Anastasia E. Kapetanakou

Development of a model describing the effect of temperature, water activity and (gel) structure on growth and ochratoxin A production by Aspergillus carbonarius in vitro and evaluation in food matrices of different viscosity.

The present study aimed: (i) to develop models for the combined effect of water activity (0.99, 0.94 and 0.90), microstructure expressed as 0, 5, 10 and 20% w/v gelatin, and temperature (15, 20 and 25 °C), on growth and OTA production rates by Aspergillus carbonarius; and (ii) to evaluate the performance of the developed models on food matrices (jelly, custard and marmalade) of different viscosity at pH 5.5. The square root of biomass increase rate (fungal growth rate) and OTA production rate were determined by the Baranyi model and were further modeled as a function of temperature, gelatin concentration and a(w) by applying polynomial models. Time for visible growth and the upper asymptote of the OTA production curve were also determined by the Baranyi model. Increase in gelatin concentration resulted in a significant delay in all parameters describing fungal growth and OTA production rates, at all temperatures and a(w). The effect of microstructure on fungal growth and OTA production rates was less evident at stress conditions of a(w) and temperature. Detection time for visible fungal growth was markedly influenced by a(w) and temperature. Coefficients of determination were 0.899 and 0.887 for the models predicting the square root (√μ(max)) of growth and OTA production rate, respectively. Predictions of growth rate agreed well with the recorded data of custard and marmalade, while observations of OTA production rate indicated low agreement with model predictions, in all food matrices except for marmalade. The present findings may provide a basis for reliable assessment of the risk of fungal growth and OTA production in foods of different structural and rheological properties.

Inhibition of A. carbonarius growth and reduction of ochratoxin A by bacteria and yeast composites of technological importance in culture media and beverages.

Five composites of yeast and six of bacterial isolates from fermented products were studied, in order to assess their ability to inhibit Aspergillus carbonarius growth and reduce OTA concentration in culture media and beverages. The antagonistic effect of the above composites against A. carbonarius growth was studied in synthetic grape medium of pH 3.5 and a(w) 0.98, 0.95, 0.92 after incubation at 25°C. Different combinations of initial inocula of bacteria or yeast composites and fungi were used (10(2)cfu/mL vs 10(5)spores/mL; 10(5)cfu/mL vs 10(2)spores/mL; and 10(5)cfu/mL vs 10(5)spores/mL). Regarding the OTA reduction experiment, 10(3) and 10(7)cfu/mL of the bacteria and yeast composites were inoculated in liquid media of different pH (3.0, 4.0, 5.0, and 6.1 or 6.5) and initial OTA concentration (50 and 100μg/L) and incubated at 30°C. Moreover, grape juice, red wine, and beer were supplemented with 100μg/L of OTA and inoculated with composites of 16 yeasts (16YM) and 29 bacterial (29BM) strains (10(7)cfu/mL) to estimate the kinetics of OTA reduction at 25°C for 5days. Fungal inhibition and OTA reduction were calculated in comparison to control samples. None of the bacterial composites inhibited A. carbonarius growth. The high inoculum of yeast composites (10(5) cfu/mL) showed more efficient fungal inhibition compared to cell density of 10(2) cfu/mL. All yeast composites showed higher OTA reduction (up to 65%) compared to bacteria (2-25%), at all studied assays. The maximum OTA reduction was obtained at pH 3.0 by almost all yeast composites. For all studied beverages the decrease in OTA concentration was higher by yeasts (16YM) compared to bacteria (29BM). The highest OTA reduction was observed in grape juice (ca 32%) followed by wine (ca 22%), and beer (ca 12%). The present findings may assist in the control of A. carbonarius growth and OTA production in fermented foodstuffs by the use of proper strains of technological importance.

Comparison of polymerase chain reaction methods and plating for analysis of enriched cultures of Listeria monocytogenes when using the ISO11290-1 method

Analysis for Listeria monocytogenes by ISO11290-1 is time-consuming, entailing two enrichment steps and subsequent plating on agar plates, taking five days without isolate confirmation. The aim of this study was to determine if a polymerase chain reaction (PCR) assay could be used for analysis of the first and second enrichment broths, saving four or two days, respectively. In a comprehensive approach involving six European laboratories, PCR and traditional plating of both enrichment broths from the ISO11290-1 method were compared for the detection of L. monocytogenes in 872 food, raw material and processing environment samples from 13 different dairy and meat food chains. After the first and second enrichments, total DNA was extracted from the enriched cultures and analysed for the presence of L. monocytogenes DNA by PCR. DNA extraction by chaotropic solid-phase extraction (spin column-based silica) combined with real-time PCR (RTi-PCR) was required as it was shown that crude DNA extraction applying sonication lysis and boiling followed by traditional gel-based PCR resulted in fewer positive results than plating. The RTi-PCR results were compared to plating, as defined by the ISO11290-1 method. For first and second enrichments, 90% of the samples gave the same results by RTi-PCR and plating, whatever the RTi-PCR method used. For the samples that gave different results, plating was significantly more accurate for detection of positive samples than RTi-PCR from the first enrichment, but RTi-PCR detected a greater number of positive samples than plating from the second enrichment, regardless of the RTi-PCR method used. RTi-PCR was more accurate for non-food contact surface and food contact surface samples than for food and raw material samples especially from the first enrichment, probably because of sample matrix interference. Even though RTi-PCR analysis of the first enrichment showed less positive results than plating, in outbreak scenarios where a rapid result is required, RTi-PCR could be an efficient way to get a preliminary result to be then confirmed by plating. Using DNA extraction from the second enrichment broth followed by RTi-PCR was reliable and a confirmed result could be obtained in three days, as against seven days by ISO11290-1.

Storage of pork meat under modified atmospheres containing vapors from commercial alcoholic beverages.

The present study aimed to evaluate the effect of AB vapors on microbial, physicochemical, and sensory profile of pork meat stored in different MAP conditions. Pork pieces (10g) and cotton/cellulose absorbent cloths (2×2cm) were placed into compartmentalized Petri-dishes in two sections. Aliquots (1mL) of water (control), 30% v/v and 40% v/v ethanol, whisky, brandy, tsipouro, raki, and ouzo were added separately to the cotton/cellulose absorbent cloths. Each pork sample was placed in one compartment and cotton/cellulose absorbent cloths supplemented with different ABs were placed in a separate compartment of each Petri-dish. Samples were packaged in 40% CO2: 30% O2: 30% N2 and 80% O2: 20% CO2 and stored at 4 and 10°C. Total viable counts, Pseudomonas sp., Brochothrix thermosphacta, lactic acid bacteria, yeasts and molds, and Enterobacteriaceae, were enumerated during storage. Changes in pH, color (L*, a*, b*), odor, taste, and overall appearance of pork meat were also evaluated along with changes in organic acid levels via HPLC. At 4°C, lactic acid bacteria and B. thermosphacta were the dominant organisms under 40% CO2: 30% N2: 30% O2 and 80% O2: 20% CO2, respectively, while at 10°C, lactic acid bacteria dominated in both MAP conditions. All applied ABs were effective (p<0.05) against lactic acid bacteria, pseudomonads, and B. thermosphacta. The inhibitory effect of ABs was also reflected through lower levels of glucose consumption or accumulation of lactic, acetic, succinic, and formic acid compared to controls. Moreover, packaged samples in 40% CO2: 30% O2: 30% N2 exhibited a significant increase (p<0.05) of acetic acid during storage at 4°C, but the concentrations of acetic acid in samples exposed to AB vapors were lower than those in controls. Both antimicrobial active MAPs extended the shelf-life of pork meat by ca. 2-fold, while samples exposed to alcoholic beverages (especially ouzo) under 80% O2: 20% CO2 resulted in better (p<0.05) sensory properties compared to the respective samples under 40% CO2: 30% O2: 30% N2. Overall, vapor action of ABs in combination with MAP may constitute a promising, antimicrobial packaging technology for extending the shelf-life of pork meat.

Assessing the capacity of growth, survival, and acid adaptive response of Listeria monocytogenes during storage of various cheeses and subsequent simulated gastric digestion

Different physicochemical and microbiological characteristics of cheeses may affect Listeria monocytogenes potential to grow, survive, or exhibit an acid adaptive response during storage and digestion. The objectives of the present study were to assess: i) the survival or growth potential of L.monocytogenes on various cheeses during storage, ii) the effect of initial indigenous microbiota on pathogen growth in comparison to expected growth curves retrieved by existing predictive models, and iii) the impact of habituation on/in cheeses surfaces on the subsequent acid resistance during simulated gastric digestion. Portions of cream (Cottage and Mascarpone), soft (Anthotyros, Camembert, Mastelo®, Manouri, Mozzarella, Ricotta), and semi-hard (Edam, Halloumi, Gouda) cheeses were inoculated with ca. 100CFU/g or cm2 of L.monocytogenes and stored under vacuum or aerobic conditions at 7°C (n=4). The impact of varying (initial) levels of starter culture or indigenous spoilage microbiota on pathogen growth was evaluated by purchasing cheese packages on different dates in relation to production and expiration date (subsequently reflecting to different batches) mimicking a potential situation of cheese contamination with L.monocytogenes during retail display. Values of pH and aw were also monitored and used to simulate growth of L. monocytogenes by existing models and compare it with the observed data of the study. Survival in simulated gastric fluid (SGF) (pH1.5; HCl; max. 120min) was assessed at three time points during storage. Mascarpone, Ricotta, Mozzarella, Camembert, and Halloumi supported L.monocytogenes growth by 0.5-0.8logCFU/g or cm2per day, since low initial levels of total viable counts (TVC) (1.8-3.8logCFU/g or cm2) and high pH/aw values (ca. 6.23-6.64/0.965-0.993) were recorded. On Cottage, Anthotyros, Manouri, Mastelo®, Edam, and Gouda, the pathogen survived at populations similar or lower than the inoculation level due to the high reported competition and/or low pH/aw during storage. L. monocytogenes growth was significantly suppressed (p<0.05) on samples purchased close to expiration date (bearing high TVC), compared to those close to production date, regardless of cheese. Cheeses which supported growth of L.monocytogenes enabled higher survival in gastric acidity along their shelf-life compared to cheeses which did not support growth. However, even in the latter cheeses (i.e., Cottage, Mastelo®, Gouda), total elimination of a persisting low initial contamination was not always achieved. Such findings may provide useful evidence for assessing the risk posed by various cheeses types in relation to their compliance with food safety regulations.

Alginate-Based Edible Films Delivering Probiotic Bacteria to Sliced Ham Pretreated with High Pressure Processing

The aim of the present work was to evaluate the efficacy of Na-alginate edible films as vehicles for delivering probiotic bacteria to sliced ham with or without pretreatment using high pressure processing (HPP). Three strains of probiotic bacteria were incorporated in Na-alginate forming solution. Ham slices (with or without pretreatment using HPP at 500 MPa for 2 min) were packed under vacuum in contact with the films and then stored at 4, 8 and 12 °C for 66, 47 and 40 days, respectively. Microbiological analysis was performed in parallel with pH and color measurements. Sensory characteristics were assessed, while the presence and the relative abundance of each probiotic strain during storage was evaluated using pulsed field gel electrophoresis. In ham slices without HPP treatment, probiotic bacteria were enumerated above 106 CFU/g during storage at all temperatures. Same results were obtained in cases of HPP treated samples, but pH measurements showed differences with the latter ones exhibiting higher values. Sensory evaluation revealed that probiotic samples had a more acidic taste and odor than the control ones, however these characteristics were markedly compromised in samples treated with HPP. Overall, the results of the study are promising since probiotic bacteria were successfully delivered in the products by edible films regardless of the HPP treatment.

Sodium alginate–cinnamon essential oil coated apples and pears: Variability of Aspergillus carbonarius growth and ochratoxin A production

The scope of the present study was to use selected fruits as model foods (wounded skin or slices of apples and pears), for the in situ assessment of the potential of natural antimicrobials to control fungal growth and OTA production and the investigation of alternative ways of their application, e.g., via edible coatings. Fresh fruits were cut: i) in halves or ii) across in round slices of ca. 1 cm thickness. Wounds were introduced into the skin and the center of the slice (5 mm deep; 4 mm diameter) and inoculated with a range of 2.0-7.5 × 103spores per wound of Aspergillus carbonarius. Following inoculation, samples were coated with Na - alginate supplemented with 0.3 (0.3% ECC) and 0.9% v/v (0.9% ECC) cinnamon ΕΟ. Inoculated samples without edible coating and EO (C) or with edible coating and without EO (EC) were used as negative or positive controls, respectively. All samples were stored under aerobic conditions at 15, 20, and 25 °C. Fungal growth was estimated by colony diameter measurements (n = 30), while OTA production was determined by HPLC (n = 4). Antimicrobial treatment with 0.9% EO was more effective on fungal growth when the inoculation took place on slices than in wounded skin (p < .05), regardless of storage temperature and fruit. The variability of μmax increased with EO concentration, except for the coated slices of apples with 0.9% v/v EO (at all temperatures), or pears with 0.3% and 0.9% v/v EO (at 15 and 20 °C), where no growth was observed. OTA was below the detection limit (1 ppb) on the majority of 0.9% ECC apples slices and in 0.3% ECC and 0.9% ECC pears slices, stored at 20 and 25 °C. However, the sample to sample variation in the produced amounts of OTA was remarkable. Thus, considering that inhibition of growth and toxin production do not always concur, the present study provided quantitative information on the variability in A. carbonarius growth and OTA production in real model foods in response to antimicrobial coating with natural active compounds. Such data could be of relevance to risk assessment and assist in designing effective control strategies for limiting OTA levels in foods and thus, protecting consumer health.

Using the gamma concept in modelling fungal growth: A case study on brioche-type products

Fungi are common spoilers of intermediate moisture foods such as bakery products. Brioche are bakery products prone to fungal spoilage due to their pH (5.8-6.2) and water activity (aw) (0.82-0.84). The aims of the present study were: (i) the identification of fungal species occurring in brioche products, (ii) the in vitro assessment of their growth potential, and (iii) the development of a validated growth model following the gamma concept. A total of 102 fungal strains were isolated, with Penicillium sp., Cladosporium sp., and Aspergillus sp. being the main genera, representing 90% of the isolates. Given the isolation frequency, any potential fungal prevalence throughout the bakery processs and/or the results of in vitro assessment of fungal growth potential under conditions mimicking brioche (pH, aw, temperature), Aspergillus flavus, Aspergillus fumigatus, and Penicillium sp. were selected for the development of the gamma model. According to in vitro validation, the model successfully predicted fungal growth, while on in situ experiments, the intrinsic parameters (aw and/or level of used preservative) of brioche in combination with packaging conditions (modified atmosphere) did not allow fungal growth.