Eva Van Derlinden

Effect of cell immobilization on heat-induced sublethal injury of Escherichia coli, Salmonella Typhimurium and Listeria innocua

The occurrence of sublethally injured cells in foods poses major public health concerns and is an essential aspect when assessing the microbial response to food preservation strategies, yet there is limited research dealing with its specific implications for mild heating. All available studies so far have been performed in broths colonized by planktonic cells, although their susceptibility to lethal agents has often been reported to be markedly different to the stress tolerance of cell colonies developed in solid foods. In this work, the effect of planktonic and colony growth, as well as the influence of colony density on sublethal injury induced by mild heating of Escherichia coli, Salmonella Typhimurium and Listeria innocua were assessed in food model systems. Detection of injured survivors relied on their inability to form visible colonies on salt-based selective media, which do not affect the growth of healthy cells. Sublethal injury (SI) increased rapidly with shorter exposure times and afterwards, decreased progressively, suggesting a mechanism of cumulative damage triggering lethal instead of SI. Cell arrangement affected the degree of SI, higher values being generally found for gelified systems, although the effect of colony density depended on the target microorganism. This information is essential for optimizing the design of food safety assurance systems.

Effect of cell immobilization on the growth dynamics of Salmonella Typhimurium and Escherichia coli at suboptimal temperatures

Predictive microbiology has recently acknowledged the impact of the solid(like) food structure on microbial behavior. The presence of this solid(like) structure causes microorganisms to grow as colonies and no longer planktonically as in liquid. In this paper, the growth dynamics of Salmonella Typhimurium and Escherichia coli were studied as a function of temperature, considering different growth morphologies, i.e., (i) planktonic cells, (ii) immersed colonies and (iii) surface colonies. For all three growth morphologies, both microorganisms were grown in petri dishes. While E. coli was grown under optimal pH and water activity (aw), for S. Typhimurium pH and aw were adapted to 5.5 and 0.990. In order to mimic a solid(like) environment, 5% (w/v) gelatin was added. All petri dishes were incubated under static conditions at temperatures in the range [8.0°C-22.0°C]. Cell density was determined via viable plate counting. This work demonstrates that the growth morphology (planktonic vs. colony) has a negligible effect on the growth dynamics as a function of temperature. The observation of almost equal growth rates for planktonic cultures and colonies is in contrast to literature where, mostly, a difference is observed, i.e., μplanktonic cells≥μimmersed colonies≥μsurface colonies. This difference might be due to shaking of the liquid culture in these studies, which results in a nutrient and oxygen rich environment, in contrast to the diffusion-limited gel system. Experiments also indicate that lag phases for solid(like) systems are similar to those for the planktonic cultures, as can be found in literature for similar growth conditions. Considering the maximum cell density, no clear trend was deducted for either of the microorganisms. This study indicates that the growth parameters in the suboptimal temperature range do not depend on the growth morphology. For the considered experimental conditions, models previously developed for liquid environments can be used for solid(like) systems.

Role of growth morphology in the formulation of NaCl-based selective media for injury detection of Escherichia coli, Salmonella Typhimurium and Listeria innocua

Sublethal injury (SI) poses major public health concerns since injured cells are responsible for serious limitations in food diagnostics and are susceptible to recovery, often developing adaptive stress responses. Detection of SI is based on the difference in plate counts between non-selective media, which represent the total cell population, and selective media, to which injured cells become sensitive. Selective media for detection of sublethal membrane damage are often based on NaCl supplement, although there is a lack of consensus in the literature about appropriate levels. Planktonic cells are generally used to investigate SI mechanisms, although they often exhibit different stress tolerance than cell colonies in/on solid food (model) systems. In this work, the effect of growth morphology, colony size and concentration of the gelling agent in the growth media, on the maximum non-inhibitory NaCl concentration in the plating medium was assessed for Escherichia coli, Salmonella Typhimurium and Listeria innocua. Stationary phase cultures of planktonic cells and large and small colonies grown in either 1.5% (w/v) xanthan gum-based system or 2.5% (w/v) xanthan gum-based system exhibited significantly different viable counts and osmotolerance. The effect of cell arrangement and xanthan gum percentage in the growth media depended on the microorganism under investigation. Additionally, differences in the maximum non-inhibitory concentration were evident, with 5.0% (w/v) NaCl for the Gram-negative bacteria and 6.5% (w/v), for L. innocua. Different extent of colony shrinkage and morphological damage was observed as NaCl concentration in the plating medium increased. This information will contribute to define NaCl-based selective media for accurate SI detection under realistic scenarios.

Robust Optimal Experiment Design: A Multi-Objective Approach

Abstract Optimal Experiment Design (OED) is an indispensable tool in order to reduce the amount of labour and cost intensive experiments in the modelling phase. The unknown parameters are often non-linearly present in the dynamic process models. This means that the Fisher Information Matrix also depends on the current guess for the parameters. In the early stage of the modelling phase these estimates are often highly uncertain. So designing an optimal experiment without taking this uncertainty into account is troublesome. In order to obtain an informative experiment, a robust optimisation approach is necessary. In recent work a formulation using an implicit weighted sum approach is proposed where the objective function is split in a nominal optimal experiment design part and a robust counterpart. This weighted sum has well known drawbacks in a Multi-Objective Optimisation approach. In this work these objectives are studied using advanced methods like the Normal Boundary Intersection and the Normalised Normal Constraint. In this way, the experimenter gets an overview of the different experiments possible. Furthermore, in past work the necessary third order derivatives are approximated using a finite different approach. The results in this work are obtained using exact third order and fourth order derivatives by exploiting the symbolic and automatic derivation methods implemented in the ACADO-toolkit.

Growth kinetics of listeria isolated from salmon and salmon processing environment: single strains versus cocktails.

The growth dynamics of Listeria monocytogenes strains isolated from salmon or a salmon processing environment and two reference Listeria innocua strains were investigated at refrigerated and close-to-optimal growth temperatures. Estimates for the growth rates and the lag-phase duration at 4, 8, 12, and 30°C were obtained for optical density measurements by using different growth parameter estimation methods, i.e., the serial dilution (SD) method and the relative rate to detection (RRD) method. Both single L. innocua and L. monocytogenes strains and mixtures of L. monocytogenes strains (cocktails) were studied. Both methods show an increase in maximum growth rate (μ(max)) of Listeria with increasing temperatures. Generally, single-strain growth rate estimates were quite similar for both species, although L. monocytogenes showed slightly higher μ(max) estimates at 4°C. The SD method gave the highest estimates for the growth rate, i.e., the estimates from the RRD method were 10 to 20% lower. This should lead to caution when using the latter method for Listeria, particularly at lower temperatures. Overall, the SD method is preferred as this method yields μ(max) estimates close to the biological value and provides estimates for the duration of lag time (λ). For discrimination between different strains, λ appeared to be a more suitable parameter than μ(max). This effect was most prominent for L. innocua. Significant differences were observed between μ(max) and/or λ of L. monocytogenes cocktails and single strains at all temperatures investigated. At 4°C, the average growth rate of cocktails was higher than that of single strains. At 8 and 30°C, this trend was reversed. The average λ of single strains were more than twice as long as those of cocktails at 4°C. At 8 and 30°C, the λ of cocktails were significantly slower than those of single strains, but the variation was considerably less and the differences were less pronounced.

Multi-objective optimisation approach to optimal experiment design in dynamic bioprocesses using ACADO toolkit

Abstract Mathematical models are valuable tools for optimizing dynamic biochemical processes. However, experimental data collection is often labour and cost intensive and can give rise to production losses. The current paper studies the trade-offs between objectives for production and optimally designing experiments in view of parameter estimation for a bioreactor benchmark. Recent deterministic multi-objective optimal control approaches (i.e., the freeware toolkit ACADO Multi-objective www.acadotoolkit.org ) are used to efficiently produce the set of trade-off or so-called Pareto optimal solutions. These trade-offs are clearly reflected when the obtained optimal control solutions are exploited to estimate the parameters from virtual experiments, while also trying to maximise the biomass production.

Approximate robust Optimal Experiment Design in dynamic bioprocess models

In dynamic bioprocess models parameters often appear in a nonlinear way. When designing optimal experiments to calibrate these models, the Fisher Information Matrix explicitly depends on the current parameter estimates. Hence, it is advisable to take this parametric uncertainty into account in the design procedure in order to obtain an experiment which is robust with respect to changes in the parameters. The current paper applies computationally efficient approximate robustification strategies based on a worst case scenario. Both methods exploit linearisation techniques to avoid the hard to solve max-min optimisation problems. The methods will be illustrated on a predictive microbiology case study.

Evaluating The Trade-offs In Optimal Experiment Design Using A Multi-Objective Optimisation Approach

Abstract Dynamic process models are widely used for operating, controlling and optimising important bioprocesses, e.g., pharmaceuticals, enzyme production and brewing. After selection of an appropriate process model structure, parameter estimates have to be obtained based on real-life experiments. To reduce the amount of labour and often cost intensive experiments Optimal Experiment Design (OED) is an indispensable tool. In Optimal Experiment Design for parameter estimation a scalar measure of the Fisher Information Matrix is used as an objective function. Over the years different criteria have been developed. These criteria may be competing as they each have a slightly different objective. For systematically evaluating the competing nature and to improve the parameter estimation procedure, a multi-objective optimisation approach is selected. To solve the multi-objective dynamic optimisation problems efficiently ACADO Multi-Objective ( www.acadotoolkit.org ) has been employed, which is a flexible toolkit for solving dynamic optimisation or optimal control problems with multiple and conflicting objectives.

Parameter accuracy vs. decorrelation in optimal experiment design: a multi-objective point of view

Abstract Dynamic process models are useful tools for operating, controlling and optimising bioprocesses. To obtain accurate parameter estimates often labour and cost intensive experiments have to be performed. Optimal Experiment Design (OED) techniques enable limiting the experimental burden. In OED, typically a scalar measure of the Fisher information matrix is used as objective to be optimised. Although several criteria exist, most fail to take correlation into account. To tackle this issue several criteria have been introduced recently. The approach used was to set, e.g., the information content as a constraint when correlation is minimised or vice versa. However, this procedure does not allow for a systematic evaluation of intrinsic trade-offs. In this work the trade-off between correlation and information content is studied systematically in a multi-objective approach. This approach is illustrated on a fed-batch benchmark problem.