Community Research and Development Information Service - CORDIS

FP7

PROBIOLIVES Report Summary

Project ID: 243471
Funded under: FP7-SME
Country: Greece

Final Report Summary - PROBIOLIVES (Table olive fermentation with selected strains of probiotic lactic acid bacteria. Towards a new functional food.)

Executive Summary:
The PROBIOLIVES is a European project that brings together 14 beneficiaries from 4 EU countries (Greece, Italy, Spain, Portugal) plus 1 from Tunisia to study the use of probiotic starter cultures in olive fermentation. The main goal was to provide to the SME-AGs and their members knowledge and tools to increase their technological level, competitiveness and profits by the production of a functional product, containing probiotic bacteria in adequate amounts to improve consumer’s health, without altering the quality characteristics of fermented olives.
It is well known that lactic acid bacteria (LAB) originated from the olive natural microbiota are the dominant microorganisms in natural fermentations and one of the key project objectives was to investigate their probiotic potential. Indeed, during the implementation of PROBIOLIVES, a great number of LAB were isolated from the olive's microbiota from different stages of fermentation, from several olive cultivars, countries and preparation styles. A total of 597 strains of LAB were screened for probiotic properties by in vitro tests and 38 strains found to possess probiotic characteristics, specifically 9 strains belonging to Lb. pentosus, Lb. plantarum and Lb. paracasei subsp. paracasei in Greece, 4 Lb. pentosus strains in Spain, 17 Lb.plantarum & Leuc. mesenteroides in Italy, 3 Lb. plantarum, Lb. paraplantarum , Lb. pentosus in Portugal and 5 Lb.plantarum strains in Tunisia. Chemometric analysis of the probiotic properties was conducted with an aim to cluster the most promising strains.
The technological performances of the selected isolates as starter cultures in simulated conditions of fermentation were monitored through a multidisciplinary approach. Traditional microbial analysis coupled with molecular assays were performed in order to emphasize the effective presence of the inoculated starters. Moreover, chemical analysis of the organic acids, sugars, phenols, volatiles present in brines and the sensory evaluation of olives were carried out to define the contribution of the starters in the definition of final characteristics of olives. Based on the complete probiotic testing as well as on the performance of the isolates in simulated conditions of fermentation and in laboratory trials, a further selection has been performed to the most appropriate starters in the large scale fermentation. In Greece the Lb. pentosus 281 and E97 and Lb. plantarum 282 have been selected, in Spain the Lb. pentosus strains 41 and 222, in Portugal the Lb. plantarum, Lb. paraplantarum, Lb. pentosus with Lb. plantarum being the most preferred, in Italy two Lb.plantarum strains (O1T90C and S1T10A) and in Tunisia the Lb. plantarum LAM118 and LAM120.
The kinetics of fermentation with the selected starters has been modeled and results showed that in most cases they resulted in normal olive fermentation. The subsequent PLS analysis allowed sensory attributes to be predicted as a function of the biochemical parameters.
Effective packaging systems have been tested and the survival of the probiotic strains in adequate numbers has been confirmed, while the overall quality of the final product, was assessed.
The collected data for pathogens surveillance as well as predictive models applied for pathogens survival, are very useful for risk assessment studies and contribute to EC regulation 2073/2005 for the safety of table olives.
Consumer perception and market studies showed that the overall acceptability of the probiotic olives exhibited very high values and the majority of the consumers declared their intention to buy it.
Successful scaling up from laboratory to industrial level was achieved for certain starters like L. pentosus B281 while significant scientific and technical knowledge obtained regarding the existence of the inoculated starter in olive biofilm as well as the improvement of inoculation system at industrial scale, facilitating in this way the utilization of any starter culture.
Dissemination activities included publications in SCI scientific journals, presentations in workshops and scientific conferences, posters, flyers, press releases, articles, project website (www.probiolives.eu), Theses (MSc & PhD). Exploitation activities of the results included the applications for 2 patents and training activities were performed through the organization of workshops by the associations.
Project Context and Objectives:
Table olives are one of the major products that are consumed fermented. The primary purpose of table olive fermentation is to achieve a preservation effect and enhance the sensory attributes of the processed product. Diverse microbial groups are involved throughout olive fermentation determining the quality and sensory properties of the final product but it is generally accepted that LAB and yeasts are the most relevant microorganisms dominating the process. Among the traditional fermented foods, table olives could be a promising probiotic food through the use of functional probiotic starter cultures. Functional starter cultures contribute to microbial safety and offer organoleptic, technological, nutritional or health advantages.
Thus the concept of the PROBIOLIVES project was to provide to the SME Associations and their members with knowledge and tools to increase their technological level, competitiveness and profits by the production of olives, fermented with probiotic bacteria, originated from the olive microbiota. As the LAB from the olive microbiota are the dominant microorganisms in natural fermentations, the study of the probiotic potential of LAB isolates is needed initially. The selected probiotic bacteria have been introduced into the brines at the onset of fermentation, to act as starters, and to investigate if they are able to dominate and ensure a proper fermentation inhibiting the growth and survival of undesirable microorganisms. The isolation and use of suitable LAB as starters can contribute to better fermentation characteristics microbiological, physiological and biochemical. The ultimate goal is the production of a functional product, containing probiotic bacteria in adequate amounts to improve consumer’s health, without altering the quality characteristics of fermented olives. Consumer acceptance studies were essential for the exploitation and the introduction of the new food into the EU and international market. At the same time a better control of the fermentation process, early detection of faulty fermentation and spoilage and assessment of the time needed for fermentation completion have been achieved by monitoring the biochemical indices throughout the process with the use of advanced instruments and mathematical tools. Risk assessment studies were carried out to ensure the safety of the new product as well as storage experiments indicated its shelf life stability. Mathematical models have been developed for the prediction of its shelf life and consumer acceptance studies to ensure the consumer perception of the final product.
The PROBIOLIVES project is built up to focus on:
? the selection of lactic acid bacteria from olive natural microbiota possessing probiotic properties
? the fermentation process:
o development, improvement and use of probiotic starters for proper fermentation
o characterization of metabolic products (acid profile, aroma)
o modelling the process and prediction of shelf life of the final functional food
o microbiological risk assessment of the final product
? increasing the shelf life in combination with new packaging technologies
The innovative part of the proposed project lies in the fermentation of table olives with probiotic strains of lactic acid bacteria, not just the fortification of the final fermented product with probiotic lactic acid bacteria that has been already a patent. The multiple benefits of this approach would be: (a) enrichment of human diet through the development of a wide diversity of flavours, aroma and textures in food, (b) preservation of substantial amounts of food through lactic acid fermentation, (c) enrichment of food substrates with essential amino acids, fatty acids and vitamins, (d) detoxification during food fermentation processing, and (e) safety from food borne pathogens.
By these means the PROBIOLIVES project will offer to members of SME-AGs i) advancement in the technological process of olive fermentation (properly controlled, more predictable, and reliable process) by using starter cultures and ii) innovation in the production of a functional and healthy product for the consumers.
In the context characterized by several food crises in the last years, consumers are more and more aware of the quality of products and their impact on health. The improvement of citizen’s health through the availability of new generation of healthy foods is one of the main perspectives of the project in terms of public health. On this side, the project will allow members of SME-AGs to improve their products in the market and thus reinforce their standard of competitiveness by putting at the disposal of European citizens some scientific data for a better olive fermentation control and for the development of probiotic foods.

Specific objectives of each WP

WP1: The co-ordination and implementation of the project components, including the administrative, legal, IPR related issues arising from the project and reinforcement of the project team in relation to the research requirements; Monitoring, review and Quality Assurance. management the Audit certificates for the financial status; Co-ordination of the Knowledge Management activities, including the technical, financial reporting to the REA, submission of deliverables (D1.1, D1.2), maximization of interactions and flow of information among the participants; Establishment of common operational procedures respecting ethical and gender issues, organization of meetings.
WP2: The main objectives of this WP were: to test lactic acid bacteria from olive microbiota for probiotic potential, to test the responses of the selected probiotic bacteria to simulated conditions of fermentation (assessment of their metabolic activity, growth kinetics), to select and identify the most suitable strains for starters in olive fermentation.
Task 2.1 Selection of lactic acid bacteria with probiotic potential from the autochthonous olive microflora: Lactic acid bacteria will be isolated from the olive’s microbiota, from different stages of fermentation, from several olive cultivars, countries (Spain, Greece, Italy, Portugal, Tunisia) and preparation styles. Isolates will be screened for in vitro properties indicating probiotic potential according to the EFSA guidelines. These screening techniques include safety and functional aspects like: ability to survive the entire digestion process and conditions of gastrointestinal transit by screening for tolerance to a low pH and resistance to high bile salt concentrations, production of antimicrobial substances to suppress pathogenic bacteria and help to maintain the intestinal balance, cell surface hydrophobic and auto aggregation properties that may be correlated to adhesion to intestinal epithelial cells, antibiotic resistance (D2.1).
Task 2.2 Responses of selected probiotic lactic acid bacteria to simulated conditions of fermentation: Starter cultures can ensure a proper and complete fermentation. If the fermentation process is adequate, the physicochemical characteristics of the drupes, particularly pH, titratable acidity and salt content, can ensure the microbiological safety and good keeping quality of the commodity. The minimum requirements for these characteristics have been specified in detail while the product should comply with the microbiological criteria established by the CODEX and IOC for fermented olives held in bulk. In this task the following will be performed for the selection of potential probiotic starter cultures: a) Evaluation of growth temperature on potential probiotic starter bacteria, b) Determination of acid production and minimal pH of potential starters, c) Biochemical profile; monitoring the microbial activities of selected strains in vitro (gel cassete system), in vivo (sterile olives) and in situ (naturally processed), d) Aroma profile, e) Tolerance to phenolic compounds (D2.2a, D2.2b).
Task 2.3 Selection of most suitable starter strains & Task 2.4 Identification of selected strains on genus and species level: The most suitable strains will be selected and will be identified on genus and species level (D2.1a, D2.1b, M2.1).
WP3: The objectives in this WP are to test the ability of the selected LAB with probiotic potential to improve the microbial ecosystem in conditions of small scale fermentation. The improvement of physicochemical, microbiological and organoleptic quality will be emphasized. Confirmation of the survival of the starters strains used at the end of the fermentation to ensure that the probiotic bacteria exist in adequate numbers on the final product. Establishment of optimal growth conditions for fermentation process.
Task 3.1 Assessment of physicochemical -biochemical parameters in fermentations with selected probiotic starter cultures:The effect of intrinsic (physical, chemical) and extrinsic parameters (temperature, microbial association) on the formation of microbial metabolites will be determined. In parallel with the microbiological analyses, the physicochemical characteristics of the inoculated fermentations with probiotic starter cultures will be monitored (pH, acidity, combined acidity, reducing sugars, organic acids profile and volatiles). Two alternative approaches will be used: (a) Inoculation of “heat socked” table olives with selected probiotic LAB. The purpose of this treatment is to eliminate the presence of other interfering and competitive microbial groups and define thus the effect of the added starter itself on the fermentation process. (b) Direct brining of table olives with additional inoculation of probiotic lactic acid bacteria. In this case, olives will be fermented with the indigenous microbiota together with the added starter culture (D3.1).
Task 3.2 Assessment of organoleptic profile of fermentations with selected probiotic starter cultures: The organoleptic profile of the final fermented product will be assessed by a trained expert panel, according to the protocol developed by the International Olive Council (Document OT/WG1-01). The controlled fermentation should produce desirable and consistent aroma compounds as found in natural fermentations. Alcohols and other volatiles in the brines will be identified and correlation between sensory attributes and biochemical indices will be performed (D3.2, D3.3).
Task 3.3 Subselection of most suitable starter strains: The probiotic strains that showed improved performance in the production of table olives not only on the biochemical and microbial level but also on the organoleptic level will be finally selected for application in large scale (M3.1).
WP4: To determine the shelf-life period of the final fermented product, (a) under different storage temperature and (b) packaging systems e.g. VP/MAP, to select the most suitable packaging and to evaluate the survival of the starters used at the end of storage to ensure the presence of the probiotic bacteria.
Task 4.1 Evaluation of shelf life of the final fermented product under different storage temperatures: The fermented olives are products relatively stable due to their low pH. For the evaluation of the shelf life of a functional product a study on the adherence and survival of probiotic LAB strains during storage is essential. In the meantime the quality of the final product will be tested during storage temperatures (4 to 20°C) and the optimum temperature conditions for preservation will be determined (D4.1a, D4.1b).
Task 4.2 Packaging of probiotic fermented olives: The performance of different packaging systems will be tested like modified atmosphere packaging (MAP) methods which can extend the shelf-life considerably without affecting the product’s characteristics. Special attention will be given in new packaging systems without brine in pouches or punnets flushed with modified atmospheres (e.g. 30% CO2/70% N2) (M4.1).
WP5: The main objectives of this WP are: to apply models in order to describe the survival /growth kinetics of the selected probiotic strains in olive fermentation, to develop and comparatively evaluate alternative modelling approaches, and a reliable model to predict the shelf life of the final fermented product and to evaluate the survival of the probiotic strains during storage (D5.1a, D5.1b, M5.1).
Task 5.1 Development of mathematical models for growth kinetics during fermentation: The population dynamics of the different probiotic strains of LAB during fermentation and the determination of the respective kinetic parameters will be based on different primary kinetic models. Growth kinetic models will be comparatively evaluated with artificial neural networks based on Multi-Layer Perceptron (MLP) architecture.
Task 5.2 Development of mathematical models for the survival of probiotic lactic acid bacteria during storage: The survival of probiotic LAB during olive fermentation will be simulated with different primary models and their goodness-of-fit will be estimated and compared using statistical indices.
Task 5.3 Development of mathematical models for shelf life prediction of the fermented probiotic product: The changes in physicochemical characteristics (e.g. colour, pH, firmness, etc) during storage of fermented olives will be monitored and modelled in order to make estimations about the expected shelf life of the product.
WP6: The main objectives are to validate existing mathematical models for the prediction of safety of the final product, and to contribute to risk assessment studies.
Task 6.1 Survival studies of L. monocytogenes, Salmonella sp. and E. coli O157:H7 in the probiotic fermented olives: The fate of pathogens S. Enteritidis, L. monocytogenes and E. coli O157:H7 inoculated in probiotic fermented olive brine will be determined contributing to studies needed by the EC regulation No2073/2005 for the microbiological criteria of foodstuffs (Annex II) (D6.1a, D6.1b).
Task 6.2 Quantitative microbiological risk assessment: The already existing predictive models will be used to simulate the growth/survival of the above mentioned pathogens in the brine environment and determine their kinetic patterns (D6.2a, D6.2b, M6.1).
WP7: The main objectives will be to explore consumer’s perception and acceptance of functional food as well as the market potential of the new product.
Task 7.1 Organoleptic assessment of probiotic fermented olives by consumers: A special scoring sheet will be used for the organoleptic assessment of the final fermented product by consumers according to the IOC (OT/WG 1-01/Doc. no. 5-3) in which table olives are scored on the basis of three general descriptors, namely negative attributes, gustatory attributes, and kinaesthetic attributes. Negative attributes will include the following descriptors: abnormal fermentation, musty, and rancid effect. Other negative effects such as soapy, earthy and cheese taste will be taken into account. As gustatory attributes, the main descriptors of bitter, salty, and acid taste will be used. Finally, in the case of kinaesthetic sensations, hardness and fibrousness will be the main sensorial descriptors (D7.1, M7.1). In order to study the market potential of the functional product a survey will be conducted to assess the consumer perception and intention to buy such a product (D7.2).
WP8: The main objective of this WP is to apply the finally selected probiotic LAB as starters in controlled fermentations of medium/large scale in collaboration with the participating SMEs including microbiological and biochemical monitoring of the fermentation in order to demonstrate the use of the probiotic starters by the enterprises (D8.1, M8.1).
WP9: This WP includes exploitation and dissemination activities for the project’s results as well as training of SMEs to transfer the RTD knowledge.
Task 9.1 Exploitation activities: Exploitation activities will be aimed to apply the knowledge management and IPR management strategy, defined in the CA signed among the partners. A careful analysis of any IPR issues, access to previous know-how, and protection of any know how generated within the project will be carried and regulated within the CA (D9.3, D9.4).
Task 9.2 Dissemination activities: Dissemination activities will mainly be carried out by the SME-AGs and SMEs in collaboration with the RTD performers and will include workshops, courses, participation in exhibitions. Scientific results not subjected to protection will be externally disseminated through presentations in conferences and publications in scientific journals during and after the end of the project (D9.1).
Task 9.3 Training activities: Training will be aimed to transfer the know-how generated from the RTD performers to the SME-AGs and their members: each SME-AG will organize, in collaboration with the RTD performers, internal workshops in its country enhancing the mobility of researchers (D9.2, M9.1).
Project Results:
Background
Fermentation is a promising food biopreservation process which refers to the extension of shelf life and improvement of the safety of a food commodity using microorganisms and their metabolites (Ross et al., 2002). Table olives are one of the major products that are consumed fermented. The primary purpose of table olive fermentation is to achieve a preservation effect and enhance the sensory attributes of the processed product (Nout and Rombouts, 1992; Tassou et al., 2010, Sánchez Gómez et al., 2006). The most important industrial preparations are the Spanish or Sevillian style for green olives, the Californian style for black oxidized olives and the Greek style for naturally black olives (Sánchez Gómez et al., 2006). Spanish-style green olive fermentation is the most economically important olive processing (Aponte et al., 2012). Traditionally, this technique consists of a treatment with alkaline lye to hydrolyse the bitter glucoside oleuropein, followed by a washing step to remove the excess alkali. Brine is then added and a spontaneous fermentation takes place (Garrido-Fernández et al., 1995). The fermentation time depends on many factors (NaCl concentration, temperature, etc.) and can last from 3-5 months.
Diverse microbial groups are involved throughout olive fermentation determining the quality and sensory properties of the final product but it is generally accepted that LAB and yeasts are the most relevant microorganisms dominating the process (Arroyo-López et al., 2008; Bautista-Gallego et al., 2011; Hurtado et al., 2012). However, the indigenous microbiota of the fruits varies as a function of the quality of the raw material, harvesting conditions and post-harvest treatments and may thus lead to variations in the sensory and organoleptic characteristics of the final product (Garrido Fernández et al., 1997). LAB influence fermentation in a variety of ways, the most important being the production of lactic acid from fermentable substrates resulting in pH decrease with a concurrent increase in acidity, ensuring thus the microbiological stability during storage even at ambient temperature for extended periods of time. Inoculation of the brine with an appropriate starter culture of LAB reduces the probability of spoilage and helps to achieve an improved and more predictable fermentation process (Panagou and Tassou, 2006). Today, pure starter cultures of LAB are available in the market and used in several vegetable fermentations (Leroy and De Vuyst, 2004; Hurtado et al., 2012; Di Cagno et al., 2013) but their use in table olive processing is still limited despite the benefits they provide. The preparation of improved commercial starter cultures specifically for table olives has been reported previously (Roig and Hernandez, 1991) with main focus on Spanish-style processing (Panagou et al., 2008). Depending on the geographical location or the olive production process, different LAB strains can be used as starter cultures. However, the majority of LAB preparations consist of L. plantarum, L. pentosus or both (Hurtado et al., 2010). For the selection of LAB as starter cultures in table olive processing certain technological properties must be assured including fast and predominant growth, antimicrobial activity, homofermentative metabolism, high acidification rate and fast consumption of fermentable substrates, utilization of non-digestible a-galactosidase sugars, tolerance to bile salts and acidic pH, screening of enzymes with biotechnological potential (Abriouel et al., 2012).
Probiotic food products are in general fermented foods containing an amount of viable and active microorganisms large enough to reach the intestine and exert an equilibrating action on the intestinal microflora (FAO/WHO, 2002). Intake of probiotics is considered to stimulate the growth of beneficial microorganisms, reduces the amount of pathogens and help boost the immune system, lowering thus the risk of gastro-intestinal diseases (Cross, 2002; Reid, 2008). To deliver the health benefits, probiotic foods need to contain an adequate amount of live bacteria (at least 6-7 logcfu/g) (Oliveira et al., 2001; Boylston et al., 2004), although there are recent convincing data on beneficial immunological effects derived from dead cells (Vinderola and Reinheimer, 2003; Mottet and Michetti, 2005). Throughout the past two decades, probiotic health-promoting microorgansims have been increasingly included into commercial products in a response to the consumer demand for healthy food options that improve overall health, intestinal function, and digestion (Menrad, 2003). Most of these microorganisms are LAB and, among them, lactobacilli represent one of the fundamental microbial groups which have been introduced in a wide range of food products. Until recently, probiotic foods have been restricted almost exclusively to dairy fermented products, such as fermented milks and yogurt (Saad et al., 2013). However, these foods cannot be consumed by certain groups of the population who suffer from lactose intolerance or need a diet based on non-milk derived products. On the other hand, nowadays there is a need for novel and nondairy probiotics and it has been found that traditional fermented foods may constitute a good working base for the development of probiotic-type functional foods (De Vuyst et al., 2008; Ruiz-Moyano et al., 2008, 2011). A window of opportunity for the development of non-dairy probiotic products has arisen from the increasing number of lactose intolerance cases occurring in the world population, coupled with the unfavourable effect of cholesterol contained in fermented dairy products (Granato et al., 2010). In the last years, fermented foods of plant origin have been increasingly considered as vectors for incorporation of probiotic cultures following the well-established procedure of vegetable fermentation (Gupta and Abu-Ghannam, 2012). In this way, fruits and vegetables already containing high levels of beneficial substances (e.g., antioxidants, vitamins, dietary fibers, minerals) can be reinforced with probiotic bacteria that can bring about additional health promoting features (Soccol et al., 2010; Peres et al., 2012).
Among the traditional fermented foods, table olives could be a promising probiotic food through the use of functional probiotic starter cultures. Functional starter cultures contribute to microbial safety and offer organoleptic, technological, nutritional or health advantages. In contrast to well-adapted industrial starters, wild-type strains that naturally dominate traditional fermentations tend to have higher metabolic capacities, which can beneficially affect product quality, for instance with regard to aroma formation and/or food safety. Natural selection is likely to have forced such strains to be more competitive by endowing them with ecological advantages (Ayad et al., 2000; Maldonado et al., 2002). The information provided from traditional fermented foods and scientific research could help develop new probiotic products for the food industry (Rivera-Espinoza and Gallardo-Navarro, 2010).
The first reported studies on olive fermentation using probiotic starter culture were by Saravanos et al. (2008) and De Bellis et al. (2010), where probiotic LAB of dairy origin (L. plantarum ACA-DC 146 and L. paracasei ACA-DC 4037) or a probiotic L. paracasei IMPC2.1 strain isolated from human intestine were used respectively. Research has focused on the exploitation of the micro-structure of the olive surface as a carrier of probiotic strains of LAB confirming the suitability of the olive surface for this purpose (Lavermicocca et al., 2005; Saravanos et al., 2008; de Bellis et al., 2010; Arroyo-López et al., 2012; Rodriguez-Gómez et al., 2013).
Most of the studies published today about physiological properties of strains intended to be used as probiotics are performed on strains from human or animal internal cavities, considering that strains of these origins would be better adapted and colonize the human/animal gastrointestinal tract (Johansson et al., 1993; Prasad et al.,1998; Xanthopoulos et al., 2000; Ouwehand et al., 2002; Ruiz-Moyano et al., 2009; Zacarías et al., 2011). On the other hand, research has started to increase on probiotic functions of lactic acid bacteria isolated from foods like dairy products (Maragkoudakis et al., 2006; Bao et al., 2010; Monteagudo-Mera et al., 2012; Espeche et al., 2012), dry sausages (Papamanoli et al., 2003; Pennacchia et al., 2004; De Vuyst et al., 2008), foods of plant origin (Husmaini et al., 2011), fruits, cereals, meat or fish (Rivera-Espinoza and Gallardo-Navarro, 2010). Traditional fermented foods are a plentiful source of microorganisms and some of them show probiotic characteristics, although the research of these matrices as raw material for probiotic microorganisms is still scarce compared with their dairy counterpart (Rivera-Espinoza and Gallardo- Navarro, 2010).
Thus the main context of the PROBIOLIVES project was to investigate the probiotic potential of the LAB colonizing the olives with the aim to select possible probiotic strains that could be used as starters in table olive fermentation. The microbiological, physiological and biochemical characteristics of the fermentation was studied in parallel as well as the storage conditions of the final product. The ultimate goal was the production of a functional product, containing probiotic bacteria in adequate amounts to improve consumer’s health, without altering the quality characteristics of fermented olives. Consumer acceptance studies were performed essential for the exploitation and the introduction of the new food into the EU and international market. The main S & T results/foregrounds in each Work package are described below:
WP2: Isolation and characterization of probiotic lactic acid bacteria as starter cultures in green olive fermentation

Lactic acid bacteria (LAB) have been isolated from the olive's microbiota from different stages of fermentation, from several olive cultivars, countries and preparation styles. All these isolates were subjected to probiotic characterization by the means of survival in simulated gastrointestinal tract, resistance to antibiotics, hydrophobicity, autoadhesion, production of bacteriocins and haemolytic activity. A total of 597 strains of lactic acid bacteria were screened for probiotic properties (111 in Spain, 71 in Greece, 365 in Italy, 14 in Portugal and 36 in Tunisia). A total of 38 strains showed to have the probiotic characteristics and according to identification tests they belong to the Lb. plantarum group (Lb. plantarum, Lb. paraplantarum, Lb. pentosus). Specifically 9 strains were pre-selected in Greece belonging to Lb. pentosus, Lb. plantarum and Lb. casei group. In Spain 4 strains selected belonging to Lb. pentosus, in Italy 17 Lb.plantarum & Leuc. mesenteroides, in Portugal 3 Lb. plantarum, Lb. paraplantarum, Lb. pentosus, and in Tunisia 5 Lb.plantarum. Most of the probiotic tests had been included in Deliverable 2.1a. Deliverables 2.1b (18M), 2.2a (M24) and 2.2b (M30) include additional tests for testing the probiotic potential like Caco2 cell adhesion, Gut model resistance including peristaltic/mechanical stress and an artificial process of human digestion with fermented olives as meal. Experiments to highlight the interaction between selected LAB and human gut model were performed also using intestinal epithelial and monocyte/ macrophages derived cell lines of human origin, named respectively H4 and TLT. Indeed 17 potential probiotic strains of L. plantarum and Leuc. mesenteroides isolated in Italy were studied using functional cell modes, in order to confirm and support their probiotic potential. In collaboration with the Medicine Faculty of the University of Maribor (Slovenia) the cytotoxicity and the adhesion ability onto undifferentiated and differentiated human epithelial cell lines were evaluated. The attachment capability analyzed on differentiated and undifferentiated cells, showed significant differences in relation to the strains and to the two cell models. Overall, the strain S11T3E belonging to the species L. plantarum showed the better probiotic performances, increasing the mitochondrial activity of the epithelial cells, their transepithelial resistance when polarized and especially, reducing significantly the L. monocytogenes infection in undifferentiated epithelial cells and in lesser amount when tested in functional model. The latter characteristic was also shown by the strains S3T60C and S2T10D, respectively belonging to L. pentosus and L. plantarum species. The potential capability of L. plantarum to inhibit the invasion of L. monocytogens was already reported in literature (Nakamura et al. 2012), but to our knowledge, no similar studies were performed on strains L. pentosus species.
Multivariate analysis has been performed to check the suitability of the isolates to be analyzed by chemometric techniques. Principal Component Analysis (PCA) was used to discriminate among lactobacilli isolates, using a varimax rotation. Indeed chemometric analysis of the probiotic properties of all the tested strains was conducted by AUA with an aim to cluster the most promising strains as derived from the studied probiotic tests. Using the viable counts of LAB strains derived from the resistance tests to low pH and bile salts, hierarchical cluster analysis (HCA) revealed the presence of three well-differentiated groups (clusters). The lower cluster (Group I) contained strains that exhibited the highest resistance to low pH and bile salts, whereas strains with little or no resistance to low pH and bile salts were located in the upper cluster (Group III). Four strains that showed intermediate performance were located in a small cluster (Group II) between the two major groups. It is characteristic that the strains that presented the highest survival to low pH and bile salts were located in the same cub-cluster in Group I, including the 9 strains that initially had been selected: 4 strains of Lb. pentosus (E281, E97, E104, E108), 3 strains of Lb. plantarum (B282, E10, E69), 2 strains of Lb. paracasei subsp. paracasei (E93, E94).
Principal Components Analysis performed by INRB to aid in unfolding relationships among the data pertaining to the various probiotic parameters. The PCA biplot of probiotic parameters in LAB strains illustrates specific differences among them and grouping them in four groups. In general, strains didn’t show a great variability respecting the GS probiotic parameter. Only strain B95 was distant from the others seems to be different.

Responses of selected probiotic lactic acid bacteria to simulated conditions of fermentation.
The strains that showed the best probiotic characteristics were used as starters in simulated fermentation experiments. AUA and NAGREF have studied the responses of selected probiotic lactic acid bacteria to simulated conditions of fermentation including: resistance to brine environment, monitoring the attachment of Lb. pentosus B281 on olive fruit under different sterile brine solutions, monitoring the microbial interactions of selected strains in vitro (gel cassete system), monitoring the microbial activities and the biochemical profile in vivo (heat shocked olives) and in situ (naturally processed) (starters: Lb. pentosus B281, Lb. plantarum B282, fermentation season of 2010-11) [D3.1 (M18)] - monitoring the microbial activities of the inoculated strains during the second fermentation period (2011-2012) (starters: Lb. pentosus E97, Lb. plantarum E10, Lb. paracasei subsp. paracasei E93 & Lb. casei Shirota), -monitoring the aroma profile with e-nose.
Resistance to pH, salt, biofilm formation on olive epidermis performed by IG-CSIC and because no differences among products obtained with the selected probiotic LAB and the control, the final selection of probiotic strains was mainly based on their imposition in the olive brine and formation of biofilm on olive surface. In principle, strains 41 and 222 were selected because they were able to establish biofilms in the olive epidermis. i) Tolerance to phenolic compounds - tests on gene detection, oleuropein tolerance, oleuropein degrading and confirmation of oleuropein degradation by LAB by using HPLC methodology; ii) Esterase and ß-glucosidase genes detection, glucolytic activity and oleuropein degradation; iii) Aflatoxin binding in vitro and in situ; iv) Proteolytic activity performed by INRB as well as starter inoculation of heat shocked olives. Other tests of desirable characteristics for enzymatic activities and evaluation of some benefit to the host were performed. The effect of fermentation conditions, including the parameters pH and salt on the growth of Lb. plantarum (LAM 118) was studied using a response surface methodology (RSM) by INRB in collaboration with the Tunisian team.
Moreover, fermentations conducted in the last campaign (2011-2012) using two potentially probiotic strains from Lactobacillus genus, LAM 120 and LAM 118 (Lb. plantarum), were assessed for the microbial diversity that was determined by a culture-independent approach based on PCR-DGGE analysis. Finally, to continue selection in simulated conditions, the effect of NaCl concentration, pH and temperature on the growth of Lb.plantarum (LAM 118) was studied using a response surface methodology (RSM).
The technological performances of 10 selected isolates as starter cultures in simulated conditions of fermentation during the campaigns of 2010 and 2011, were monitored through a multidisciplinary approach by UNITO. Traditional microbial analysis coupled with molecular assays were performed during whole fermentation process in order to emphasize the effective presence of the inoculated starters. Moreover, chemical analysis of the organic acids present in brines and the sensory evaluation of olives were carried out to define the contribution of the starters in the definition of final characteristics of the product. In the campaign of 2011 five strains (O2T60C, S2T10D, S1T3B, O3T15B and FS50Q) were used as starter cultures in laboratory trials, while two strains (O1T90C and S1T10A), had already tested in laboratory during the 2010 campaign.
In traditional processes, as soon as the fruits are submerged in the brine, a spontaneous fermentation takes place, based mainly on the competitive activities of the indigenous micro?ora of the fruits. Generally, the initiation of a spontaneous fermentation process takes a relatively long time (24–48h), with a high risk of spoilage. During this early phase, which is associated with the lag phase of microbial growth, contaminating micro-organisms (mainly enterobacteria) increase in number and compete for nutrients in order to produce metabolites. The length of this phase can be signi?cantly reduced by inoculation with a selected strain of pure lactic culture. In our study it was evident that the population of lactic acid bacteria was considerably higher in the brines inoculated with the starter from the onset of fermentation. This difference had a pronounced effect on the survival period of Gram-negative bacteria (enterobacteria and pseudomonads), which were inactivated faster in the inoculated processes than in the control. This effect, although it has been reported previously in the literature for Spanish style green olives, is observed for the ?rst time in the natural fermentation of Conservolea green olives. The addition of L pentosus justi?es the advantage of using lactic starter cultures in order to decrease the possibility of spoilage, particularly during the ?rst stage of fermentation when the risk of gas pockets and butyric acid spoilage is highest. However, it must be emphasized that differences between inoculated and control processes were diminishing with time and were not detectable after 30 days of fermentation. This means that the bene?cial effects of inoculation are more pronounced during the early phases of fermentation. Once the lactic acid fermentation is well established, there is no difference between inoculated and control processes.
The beneficial effect of the inoculated LAB strains was also exemplified in the decrease of the Enterobacteriaceae population compared to spontaneous process, where this microbial group reached the lowest populations depending on the inoculated LAB strain and salt level in the brines. An important issue in inoculated olive fermentation is to control the survival of the specific inoculated strain during the process using molecular identification techniques, a fact that has received the attention of the scientific community only recently in similar studies (Saravanos et al., 2008; De Bellis et al., 2010; Rodriguez-Gómez et al., 2013). In this work, identification of LAB isolates using PFGE analysis revealed several changes in the survival of the inoculated LAB strains. Specifically, L. pentosus B281 was able to survive at high rates in both salt brines in contrast to L. plantarum B282 that could not be recovered from 10% salt bines at the end of the process. It seems that L. plantarum B282 was sensitive to high salt concentration and was not able to impose over the native LAB microbiota on the surface of olives. Interestingly, the same LAB strain could not be recovered at all in the case of mixed inoculum fermentation from the middle stage of the process (day 69), indicating a dominance of L. pentosus B281 making thus this strain a better selection for inoculated fermentation as it can survive at high rates in diverse salt concentrations and can also impose over the wild LAB strains on the surface of olives.
The responses of the brines with e-nose analysis were subjected to Discriminant Function Analysis (DFA) in an attempt to discriminate the fermentation process time according to the sampled profile of volatile compounds. Results showed that it was possible to achieve good differentiation of fermentation time based on e-nose measurements. This differentiation was clearer in the early stages of fermentation (1-20 days), whereas at later sampling times (32-53 days) this difference was less evident. However, no difference could be established among the inoculated and spontaneous processes and hence the e-nose approach is not effective in discriminating the different fermentation treatments, but rather in predicting the time of the process.
Selection of most suitable starter strains & Identification of selected strains on genus and species level
Different molecular methods were used in order to identify the isolates obtained from the olive fermentations. Multiplex PCR analysis with species specific primers, PCR, and Rep-PCR, DGGE, PCR-ARDRA and PFGE were the main methods used to reach the species level. Interestingly the LAB strains identified show to belong to different genera and species based on the country. While a pretty big diversity was found in Greece constituted by Leuc. mesenteroides and paramesenteroides, plus Lb. plantarum group including Lb. plantarum, Lb.pentosus, Lb. paraplantarum and Lb. casei group (including Lb. casei, Lb. paracasei), in Spain almost all the isolates were Lb. pentosus while in Italy Lb. plantarum and Leuc. mesenteroides. For Portugal and Tunisia the results underlined that the isolates belong to the Lb. plantarum. Results showed that from the Lactobacilli strains submitted to species molecular identification (by UNITO), 54, 36, and 9% were accomplished to Lb. plantarum, Lb. paraplantarum and Lb. pentosus, respectively. Spain and Italy also completed the genetic characterization of the isolates by the mean of the REP-PCR. In Spain a reduced LAB inter-specific biodiversity was noticed, with 94% of the isolates belonging to the Lb. pentosus species.
A total of 38 strains showed initially to have the probiotic characteristics and according to identification tests they belong to the Lb. plantarum group (Lb. plantarum, Lb. paraplantarum, Lb. pentosus). Specifically 9 strains were pre-selected in Greece belonging to Lb. pentosus, Lb. plantarum and Lb. casei group. In Spain 4 strains selected belonging to Lb. pentosus, in Italy 17 Lb.plantarum and Leuc. mesenteroides, in Portugal 3 Lb. plantarum, Lb. paraplantarum, Lb. pentosus, and in Tunisia 5 Lb. plantarum.
Based on their complete probiotic testing as well as on their performance in simulated conditions of fermentationd and in laboratory trials, a further selection has been performed to the most appropriate strains for use as starters in the large scale fermentation by the SMEs. In Greece the Lb. pentosus 281 and E97 and Lb. plantarum 282 have been selected as the most promising starter with the first being selected to be applied in industrial scale, in Spain the strains 41 and 222 both Lb. pentosus, with strain 41 being the most appropriate, in Portugal the Lb. plantarum, Lb. paraplantarum, Lb. pentosus with Lb. plantarum being the most preferred, in Italy two Lb.plantarum strains (O1T90C and S1T10A) and in Tunisia the Lb. plantarum LAM118 and LAM120.

WP3: Application of the selected probiotic lactic acid bacteria as starter cultures in small-scale controlled fermentations

Small scale fermentation experiments were performed in each country with the selected isolates possessing probiotic potential as starters improving the fermentation conditions according to the guidance of the local participating SMEs. Microbiological, physicochemical and organoleptic evaluation have been performed (HPLC of organic acids or sugars, GC of volatiles, phenolics). Results showed that in most cases the inoculated strains resulted in normal olive fermentation with satisfactory microbiological and physicochemical characteristics with some strains being better and more appropriate as starters. The period needed for the fermentation was almost 3 months. The “normal period of 3 months” established from our experiments should be considered as a reduced period with respect to green olive fermentation when no starter cultures are applied and low season temperatures exist, a situation that occurs in many small companies of the sector. Therefore the fixation (standardization) of the period of a normal fermentation for 3 months with the use of starter culture is a real progress for these industries.
Results showed also that the major organic acid detected in the brines was lactic acid, followed by acetic acid to a lesser extent. Other acids detected in lower concentrations were succinic, citric, malic, and formic acid. It needs to be noted that in the spontaneous fermentation, there was a decrease in the concentration of lactic acid after 20 days of the process with a concurrent increase in propionic acid. After 55 days, no lactic acid could be detected in the brines and at the same time the concentration of propionic acid was maintained at 20-21 mM. This event is a clear deviation from normal fermentation, as these olives received the lowest scores in the sensory assessment. This observation justifies the importance of using starter cultures in Spanish-style green olive fermentation. No propionic acid was detected in any other fermentation process. Ethanol and methanol were the major volatiles in brines, whereas ethyl acetate, methyl acetate, acetaldehyde, propanol, and 2-butanol were detected in lower amounts. Ethanol, the most abundant of the volatile compounds was present in high levels and its concentration varied greatly between the different fermentations, reflecting varying degrees of activity by yeasts and hetero-fermentative lactic acid bacteria. The major phenolic compounds in the brines were hydroxytyrosol and tyrosol in comparable concentrations regardless of NaCl initial level in the brines. These compounds can be considered as the main products of oleuropein degradation. Vanillic acid was also detected but at much lower concentrations. The most exhausted sugars were fructose and glucose while a residual proportion of sucrose was left even at the end of the process. Mannitol was poorly used by the microorganisms.
Assessment of organoleptic profile showed that overall, higher acceptability scores were given by the panellists to inoculated processes with the probiotic starters except of the case in Italy where the inoculated strains were not detected at the end of the fermentation. Molecular characterization (PFGE) of isolated colonies during the fermentation defined the survival rate of the starters used and electron microscopy the formation of biofims. All the probiotec strains selected had improved characteristics with respect to the reference strains with well recognized probiotic characteristics.
At the end of fermentation, olives were assessed organoleptically by a taste panel of 9-11 members according to the organoleptic protocol developed by the International Olive Council (IOC) (D3.3). The attributes taken into account were divided into three classes, namely negative attributes (abnormal fermentation with main focus on the presence of off-odours), positive attributes (saltiness, bitterness, acidity), and kinaesthetic sensations (firmness, crunchiness). Moreover, an additional descriptor was taken into account expressing the overall acceptability of the final fermented product by the taste panel. Results were presented as spider plots containing the median value for each attribute based on the scores provided by each member of the taste panel. Results for olives of Halkidiki variety fermented in Greece showed that inoculated fermentations with the probiotic LAB presented better consistency in terms of sensory characteristics. Olives fermented with the indigenous microbiota in 8% salt presented off-odour development and received lower scores by the panelists.
The use of probiotic starter culture in Spanish green table olive processing, using Manzanilla fruits, does not lead to any abnormal or anomalous fermentation. Furthermore, gustative perceptions were similar among treatments with the only exception of some kinaesthetic sensations (hardness, mainly, and crunchiness) which scores were significantly improved with respect to the spontaneous process when the fermentation was inoculated with the potential probiotic strains.
Sensory evaluation of Portugese variety showed that the panellists tended to prefer the spontaneous fermentation product (control) followed by the fermentation with L. plantarum. This preference was in accordance with the specificity of the inoculated assays, where a pure lactic fermentation and as a result the metabolites produced by yeasts are absent influencing the sensorial quality. One of the negative attributes referred consistently (trained panellists) was the ‘cooked taste’ in all of the thermal treated fruits. In general, the kinaesthetic sensations (hardness and fibrousnesses) were referred as good independent of the samples (control and inoculated).
The results of sensory evaluation of the Italian varieties highlighted a different organoleptic profile of each fermentation. The fermentations carried out with the strains O2T60 C, S2T10 D and FS50 Q received the highest score, mainly due to low bitterness and good acidity. INSAT has noticed a huge diversity in taster responses probably due to the lack of the panel expertise.
The physico-chemical/biochemical parameters determined during the fermentation processes under were combined with sensory assessment data in an attempt to find any correlations between the two different types of data sets. The data were analyzed using Partial Least Squares Regression (PLS-R) analysis. Two different approaches were followed by NAGREF & AUA: a) Correlation of physicochemical parameters (pH, titratable acidity, salt content, combined acidity, concentration of organic acids) with sensory assessment (off-odours, acid, salty, bitter, hardness, crunchiness) and overall acceptability of fermentations and b) Correlation of volatile profile with sensory assessment of fermentations (D3.2). The PLS regression map where the sensory scores are correlated with the physicochemical characteristics has shown that the spontaneous fermentation at 8% NaCl (denoted as Control 8%) was related with off-odour. This group also received the lowest acceptability from the taste panel as it is located on the opposite quadrant of the preference map. Fermentations with 10% NaCl are correlated with the NaCl variable. The fermentations with Lb. pentosus B281 and Lb. plantarum B282 at 8% salt level are correlated with high overall acceptability. The volatile compounds analyzed by GC-FID can also be clustered in three major groups. Specifically, 2-butanol and methanol form one group, propanol alone forms another group and the remaining compounds form the third group. Spontaneous fermentation at 8% NaCl (denoted Control 8%) was correlated with 2-butanol and methanol, and also with the sensory attribute ‘off-odour’. This observation is in good agreement with PLS regression map where the sensory scores are correlated with the physicochemical characteristics, and the same treatment is correlated with off-odour. On the contrary, inoculated fermentation with Lb. pentosus B281 with 8% salt in the brine (denoted Lb. pentosus 8%) is correlated with the majority of volatile compounds analyzed. Indeed, the presence of ethanol and volatile esters (ethyl and methyl acetate) provide a pleasant sensory aromatic profile to this fermentation process.
The overall PLS analysis, considering sensory attributes or descriptors as variables to be predicted as a function of the biochemical parameters, were useful for studying the relationships among them, segregate treatments (inoculated olives with different potential probiotic strains and spontaneous fermentation) and for deducing the corresponding equations which permitted express the sensory scores as a function of the biochemical variables. The goodness of the prediction was assessed by graphing the real values obtained by the panel of assessors vs. the estimated ones, using the equations. Most of the deduced models were able to predict the sensory attributes with fairly good accuracy.

Subselection of the probiotic strains has been done in order to be used in industrial fermentations.
In Greece two strains of lactic acid bacteria have been selected for their in vitro probiotic potential, namely Lb. pentosus (B 281) and Lb. plantarum (B 282). Results showed a good response of the inoculated starters in olive fermentation. However, brines inoculated with Lb. pentosus B 281 presented higher acidity and hence lower pH values compared with other fermentations.
In Spain four strains of the most promising probiotic characteristic strains have been selected to inoculate experimental fermentations. Strains 41 and 168 may have a higher survival to the alkaline initial conditions than strains 222 and 327 however strains 41 and 222 showed better adherence to olive biofilm formation. In Portugal three strains were selected for the fermentation assays and used successfully as starters: Lb. plantarum, Lb. paraplantarum and Lb. pentosus. In Italy Four strains of Lb. plantarum and one strain of Leuconostoc mesenteroides were used as single starter cultures in small-scale laboratory fermentation Overall, the 4 strains of Lb. plantarum (O2T60 C, S2T10 D, S1T3 B, O3T15 B) and Leuc. mesenteroides strain (FS50 Q) used in small-scale laboratory fermentations while in Tunisia a Lactobacillus spp. showed good behaviour in fermentation. Only two strains (S1T10A and OL1T90C) were able to conduct the technological process throughout the 90 days of fermentation and in particular the strain S1T10A was present in every sampling of olives and brines until the end of the trial. Instead, in the other three trials, the inoculated strains (OL1T90E, S1T30B and OL1T90B) were isolated just until 30th day of fermentation and were not able to dominate the native microbiota developed during fermentation because they respectively represented the 18%, 12% and 7% of total isolated LAB strains.

WP4. Evaluation of shelf life of the final product under different storage conditions
Table olives that had been fermented successfully with the most promising probiotic strains were used to study their shelf life under different packaging systems and storage conditions. Emphasis was given in the survival of the probiotic strains during the shelf life. DEMETER and AUA (EL) studied the packaging of table olives from WP3 either in MAP (70%N2:30%CO2) without brine (for 6 months) or in brine at 4°C and 20°C (for 12 months) and the survival of the probiotic strains [Lb. pentosus (B 281) and Lb. plantarum (B 282) that have been used as starters], while the overall quality of the final fermented product, was assessed.
-Packaging and storage of fermented olives in brine at 4 and 20oC and survival of the probiotic starters: The results showed that the final population of lactic acid bacteria (after 12 months) in both brine and olive flesh reached approximately 5 log cycles, while the same bacterial group presented counts ca. 1.0-3.5 log CFU/mL and 3.5-4.0 log CFU/g in brine and olive flesh, respectively, at 4oC. PFGE analysis revealed that Lb. pentosus B281 and Lb. plantarum B282 showed a high survival rate with a recovery of 100 and 96%, respectively at 4oC, whereas the survival rate of both probiotic strains was less than 20% at higher storage temperature after 6 months of storage. At the end of storage period (357 days of storage), only Lb. pentosus B281 showed high survival rate during storage at both temperatures. Finally, in the packing that contained olives fermented by the mixture of both strains, Lb. pentosus B281 dominated over Lb. plantarum B282 throughout storage at both temperatures. Specifically, the survival rate of Lb. pentosus B281 reached approximately 96% at lower storage temperature, whereas the survival rate was 50% at 20oC.
- Packaging and storage of fermented olives without brine in MAP at 4? & 20? C and survival of the probiotic starters: The green olives fermented with lactic acid bacteria exhibiting probiotic potential were found to contain satisfactory amounts (5.5-6.0 CFU/g) of LAB after a year of storage at both 4 and 20°C. The sensory attributes of the product received higher values when stored at 20o than at 4oC, whereas at both cases it was characterized more preferable after 6 months than 12 months of storage. The PFGE analysis demonstrated that the survival of the inoculated strains depended on the storage time and temperature. In the Lb. pentosus B281 case, higher survival rate was shown after 6 months of storage compared to the end of storage period at 4°C, whereas lower survival rates were observed during storage at 20 °C. In Lb. plantarum B282 case, very high survival was shown after 6 months of storage and lower after 1 year at 4 °C, but was not detected at all at both time points at 20 °C. In the mixed culture case, the strain B281 dominated over the strain B282, and showed 100% survival at both middle and end of storage period at 4 °C. At 20°C both of the strains were detected at the end of storage, but the survival rates were low. Finally, regarding the identification of the rest microbiota isolated, different strains of Lb. pentosus and Lb. paraplantarum were found with a high biodiversity being observed, especially at 20°C.
- Packaging of olive paste in vacuum at 4o and 20o C: The olive paste from green table olives, previously fermented with lactic acid bacteria exhibiting probiotic potential was found to contain satisfactory amounts (=6.0 log cfu/g) of LAB at the end of the storage period at all cases. The strain Lb. pentosus E97 showed the highest survival rates, followed by Lb. paracasei subsp. paracasei E93. In the case of inoculated green olive paste with Lb. pentosus B281, Lb. plantarum GG, both as monoculture and co-culture at both 4°C and 10°C, low recovery rates were attained at the end of storage period, although the paste was preserved without spoilage.
Results of INRB indicated that environmental temperature and storage in pouches with brine seem to be the best conditions for storage during 90 days, for all strains and all treatments. The best results seems to be showed in olives in brine, at 22 ºC storage temperature. They could be related with brine characteristics. In conclusion, it must be emphasized that the population remained at or above 5 -6 log CFU/ml in the end of storage time (90 days). Thus, it can be stated that application of refrigeration in storage is not necessary for maintaining the characteristics both physic-chemical and sensorial of the product during storage time (90 days). Room temperature is, in general, more favourable (22 ± 2 ºC). Statistical studies showed that there are no differences between samples of raw material and between samples from different treatments after the shelf life. At 4ºC storage temperature there is a high sensibility of LAB to low temperature.
On the other hand IG-CSIC studies showed that no significant changes in pH, free acidity and combined acidity of cover brines during storage at 7ºC and 20°C are observed in all bags and bottles. According with this stability it can be said that all packed olives were stable without alterations for the studied time (8 months). A high amount of information on LAB and yeast survival in brine and on olive surface has been obtained and an overview of the expected survival over this period derived. The largest population of lactic acid bacteria with probiotic characteristics of the olives surface was when fruits were packed in glass jar with cover brine or in plastic bag with N2 atmosphere and containers were stored at ambient temperature 20ºC. Maintaining low temperatures (cooling) implies a decrease in desired population of probiotic lactic acid bacteria. Also these packaging conditions and storage temperature produced a slight loss of superficial colour and no influence on the firmness. In summary, it can be stated that the storage of probiotic olives in diverse containers was very useful to estimate the changes in the main physicochemical parameter and in the lactobacilli and yeast population on the olive surface. Apparently, packing in plastic bags, plastic bags with brine, and glass jars with brine lead to similar end characteristics. With respect to the lactobacilli population that remained on the olive surface, there were only slight differences and ranged around 6 logcfu/mL. Approximately, final products may contain about 6 logcells/olive. In order to improve storage and preservation it is advisable to correct the combined acidity at the end of fermentation to values around 110 or 25-50 mN after packing. Similarly, it is recommendable to use slightly higher percentages of titratable acidity (0.6%) and NaCl (6.0%) in packing that those usually applied (0.5% and 5.0%) to improve stability and prevent pH and acidity decrease and the subsequent possible spoilage. Then, in this case, the shelf life would also be enlarged. Finally the overall study has demonstrated that the best conditions for probiotic fermented olives were the following:
Physicochemical conditions: NaCl concentration at the equilibrium: 5.5%, Titratable acidity at the equilibrium: 0.6%
Container’s type: Glass jars, when brine was used as cover solution or Plastic bags in Nitrogen atmosphere, when olives were packaged in dry
Storage temperature: Room temperature (around 22ºC)

WP5: Modelling the fermentation kinetics and survival during storage of the selected probiotic lactic acid bacteria
The growth and survival of lactic acid bacteria (LAB) during green olive fermentation was simulated using different fitting models. More specifically, the counts (log cfu/g) of the LAB on olives, during the fermentation were modelled as a function of time using the primary model of Baranyi and Roberts (1994), and the kinetic parameters were estimated. The data fitted to these models originated from the 2 fermentation sets, of the years 2010-2011 and 2011-2012. In order to fit the dynamic changes of the inoculated LAB or the LAB from the indigenous flora during the fermentation of green table olives, the actual growth-survival curve of microbial population has been divided in 2 time periods; the “growth phase” and the “survival phase”. The estimates by the primary model of Baranyi and Roberts for the initial, maximum and final population, lag phase, growth rate and reduction rate and stationary phases were summarised and the goodness of fit as well as R2 were calculated to evaluate and compare the models. The majority of the models were shown to have a good fit (R2 0.90-1.00) especially in the “growth phase” at both fermentation seasons (2010-2011 and 2011-2012). In the case of the “survival phase”, the models of 2010-2011 season showed higher goodness of fit in comparison with 2011-2012 season. Nevertheless, at both seasons, the “growth’’ models exhibited higher accuracy in comparison with the “survival’’ models. The survival of the inoculated probiotic strains during fermentation has been assessed by PFGE and their identification with specific PCR.
The potential of Artificial Neural Networks (ANN) to simulate the growth/survival kinetics of the population of LAB and yeasts during fermentation was also investigated. Initially, the type of network chosen was a multilayer perceptron (MLP) that was proved quite efficient in simulating the growth/survival profile of probiotic lactic acid bacteria in the brines together with the evolution of the population of yeasts. The feasibility of using Radial-basis function networks (RBFNs) has been explored for the same reason. Results indicated that artificial neural networks, both multilayer perceptron (MLP) and radial basis function (RBF), although considered as data driven approaches, can be effectively employed for non linear regression purposes in small datasets.
The consequent storage and evaluation of the different fermented olive products’ shelf life, has been performed. An excess amount of data were collected for olive products stored in brine, MAP or vacuum at different temperatures. The survival of the LAB was not modelled for the storage of these products, since the changes of the specific microbial group were not dramatic. However, it was considered crucial to assess the survival of the inoculated strains during storage through the use of PFGE, since the reference of a stable population above 6 logs is not sufficient to guarantee the presence of these strains.
Moreover the effect of temperature, pH and salt on growth kinetics on different LAB strains with probiotic characteristics has been modelled. The growth was monitored by means of optical density (OD) measurements, and the biological growth parameters (µmax and ?) (primary modelling) estimated by means of the re-parameterized Gompertz equation proposed by Zwietering et al. (1990). Then, the secondary modeling was carried out using the cardinal model with inflection (CMI) proposed by Rosso et al. (1993). It describes the µmax changes of LAB as a function of environmental variables. In this study the approach consisting of estimating the area under the respective growth curves. For the mathematical modeling of the growth of LAB and yeasts during fermentation Pruitt and Kamau (1993) and Bello and Sánchez Fuertes (1995) models were used. Consequently, according to Pruitt and Kamau, the maximum population (maximum asymptote), the growth rate, the time at which the Nmax/2 is reached, the part of the population fatally damaged, and the death rate were calculated, whereas Bello and Sánchez Fuertes model calculated the initial population of micro-organisms, the increase of micro-organisms from the initial level to the maximum level, the relative growth rate of the microorganism, the time at which growth rate is maximum, the decrease of micro-organisms from the maximum level to a minimum level, the relative death rate of micro-organisms, the time at which death rate is maximum. Finally the performance of the models was compared using R2.
The application of cardinal parameter models to the preselected potential probiotic LAB strains led to a proper modelling of the diverse growth parameters and the estimation of these important characteristics, useful for the further selection of strains.
The growth (inoculated or spontaneous) of LAB and yeasts populations were successfully modelled in the cover brines during fermentation. In case of inoculation, the most appropriate model for LAB was Pruitt and Kamau (1993) due to the initial death phase observed in the added cultures; in case of spontaneous fermentation, the model of Bello and Sánchez Fuertes (1996) was more convenient. Unfortunately, changes in the LAB and yeast populations during packaging in both cover brines and olive surface could not be modelled with the equations available in the literature. In spite of this drawback, a high amount of information on LAB and yeast survival in brine and on olive surface has been obtained and an overview of the expected survival over this period derived. According to this data, the best packaging system which allowed the survival of LAB population for longer period of time was plastic bag with nitrogen at 20ºC.
The survival of the LAB population in the experiments carried out at industrial fermentation scale was different depending of the inoculation date (September or October). High populations of the LAB2 strain (around 7 log10 cfu/cm2) were obtained in all inoculated fermenters, while the population of yeast was lower (between 5-6 log10 cfu/cm2). Both adapted and laboratory LAB2 starters showed a good percentage of imposition during validation at industrial scale.

WP 6: Risk assessment and predictive modelling of potential pathogenic bacteria on the final product
It is only recently that consumer concern for food safety and high demand for traditional food products is becoming an important challenge for the food industry. However, safety issues about the hygienic condition and the prevalence of foodborne pathogens for a diversity of popular traditional fermented foods of vegetable origin is rather limited. Until recently, the production of these food commodities presented heterogeneity due to geographical area and local practice resulting in a final product with diverse microbiological, physicochemical and sensory characteristics (Frankel, 2011). The condition has drastically changed over the last years where the production has shifted from artisanal practice to industrial level under strict processing and hygienic conditions (Panagou et al., 2013). This is the case with table olives in which the microbiological criteria set for these products stipulate that the final product shall be free from microorganisms and parasites in amounts which may represent a hazard to health and shall not contain any substance originating from microorganisms in amounts which may represent a hazard to health (Anonymous, 1987; IOC, 2004). In addition, fermented olives held in bulk in a covering liquid may contain lactic acid bacteria and/or yeasts used for fermentation, the number of which in a selective culture medium may be for each one up to 9log cfu/ml of brine or per gram of flesh depending on the level of fermentation. In many countries, the main volume of processed olives is marketed in bulk where the product is freely available to consumers in retail outlets in open containers, stored at ambient temperature, exposed thus to potential post- processing contamination from the environment due to uncontrolled hygiene and unsanitary procedures by the personnel as well as inadequate cleaning and sanitizing of storage containers and other utensils. On the other hand, olives are considered as ready-to-eat products that are consumed without prior cooking making thus contamination with pathogens a potential public health concern. It is thus necessary to provide scientific data about pathogen responses on fermented olives in an attempt to increase our knowledge on the microbial ecosystem of the product and provide a pool of data for future risk assessment studies. So far, the behaviour of different pathogens in olive fermentation has been reported for Listeria monocytogenes (Randazzo et al., 2012), Escherichia coli O157:H7 (Spyropoulou et al., 2001; Skandamis and Nychas, 2003), Bacillus cereus (Panagou et al., 2008), and Salmonella enterica ser. Enteritidis PT4 (Panagou et al., 2013). The results of these studies showed rapid decline of the population of the pathogens during fermentation indicating that the brine environment does not support pathogen growth provided that proper process control is ensured. Table olives have also been suspected for botulism intoxication caused by the growth of Clostridium botulinum type B in Italian fermentations of both green and black olives (Fenicia et al., 1992; Cawthorne et al., 2005). On the other hand, market surveys have been undertaken in Spain (López-López et al., 2004), Italy (Caggia et al., 2004; Franzetti et al., 2011; Tofalo et al., 2012), Portugal (Pereira et al., 2008), and Greece (Panagou et al., 2006) to determine the microbiological profile and the presence of pathogens in commercialized table olive sampled directly from retail outlets. Results showed that proper fermentation conditions together with good hygiene practices during and after processing did not allow the presence of pathogenic bacteria on olives. However, there is little information about the behaviour of post-processing contaminating pathogenic bacteria on olives packed with or without brine.
The fate of three major foodborne pathogens inoculated in fermented green olives has been studied. Green olives in brine 6% (w/v, NaCl) previously fermented by a strain that showed probiotic properties (Lactobacillus pentosus B281) and olives previously fermented by the indigenous microbiota (control), were inoculated with 3 different pathogens Escherichia coli O157:H7, Salmonella Enteritidis, and Listeria monocytogenes (co-culture of 5 strains at each case) and stored at 20o C. In the case of olives previously fermented by Lb. pentosus B281, the population of all pathogens reduced gradually during storage, with E. coli being detected even on the 27th and 19th day of storage in the brine and olive fruits respectively, S. Enteritidis detected on the 21st day of storage in both brine and olive fruits and L. monocytogenes declined slowly reaching 1.5 log cfu/ mL and 1.0 log cfu/ g in the brine and olive fruits respectively, at the end of storage period (i.e. 48 days).
In black fermented olives, a strain of Listeria monocytogenes survived at 4oC beyond the end of storage (20 days) while at 20oC, only B-129 strain managed to survive until day 15th. Almost all strains of Escherichia coli O157:H7 were enumerated after 24h of the experiment. From the second day and then until the end of storage the levels of the pathogen were below the detection limit and all samples were found enrichment negative. A similar trend was observed for Salmonella Typhimurium and Salmonella Enteritidis. At 4oC, Staphylococcus aureus ATCC 6538 and B-135 strains as well as the mixed culture were enumerated until the 2nd day of storage while at 20oC the counts were at the same time below the detection limit.
In situ studies on surveillance of pathogens in sterile fermented brines: Assays were developed in black olive brines previously fermented with potencial probiotics (B96, A1, 614) and with natural wild microbiota (control) in the end of fermentation process from WP3. The culture medium consisting in sterilized brines previously characterized according pH, titrable acidity, and sodium chloride level and after inoculated in co-culture of each potencial probiotic LAB with each selected pathogens (Salmonella Enteritidis, Listeria monocytogenes and Escherichia coli ATCC 8739) simultaneously. Bacterial growth was monitored daily along 2 months, at room temperature (20-22 ºC). Results indicated the non survival of pathogens in olive brine conditions.
A literature review about the prevalence of pathogens in fermented olives revealed that a survey in Italy resulted that green olives are unfavorable environment for the survival of pathogenic microflora on olives while in a survey of green olives in Spain, spore-forming mesophilic aerobes (e.g. Bacillus cereus), were present in only some samples and were always at low levels. In a similar survey in Greece no enterobacteriaceae, pseudomonads, Bacillus cereus, or Clostridium perfringens were detected in any of the olive samples. Studies with inoculated pathogens showed that although they declined rapidly they managed to survive even in low numbers. In another survey in Italy from a total of 69 samples of olives, 36 were contaminated with Listeria from which 26 were identified as L. monocytogenes. Overall, the existing data from the literature suggest that the brine environment seems to be hostile for the survival of foodborne pathogens in case of post processing contamination scenario. In addition, AUA has investigated the case of post-processing contamination of natural black olives with different strains of L. monocytogenes, E. coli O157:H7, Salmonella spp. and S. aureus. Olives were stored without brine at 4 and 20°C and the obtained results indicated rapid decline of all pathogens from the 2nd day of storage at both storage temperatures, suggesting that the olive surface itself is an adverse place for the attachment an survival of pathogenic bacteria.
A survey was conducted in Greece for several types of fermented olives available, to evaluate their safety regarding several pathogens such as L. monocytogenes, E. coli O157:H7, and Salmonella spp., Clostridium perfrigens and coagulase positive Staphylococci. Seventy eight (78) samples of olives and olive products of different origin were tested in total, for the presence of the aforementioned pathogenic bacteria. It was demonstrated that none of the pathogens could be detected in the total of the 78 samples.
In an attempt to apply predictive models on pathogen survival, the modified Weibull model was fitted to the experimental data to describe the survival kinetics of the pathogens Listeria monocytogenes, Escherichia coli O157:H7, and Salmonella Enteritidis in the brine during storage of olives. Indeed the modified Weibull model provided a good fit to the survival data of all pathogens. This model was selected on the basis that it describes both the decrease of pathogen population during storage and the residual bacterial concentration when a tailing effect is observed. Convex (p>1) survival curves were obtained for E. coli O157:H7 and S. Enteritidis throughout storage in both covering brine and olive fruits indicating a shoulder effect in the obtained curves. It can be concluded that S. Enteritidis showed the highest survival as it needed 3.0 days for a 1 log reduction in the population of the pathogen in the covering brine of the pouches, followed by E. coli O157:H7 and L. monocytogenes with 1.96 and 0.28 days, respectively. A similar pattern was observed for the survival of pathogens on olive fruits, but in this case the values of the d parameter were lower, especially for E. coli O157:H7 and S. Enteritidis, indicating lower survival ability of these pathogens on olive fruits compared to the brine.
In conclusion, the results demonstrated that even though the growth of the studied pathogenic strains was not supported, they may survive for a long period in a stressful environment of a fermented product with low pH value (4.2) and high salt concentration (6.0%) and thus they are a valuable contribution to risk assessment studies related to ready to eat foods in general. Moreover, since the application of appropriate thermal treatment cannot be applied in the case of a functional/potential probiotic product to ensure the safety of the product, the application of strictly good manufacturing practices to reduce the possibility of cross contamination during packaging or post processing, is a major priority.
Monitoring the Enterobacteriaceae during the industrial fermentation, the inoculation systems developed for the industrial scale application of starter cultures has shown that Enterobacteriaceae were never found in the olive biofilm in the two experiments carried out in the 2012/2013 season. This means that the new inoculation procedures used for inoculation, in addition to represent an efficient system for assuring the implantation of the added starter culture and its specific LAB strain, may also contribute efficiently to prevent safety risks involved in table olive fermentation. Therefore, the project not only has led to the development of new functional table olives but also to the production of safer olives.
In a study where the enterohaemorragic pathogen Escherichia coli O157:H7 strain subjected to an artificial process of human digestion with fermented olives as meal has shown that the strain was very sensible to the low pH of stomach, while it was less affected during its transit during duodenum. On the contrary, the microorganism grew in the jejunum and ileum because of the higher pH in both compartments. In this way, the overall survival of the strain to all digestive process was 117.5%, which is indicative that this strain could grow during real digestion with olives as meal.

WP 7: Consumer acceptance studies
The objective of this task was to explore consumer’s perception and acceptance of functional food in relation to the traditional (conventional) product. It is essential that the organoleptic characteristics of the new fermented probiotic olives must be as similar as possible to the traditional (spontaneous) product. Then, a comparative evaluation of the new developed product against the traditional (spontaneous) one is necessary. To achieve this task, the olives were first tasted by a trained panel, fully described in D3.2 for olive classification and later they were subjected to the consumer’s test. This first classification was carried out with the objective of avoiding the use of spoiled or unmarketable olives in the consumer’s tests.
As conclusion of the consumer’s test conducted in different countries, it can be stated that the fermentation of olives with probiotic starter cultures did not introduce any changes in the sensory characteristics of the traditional table olives. The overall acceptability of the product exhibited very high values and the majority of the consumers declared that they would buy this product. Therefore it can be assumed that the new developed functional product could be commercialized without any trouble. Furthermore, possibly, it will cover a different niche among the consumers interested in the relationship between diet and health that is unattainable by the traditional product which is mainly identified as appetizer and as a complement for certain dishes. Then, the combination of probiotic characteristics and lower levels of salt (which may increase the survival during bulk storage and shelf life) can lead to a new functional food with a renewed table olive image fairly appropriate for specific populations from developed countries concerned for the convenience of a healthy diet.

WP 8: Test Field
The selected potential probiotic starters had been used in laboratory scale fermentations of olives. Some of the starters had been proved to be very effective to complete the fermentation and to produce green olives with acceptable sensory characteristics similar or better to those of a spontaneous fermentation. Moreover these starters had been able to survive on olives until the end of the fermentation in high numbers sufficient to characterize the product as probiotic. The survival of the potential probiotics had been successful also during packaging that followed the fermentation. The next step to establish the capacity of these strains to be used as starters by the industry was to apply them in medium or large scale fermentations with the active involvement of the participating industries.
The results obtained from the large scale fermentation process showed that the selected strain of Lb. pentosus B281 was able to dominate the process and maintain high population levels above 6.0 log CFU/g of olives after 120 days of fermentation, indicating successful scale-up of the process from laboratory to industrial level. The strain was adapted successfully to the salt environment of the brine and the attained acidity in the brine was also satisfactory. Finally, molecular analysis showed high recovery rates of the starter at both the beginning and final stages of fermentation.
The experiments carried out at pilot and industrial scale have also demonstrated by the first time the possibility of fermenting olives with specific starter cultures. In the past, the inoculation was not possible to be driven with respect to the imposition of the added strain and in fact it is very likely that the initial strain introduced, initiated the process but, later on, it were substituted by the wild LAB microbiota. This possibility might have been facilitated by the inoculation procedures. Furthermore, it could have been also possible that the starter cultures were completely substituted from the very beginning by the wild LAB and that the inoculation were absolutely ineffective.
The progressive inoculation scaling planned in the project has shown to be very convenient to study in detail the starter culture imposition process and disclose the difficulties and uncertainty involved in the past inoculation procedures. Then, the project, on the one side, has markedly improved the inoculation system at industrial scale, facilitating in this way the utilization of any starter culture and, on the other hand, has proved that, when the starter culture is a probiotic LAB, this can be imposed over the wild strains not only in the brines but also on the olive surface. Thus, the objectives proposed in the project have surpassed all expectations and have established the bases for probiotic olive production.
Furthermore, the association of LAB and yeasts in brine was well known but their association on the olive biofilm not only in small scale fermentations but also in industrial fermentation. The possibility of controlling the LAB strain on the olive biofilm is a complete novelty generated in this project. As result, the project has also open new and completely unexplored research lines that have expanded the frontiers of the olive fermentation further away the current basis of the table olive investigation.
Finally the Association of Greek Food Industries (SEVT) has organized for the second consecutive year the national ECOTROPHELIA 2012 contest for the creation of new eco-innovative food products. The competition was held in Athens, on September 10, 2012, and it was part of the wider ECOTROPHELIA EUROPE contest. The aim of the competition is to motivate Universities and Research Foundations engaged in food science and technology to develop innovative food products with high ecological impact that will provide a pool of innovative ideas for the food industry. The Agricultural University of Athens (AUA) participated in the contest with the product ‘Probiotic green olives’ that was qualified for the final stage of the competition and it was awarded the second prize (silver award). The product has been developed through the PROBIOLIVES project.



Potential Impact:
The Consortium expected a number of potentially exploitable results which are summarized together with their type of exploitation by each association or company in the next tables, as described in the Annex I –“Description of work” the grant agreement 243471. The exploitation of the project’s results after the training of the SME-AGs key personnel, and after the training of SMEs involved is the most significant impact of this project. Our first aim was to protect the IPR owned by SME-AGs by patent applications, as well as to release non-confidential research findings from this project into the public domain, though oral presentations at conferences and in peer reviewed journal articles or in the website. Making the project outputs available in this way, this will maximize the impact of the project.

Management of knowledge and intellectual property: The Consortium has chosen the default Intellectual Property Rights regime for projects for the benefit of SME-AGs. This means, giving full ownership of all project results and intellectual property rights to the SME-AGs participants which will use them for their benefit while their SMEs will have license to use them. The RTD performers are remunerated for their work. The Consortium agreement precisely states how the partners associations share or jointly own these results according to the following rules as described in the CA.
The following activities have been performed for the exploitation of results:
• Development of the new effective and reliable quality assurance and optimization management system for the olive fermentation process in the SMEs.
• Collection of data regarding the training and knowledge transfer needs of the participating SMEs. Training activity regarding the personnel of the participating SMEs.
• Activities for the development and launch of new products that will use the new probiotic starter technology. These activities, intending to increase firm competitiveness in domestic and foreign markets included: (a) A chain of meetings-workshops with the directors of the participating firms. (b) A chain of communications with the principal clients/ agents of the participating firms in their export countries.
• Attempt to incorporate information about the new products in the marketing and advertising campaigns of the participating companies to ameliorate the participating-companies-food-products’ safety and innovativeness profile across the wider public.
• A public relations / press campaign targeting the food sector.
• Training actions to educate young & experienced food scientists on the new product technologies.

Exploitable results of the project and their impact:

Project Result 1: Lactic acid bacteria with probiotic properties
A significant number of different lactic acid bacteria has been isolated from the olive natural microbiota in Greece, Spain, Italy and Portugal and found to possess probiotic properties in vitro belonging mainly to Lactobacillus species. The isolation of lactic acid bacteria with probiotic properties from olive microbiota has a great impact not only for olive products but also on other food commodities as they can be introduced in other foods as supplements. These bacteria also enrich the bank of probiotic cultures available for further tests and for use as heath promoting live cultures. Their successive use as starters in olive fermentation offers the opportunity to achieve a better controlled fermentation with a final product of good quality that is functional with added value for the consumers. Their use as starters in olives will develop functional olives.
Exploitation form: The isolated bacteria belong to the culture collections of the Research organization in each country and are available for further use by the corresponding associations. After the completion of the olive fermentation experiments the selected bacteria that were able to survive and complete the fermentation were characterized and decided by the GA to be patented for the benefit of the associations (see below). Scientific results not subjected to protection had been externally disseminated by the RTDs via presentations in scientific conferences (national, European or international) and publications in scientific journals. However since 14 December 2012 the EU regulation 432/2012 entered into force. According to this regulation “no food can have on its label the term «probiotic»”. Since then the EFSA has rejected many heath claims for foods related to probiotics benefits as there were not enough reliable clinical studies to substantiate scientifically the claims. Considering this situation there is a major concern by the participating RTDs to search for funding to continue research in order to establish the probiotic properties of the selected strains in vivo and moreover with clinical studies in human. Our aim is to gather more data to contribute in the scientific substantiation of the particular health claims of “our” probiotic isolates.

Project Result 2: Probiotic bacteria used successfully as starters in small scale fermentation
Certain lactic acid bacteria that found to possess probiotic properties have been tested successfully as starters in small scale fermentations. They were able to survive until the end of the fermentation in high populations ac.ceptable for a probiotic food and they were able to produce fermented olives with acceptable physicochemical and organoleptic characteristics in order to be able to produce functional table olives
Exploitation form: Two applications for patents have been done, in Spain by the Spanish SME-association - ASEMESA [patent No. P-201131011 “Use of autochthonous lactic acid bacteria with probiotic features in the preparation of food products”, (date of application 16-6-2011)] and in Greece by the Greek Association - PEMETE [patent No. 20110100600 “Functional table olives fermented with lactic acid bacteria exhibiting probiotic properties” (date of application 14-10-2011)]. A patent sharing agreement has also been signed between the associations after the patent application according to the CA. Relevant scientific publications followed after the application for patents. Scientific results not subjected to protection had been externally disseminated by the RTDs via presentations in scientific conferences (national, European or international) and publications in scientific journals. This kind of dissemination will continue well after the end of the project.
Moreover, the knowledge acquired has a great impact and plays a major role in offering a better understanding of the olive fermentation process as it involves a series of microbiological assessments of the starter, along with monitoring of metabolic profiling throughout the fermentation. This would have a significant impact on the food industry, as it will provide a more detailed scientific understanding of the fermentation process as well as tools to predict faulty fermentation. Furthermore, the association of LAB and yeasts in brine was well known but their association on the olive biofilm not only in small scale fermentations but also in industrial fermentation. The possibility of controlling the LAB strain on the olive biofilm is a complete novelty generated in this project. As result, the project has also open new and completely unexplored research lines that have expanded the frontiers of the olive fermentation further away the current basis of the table olive investigation. These results, as it has been stated above, have been or will be published after the patent application and will be available to the scientific and industrial community. These results have also been disseminated to the Associations members through the several organized workshops (see Deliverable 9.1).
Another important issue in this result is the fact that the sensory assessment of table olives [recently adopted by the IOC (COI/OT/MO No 1/Rev.2 November 2011)] has been well established and used by the researchers during the experiments and the consumers during the evaluation of the consumer preference of probiotic table olives.
Future plans: Research efforts will also continue to clarify the role of yeast in table olive fermentations. Moreover it has already started a training campaign by the IOC in which national panel groups in each table olive producing country will be selected and trained for the sensory assessment of table olives. Some of the project participants are involved in this campaign (e.g. PEMETE, AUA, DEMETER, IG-CSIC, APABI, INRB).

Project Result 3: Probiotic bacteria that can be used as starters in medium or large scale fermentation
Selected strains of Lactobacillus were able to dominate the fermentation process and maintain high population levels above 6.0 log CFU/g of olives after 120 days of fermentation, indicating successful scale-up of the process. General advancement of knowledge has also been obtained for a successful scaling up of the starter technology.
Exploitation form: The results obtained from the large scale fermentation process showed that the selected strain of L. pentosus B281 was able to dominate the process and maintain high population levels above 6.0 log CFU/g of olives after 120 days of fermentation, indicating successful scale-up of the process from laboratory to industrial level. The strain was adapted successfully to the salt environment of the brine and the attained acidity in the brine was also satisfactory. Finally, molecular analysis showed high recovery rates of the starter at both the beginning and final stages of fermentation The experiments carried out at pilot and industrial scale have also demonstrated by the first time the possibility of fermenting olives with specific starter cultures. In the past, the inoculation was not possible to be driven with respect to the imposition of the added strain and in fact it is very likely that the initial strain introduced, initiated the process but, later on, it were substituted by the wild LAB microbiota. This possibility might have been facilitated by the inoculation procedures. Furthermore, it could have been also possible that the starter cultures were completely substituted from the very beginning by the wild LAB and that the inoculation were absolutely ineffective. The progressive inoculation scaling planned in the project has shown to be very convenient to study in detail the starter culture imposition process and disclose the difficulties and uncertainty involved in the past inoculation procedures. Then, the project, on the one side, has markedly improved the inoculation system at industrial scale, facilitating in this way the utilization of any starter culture and, on the other hand, has proved that, when the starter culture is a probiotic LAB, this can be imposed over the wild strains not only in the brines but also on the olive surface. Thus, the objectives proposed in the project have surpassed all expectations and have established the bases for probiotic olive production. Furthermore, the association of LAB and yeasts in brine was well known but their association on the olive biofilm not only in small scale fermentations but also in industrial fermentation. The possibility of controlling the LAB strain on the olive biofilm is a complete novelty generated in this project. As result, the project has also open new and completely unexplored research lines that have expanded the frontiers of the olive fermentation further away the current basis of the table olive investigation.
Finally the Association of Greek Food Industries (SEVT) has organized for the second consecutive year the national ECOTROPHELIA 2012 contest for the creation of new eco-innovative food products. The competition was held in Athens, on September 10, 2012, and it was part of the wider ECOTROPHELIA EUROPE contest. The aim of the competition is to motivate Universities and Research Foundations engaged in food science and technology to develop innovative food products with high ecological impact that will provide a pool of innovative ideas for the food industry. The Agricultural University of Athens (AUA) participated in the contest with the product ‘Probiotic green olives’ that was qualified for the final stage of the competition and it was awarded the second prize (silver award). The product has been developed through the PROBIOLIVES project. This is a very promising indication that shows the impact that such a product could have in the market.
Future plans: It is well meant that due to the seasonal character of this product, it was not feasible to apply the selected starters more than one harvesting period. Therefore, trials will continue by the participating companies even after the end of the project to incorporate the starters in industrial scale. The IG-CSIC plan to continue the experimentation at industrial scale in order to compare the results obtained with the two sub-selected strains because after the studies in relation to the gastrointestinal passage LAB4 has also shown interesting potential probiotic characteristics. The same institution will carry out experiments to clarify the causes of the sharp decrease of LAB 2 on olive surface after fermentation. Furthermore, the role played by the yeast at industrial scale is still unknown. The number of species found were very limited; so it would be interesting to clarify the reasons for this selective colonization of the olive surface.

Project Result 4: Predictive models for risk assessment
The modified Weibull model was fitted to the experimental data to describe the survival kinetics of the pathogens Listeria monocytogenes, Escherichia coli O157:H7, and Salmonella Enteritidis in the brine during storage of olives.
It is a valuable contribution to risk assessment studies related to ready to eat foods in general. Moreover, since the application of appropriate thermal treatment cannot be applied in the case of a functional/potential probiotic product to ensure the safety of the product, the application of strictly good manufacturing practices to reduce the possibility of cross contamination during packaging or post processing, is a major priority. This is valuable information for the table olive producing companies.
In the Deliverables 6.2a &b there is a gathering of data that are useful for risk assessment of fermented olives. These are coming from A) a literature review about the prevalence of pathogens in fermented olives, B) a survey for the presence of pathogens in different types of olives and olive products in the market, C) the application of predictive models on the survival of certain pathogens in olives and brine, D) the monitoring the Enterobacteriaceae during the industrial fermentation with probiotic starters and E) testing the fate of E. coli O157:H7 after subjecting in an artificial process of human digestion with fermented olives as meal.
In an attempt to apply predictive models in the experimental data of Del. 6.1b, the modified Weibull model was fitted to the experimental data to describe the survival kinetics of the pathogens Listeria monocytogenes, Escherichia coli O157:H7, and Salmonella Enteritidis in the brine during storage of olives. Indeed the modified Weibull model provided a good fit to the survival data of all pathogens. This model was selected on the basis that it describes both the decrease of pathogen population during storage and the residual bacterial concentration when a tailing effect is observed.
Exploitation form: In conclusion, this result demonstrated that even though the growth of the studied pathogenic strains was not supported, they may survive for a long period in a stressful environment of a fermented product with low pH value (4.2) and high salt concentration (6.0%) and thus they are a valuable contribution to risk assessment studies related to ready to eat foods in general. Moreover, since the application of appropriate thermal treatment cannot be applied in the case of a functional/potential probiotic product to ensure the safety of the product, the application of strictly good manufacturing practices to reduce the possibility of cross contamination during packaging or post processing, is a major priority. This a valuable information for the table olive producing companies. Relevant publications on this subject have been sent for publication in journal and will be published soon in the upcoming International Conference on Predictive Modelling in Foods (ICPMF 2013, http://www.icpmf8.org/).

Project Result 5: Shelf life prediction models
This result has a profound impact on the food microbiology sector, in terms of the discovery of new biological knowledge of interest to industry and academia. The knowledge generated on the use of predictive models is available to the associations for use of any interested company.
The growth and survival of lactic acid bacteria (LAB) during green olive fermentation was simulated using different fitting models that of Baranyi and Roberts (1994), the Artificial Neural Networks (ANN), the Radial-basis function networks (RBFNs). Results indicated that artificial neural networks, both multilayer perceptron (MLP) and radial basis function (RBF), although considered as data driven approaches, can be effectively employed for non linear regression purposes in small datasets.
Moreover the effect of temperature, pH and salt on growth kinetics of LAB was modelled. The application of cardinal parameter models to the preselected potential probiotic LAB strains led to a proper modelling of the diverse growth parameters and the estimation of these important characteristics, useful for the further selection of strains.
The consequent storage and evaluation of the different fermented olive products’ shelf life, was performed. The survival of the LAB was not modelled for the storage of these products, since the changes of the specific microbial group were not dramatic. However, an excess amount of data were collected for olive products stored in brine, modified atmosphere packaging or vacuum at different temperatures and the survival of the inoculated strains during storage was assessed with PFGE, since the reference of a stable population above 6 logs is not sufficient to guarantee the presence of these strains.
Exploitation form: This result deals with the mathematical analysis of the data collected from the fermentation and storage studies and has a profound impact on the food microbiology sector, both in terms of the discovery of new biological knowledge of interest to industry and academia, and also in terms of demonstrating to the community the benefits of multivariate data analysis in their field. Traditionally, microbiologists (and biologists in general) have taken a univariate approach to interpreting their data – looking at a single marker or characteristic at a time. Given the much more powerful analytical approaches now available (and used in this project) multivariate techniques are essential if maximum information is to be extracted from the data. The particular knowledge provides insights into the efficacy of survival of the probiotic starters in olive fermentation which is very important knowledge for the companies that will apply them. Encouraging the food microbiology community to move towards multivariate analysis is very important if maximum information content is to be extracted from the data that they acquire. This project provides a demonstration of the benefits of such approaches in an area of great significance to enterprises and citizens, and because this is a European project it is an ideal platform that demonstrates the benefits of this approach to a wide audience.
The knowledge generated on the use of predictive models is available to the associations for use of any interested company. The RTDs are willing to demonstrate and help any time each of them will need help to apply such models in his own conditions. Although it was not feasible to apply predictive models for the shelf life of fermented olives, a high amount of information on LAB and yeast survival in brine and on olive surface has been obtained and an overview of the expected survival over this period derived which can be of interest and help the table olive producing associations and companies. These results have been disseminated to the Associations members through the several organized workshops.

Project Result 6: Effective packaging systems for the preservation of the final product
Effective packaging systems for the preservation of table olives previously fermented with probiotic starters included packaging of table olives either in MAP (70%N2:30%CO2), N2 or vacuum without brine or in brine at 4-7°C and 20-22°C. It will offer to SMEs the opportunity to increase their knowledge about packaging systems to extend the shelf life of probiotic olives while maintaining a high probiotic population 6 logcfu/g, the original appearance, smell and taste of the product.
Exploitation form: It will offer to SMEs the opportunity to increase their knowledge about packaging systems to extend the shelf life of probiotic olives while maintaining a high probiotic population, the original appearance, smell and taste of the product, using different packaging solutions, enabling packaging to be optimized according to the needs of the packed product and the supply chain. Result found in the project have also shown interesting conditions for olive stabilization without the use of pasteurization. This achievement would be of interest for the small industries which resources do not allow them the use of thermal treatment for the olive preservation. These results have been disseminated to the Associations members through the several organized workshops, they have been used for the training of technical staff and are going to be published also in scientific papers.

CONTRIBUTION TO REGULATIONS
A. The EC regulation No2073/2005 for the microbiological criteria of foodstuffs (Annex II), includes the need for studies to evaluate the growth or survival of certain microorganisms of concern that may be present in a product during its shelf life under reasonably foreseeable conditions of distribution, storage and use. Project Result 4 contributes with data to the application of this regulation in order to provide to consumers safe food.
B. In December 2006, Regulation on nutrition and health claims made on foods was adopted by the Council and Parliament (EU Regulation 1924/2006). The COMMISSION REGULATION (EU) No 432/2012 of 16 May 2012 established a list of permitted health claims made on foods, other than those referring to the reduction of disease risk and to children’s development and health. The COMMISSION IMPLEMENTING DECISION of 24 January 2013 adopted guidelines for the implementation of specific conditions for health claims laid down in Article 10 of Regulation (EC) No 1924/2006 of the European Parliament and of the Council. Article 10 lays down specific conditions for the permitted use of authorised health claims. It is important to note that even authorised health claims may not be used unless their use fully complies with all the requirements of the Regulation.
The regulation 432/2012 entered into force since 14 December 2012. Since then no food can have on its label the term “probiotic”. So far a great number of health claims related to probiotics and gut health have been rejected by EFSA. EFSA has rejected the total 125 health claims related to the genus Lactobacillus (EFSA Journal 2012;10(8):2858). The 75% of those rejected due to inadequate taxonomic characterization of the microorganisms. They have been resubmitted with complete data on strain characterization and this time EFSA approved the strain identification but rejected all the health claims associated to them due to inadequate scientific substantiation. The reasons for non authorization has been either a) the quantity of the submitted studies was not enough, b) they did not prove the allegation beyond a reasonable doubt, c) were not made in the target population, d) failed to have consistent results in studies. The conclusion was that there are not enough reliable clinical studies to substantiate scientifically the claims (http://ec.europa.eu/nuhclaims/).
Currently hopes are rising that the use of the term ‘probiotic’ will not have to disappear on products following the failure of the European Food Safety Authority (EFSA) to give approval for a generic health claim for the ingredient last year (http://www.nutraingredients.com/On-your-radar/Health-claims/EU-working-group-to-discuss-probiotic-as-a-general-descriptor).
It is also well approved that technology and functionality data need to be linked in order to substantiate the health claim of a probiotic culture (Sanchez et al. 2012). Our Project Results 1 and 2 contribute with data on the in vitro functionality of the selected bacteria combined with the application technology and survival of them in olive fermentation. Our aim is to continue with clinical studies in order to fully support a future health claim for probiotic table olives.

Dissemination activities aimed to widely diffuse awareness on the scientific achievements of the project and raise interest on the application possibilities it opens in fermentation process. The objective was to disseminate the learned outcomes about the probiotic bacteria of olives and the effective and reliable quality optimization of fermentation process, including the use of probiotic starters, that allow SMEs to introduce into the market the new functional product and apply the different models developed and validated within the project. Our aims included the transfer of knowledge and best practice to industry and the scientific community for the new functional product as well as its effective and reliable safety assurance and quality optimization management system ultimately targeting the development and launch of new food products and packaging systems implementing the new learned outcomes. The main results of this research are expected to improve the competitiveness of the European olive industry.

Dissemination activities
A. Publications in scientific journals
Dissemination activities including but not restricted to publications and presentations were governed by Article II.30 of the Grant Agreement with the additions described in art. 8.3.1 of Consortium Agreement. According to these all the deliverables of the project were identified and the knowledge generated was assessed. Indeed after the successful isolation and characterization of potential probiotic bacteria and the subsequent testing of their performance in olive fermentation, two patents have been applied in Spain by the Spanish SME-association - ASEMESA [patent No. P-201131011 “Use of autochthonous lactic acid bacteria with probiotic features in the preparation of food products”, (date of application 16-6-2011)] and in Greece by the Greek Association - PEMETE [patent No. 20110100600 “Functional table olives fermented with lactic acid bacteria exhibiting probiotic properties” (date of application 14-10-2011)]. Relevant scientific publications followed after the application for patents. Scientific results not subjected to protection had been externally disseminated by the RTDs via presentations in scientific conferences (national, European or international) and publications in scientific journals. The publications in SCI scientific journals from all the participating RTD groups (17 published, 6 submitted, 4 under preparation) and more than 100 other publications in workshops, scientific conferences (oral presentations or posters), project poster, flyers, press releases, articles, project website etc. as well as Theses masters & PhDs. All the publications were identified as the results of a research project that was supported by European Commission (FP7-SME) under grant agreement 243471. This kind of dissemination will continue well after the end of the project. Dissemination activities included also the update of the website, the production of a flyer, press releases followed each meeting or workshop. A wider diffusion of the project has been achieved thanks to the webpage of the project (http://www.probiolives.eu). The construction of this page made possible world diffusion through Internet, the most powerful diffusion tool. Links with this web page has been inserted in all the organizations participating in the project.

B. Other dissemination activities
It was foreseen that the dissemination activity of Probiolives will be primary promoted by the SME-AGs. SME-AGs will be in charge of spreading within the local Table Olive Sector the information gathered during the development of the project and to allow and speed up the take up of the results by SMEs and by the market. To carry out such activities they will use the usual news-sheets that are distributed periodically among their members. In addition, they will organize specific meetings to inform their members of the project evolution and possibilities as well as workshops in which the experts from the different participating countries may communicate the results of their respective teams. They could also use the diverse annual promotion campaigns to inform the consumers about the eventually developed products. These may be an effective way for the simultaneous spreading of the products without competition among the developers since these networks have diverse structures and have their own consolidate clients. Furthermore, the commercial networks of their members could be very useful for the performance of consumer’s panels that could collect the public opinion about the characteristics of the developed products.
Indeed such dissemination activities were successfully carried out by the SME-AGs of the project in collaboration with the RTD performers so that not only the SME involved in the project had been informed of the project progresses but also the rest of their members, and included press releases, TV or radio interviews, articles in the popular press, flyers, presentations in workshops, organization of workshops particularly for the PROBIOLIVES project.
Website: A website (www.probiolives.eu ) has been constructed for the project with regular updates
Poster and flyer for the project: The Poster and the flyer has been presented in many national and international conferences in different countries by all the RTDs.
Press releases: Several press releases have been announced at the start and during the project before and after the organization of each project meeting. These activities mainly carried out by the SME associations through their websites, magazines, newsletters. There has been such a great impact of the subject that numerous other announcements in other relative websites followed the initial.
Workshops for the members of SME associations and other: Workshops have been organized by the SME associations to inform their members for the progress of the project. These workshops usually followed the progress meeting in each country e.g. in Spain after the 6th month meeting (8-9-2010), in Rome after the 12th month meeting (25-3-2011), in Portugal after 18th month meeting (9-9-2011), in Spain after the final meeting (19-2-2013) but also were organized the rest of the year during the project e.g. in Greece (27-2-2010, 26-2-2011, 10-3-2012, May 2013). RTD performers were invited in each country to present the project and their results. RTDs were also invited to present the project and the results in other relevant workshops organized during a scientific conference or not. In these conference workshops researchers of the project group from other countries were invited as speakers [like Dr. Garrido in OLIVEBIOTEQ (November 2011), or Dr. Tassou in Table olive conference in Cordoba (February 2012)]. Regardless of these presentations, SME-AGs will be in charge of providing detailed information of results generated by RTDs to all participating SMEs even after the end of the project.
Theses: A significant number of Theses for Bachelor, MSc and PhD have been implemented during the Probiolives project in the participating Universities and Research Organisations.
List of Websites:
www. probiolives.eu

Related information

Reported by

HELLINIKOS GEORGIKOS ORGANISMOS - DIMITRA (HELLENIC AGRICULTURAL ORGANIZATION - DEMETER)
Greece
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top