their ability to produce volatile phenols in synthetic medium. All strains were able to produce 4-ethylcatechol (4-EC), 4-ethylphenol (4-EP) and 4-ethylguaiacol (4-EG) from caffeic, p-coumaric and ferulic acids, respectively. Interestingly, D. anomala and L. collinoides were also able to produce 4-EC, 4-EP and 4-EG in cider conditions. The quantities of ethylphenols produced by these two species were similar in both tested ciders. The impact of precursor quantities was studied and it showed that the addition of caffeic and p-coumaric acids in ciders allowed for higher 4-EC and 4-EP production by D. anomala and L. collinoides. In parallel, D. anomala and L. collinoides strains were isolated from a phenolic off-flavour defective bottled cider after ethylphenol production hence confirming the implication of these two species in this cider spoilage. Finally, detection thresholds of the main ethylphenols were determined in ciders by orthonasal and retronasal sampling. The 4-EC and 4-EP detection thresholds (close to 20-25mg/l and 1.5-2.0mg/l, respectively) were matrix dependant.     

Diversity and assessment of potential risk factors of Gram-negative isolates associated with French cheeses

The goal of this study was to identify at the species level a large collection of Gram-negative dairy bacteria isolated from milks or semi-hard and soft, smear-ripened cheeses (cheese core or surface samples) from different regions of France. The isolates were then assessed for two risk factors, antibiotic resistance and volatile and non-volatile biogenic amine production in vitro. In total, 173 Gram-negative isolates were identified by rrs and/or rpoB gene sequencing. A large biodiversity was observed with nearly half of all Gram-negative isolates belonging to the Enterobacteriaceae family. Overall, 26 different genera represented by 68 species including potential new species were identified among the studied Gram-negative isolates for both surface and milk or cheese core samples. The most frequently isolated genera corresponded to Pseudomonas, Proteus, Psychrobacter, Halomonas and Serratia and represented almost 54% of the dairy collection. After Pseudomonas, Chryseobacterium, Enterobacter and Stenotrophomonas were the most frequently isolated genera found in cheese core and milk samples while Proteus, Psychrobacter, Halomonas and Serratia were the most frequently isolated genera among surface samples. Antibiotic resistance profiles indicated that resistances to the aminosid, imipemen and quinolon were relatively low while more than half of all tested isolates were resistant to antibiotics belonging to the monobactam, cephem, fosfomycin, colistin, phenicol, sulfamid and some from the penam families. Thirty-six% of isolates were negative for in vitro biogenic amine production. Among biogenic amine-producers, cadaverine was the most frequently produced followed by isoamylamine, histamine and putrescine. Only low levels (<75 mg/l) of tyramine were detected in vitro.     

Synthesis and characterizations of poly(ethylene-co-vinylalcohol)-grafted-poly(3-hydroxybutyrate-co-hydroxyvalerate) copolymers

Poly(ethylene-co-vinylalcohol)-grafted-poly(3-hydroxybutyrate-co-hydroxyvalerate) (EVOH-g-PHBV) was prepared by a catalyzed grafting of PHBV onto immiscible EVOH. These reactions were performed at high temperature in the molten state. The choices of Tin(II) bis(2-ethylhexanoate) (Sn(Oct)₂) and 1,5,7-Triazabicyclo[4.4.0]dec-5-ene (TBD), as catalyst, were justified by a first study in an internal mixer. The catalyst quantity, reaction temperature and reaction time were optimized to apply them in the conditions of a reactive extrusion process with a co-rotating twin screw extruder. High reaction temperature, from 180 °C to 220 °C, associated to intensive mixing led to efficient grafting of PHBV onto EVOH, compatible with short residence times required.Rheological, thermal and morphological analyses were conducted to characterize the obtained copolymers. Molecular weights were determined by SEC. SEM imaging were made on cryo-fractured surfaces of blends extruded with and without catalyst. In most of reaction conditions, a compatible blend was obtained with a fine micro phase separation. Grafted copolymers were only obtained when well defined and controlled conditions are applied.     

Evaluation of a Microwave-Assisted Extraction Method for Lignan Quantification in Flaxseed Cultivars and Selected Oil Seeds

An optimized microwave-assisted extraction (MAE) method was evaluated through repeatability, recovery and efficiency testing. The repeatability tests, performed by three users over time, were not significantly different. The recovery of lignan throughout the extraction, preparation and analysis steps is 97.5% with a coefficient of variation <1%. The MAE method is efficient for extracting lignans from the plant matrix, and it achieves significantly higher extraction yields than the two established reference methods. The applicability of the MAE method was demonstrated by extracting lignans from a variety of plant samples. The secoisolariciresinol diglucoside (SDG) content of seven flaxseed cultivars grown in Saint-Mathieu-de-Beloeil, QC, in 2009 ranged from 14.6 to 18.9 mg SDG per gram of seed. Flax hulls produced in Winchester, ON, in 2010 were very rich in lignan; their SDG content was 40.0 mg/g of flax hull. Sesame seeds contained 0.18–1.89 mg SECO (aglycone of SDG) per gram of seed, with significant differences among black, white and brown sesame seed. Chia seeds contained 0.99–1.29 mg SECO per gram of seed, with significant differences among batches of seeds. Sunflower seeds had 0.046 mg SECO per gram of sample and almonds had 0.029 mg SDG per gram of sample. The optimized and evaluated MAE method is recommended for the general analytical quantification of lignans in plant samples.     

Crystallization of fluorescent quantum dots within a three-dimensional bio-organic template of actin filaments and lipid membranes

Biological molecules and molecular self-assemblies are promising templates to organize well-defined inorganic nanostructures. We demonstrate the ability of a self-assembled three-dimensional crystal template of helical actin protein filaments and lipids bilayers to generate a hierarchical self-assembly of quantum dots. Functionnalized tricystein peptidic quantum dots (QDs) are incorporated during the dynamical self-assembly of this actin/lipid template resulting in the formation of crystalline fibers. The crystal parameters, 26.5×18.9×35.5 nm3, are imposed by the membrane thickness, the diameter, and the pitch of the actin self-assembly. This process ensures the high quality of the crystal and results in unexpected fluorescence properties. This method of preparation offers opportunities to generate crystals with new symmetries and a large range of distance parameters.     

Original features of cell-envelope proteinases of Lactobacillus helveticus. A review

Lactobacillus helveticus is a lactic acid bacterium very used in fermented milks and cheese. The rapid growth of L. helveticus in milk is supported by an efficient cell envelope proteinase (CEP) activity, due to subtilisin-like serine proteases. These enzymes play also crucial roles in texture and flavor formation in dairy products as well as in generating in situ bioactive peptides. In L. helveticus, several genes encoding putative CEPs were detected and characterized by a large intraspecific diversity; little is known about regulation of expression of CEP-encoding genes. Anchored at the bacterial surface, CEPs are large-sized enzymes (> 150 kDa) hydrolyzing β- and α_s1-casein as well. Substrate cleavages occur after almost all types of amino acids residues, but mass spectrometry analysis revealed L. helveticus strains with specific profiles of substrate hydrolysis, which could explain identification of strains associated with interesting technological properties. In this review, the most recent data regarding CEP-encoding genes, CEP activities toward caseins and L. helveticus strain diversity are discussed.     

Kullback-Leibler Approach to CUSUM Quickest Detection Rule for Markovian Time Series

Abstract

Optimality properties of decision procedures are studied for the quickest detection of a change-point of parameters in autoregressive and other Markov type sequences. The limit of the normalized conditional log-likelihood ratios is shown to exist for Markov chains satisfying the ergodic theorem of information theory. Then closed-form expressions for this limit are derived by making use of the time average rate of Kullback-Leibler divergence. The good properties of the detection procedures based on a sequential analysis approach are proven to hold thanks to geometric ergodicity properties of the observation processes. In particular, the window-limited CUSUM rule is shown to be optimal for detecting the disruption point in autoregressive models. Sparre Andersen models are specifically studied.