Shelf life definition for italian anchov

Innovative Food Science and Emerging Technologies 16 (2012) 171–180 Contents lists available at SciVerse ScienceDirect Innovative Food Science and Emerging Technologies journal homepage: www.elsevier.com/locate/ifset Shelf life definition for Italian anchovies inoculated with Lactobacillus plantarum and Bifidobacterium animalis subsp lactis Barbara Speranza, Antonio Bevilacqua, Milena Sinigaglia, Maria Rosaria Corbo ⁎ Department of Food Science, Faculty of Agricultural Science and Food Quality and Health Research Center, University of Foggia, Via Napoli 25, 71100 Foggia, Italy a r t i c l e i n f o Article history: Received 26 January 2012 Accepted 22 May 2012 Editor Proof Receive Date 15 June 2012 Keywords: Non dairy functional foods Probiotics Fish products Shelf life a b s t r a c t This paper reports on the optimization of marinated anchovies, inoculated with two probiotic strains (Lactobacillus plantarum and Bidifobacterium animalis subsp lactis) The research was divided into three phases: a) choice of the hurdles for the marinating brine; b) brine optimization; and c) production of functional anchovies After studying the effects of pH and NaCl as typical hurdles of brine, the effect of citrus extract towards the probiotics was assessed; this antimicrobial agent was chosen as an additional hurdle for product storage Process optimization, conducted through the use of a centroid design, revealed that the optimal composition for marinating brine was the following: 2% of acetic acid, 10% of NaCl and 200 ppm of citrus extract Then, anchovies were inoculated with probiotics, marinated, packed under different conditions (air, vacuum, in oil and in a diluted brine) and stored at °C During 21 days, analyses were performed to evaluate quality attributes (odor, color, texture and overall quality) as well as the survival of the tested strains The limiting factors for shelf life definition resulted to be sensory scores (odor in particular); the best sample was that packed in diluted brine, as it retained the sensory quality at the maximum level for at least 6–10 days and showed a shelf life longer than 21 days Industrial relevance: This paper proposes a new kind of functional food, based on marinated anchovies The new non-dairy functional food may both contribute to a regular assumption of probiotics in individuals with lactose intolerance or with a diet lacking milk-derived products and to expand the range of health-promoting foods By increasing the amount of functional foods in the diet, will be easier for consumers to achieve the optimal supply of probiotics able to guarantee a healthy bacterial gut flora Moreover, anchovy is a low cost product in Southern Italy; therefore the preparation of a functional dish with this kind of raw material could be a promising way for small-scale producers and very attractive for consumers, due to mild treatments used for its stabilization © 2012 Elsevier Ltd All rights reserved Introduction Nowadays, the development and consumption of functional probiotic foods have been increasing alongside awareness of their beneficial effects in promoting gut health and preventing diseases The contribution of probiotic bacteria, mainly lactobacilli and bifidobacteria, to provide health effects is well documented and numerous investigations have underlined the benefits deriving from a regular intake of foods fortified with this kind of microorganisms (Cross, 2002; Gotcheva et al., 2002; Imasse et al., 2007; Marteau, de Vrese, Cellier, & Schrezenmeir, 2001; Nomoto, 2005; Shah, 2007) These aspects have raised consumer interest in foods enriched with health-promoting bacteria so stimulating food manufacturers to include these strains in foods which are part of a normal diet in order to provide health defense while enjoying meals and to differentiate such functional products from concentrated probiotic preparations available as capsules, powders, or liquids On the market, these health-promoting foods are mainly produced with ⁎ Corresponding author Tel: + 39 0881 589232; fax: + 39 0881 589231 E-mail address: m.corbo@unifg.it (M.R Corbo) 1466-8564/$ – see front matter © 2012 Elsevier Ltd All rights reserved doi:10.1016/j.ifset.2012.05.009 fresh milk or milk derivatives such as yogurt, cheese, ice cream, and desserts Since to colonize the gastrointestinal tract, probiotic strains need to be ingested as large populations and on a daily basis (Reid, Howard, & Gan, 2001), functional food industries are now focusing on new non-dairy foods that can both contribute to a regular assumption of probiotics in individuals with lactose intolerance or with a diet lacking milk-derived products and to expand the range of health-promoting foods By increasing the amount of functional foods in the diet, it will be easier for consumers to achieve the optimal supply of probiotics able to guarantee a healthy bacterial gut microbiota At present, there are already some functional non-dairy products available on the market, generally probiotic fruit juices or drinks (Rivera-Espinoza & Gallardo-Navarro, 2010) However, there are several studies on the development of alternative probiotic foods such as sausages fermented by Lactobacillus rhamnosus strains, fruit pieces containing Bifidobacterium spp., dried fruits vacuum-impregnated with Saccharomyces cerevisiae and Lactobacillus casei subsp rhamnosus, soy milk fermented with Bifidobacterium breve, an oat-based cereal bar including Bifidobacterium lactis, table olives enriched with L rhamnosus, Lactobacillus paracasei, Bifidobacterium bifidum, and Bifidobacterium longum (Rivera-Espinoza & Gallardo-Navarro, 2010) Probiotic bacterium L paracasei IMPC2.1 172 B Speranza et al / Innovative Food Science and Emerging Technologies 16 (2012) 171–180 applied to table olives or artichokes has even received marketing authorization from the Italian Ministry of Health (Lavermicocca et al., 2009) Despite the aforementioned research about the development of new probiotic non-dairy foods, at the best of our knowledge, the suitability of fish products as a substrate for growing probiotic bacterial strains has not yet been investigated The incorporation of probiotic bacteria into fish products, apart from being a novelty, would add functional features to their already high nutritional value In fact, in addition to proteins and trace elements (particularly selenium), the high nutritional value of fish is mainly due to their lipid composition: fish contain high levels of polyunsaturated fatty acids (PUFAs), mainly eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), which are recognized to protect against cardiovascular diseases, to prevent or delay the clinical manifestations of certain cancers and to alleviate some affective and psychiatric disorders (Sidhu, 2003) Among fish products, marinades consist in portions of fish processed by treatment with an edible organic acid, usually acetic acid, and salt and put into brines, sauces, creams or oil (Meyer, 1965) Marinades are semipreserves; acid and salt retard the action of bacteria and enzymes, resulting in a product with a characteristic flavor and an extended but limited shelf life (McLay, 1972) Besides the prevention of microbial growth, marination is also used to tenderize or to change taste, textural and structural properties of raw material These marinades are typically consumed as ready-to-eat products with no heat treatment, thus representing a potential vehicle for probiotics In particular, the anchovy (Engraulis encrasicholus) is a pelagic species belonging to the Engraulidae family, easily found in the Mediterranean off the coasts of France, Italy and Spain, as well as off the Atlantic coasts of Europe, generally consumed marinated and sold packed in oil or brine This study was aimed to broaden the range of functional food types by the development of a tasty functional food consisting of marinated anchovies fortified with probiotic lactic acid bacteria In a first phase, assays were conducted to optimize brine composition designed to vehicle microorganisms, then functional fish products were realized and microbiological and sensorial analyses were performed to evaluate their quality attributes as well as the survival of the tested strains Materials and methods 2.1 First experimental phase: choice of the hurdles of the marinating brine 2.1.1 Microorganisms The strains used in this study were Lactobacillus plantarum (c19), isolated from Italian table olives Bella di Cerignola and belonging to the Collection of the Department of Food Science of University of Foggia (Bevilacqua, Altieri, Corbo, Sinigaglia, & Ouoba, 2010) and Bifidobacterium animalis subsp lactis (10140), purchased from a Public Collection (Deutsche Sammlung von Mikroorganismem und Zellkulturen's collection, Braunschweig, Germany, DSMZ) Before each experiment, the microorganisms were grown in MRS broth (Oxoid CM359B, Milan, Italy) and incubated at 37 °C for 48 h under anaerobic conditions, in order to attain a cell concentration of log cfu/ml 2.1.2 Effect of acetic and lactic acids These preliminary assays were performed in MRS broth, adjusted at different pH values (3.5, 4.0, 4.5) through acetic acid (Baker, Milan, Italy) or lactic acid (Baker, Milan, Italy) The samples were inoculated with log cfu/ml of each strain separately and incubated at 37 °C in aerobic conditions Aliquots of not-modified MRS broth (pH 5.8), also inoculated and incubated a 37 °C, were used as controls Microbial population was evaluated after 24 and 48 h through direct count on appropriate media: MRS Agar (MRSA, Oxoid) for L plantarum c19, and MRSA plus filter-sterilized NPNL antibiotic solution (NPNL– MRSA) for B animalis 10140, both incubated at 37 °C for 48 h under anaerobic conditions The antibiotic solution contained mg/l neomycin sulfate (Sigma-Aldrich, Milan, Italy), 10 mg/l paromomycine sulfate (Sigma-Aldrich), 750 mg/l nalidixic acid (Sigma-Aldrich) and 150 mg/l lithium chloride (Sigma-Aldrich) (Vinderola & Reinheimer, 1999) Experiments were performed twice; results were expressed as decrease of cell number referred to the control and plotted as a function of the pH of the medium 2.1.3 Antimicrobial activity of citrus extract The assays were performed in MRS broth, added with different concentrations of a natural antimicrobial agent (100, 200, 300, 400 and 500 ppm) The antimicrobial compound chosen was a citrus extract (GFSE, Probena s.l, Zaragoza, Spain) that showed a good effectiveness in slowing down the microbial quality decay of refrigerated fish products (Corbo et al., 2008, 2009; Del Nobile et al., 2009) The samples were inoculated with log cfu/ml of each strain separately and incubated at 37 °C under aerobic conditions Aliquots of not-modified MRS broth (0 ppm of citrus extract), also inoculated and incubated a 37 °C, were used as controls Microbial population was evaluated after 24 and 48 h through direct count on appropriate media, as described previously Experiments were performed twice; results were expressed as decrease of cell number referred to the control and plotted vs the concentration of the antimicrobial agent 2.2 Second experimental phase: brine optimization 2.2.1 Experimental design The optimization of the conditioning brine/liquid for the storage of functional fish products was performed through a mixture design, called simplex centroid design: this kind of design involves different variables; each variable is usually set at different levels, identified with the code (minimum), (maximum), 0.5 (half point of the range) and 0.33 (1/3 of the maximum) (Bevilacqua, Corbo, & Sinigaglia, 2010a; Bevilacqua & Sinigaglia, 2010) For the purpose of this research, the independent variables of the design were pH, NaCl and citrus extract; NaCl and citrus extract were included in the range 0–12% and 0–300 ppm, respectively On the other hand, the pH was adjusted from 3.5 to 5; only for this variable an inverse correlation real vs coded values was used, i.e the code (minimum) was imposed to 5.0 and the code (maximum) to 3.5 Table reports the combinations of the centroid and the control, i.e a further combination where the pH was set to 6.0 and NaCl and citrus extract to 0.0% and ppm, respectively In particular, two mixture designs were prepared, as the pH was adjusted with either acetic acid or lactic acid 2.2.2 Samples preparation Agar diskettes (2% w/v) were used as model system to simulate fish filets Prior to use, cultures of the target probiotic bacteria were grown in MRS broth at 37 °C under anaerobic conditions After 24 h, Table Combinations of centroid Coded values Control Real values pH NaCl Citrus extract (ppm) pH NaCl (%) Citrus extract (ppm) 0 0.50 0.50 0.33 – 0.50 0.50 0.33 – 0 0.50 0.50 0.33 – 3.5 5.0 5.0 4.25 4.25 5.0 4.0 6.0 12 6 0 300 150 150 100 B Speranza et al / Innovative Food Science and Emerging Technologies 16 (2012) 171–180 a cocktail of the two strains was prepared by mixing 50 ml of each culture Inoculum for experiments was prepared by centrifugation of the obtained culture in an ALC 4239R centrifuge at 1000 ×g for 15 at °C The pellet was re-suspended in 100 ml of sterile physiological solution (0.9% NaCl) temperated at °C and the resulting cell suspension solution (9 log cfu/ml for each microorganism) was used as inoculum for test tubes containing agar solution (2%), in order to achieve a final concentration of log cfu/ml After agitation, the inoculated agar was put into a Petri plate and allowed to dry for h Then, to each inoculated diskette was added a different marinating brine (1:1.5), whose composition varied according to the experimental design (Table 1) After 24 and 48 h, viable probiotic population was evaluated through direct count on appropriate media: more specifically, agar diskettes (20 g) were diluted with 180 ml of sterile physiological solution (0.9% NaCl) in a Stomacher bag (Seward, London, England) and homogenized for in a Stomacher Lab Blender 400 (Seward) After opportune serial dilutions of homogenates, MRSA (Oxoid) for L plantarum c19 and NPNL–MRSA for B animalis 10140 were used, both incubated at 37 °C for 48 h under anaerobic conditions Each experiment was performed twice 2.2.3 Modeling For each time of analysis (24 and 48 h), microbiological data were modeled as follows: Δ log N ¼ Nc −Ns where Nc and Ns are cell numbers (log cfu/g) in the control and in each of the combination of centroid, thus attaining two sets of data (i.e ΔlogN24 and ΔlogN48, corresponding to the decrease of cell counts after 24 and 48 h) The standardized indices ΔlogN24 and ΔlogN48 were used as the input values to build a polynomial equation, through the software Statistica for Windows: y¼ X i¼1 i xi ỵ X X i j ij xi xj ỵ X X X i j ijk xi xj xk ỵ 1ị k where βi, βij and βijk are the coefficients of the individual (xi) and interactive effects (xixj − xixjxk) of the independent variables (pH, NaCl and citrus extract); ε is the standard error of the model Pareto chart of standardized effects was used to point out if a variable was significant or not Standardized effects were evaluated as the ratio of the mathematical coefficient of each term of the equation vs its standard error; if the effect overcame the significance breakpoint (P b 0.05), corresponding to the vertical line, it was regarded as significant and included into the equation The effect of each independent variable on the decrease of cell count was evaluated through the individual desirability functions, estimated as follows: y ≤ y < 0; d ¼ ðy À y Þ=ðy max Ày Þ ymin ≤ y≤ y max : 1; y ≥ y max ð2Þ where ymin and ymax are the minimum and maximum values of the dependent variable, respectively For each value of the pH, NaCl and citrus extract, the desirability was included in the range 0–1 The desirability function was set to (the worst result), when the decrease of cell number was maximum; otherwise the desirability function was set to 1, when no effect of the conditioning solution was observed The desirability of each variable was evaluated through a mathematical device, i.e the evaluation of each variable at a time and the exclusion of the other two 173 factors by imposing some constant values as follows: pH, 3.5; NaCl, 0%; and citrus extract, ppm 2.3 Third experimental phase: functional fish product realization 2.3.1 Raw material and microorganisms Anchovy from Adriatic Sea (E encrasicholus) was purchased from fisherman in the Gulf of Manfredonia (Foggia, Italy) The fishes were directly transferred to the laboratory in polystyrene boxes within h after purchase Then, fishes were decapitated, eviscerated, fileted and washed The probiotic strains used were L plantarum (c19) and B animalis subsp lactis (DSMZ10140) 2.3.2 Sample preparation All the fish samples were immersed into a marinating brine consisting of 2% (v/v) acetic acid, 10% NaCl (w/v) and 200 ppm of citrus extract The ratio of fish to solution was 1:1.5 (w/v) The immersing process was performed at °C and completed within 30 h; then, the marinated fish was removed from the solution, washed with tap water and packed In particular, four different packaging conditions were performed, as detailed in the following: • PB1, fish filets were inoculated with probiotics and packed in plastic jars (with lid), under air, without brine; • PB2, fish filets were packed in plastic jars filled with extra virgin olive oil inoculated with probiotics; • PB3, fish filets were packed in plastic jars filled with diluted brine (1:10; 0.2% acetic acid, 1% NaCl and 20 ppm of citrus extract) inoculated with probiotics; • PB4, fish filets were inoculated with probiotics and packed under vacuum in high-barrier plastic bags [Nylon/Polyethylene, 102 μm (Tecnovac, San Paolo D'Argon, Bergamo, Italy)] by means of S100Tecnovac equipment The bags were 170 mm×250 mm long with properties specified by the manufacturer as follows: CO2 and O2 permeability of 3.26×10− 19 mol m/m2 s Pa and 9.23×10− 19 mol m/m2 s Pa respectively and water vapor transmission rate of 1.62×10− 10 kg/m2 s For PB1 and PB4 samples, the probiotic cocktail was prepared as previously described: more specifically, a cocktail of the two strains was prepared by mixing 50 ml of each 24 h-culture and inoculum was obtained by centrifugation The pellet was re-suspended in 100 ml of sterile physiological solution (0.9% NaCl) temperated at °C and the resulting cell suspension solution (9 log cfu/ml for each microorganism) was used as inoculum for PB1 and PB4 samples (~7 log cfu/g) For PB2 and PB3 samples, the microbial pellet was instead re-suspended in oil and diluted brine, respectively The obtained solutions were used to fill jars containing anchovies (ratio fish/solution, 1:1.5), to reach a final probiotic concentration of log cfu/ml For each sample a control was also prepared, i.e fish product without probiotics The samples were stored at °C for 21 days during which microbiological, sensorial analyses and determination of pH were periodically made; details of analyses are given below All experiments were performed twice 2.3.3 Microbiological analyses and determination of pH For PB2 and PB3 samples, anchovy filets were aseptically removed from oil and brine, respectively; the excess liquid was eliminated For microbiological analyses, filets (25 g) were diluted with 225 ml of sterile physiological solution (0.9% NaCl) in a Stomacher bag (Seward) and homogenized for in a Stomacher Lab Blender 400 (Seward) Serial dilutions of fish homogenates were plated on the surface of the appropriate media in Petri dishes The media and the conditions used were: Plate Count Agar (PCA) incubated at 30 °C for 48 h and at °C for a week under aerobic conditions, for aerobic plate count (APC) and psychrotrophic bacteria (PB); pour plated Iron Agar (IA), incubated at 25 °C for three days in aerobic conditions, for mesophilic hydrogen sulfide producing bacteria (mSSO); spread plated chilled IA, supplemented B Speranza et al / Innovative Food Science and Emerging Technologies 16 (2012) 171–180 with g/l NaCl and incubated at 15 °C for days in aerobic conditions, for psychrotolerant and heat labile aerobic bacteria (pSSO); MRSA, incubated at 37 °C for 48 h under anaerobic conditions, for c19 strain; NPNL–MRSA, incubated at 37 °C for 48 h under anaerobic conditions, for 10140 strain All media were supplied from Oxoid (Milan, Italy) Microbiological data were log-transformed and expressed as the average of two replicates The measurement of pH, conducted in duplicate, was performed on the first homogenized dilution of the fish samples with a pH meter (Crison, Barcelona, Spain) 2.3.4 Sensory analysis The sensory evaluation panel consisted of 10 panelists aging between 22 and 38 years (students and researchers of the Department of Food Science, Faculty of Agricultural Science, University of Foggia) The panel training consisted of different sessions during which different samples were examined by the panel to define the evaluation techniques and to familiarize with the off-odor, texture, and color attributes of anchovies During this stage, triangle tests were performed to evaluate the reproducibility of the judge's answer and their capability in discriminating among samples During the test sessions, filet samples were coded by a letter and presented individually to each panelist in plastic cups covered with a lid in random order Using a scale ranging from to (where = very poor and = excellent), the sensorial overall quality of the marinated filets was determined Panelists were asked to base their decision evaluating color, odor and texture attributes: therefore, sample overall quality has to be considered an average of the three sensory attributes Each attribute was defined as unacceptable (corresponding to a score equal or b2) after development of first off-odor or under application of finger pressure, muscle returns more than half way (muscle texture) or slime production or after the 1st discoloration (muscle color) 2.3.5 Modeling Sensory scores were modeled through the shoulder/tail model, reported by Geeraerd, Valdramidis, and Van Impe (2005), reading as follows: Àkmax t Ntị ẳ N0 Nres ịe kmax SL e ! þ Nres : Á þ ekmax SL À1 eÀkmax t This function is usually used to model microbial inactivation, showing three different phases: a shoulder phase (SL), i.e time with a constant cell number; an exponential death kinetic (kmax), i.e the phase when the decrease of cell number follows an exponential trend; a tail (Nres), i.e a residual sub-population The parameters of this function were given a different meaning to fit sensory analyses; therefore, the shoulder length was named “no decay time” (NDT), i.e the time when the sensory scores remained at their maximum level; kmax was named rate of sensory decay (SD) (rate of decrease of sensory scores) Finally, Nres was re-named residual sensory score (RSS), i.e the residual score at the end of running time Using these new parameters, the equation of Geeraerd et al (2005) reads as follows: SDt Stị ẳ S0 RSSịe ! SDNDT e 1ỵ e SDNDT SDt e NDT values for odor, color, texture and overall quality were used as input data to run a Principal Analysis through XLSTAT component (Add-In-Soft, Paris, France) Results 3.1 Preliminary assays Preliminary assays were conducted as background for brine optimization and production of functional fish products Concerning the effect of pH (Fig 1), after 48 h L plantarum c19 was decreased by ca 3.5 log cfu/ml in the presence of acetic and lactic acids but only at pH 3.5; at the same pH value the reduction of B animalis 10140 was ca log cfu/ml Fig reports the effects of citrus extract towards L plantarum c19 and B animalis 10140 after 48 h of incubation; results were expressed as decrease of viability referred to the control and plotted vs the concentration of the antimicrobial agent Both L plantarum c19 and B animalis 10140 decreased by less than log cfu/ml at the lowest concentrations of citrus extract (100 and 200 ppm); otherwise, the decrease of viable count of probiotics was significant at 300 ppm (ca 1–1.2 log cfu/ml) and increased with increasing the concentration of the extract The increase was strong in the case of L plantarum c19, that experienced a reduction of viable count of log cfu/ml with 500 ppm of citrus extract added; otherwise, the reduction of cell number appeared slight for B animalis 10140 Following these results, a critical threshold of citrus extract was imposed at 300 ppm, due to the strong decrease of lactobacilli after this amount 3.2 Brine optimization Following the results of the preliminary assays, in the second step a simplex centroid was used to point out the optimal conditions able to enhance probiotic survival in a brine intended for fish marinating; the independent variables were citrus extract, NaCl and pH, adjusted at various levels (from 3.5 to 5.0) either with acetic acid or lactic acid A step-by-step procedure was used to model data; cell counts after 24 and 48 h of incubation were modeled as difference/decrease referred to control (primary model) and then used as input values to build a polynomial equation (secondary model) Concerning the viability of L plantarum c19, Fig reports the results observed after 48 h in the brine acidified through acetic acid The corresponding polynomial equation was the following:   log N48 ẳ 2:01ẵcitrus2:96ẵNaClẵcitrus adjustedR ; 0:994 logNc-logNs (log cfu/ml) 174 þRSS where S(t) is the dependent variable (sensory score within the time) and S0 the sensory score at the beginning of the experiment Data were fitted through the GiNaFiT component (Geeraerd et al., 2005); all the replicates of each point were used for model development c19-A c19-L 10140-A 10140-L -1 3,5 4,5 pH Fig Effect of pH on the survival of L plantarum c19 and B animalis subsp lactis DSMZ 10140 after 48 h (data are expressed as difference referred to control) A, pH adjusted through acetic acid; L, pH decreased through lactic acid; Nc and Ns, cell numbers (log cfu/ml) in the control and in each acidified sample B Speranza et al / Innovative Food Science and Emerging Technologies 16 (2012) 171–180 Fig Effect of citrus extract added at various levels in MRS broth after 48 h of incubation on the survival of L plantarum c19 and B animalis subsp lactis DSMZ 10140 Data are expressed as decrease of cell counts referred to the control Nc and Ns, cell numbers (log cfu/ml) in the control and in each sample added with citrus 175 As it can be inferred, the viability of L plantarum c19 was affected in a positive way by citrus extract, i.e cell number decreased with increasing the extract in the brine; otherwise, the interaction NaCl vs citrus extract was negative, thus highlighting that from a theoretical point of view a decrease of cell count could be achieved through a decrease of citrus and an increase of NaCl or by an increase of citrus with a contemporary decrease of NaCl NaCl and pH, as individual terms, were not significant What is the real impact of this equation? The answer is reported in Fig 3a, showing the ternary plot and the real decrease of cell count for each level of the independent variables The picture highlights that a real practical effect on viability of L plantarum c19 was exerted only by citrus extract, as the maximum decrease of cell count (2 log cfu/g or more) was observed at the highest level of the antimicrobial agent (300 ppm, corresponding to the coded level 1) This ternary plot highlights the effects of each variable in the mixture (i.e when all the variables play a role); however, for a process optimization it would be advisable to know what is the effect of each variable alone, without the others This information could be obtained through the desirability Fig Effects of citrus extract (0–200 ppm), NaCl (0–12%) and pH (3.5–5.0) on the survival (decrease of cell count referred to control) of L plantarum c19 after 48 h The acidification of the medium was performed through acetic acid a) Ternary plot; b) desirability profile 176 B Speranza et al / Innovative Food Science and Emerging Technologies 16 (2012) 171–180 function (Fig 3b) The desirability highlighted that pH did not exert a significant effect on cell viability after 48 h of incubation, as ΔlogN48 was always 0, for each coded value (from 0, corresponding to a pH of 5.0, to 1, pH 3.5); surprisingly, the polynomial equation predicted a quadratic profile for the NaCl with a point of minimal desirability corresponding to a concentration of 6% Finally, the effect of citrus extract was linear and negative, as the desirability decreased with increasing the amount of the extract The desirability approach was used to try a qualitative approach to process optimization and to point out the conditions (i.e the composition of brine) able to maintain the viability of probiotics at acceptable levels Desirability profiles were time-dependent, therefore the first step for process optimization is the choice of a time: we chose 48 h, to simulate the most unfavorable conditions (prolonged exposure of probiotics in brine); then, the following step is the setting of a critical break-point for desirability, i.e a value after which the decrease of desirability is not acceptable This critical threshold was imposed to 0.5, corresponding to a decrease of cell count of 1.5 log cfu/g of L plantarum c19; in the desirability function, desirability is the dependent variable and is related to each factor of design (independent variables) In our conditions, a desirability of 0.5 was related to an amount of citrus extract of 200 ppm; therefore this concentration of the extract was the result of process optimization: higher amounts of citrus extract could not be used as they caused a decrease of desirability below the critical value (0.5) The same procedure and data modeling was used to predict the effect of the independent variables on the survival of L plantarum c19 in the brine acidified through lactic acid For this brine, the polynomial equation reads as follows: logN48 ẳ 0:33ẵNaCl ỵ 2:76ẵcitrus1:94ẵpHẵcitrus   5:11ẵNaClẵcitrus adjustedR2 ; 0:999 : The obtained results showed that the citrus extract was the most significant term, but it was not the only significant variable A major role was exerted also by the interactions citrus extract vs NaCl and citrus extract vs pH; NaCl alone played a minor role As observed for brine containing acetic acid, desirability function highlighted the major role of citrus extract and the achievement of the critical break-point (desirability of 0.5) for amounts of citrus extract of 200 ppm (data not shown) Concerning the effects of citrus extract, NaCl and pH on the survival of B animalis 10140, data modeling highlighted that cell number of the probiotic was affected only by the antimicrobial agent, both with acetic (Fig 4a) and lactic acid added (Fig 4b); concerning desirability, the critical break-point (decrease of cell number of 1.5 log cfu/g) was achieved for a concentration of citrus extract of 200 ppm (data not shown) Following the results of desirability approach, a process optimization was conducted keeping in mind that the limiting factor for probiotic survival was citrus extract; as aforementioned, results showed that a critical break-point for probiotic survival was about 200 ppm, thus a conditioning brine for fish marinating containing this amount of extract could be proposed NaCl and pH were not limiting factors and they could be set also at level (NaCl 12% and pH 3.5); however, the choice of their values, as well as the kind of acid, relies upon different elements, such as the cost, the necessity of avoiding possible interactive effects with citrus extract towards probiotics and the effect on fish 3.3 Application on fish products In the last step of the research, functional fish products were prepared and packed under air (PB1), in oil (PB2), in the diluted brine (PB3) or Fig Pareto charts of standardized effects of citrus extract (0–200 ppm), NaCl (0–12%) and pH (3.5–5.0) on the survival (decrease of cell count referred to control) of B animalis subsp lactis DSMZ 10140 after 48 h A, pH; B, NaCl; C, citrus extract; AB, pH∗ NaCl; AC, pH∗ citrus extract; BC, NaCl ∗ citrus extract The acidification of the medium was performed either through acetic (a) or lactic acid (b) under vacuum (PB4); for each sample a control was prepared, i.e fish product without probiotics Before processing, fish was characterized by APC and PB counts of 6.93 and 5.57 log cfu/g respectively, whereas the specific spoilage microorganisms (both mSSO and pSSO) were ca log cfu/g; initial pH was 6.54 (data not shown) After processing and fish marinating, total viable count was log cfu/g and SSOs were at the undetectable level (b2 log cfu/g); the pH was 4.29 Table reports cell counts of probiotics under the different conditions throughout the storage at °C for 21 days L plantarum c19 was inoculated at ca log cfu/g and remained constant for the entire running time or underwent to a slight increase at the end of storage; on the other hand, in the sample PB2 a lower count was recovered, due probably to some difficulties encountered in cell recovery for oil presence Concerning B animalis 10140, the probiotic was recovered within 11 days, then it was at the undetectable level for the remaining running time; in all the samples SSOs were always lower than the detection limit, whereas the pH of fish remained constant within the storage (data not shown) Functional fish products, as well as controls, were analyzed by a sensorial panel to assess color, odor, texture and overall quality Unfortunately, it was not possible a classical evaluation of shelf life (best before time) (Speranza, Bevilacqua, & Corbo, 2010), defined as the time to attain the critical threshold of Some samples, in fact, showed a residual sensory score (RSS) higher than this break-point; B Speranza et al / Innovative Food Science and Emerging Technologies 16 (2012) 171–180 2.5 Table Cell numbers (log cfu/g) of L plantarum c19 (Lp) and B animalis subsp lactis 10140 (Ba) in fish products, marinated and packed under different conditions PB1, samples packed in air; PB2, samples filled with oil; PB3, samples filled with diluted brine; PB4, sample packed under vacuum Values are the average of two replicates PB1 PB2 PB3 PB4 Lp Ba Lp Ba Lp Ba Lp Ba 11 14 21 6.96a 7.20a 6.98a 7.00a 6.91a 7.01a 8.20b 6.89a 6.80a 6.65a 6.90a 6.88a – – 6.96a 6.11a,b 5.83b 6.66a 5.59b 5.40b 6.20a,b 6.89a 6.77a 4.85b 5.80c 4.74b – – 6.96a 7.23a 6.96a 7.30a 7.61a,b 7.63a,b 7.84b 6.89a 6.75a 6.67a 7.32b 7.51b – – 6.96a 7.40a,b 6.94a 7.22a 7.17a 7.66b 8.01c 6.89a 6.82a 6.69a 7.04a,b 7.07a,b – – a,b Values in the same columns with different letters are significantly different (one-way ANOVA and Tukey's test) (P b 0.05) PB3 color 1.5 component (30% ) > Time (days) 177 C4 0.5 overall quality C2 C3 PB1 texture odor -0.5 PB2 C1 -1 -1.5 PB4 -2 therefore a new approach was used, focusing on the parameter NDT (no decay time), corresponding to the time (days) when the sensory scores remained at their maximum level The concept of NDT was quite similar to the most common idea of HQL (high quality life), i.e the time corresponding to a period when food characteristics remain at the highest level, and it is almost different to shelf life, that is a practical quality life of food Table shows NDT time for the various samples and the respective controls, along with the standard errors around the mean values A paired comparison (t-test) between each functional product vs its respective control revealed that the presence of the probiotic did not affect negatively sensory scores of fish; on the other hand, in PB3 sample the presence of probiotics seemed to improve sensory attributes Concerning shelf life limiting element, color showed the lowest NDT, with some exceptions to this generalized statement After this preliminary modeling, NDT values were used as input parameters for a principal component analysis (PCA), in order to highlight the best and the worst samples; odor, color, texture and overall quality were used as discriminative variables The results of PCA are reported in Fig 5; as showed by the sum of explained variability by factor and factor (77%), the approach fitted well the data Concerning the correlation of discriminative variables with factors, odor and texture were related with factor (the correlation coefficients were 0.771 and 0.896, respectively), while color and overall quality were mostly related to factor (with correlation coefficients of ca 0.650) Amongst the probiotic-inoculated samples, PB2 was the worst one, as situated in a place opposite to the lines of the discriminative variables; this result was not surprising, as panelists always referred to this sample as the most disliked for the color of oil that appeared strange Otherwise, the best sample was the one packed in the diluted brine (PB3), as it showed the best overall quality and color and a good value of NDT for odor A good compromise for its sensorial quality was also the sample PB4 (fish packed under vacuum), which showed the highest results for odor scores Table No decay time (days) for the sensory scores (mean±standard error) PB1, samples packed in air; PB2, samples filled with oil; PB3, samples filled with diluted brine; PB4, sample packed under vacuum The samples C1/C4 are the controls (fish products without probiotic) Sample Odor Color Texture Overall quality PB1 PB2 PB3 PB4 C1 C2 C3 C4 5.05 ± 0.67 3.24 ± 1.45 6.04 ± 1.10 9.93 ± 0.50 5.87 ± 0.59 3.24 ± 1.45 2.50 ± 0.24 2.55 ± 0.25 7.19 ± 1.24 4.15 ± 1.57 9.89 ± 1.98 6.86 ± 1.33 5.83 ± 1.79 6.40 ± 1.23 6.64 ± 1.24 7.65 ± 1.20 7.12 ± 1.78 7.75 ± 2.51 11.21 ± 0.56 10.90 ± 0.28 10.66 ± 0.95 7.75 ± 2.51 7.75 ± 2.95 2.71 ± 0.58 5.57 ± 0.97 6.40 ± 1.23 10.00 ± 0.52 4.49 ± 1.90 5.63 ± 0.80 6.63 ± 1.26 5.92 ± 1.47 4.49 ± 1.63 -2.5 -4 -2 component (48% ) > Fig Principal component analysis performed on the values of no decay time of sensory scores (odor, color, texture, overall quality; discriminative variables) PB1, samples packed in air; PB2, samples filled with oil; PB3, samples filled with diluted brine; PB4, sample packed under vacuum The samples C1/C4 are the controls (fish products without probiotics) Discussion Lactic acid bacteria (LAB) in fish flesh have long been disregarded because they are not the predominant microora of seafood (Franỗoise, 2010) However, the changes in culinary habits and consumer demand towards ready-to-eat and convenience foods caused a change in processing, thus some microorganisms are usually inhibited (Gram negative bacteria), whereas the growth of other genera (e.g those belonging to the group of lactic acid bacteria) could be promoted (Franỗoise, 2010) Lactobacilli and carnobacteria were recovered in Atlantic salmon, pollock, Arctic char, cod and rainbow trout (Franỗoise, 2010; Huber et al., 2004); the strains isolated belong to the species L plantarum, Carnobacterium maltaromaticum, C divergens, C gallinarum, Leuconostoc mesenteroides, Weissella spp (Franỗoise, 2010; Liu, Li, Ji, & Yang, 2009; Matamoros, Pilet, Gigout, Prévost, & Leroi, 2009) Despite the role of some species in fish disease (like Lactococcus garviae and C maltaromaticum), in spoilage and amine production (Franỗoise, 2010), many authors reported on the benecial role of LAB in marine products for at least reasons: a) biopreservation, antagonism towards some pathogens and shelf life prolongation (Ma, Cho, & Oh, 2009; Matamoros, Pilet, et al., 2009; Matamoros et al., 2009; Mejlholm & Dalgaard, 2007); b) production of fermented fish products (Gelman, Drabkin, & Glatman, 2000; Glatman, Drabkin, & Gelman, 2000; Taira, Funatsu, Satomi, Takano, & Abe, 2007); and c) production of functional fish-based foods An increasing trend in functional product market is the production of non-dairy probiotic foods, like vegetables (olives and artichokes), cereal foods, soja and meat products containing probiotic microorganisms (Rivera-Espinoza & Gallardo-Navarro, 2010); however there are no reports on the use of probiotics in fish-based foods intended for human use Therefore the background of this research was the development of a functional-fish based product, containing probiotic microorganisms; LPFP anchovies (i.e lightly preserved fish products) (marinated fish) were chosen as carriers for probiotics for at least ideas: Anchovy is a low cost product in Southern Italy; therefore the preparation of a functional dish with this kind of raw material could be a promising way for small-scale producers LPFP are attractive products, due to mild treatments used for their stabilization Moreover the use of mild technologies assures probiotic survival throughout processing and storage 178 B Speranza et al / Innovative Food Science and Emerging Technologies 16 (2012) 171–180 The design of a new functional product should follow a complex design, including some points related to technology and others with the maintenance of active ingredient from production to consumer (McClements & Decker, 2009); if the functional ingredient is a microorganism (probiotic), its GRAS status, as well as survival and kind of health benefit, should be clarified and demonstrated Due to the fact that the microorganisms under investigation (L plantarum c19 and B animalis subsp lactis DSMZ 10140) are well known and their probiotic properties were assessed in the past, this paper focused on the technological evaluation of their suitability for LPFP anchovies The optimization of marinated functional anchovies was performed through a stepby-step procedure, focusing on the effects of the most important hurdles of brine (NaCl and pH) as well as on the use of a preservative to prolong anchovy shelf life (citrus extract); then, hurdles were combined to point out the optimal combination for brine preparation and anchovy marinating Finally, the impact of processing and probiotic inoculum was assessed through microbiological and sensorial analyses Marinating is one of the oldest technology for fish preservation and involves some main hurdles, i.e acidification through a dipping of fish in a diluted solution of vinegar or acetic acid, containing also salt at various levels (Gokoglu, Cengiz, & Yerlikaya, 2004); the treatment is usually performed at room temperature for 24–48 h and the products consumed uncooked Marinated fish has a limited shelf life, as the increase of traditional preservatives to levels able to suppress bacterial spoilage is not advisable for consumer preference (Kilinc & Cakli, 2005) A possibility is the use of an additional hurdle, like a natural antimicrobial agent; this approach was proposed by Gokoglu, Topuz, and Yerlikaya (2009) to prolong the shelf life of marinated anchovies They used pomegranate sauce and achieved a high oxidative and microbiological stability In this paper, citrus extract was proposed as an additional hurdle for the brining solution, as it was able to prevent the growth of the specific spoilage microorganisms of fish (Corbo et al., 2008, 2009; Del Nobile et al., 2009) pH as single hurdle (Fig 1) affected in a strong way the survival of both B animalis 10140 and L plantarum c19 when it was b4.0; this result was in agreement with some literature reports Kailasapathy, Harmstorf, and Phillips (2008) found a strong correlation between pH and the survival of B animalis subsp lactis and found a higher retention of viability at pH 4.1–4.5 more than at lower values (b4) Concerning the strain L plantarum c19, Bevilacqua, Altieri, et al (2010) found that this microorganism was able to grow in MRS broth acidified at pH 4.0, but not at lower values L plantarum is known as a species resistant to acidic conditions as well as to other kind of stresses (Molin, 2001); the strong decrease of viability after 48 h at pH 3.5 was probably due to the kind of acids used (lactic and acetic acids), able to control and/ or stop microbial growth not only for the decrease of the pH of the medium, but also for their intrinsic antimicrobial properties It is important to highlight that this sensitivity to pH was shown by L plantarum c19 and B animalis 10140 at 48 h, but not after 24 h of incubation In the combinations of DoE (Design of Experiment), in fact, only citrus extract played a major role, whereas the effect of pH was not significant, thus suggesting that its inhibitory effect towards probiotics could be recovered only after a prolonged contact time Apart from salt, whose effect was not significant, another important hurdle for brining solution was citrus extract This oil provides a natural source of antioxidants and flavonoids as well as terpenic compounds and its bioactivity could be the result of synergistic action (Fisher & Phillips, 2008; Rehman, 2006) Its effectiveness towards lactic acid bacteria was studied by Bevilacqua, Corbo, and Sinigaglia (2010b), who reported a delay of growth of L plantarum and L brevis in the presence of 40–50 ppm The results of this paper showed that higher amounts could affect significantly probiotic survival and the extent of cell decrease was related to the concentration of the extract into the medium However, the choice of an acceptable concentration of citrus extract should be performed through a balance between the risk (the decrease of probiotic) and the benefit (the inhibition of the spoiling microflora) In the preliminary assays, an initial inoculum of log cfu/ml was used and a decrease of the probiotic counts up to log cfu/ml (ca 2-log reduction) was regarded as acceptable, as this level is the minimum one required for a probiotic product in Italy (Fortina, 2007) This reduction was achieved by L plantarum c19 for an amount of citrus extract ranging between 300 and 400 ppm; therefore the critical concentration of the antimicrobial agent was set to 300 ppm and this amount was used as the level “1” in the mixture design A kind of risk/benefit analysis was used also in the mixture design to point out the optimal combination of brine solution for anchovy marinating In this case the critical break-point for the acceptability was set to a desirability level of 0.5, corresponding to a reduction of cell count of 1.2–1.5 log cfu/g and to a final level of the probiotics of 6.5– 6.8 log cfu/g; this decrease of the counts of L plantarum c19 and B animalis 10140 was achieved for a concentration of citrus extract of ca 200 ppm Concerning the effectiveness of citrus extract in the brining solution, the extent of reduction was stronger than that recovered in the preliminary assays (a higher decrease of cell count for a lower amount of extract); however, this result was not in disagreement with the data of the first phase, as the preliminary assays were conducted in a laboratory medium (MRS broth), whereas the second step was performed in brines, thus confirming literature reports on the significance of the kind of medium on the effect of extracts and essential oils (Burt, 2004) Desirability approach was a useful and friendly tool for process optimization, as it was able to convert a mathematical value (i.e the value of an independent variable, corresponding to a dependent value of 1.2– 1.5 log cfu/g) in a qualitative and simple response (how much you prefer the result? I prefer the results fifty/fifty, i.e I accept a desirability of 0.5) Moreover, desirability function highlighted that for a moderate processing time (24 h), neither salt nor pH (or better the kind of acid) played a significant role for the survival of probiotics As done for citrus extract, the choice of the levels of salt, pH and kind of acid was done through a risk/benefit decision, keeping in mind that high levels of salt and acid are not advisable for consumer preference Therefore, for the final step of this research salt was used at 10% and acetic acid was chosen, for its lower impact on the sensory characteristics of anchovies In the last step a further variable was added (i.e the use of oil or a diluted brine), as it's a common practice to sell marinated fish packed in brine or in oil As an alternative to this traditional storage, the packaging under vacuum was proposed Functional anchovies were analyzed for their microbiological and sensory characteristics in order to assess the limiting variable/factor for the definition of the shelf life Generally, the shelf life of probiotic foods relies on the survival of the probiotics throughout the storage and is defined as the time to attain a critical threshold of the probiotic (6 or log cfu/ml) (Altieri, Bevilacqua, & Sinigaglia, 2011) For the purpose of this paper, cell count of L plantarum c19 never attained a critical value as no significant changes were recovered; otherwise B animalis 10140 underwent to a drastic decrease after 11 days at °C This result suggested that L plantarum c19 was the optimal probiotic for the production of marinated anchovies, as it assured the minimal level of probiotic survival required for a functional product It should also be considered that a possible competition could have occurred between L plantarum c19 and B animalis 10140 during their mixed inoculation in fish: this event might have affected the intrinsic abilities of each single strain This is just a speculation that could be the topic of a future research; however, preliminary experiments in laboratory medium revealed that when B animalis 10140 was consociated with lactobacilli, a positive interaction occurred Many evidences are available for Lactobacillus delbrueckii subsp bulgaricus (Altieri, Bevilacqua, D'Amato, Del Nobile, & Sinigaglia, 2008) and some others for L plantarum c19 too (data not yet published) B Speranza et al / Innovative Food Science and Emerging Technologies 16 (2012) 171–180 The evaluation of NDT (no decay time) performed both for controls and inoculated samples revealed that generally odor was the limiting element for durability definition; moreover, the inoculum of the probiotics did not affect negatively sensory scores Another key-element is the kind of conditioning and storage: both conditioning in air and packing in oil were not suitable for the sensory attributes of the product Otherwise, the best sample was the one packed in the diluted brine and this result is of great concern as brine contained also low levels of some hurdles (salt and citrus extract), that combined with storage under refrigeration and low pH, could control and/or inhibit spoiling microorganism coming from a post-contamination Some authors reported that the shelf life of marinated fish (Pacific saury, sardine, pink shrimp, anchovy, cod filets) varied from 40 days to some months, depending on the quality of raw material, storage temperature, levels of preservatives (salt, amount of vinegar/acetic acid), use of additional hurdles (like pomegranate sauce) (Cadun, Cackli, & Kisla, 2005; Esaiassen et al., 2004; Gokoglu et al., 2004; Gokoglu et al., 2009; Kilinc & Cakli, 2005; Sallam, Ahmed, Elgazzar, & Eldaly, 2007) The limiting factors for shelf life definition are different (sensory scores, total volatile nitrogen, fat oxidation) and only in the paper of Sallam et al (2007) counts of spoiling microflora, in particular the cell number of psycrothrophic bacteria, were used for durability evaluation The product proposed in this paper showed a shelf life of at least 10 days (sample PB3), evaluated as the time to the maintain overall quality of the maximum level, or 21 days (running time), if the classical definition of shelf life is used Conclusion This paper proposes a new kind of functional food, based on marinated anchovies and produced as follows: Fish processing could be performed through a conditioning brine at 2% of acetic acid, 10% of NaCl and 200 ppm of citrus extract as an additional hurdle for the spoiling microflora After processing, anchovies should be inoculated with L plantarum c19, able to survive throughout the storage, without affecting sensory properties of fish Then inoculated samples should be packed in a diluted brine (the same used for fish processing, but diluted to 10) and stored at °C Shelf life definition relies upon the kind of quality that producers want to maintain: if the goal is to sell a product with the best sensory scores, than shelf life was ca (odor) or 10 days (overall quality) Otherwise, if shelf life is defined as best before date and the critical break-point is set at 2, product durability is at least 21 days The proposed food should broaden the range of functional food types making easier for consumers to achieve the optimal supply of probiotics able to provide health defense while enjoying different meals Future perspectives for this research could be process validation with other kinds of raw materials, or under thermal abuse conditions and the simulation of cross- and post-contamination of the product with some specific spoilage microorganisms of fish Finally, a consumer test should be performed to assess preference towards this kind of food References Altieri, C., Bevilacqua, A., D'Amato, D., Del Nobile, M A., & Sinigaglia, M (2008) Modelling the survival of starter lactic acid bacteria and Bifidobacterium bifidum in single and simultaneous cultures Food Microbiology, 24, 729–734 Altieri, C., Bevilacqua, A., & Sinigaglia, M (2011) Prolonging the viability of Lactobacillus plantarum through the addition of prebiotics into the medium Journal of Food Science, 76, M336–M345 Bevilacqua, A., Altieri, C., Corbo, M R., Sinigaglia, M., & Ouoba, L I I (2010) Characterization of lactic acid bacteria isolated from Italian Bella di Cerignola table olives: Selection of potential multifunctional starter cultures Journal of Food Science, 75, M536–M544 179 Bevilacqua, A., Corbo, M R., & Sinigaglia, M (2010a) Design of experiments: A powerful tool in food microbiology In A Mendéz-Vilas (Ed.), Current research, technology and educational topic in applied microbiology and microbial biotecnology (pp 1419–1429) Badajoz (E): Formatex Research Center Bevilacqua, A., Corbo, M R., & Sinigaglia, M (2010b) In vitro evaluation of the antimicrobial activity of eugenol, limonene and citrus extract against bacteria and yeasts, representative of the spoiling microflora of fruit juices Journal of Food Protection, 73, 888–894 Bevilacqua, A., & Sinigaglia, M (2010) Food shelf life and safety: Challenge tests, prediction and mathematical tools In A Bevilacqua, M R Corbo, & M Sinigaglia (Eds.), Application of alternative food-preservation technologies to enhance food safety and stability (pp 161–187) Sharjah (UAE): Bentham Science Publisher Burt, S (2004) Essential oils and their antibacterial properties and potential applications in foods—A review International Journal of Food Microbiology, 94, 223–253 Cadun, A., Cackli, S., & Kisla, D (2005) A study of marination of deepwater pink shrimp (Parapenaeus longirostris, Lucas, 1846) and its shelf life Food Chemistry, 90, 53–59 Corbo, M R., Speranza, B., Filippone, A., Conte, A., Sinigaglia, M., & Del Nobile, M A (2009) Natural compounds to preserve fresh fish burgers International Journal of Food Science and Technology, 44, 2021–2027 Corbo, M R., Speranza, B., Filippone, A., Granatiero, S., Conte, A., Sinigaglia, M., et al (2008) Study on the synergistic effect of natural compounds on the microbial quality decay of packed fish hamburger International Journal of Food Microbiology, 127, 261–267 Cross, M L (2002) Microbes versus microbes: Immune signals generated by probiotic lactobacilli and their role in protection against microbial pathogens FEMS Immunology and Medical Microbiology, 34, 245–253 Del Nobile, M A., Corbo, M R., Speranza, B., Sinigaglia, M., Conte, A., & Caroprese, M (2009) Combined effect of MAP and active compounds on fresh blue fish burger International Journal of Food Microbiology, 135, 281–287 Esaiassen, M., Nilsen, H., Joensen, S., Skjerdal, T., Carlehog, M., Eilertsen, G., et al (2004) Effects of catching method on quality changes during storage of cod (Gadhus morhua) Lebensmittel-Wissenschaft und Technologie, 37, 643–648 Fisher, K., & Phillips, C (2008) Potential antimicrobial uses of essential oils in food: Is citrus the answer? Trends in Food Science and Technology, 19, 156–164 Fortina, M G (2007) I prodotti lattiero-caseari In L S Cocolin, & G Comi (Eds.), La microbiologia applicata alle industrie alimentari (pp 289–336) Roma: Aracne Editrice Franỗoise, L (2010) Occurrence and role of lactic acid bacteria in seafood products Food Microbiology, 27, 698–709 Geeraerd, A H., Valdramidis, V P., & Van Impe, J F (2005) GInaFiT, A freeware tool to assess non-log-linear microbial survivor curves International Journal of Food Microbiology, 102, 95–105 Gelman, A., Drabkin, V., & Glatman, L (2000) Evaluation of lactic acid bacteria, isolated from lightly preserved fish products as starter cultures for new fish based food products Innovative Food Science and Emerging Technologies, 1, 219–222 Glatman, L., Drabkin, V., & Gelman, A (2000) Using lactic acid bacteria for developing a novel fish food products Journal of the Science of Food and Agriculture, 80, 375–380 Gokoglu, N., Cengiz, E., & Yerlikaya, P (2004) Determination of the shelf life of marinated sardine (Sardina pilchardus) stored at °C Food Control, 15, 1–4 Gokoglu, N., Topuz, O K., & Yerlikaya, P (2009) Effects of pomegranate sauce on quality of marinated anchovy during refrigerated storage LWT- Food Science and Technology, 42, 113–118 Gotcheva, V., Hristozova, E., Hrostozova, T., Guo, M., Roshkova, Z., & Angelov, A (2002) Assessment of potential probiotic properties of lactic acid bacteria and yeast strains Food Biotechnology, 16, 211–225 Huber, I., Spanggaard, B., Appel, K F., Rossen, L., Nielsen, T., & Gram, L (2004) Phylogenetic analysis and in situ identification of the intestinal microbial community of rainbow trout (Oncorhynchus mykiss, Walbaum) Journal of Applied Microbiology, 96, 117–132 Imasse, K., Tanaka, A., Tokunaga, K., Sugano, H., Ishida, H., & Takahashi, S (2007) Lactobacillus reuteri tablets suppress Helicobacter pylori infection— A double blind randomised placebo-controlled cross-over clinical study— Kansenshogakuzasshi Journal of the Japanese Association of Infectious Diseases, 81, 387–393 Kailasapathy, K., Harmstorf, I., & Phillips, M (2008) Survival of Lactobacillus acidophilus and Bifidobacterium animalis ssp lactis in stirred fruit yogurts Lebensmittel-Wissenschaft und Technologie, 41, 1317–1322 Kilinc, B., & Cakli, S (2005) Determination of the shelf life of sardine (Sardina pilchardus) marinades in tomato sauce stored at °C Food Control, 16, 639–644 Lavermicocca, P., Lonigro, S.L., Visconti, A., De Angelis, M., Valerio, F., & Morelli L (2009) Patent: Table Olives Containing Probiotic Microorganisms Priority date: 5.12.2003 no MI2003A002391 European Patent EP1843664 (granted 8.7.09) Liu, J Y., Li, A H., Ji, C., & Yang, W M (2009) First description of a novel Weissella species as an opportunistic pathogen for rainbow trout Oncorhynchus mykiss (Walbaum) in China Veterinary Microbiology, 136, 314–320 Ma, C W., Cho, Y S., & Oh, K H (2009) Removal of pathogenic bacteria and nitrogens by Lactobacillus spp JK-8 and JK-11 Aquaculture, 287, 266–270 Marteau, P R., de Vrese, M., Cellier, C J., & Schrezenmeir, J (2001) Protection from gastrointestinal disease with use of probiotics American Journal of Clinical Nutrition, 73, 430s–436s Matamoros, S., Leroi, F., Cardinal, M., Gigout, F., Kasbi Chadli, F., Cornet, J., et al (2009) Psychrotrophic lactic acid bacteria used to improve the safety and quality of vacuum-packaged cooked and peeled tropical shrimp and cold smoked salmon Journal of Food Protection, 72, 365–374 Matamoros, S., Pilet, M F., Gigout, F., Prévost, H., & Leroi, F (2009) Selection and evaluation of seafood-borne psychrotrophic lactic acid bacteria as inhibitors of pathogenic and spoilage bacteria Food Microbiology, 26, 638–644 180 B Speranza et al / Innovative Food Science and Emerging Technologies 16 (2012) 171–180 McClements, D J., & Decker, E A (2009) Designing functional foods Cambridge (UK): Woodhead Publishing McLay, R (1972) Marinades Torry advisor Note, N° 56, : Ministry of agriculture, fisheries and food Torry Research Station Mejlholm, O., & Dalgaard, P (2007) Modeling and predicting the growth of lactic acid bacteria in lightly preserved seafood and their inhibiting effect on Listeria monocytogenes Journal of Food Protection, 70, 2485–2497 Meyer, V (1965) Marinades In G Borgstrom (Ed.), Fish as food Processing: Part (pp 165–193) New York: Academic Press Molin, G (2001) Probiotics in foods not containing milk or milk constituents, with special reference to Lactobacillus plantarum 299v American Journal of Clinical Nutrition, 73, 380s–385s Nomoto, K (2005) Review prevention of infections by probiotics Journal of Bioscience and Bioengineering, 100, 583–592 Rehman, Z (2006) Citrus peel extract — A natural source of antioxidant Food Chemistry, 99, 450–454 Reid, G., Howard, J., & Gan, B S (2001) Can bacterial interference prevent infection? Trends in Microbiology, 9, 424–428 Rivera-Espinoza, Y., & Gallardo-Navarro, Y (2010) Not dairy probiotic products Food Microbiology, 27, 1–11 Sallam, K I., Ahmed, A M., Elgazzar, M M., & Eldaly, E A (2007) Chemical quality and sensory attributes of marinated Pacific saury (Cololabis saira) during vacuum-packaged storage at °C Food Chemistry, 102, 1061–1070 Shah, N P (2007) Functional cultures and health benefits International Dairy Journal, 17, 1262–1277 Sidhu, K S (2003) Health benefits and potential risks related to consumption of fish or fish oil Regulatory Toxicology and Pharmacology, 38, 336–344 Speranza, B., Bevilacqua, A., & Corbo, M R (2010) Food spoilage and safety: Some key-concepts In A Bevilacqua, M R Corbo, & M Sinigaglia (Eds.), Application of alternative food-preservation technologies to enhance food safety and stability (pp 17–34) Sharjah (UAE): Bentham Science Publishers Taira, W., Funatsu, Y., Satomi, M., Takano, T., & Abe, H (2007) Changes in extractive components and microbial proliferation during fermentation of fish sauce from underutilized fish species and quality of final products Fisheries Science, 73, 913–923 Vinderola, C G., & Reinheimer, J A (1999) Culture media for the enumeration of Bifidobacterium bifidum and Lactobacillus acidophilus in the presence of yoghurt bacteria International Dairy Journal, 8, 497–505