Skip to main content
European Commission logo print header

High quality meat-like products - from niche markets to widely accepted meat alternatives

Final Report Summary - LIKEMEAT (High quality meat-like products - from niche markets to widely accepted meat alternatives)

Executive Summary:
Several environmental and health-promoting issues substantiate the shift from an animal protein – in particular meat - to a vegetable protein based diet. Although meat analogues have attracted increasing industry and consumer attention, the market for these products is still very small. This is probably due to the fact that commercial available meat analogues - in particular their sensory qualities – do not meet consumer preferences. The project LikeMeat aimed at the development of meat analogues with excellent, well-accepted texture, juiciness, appearance and aroma based on plant proteins and combinations with appropriate hydrocolloids.

The LikeMeat project provides a deep understanding of manufacturing fibrous meat analogues using a high moisture cooking extrusion process. Both the influence of various ingredients and of process conditions was investigated. The gained knowledge was used to develop a set of base products that were refined to consumer products. The processing depends on commercially available raw materials and equipment that enables the transfer to full commercial scale. The industry partners developed a number of conceptual products such as coated or filled products, delicatessen products and ready-to-eat meals (e.g. schnitzel, sugo, Mexican dish, baked stripes for salads). The development of refined products was done with the vegan base material. This first vegan formulation was later on replaced by an optimised base product with a more juicy texture form. In addition nutritional values of the base products were determined and compared with equivalent meat products. Five prototype products were selected and subsequently tested in consumer studies. Beside the processing and product development, aspects on food safety and shelf life were crucial tasks. The collected experimental results helped to identify potential microbial risks in the final product and led to microbiological raw material specifications as well as to obtain valuable data for the further product development. The product development was based on a deep investigation of consumer preferences and needs. The target group flexitarians was identified as prospective consumers of LikeMeat products.

All the gathered data and the evaluation were passed to the project partners. Based on the results and based on the in the meantime optimised LikeMeat raw product the partners optimised and developed further food products. At the end of this phase, which represents the end of the entire project, several different foods with a good consumer acceptance and potential market success were available.

LikeMeat enabled SMEs to produce innovative, widely accepted meat alternative products. A better understanding of the consumers' behaviour related to meat analogue consumption, better products and a high level of food safety minimise risks for new product launches shortly after the project. Not only the meat analogue producers will participate and profit on new product launches, but also raw material suppliers, pre-mix producers, distributors and retailer due to the new networks. The meat alternative market turned out to develop during the last years, making it very profitable. Therefore the involved SMEs can significantly improve their competitiveness, increase sales and increase employment by producing and selling the new products. The LikeMeat project found already during the project period some promising interest in the food industry. Society will not only profit from new healthy and tasty products, but also from products that have the potential to replace partly meat in the daily diets. Since the alternative products have a significantly lower impact on natural resources, the LikeMeat products will contribute to sustainability.
Project Context and Objectives:
Due to a steady increase in the world population and rapid economic progress in developing and newly industrialised countries, global meat production is continuously growing. However, the production of animal protein requires a lot of resources and cannot guarantee the basic supply of food for the human population. For the production of 1 kilogram of animal protein, 3 to 10 kilograms of plant raw materials are required. As a consequence, half of the plant proteins produced worldwide is currently used as animal feed. Moreover, meat production, mostly associated with industrial livestock farming, contributes to the emission of carbon dioxide and methane. In the future, it is of utmost importance to make food production and consumption more sustainable, which could be achieved by replacing animal proteins with plant proteins. Besides environmental considerations, a diet with higher quantities of plant proteins is recommended by public dietetic associations due to the related health promoting effects [1].
The introduction of meat analogues (also termed meat substitutes, meat alternatives and meat replacement food) in Western markets is a relatively recent development, starting at the beginning of the 1960s [2; 3]. In contrast, products such as tofu and tempeh have been eaten in Asia for centuries. In addition to these traditional Asian products, the launch of meat analogues started with the production of texturised vegetable protein, which is produced by cooking extrusion of usually defatted soy meal or soy protein concentrates. However, due to the permanent GMO debate, products containing soy proteins are often considered critically and are not accepted by many consumers. Moreover, soy protein is often used in combination with wheat gluten, which must not be consumed by people having coeliac disease. Only in recent years have proteins other than those from soy been introduced as the starting materials for the production of meat analogues. In particular, combinations of different ingredients seem to be a promising approach for mimicking the fibrous and juicy texture of meat muscle, as opposed to the more “spongy” products which have been available up until now.

The number of vegetarians and number of consumers who are reducing their meat consumption has been increasing over recent years. Besides environmental considerations and health benefits, an increasing number of public issues relating to food and, especially to meat and meat products, can be held responsible for this. In order to achieve a considerable reduction in the consumption of meat, meat analogues must be competitive with meat products. Although several aspects support the shift in the diet from meat to plant proteins, the market for meat analogues is still quite small. For example in the Netherlands, the share of meat substitutes is only about 1 % of the total market for meat and meat products, probably due to the fact that present meat analogues do not meet consumer preferences with regard to sensory quality. Especially the bite, taste and juiciness score low by comparison with meat. To obtain a larger market share, meat analogues have to be developed that better meat consumer demands than the present meat analogues, and this could be achieved by means of consumer oriented approaches [4].
[1] Davis, J.; Sonesson, U.; Baumgartner, D.U.; Nemecek, T. (in press): Environmental impact of four meals with different protein sources: Case studies in Spain and Sweden. Food Research International. [2] Sadler, M. J. (2004): Meat alternatives - market developments and health benefits. Trends in Food Science & Technology 15 (5): 250-260. [3] Davies, J.; Lightowler, H. (1998): Plant-based alternatives to meat. Nutrition and Food Science 98 (2): 90-94. [4] Aiking, H.; de Boer, J.; Vereijken, J.M. (2006): Sustainable protein production and consumption: Pigs or Peas? Springer, Dordrecht, The Netherlands.

The technological and scientific problem or needs of SMEs
During recent years, the small market for meat analogues has been dominated by small and medium-sized enterprises (SMEs). With growing awareness of the health and environmental impact of food products, meat analogues have attracted growing industry and consumer attention. One consequence of this has been that multinational companies have entered the market and are strategically buying emerging small companies that have innovative products (Threat by large companies). In order to survive economically, the SMEs have to cope with this increasing market pressure and also with the rapid advancement in processing technology. In contrast to multinational companies, SMEs do not have the personnel and financial resources for targeted development of innovative, competitive products with superior textural and sensorial properties and having high consumer acceptance (Expensive research). The substitution of meat proteins and concomitantly the realisation of a comparable fibrous and juicy texture is a considerable technological challenge requiring detailed scientific knowledge for successful product development. This knowledge includes the screening and selection of appropriate raw materials facilitating a fibrous and juicy texture, the optimisation of processing operations and the assurance of microbiological safety (Lack of technological and ingredient knowledge). Additionally, a major barrier to the successful development and market launch of innovative, sustainable and healthy meat analogues is the lack of knowledge about consumer behaviour and preferences. Furthermore, they face the difficulty of getting entry to the mainstream buying channels, especially supermarkets. Therefore, important information about the positioning and marketing strategy of the products is partially missing and probably inhibits commercial success (Lack of consumer orientated research).

The combination of both technological and sensory investigations resulted in meat-like products with sensory and textural characteristics similar to meat, which promote high consumer acceptance. These issues were the focus of the LikeMeat project. Having worked closely together, the consortium partners acquired a broad range of knowledge about the formation of meat-like textures, taste and aroma and about microbiological safety. These findings are a prerequisite for consumer orientated product development, development of efficient marketing strategies and will tremendously reduce the risk for market launch. The results acquired during LikeMeat can now be transferred and used for a successful and innovative future product development. In particular due to the increasing demand for meat analogues with a juicy texture and well-adapted sensory characteristics, it can be assumed that the companies’ turnover will rise, whilst at the same time maintaining employment opportunities and strengthening their competitiveness against large enterprises.

Scientific and technological objectives
The overall objective of the project was the development of meat analogues with excellent, well accepted texture, juiciness, appearance and flavour. To achieve this, plant proteins and mixtures of plant and animal (milk, egg) proteins were converted into fibrous structures and stabilised, requiring a detailed knowledge about protein aggregation, protein interactions with other components in the recipe and the impact of processing conditions on product quality. These findings facilitated the selection of appropriate raw materials (mimicking meat proteins) and the optimisation of the recipes in order to obtain products similar to the reference meat containing goods. The fundamental evaluation of consumer preferences, acceptance and needs (PAN) allowed target-oriented food developments and the promise of well-accepted products. The scientific and technical objectives that need to be fulfilled over the course of this 28-month project can be summarised as follows:

1. The assessment of consumer preferences and needs with respect to enhancing the acceptance of meat replacers: An early evaluation of consumer preferences and needs helped to understand the barriers which hamper good consumer acceptance of meat-like foods and thereby delivered the targets for the subsequent research and development work. The developed products were tested for overall consumer appreciation and their willingness to buy the products. Using these data, the suitability of the new products for specific market segments and the particular usage (in-home or out-of home usage) was estimated and the results on the best marketing attributes was assessed in real life situations. An overview of critical success and failure factors of previous launches of meat-like products was made.

2. Identification of promising protein sources: To date, mainly soy and wheat proteins have been used as the basis for meat analogues. Within the project, alternative protein-rich raw materials that find high acceptance with consumers were indentified. These proteins were evaluated for their technological, sensory and microbiological properties.

3. Improving the texture and juiciness of meat analogues: The impact of various natural starches, fibres and hydrocolloids with high water binding capacity on the texture formation and the taste of the meat analogues were examined. Furthermore, their interactions with the proteins and the formation of the fibrous structure was investigated at a microscopic level. Based on these data, suitable ingredients were chosen and compositions for juicy well-accepted meat analogues were developed.

4. Description of the impact of processing parameters on aroma development and the adaption of aromatisation and seasoning: The aroma of meat analogue products was optimised in this project to meet the consumer expectations for high quality foods. The aroma development was monitored by identification of odour-active compounds in the raw materials and in the final extruded products. Based on these results, the influence of extrusion on aroma development was observed and strategies were developed to adapt the extrusion process or the formulations and therewith obtain products with high aroma quality.

5. Development of finished products such as convenience foods and delicatessen products: Several conceptual food products based on the three raw meat analogue formulations which come out best in the evaluation were developed. These products met the highest quality standards with regard to safety, nutritional composition and sensory quality. The nutritional composition of these products was matched with the type of meat products that they are most likely to replace in real life situations. These developments will enable the SMEs to launch first products at short notice.

6. Improvement of the shelf-life and ensuring the safety of the products: The microbiota in the raw materials and intermediate products was analysed by culture and molecular biological methods. Subsequently the basic degradation kinetics of critical microorganisms was investigated for the high temperature - short time (HTST) extrusion process. To determine the shelf-life, the final products were stored for various time periods at different temperatures and atmospheres. Furthermore, production hygiene and the impact of food preservation strategies such as post packaging heat treatment and the application of marinades or coatings containing antioxidative and antimicrobial ingredients was tested.
Project Results:
The project was subdivided into four phases and accompanied by the horizontal activities of dissemination and training.

Phase 1 started with two main activities, the consumer evaluation (WP1) and a screening of suitable proteinious raw materials for processing meat analogues (WP2). Both qualitative and quantitative consumer studies were carried out to find out which group of consumer is aiming for what kind of meat alternative products. As a point of departure, we target as prospective consumers of the LikeMeat product, people who already have an eating pattern in which meat isn't a daily component but who still eat meat: flexitarians. We assume that these flexitarians are easier persuaded to consume the LikeMeat product as they don't necessarily need to have meat on a daily basis. The first step in our consumer research consisted of an exploration of flexitarians by conducting qualitative, individual, in-depth interviews. The subsequent step was to conduct a survey to quantify insights derived from the qualitative study. In the subsequent design sessions in WP1.2 and WP6 and after consultation of the consortium partners, the drivers and barriers specified in this work were used to create specifically designed new protein product concepts for each or a selection of segments.
At the same time a range of eleven protein ingredients were screened and identified as potential candidates for such products and to characterise their processing and sensory properties. Further to identify critical mirobiota in the selected raw materials to predict adequate heat-time treatment during the cooking extrusion process for save and stable food products (WP2).

In Phase 2, three basic recipes were developed at Fraunhofer in close cooperation with the partners, in particular Vegetalia, Angermuehle, CyberColloids, Wiberg and Etol, based on the results from WP2 (WP4). The processing conditions on both a lab and pilot scale were adapted. The formulations were either vegan, vegetarian or gluten-free. For the recipe developments experiments with a mixture designs were carried out and optimised formulations calculated. All three recipes were processed both in lab (1kg/h) and pilot scale (80kg/h).
The work on formulation and processing was accompanied by detailed analysis of the degradation kinetics of critical microorganisms during the HTST process (WP3). This task was performed by BOKU.

In the course of Phase 3, the basic recipe was translated into food products. FBR and the industry partners Vegetalia, Neuburger, Radatz, Ruig and Wiberg developed a number of conceptual products such as coated or filled products, delicatessen products and ready-to-eat meals (e.g. schnitzel, sugo, Mexican dish, baked stripes for salads) (WP6). The product development started with the basic vegan recipe material that was later on replaced by an optimised base product with a more juicy texture from WP5. In addition nutritional values were determined and compared with equivalent meat products.

Phase 4 focused on consumers’ opinions and the assessment of critical success factors for the transfer
to market (WP8). A set of five different LikeMeat- products was tested in an in-home and out-of-home setting. The latter test was conducted in the Restaurant of the Future in Wageningen in The Netherlands. The products were also tested in an in-home situation; the other main location for the consumption of food products.
All the gathered data and the evaluation were passed to the project partners. Based on the results and based on the in the meantime optimised LikeMeat raw product (WP5) the partners optimised and developed further food products. At the end of this phase, which represents the end of the entire project, several different foods with a good consumer acceptance and potential market success were available.

In detail following progress and achievements were reached within the individual work packages during the project: 

WP1 Consumer preferences, acceptance and needs :
The objective in WP1 was to find out what (combination of) functional product properties are required to match the preferences of European consumers, for different product categories related to meat products. In WP1 three types of activities were implemented corresponding with three deliverables: Qualitative research (stage 1, Task 1.1); quantitative research (stage 2, Task 1.3) and creative sessions and selection (stage 3, Task 1.2 1.4 and 1.5). The aim of these studies was to identify product concepts of meat-like products for specific target groups.

Task 1.1 Qualitative consumer studies:
The first type of activity was to detect and describe target groups with respect to (non-) meat consumption in Task 1.1. The segments describe different types consumers in what meat products represent to them and what are their preferences with respect to meat – or alternative products. These functional properties can be used to create new (LikeMeat) products that serve these functions. To get these results, first qualitative interviews with consumers have been performed on location in The Netherlands and in Germany. Participants were asked to take photographs of their shoppings and in the interviews they were invited to elaborate on the products they buy, the reasons why they buy these products and not others, and what especially the meat (and/or other protein sources) represent to them. We focused on the eating pattern of the main meal of the day in so-called flexitarians: people that do not eat meat on a daily basis but are not full-time vegetarians. This study identified four segments of flexitarians with distinct behaviour and motivation patterns: health flexitarians, animal friendly flexitarians, hedonist flexitarians, and convenience flexitarians. Each of these groups differ as much as possible with respect the functional properties they see in meat products.

Task 1.2 Consumer driven product concept generation:
The consumer driven product concept generation in Task 1.2 reflects the development of different product concepts. After consultation of the consortium partners, the drivers and barriers specified in report D1.1 (Task 1.1 and Task 1.3) were used to create specifically designed new protein product concepts for each or a selection of segments in the subsequent design sessions in WP1.2 and WP6.
In stage 3 of the evaluation of consumers' preferences and needs the identified segments of consumers were used as a starting point for creative sessions to identify product concepts. In two creative sessions participants from different backgrounds and disciplines (both business and science) a total of fourteen product ideas were derived that were elaborated into product concepts. This was done through the use of very personal descriptions of fictive persons from each of the consumer segments. Thus, the participants were able to empathize directly with these potential customers and to disconnect from their specific expertises and interests. The fourteen concepts could be categorized into four categories: Ready-to-eat; Minimum preparation; Hobby cooking; and Snacks. Subsequently these ‘ideas’ were elaborated into ‘product concepts’; more complete descriptions of potential products, including additional information.

Task 1.3 Consumer market segmentation for meat analogues:
In Task 1.3 a consumer survey was applied to distinct consumer market segmentation for meat analogues, and to validate and quantify the four segments of flexitarians as outcome from Task 1.1.
The aim of the quantitative studies of Task 1.3 was to detect and describe target groups with respect to (non-) meat consumption. Based upon the results of the qualitative studies and on general (food related lifestyle) factors that are known to relate to meat and other protein-rich products, a quantitative survey was performed in Spain, Germany and The Netherlands. The aim was to try to confirm the segments that were identified, and to quantify them. Flexitarians in The Netherlands, Germany and Spain completed the survey (n~700 per country). The data showed five relevant segments of flexitarians based on food related lifestyle. Four segments resemble the outcome of the qualitative interviews. In addition, a segment of price driven consumers was identified. Group size of respondents belonging to a segment vary between 300 (health flexitarians) to about 600 (convenience flexitarians). The animal friendly flexitarians and the convenience flexitarians seem most open to new products. Especially for the animal friendly group non-meat products could represent added value. The health oriented flexitarians that want to reduce meat consumption might offer the best opportunities for new LikeMeat products. Interestingly, the reasons (not) to reduce meat were not distinctive between groups, indicating that this is not a main driver for product choice. It confirms the idea central to the consumer studies in LikeMeat, that the interest of specific consumer groups in new protein products by no means needs to be driven by a desire to consume less meat.

Task 1.4 Concept selection:
During Task 1.4 the concepts from Task 1.2 were evaluated by nine different project partners, on positive and negative aspects, interest, technical challenges, and overall preference for further development. The fourteen concepts from the creative sessions which were presented in a questionnaire to the LikeMeat partners for feedback. This evaluation was done in a survey using a pencil and paper questionnaire. Additionally, the results were discussed during the project meeting in Barcelona, November 2011. Based on the feedback of the partners and the raw materials that was selected in WP2, a subset of concepts was selected for further consumer research in WP6 and WP8 and more detailed product concept descriptions.
With respect to the ‘Consumer driven concept selection‘, the selection criteria, process, and points of discussion are described. Generally, the concepts that were most popular among the partners were the ones that were perceived not too novel or innovative and being perceived to have broad application - and market possibilities. Feasibility did not play a big role in the popularity of the concepts, as the popular concepts were both considered ‘easy’ as well as ‘difficult’. The result was a final selection of six products. These provide further input for the final selection of products to be developed within the LikeMeat project. This process continued in the activities around Task 1.5 6.1 and final testing in Workpackage 8.

Task 1.5 Identification of product requirements:
Task 1.5 concerned the identification of product requirements. This process was integrated with the process of creating recipes (Tasks 6.1 6.2 6.3) that was reported in deliverable 6.1 (see WP6). The reason is that in practice and in interaction with the SME’s, the product characteristics that need to be optimised change with each adaptation of the products, and changing the ideas about the products in discussion with the partners in the consortium. A preliminary report on the identification of product characteristics that had to be optimised for each product concept, for each intended use (Task 1.5) was prepared based on the 2nd periodical meeting (M6 – Maribor/workshop). In the report it was concluded that there are three domains of product requirements that need to be integrated to make the ideal mix of product characteristics: i) technical developments; ii) consumer research input: consumer and researcher evaluations give indications as to what products have the most potential for further development; iii) business: the gut feeling of the industrial partners is probably one of the most important indicators of the requirements for potential successful products.

WP2 Characterisation of protein-rich raw materials:

The production of juicy and fibrous meat analogue products in a cooking extrusion process relies on the network formation of the protein under the very specific process conditions. Therefore, the protein components have a crucial technological role and need to be carefully selected. Besides, the protein materials have also a high impact on the nutritional quality of the meat analogue and they determine the taste of the non-flavoured basic product. Therefore, the overall objective of work package WP2 was to characterise protein-rich raw materials that might be suitable to be used as a main ingredient for the basic meat analogue formulation with respect to techno-functional, sensorial and microbial properties.

Task 2.1 Characterisation of protein-rich raw materials:
In total 11 different protein concentrates and -isolates from grain legumes (pea, soya, lupine), cereals (wheat, rice), potato, egg (egg white) and milk (whey, casein) were selected in accordance with industry. To characterise the techno-functional properties of the protein ingredients viscosity profiles, penetrative and in situ gelling tests as well as texturisation trials were performed. Extrudates were tested in tensile strength tests using a texture analyser. Aroma and flavour of the screened raw materials were described by sensory analysis before and after processing. The comparative studies provided fund knowledge on functional properties of texturates as well as their flavour and mouth-feel. The tested protein ingredients can be grouped to three types of products. The basic proteins such as protein isolates or concentrates from soya, pea, lupine or wheat are in general easy to process and form a good texture. Their taste and perception attributes are on a low extinction and are accepted. Specific protein ingredients are egg albumine and whey protein isolate due to their high gelling power but also due to their specific sensory attributes. This limits in particular for egg white the maximum percentage in a formulation. Potato protein was the only ingredient that worked well at very low pH-values. Potato protein has as pure product a strong bitter taste that is hardly acceptable. More unspecific characteristics showed rice and milk protein. Its application would be a kind of filler in meat analog formulations that contribute to the nutritional value of the product.

Task 2.2: Isolation and identification of the raw material microbiota:
Depending on the selected raw-materials, high microbial loads can occur in the starting formulations and pre-mixtures including pathogenic species. Heat-stable spore formers could be able to survive the extrusion cooking and could be able to germinate in the intermediate products. The high moisture content and the almost neutral pH of the meat-like products cannot avoid the growth of surviving and post-process contamination related microorganisms. Therefore it is necessary to define other microbicidal hurdles for the final products.
Hence, a spectrum of 10 different plant (Lupine, Pea, Potato, Soy and Wheat) and animal (Milk, Egg) derived protein powders were analysed for the presence of critical microorganisms. Intermediate products, processed by cooking extrusion, were further examined for surviving, spoiling and potentially pathogenic and toxinogenic microorganisms. Therefore international established cultural methods (ISO standards) were applied to enumerate and detect microorganisms with relevance for spoilage or product safety. As standard methods targeting all needs exactly are not available, when necessary, the methods were modified to for practical purposes. Therefore protocols and guidance documents on the general examination of foodstuffs or meat products related to the target products were applied. This is true for the sampling strategy and homogenisation procedure of the examined intermediate products. The selection of the applied standard methods was based on the formulation, texture, moisture content and nutritional components of raw components and intermediate meat-like products. Published critical limits for the microbiological evaluation of the meat-like products are only available for tofu from soy protein.

Microbiological standard methods were evaluated for the routine detection and enumeration of relevant microorganisms in protein-rich raw materials, therefrom derived intermediates and intended protein-rich products with special consideration on potential heat stable or hazardous microbial load.
The tested raw products (protein powders) were from plant and animal origins (e.g. Soy, Pea, Lupine, Potato, Wheat, Chicken Egg, and Milk). Further, these powders were processed by cooking extrusion and the microbial and hygiene status of the resulting intermediate products was also analysed in fresh and frozen samples. The set of applied microbiological examinations is mainly based on well-established and standardized (acc. to ISO) culture-based methods. Frequently detected spoiling or (potential) pathogenic microorganisms were isolated and identified by molecular methods. Of special interest were heat-stable bacteria, which could be able to survive the cooking extrusion process. In detail, the focus was set on spore forming bacteria, such as Bacillus spp. and Clostridium spp. In this part of the project the practicability of available standard methods for this type of protein-rich powders and intermediate products was tested and modifications were applied when necessary (e.g. nutrition media, pH, temperature of incubation etc.). As official reference threshold values for the examined microbial parameters in protein-rich powders are not available, the results were assessed with data from industrial guidelines and specifications (similar products with comparable physico-chemical properties). The collected experimental results help to identify potential microbial risks up to the final product and will lead to microbiological raw material specifications as well as to obtain valuable data for the further product development. Based on the culture-based and molecular microbiological results, acceptable raw materials were chosen for further recipe developments, sensory testing and technological trials.
The occurrence of defined spoiling and potential pathogenic microorganisms was examined. Special attention was set on heat stable and spore-forming bacteria because they may pose a spoilage risk for the product, as well as a health risk to the consumer (risk-based approach). The intermediate products produced by cooking extrusion were analyzed by similar microbiological methods. Here the sampling approach was different. Special emphasis was put on the detection of foodborne pathogens, surviving microorganisms as well as re-contaminating microbiota. For both product categories special attention was given to potential occurring pathogenic bacteria, including Salmonella species, Listeria monocytogenes, Bacillus cereus, Clostridium species and Staphylococcus aureus. Officially approved microbiological methods for the detection and enumeration of microorganisms in foodstuffs were applied and evaluated for their "fit-for-purpose" capability in this food matrix. In general, among the raw materials tested, the highest microbial loads were detected in the Lupin protein isolate with an average aerobic mesophilic count of 2.4x10E4 cfu g-1. Foodborne pathogens including Salmonella species, Listeria monocytogenes, Clostridium species and Bacillus cereus were negative in all tested samples. Presumptive isolates of potential pathogenic bacteria were further characterized and identified by sequencing the 16S rRNA gene.
A total of 10 samples of raw protein powders and 10 intermediate products after cooking extrusion were examined with the aim to assess compliance with the microbiological quality guidelines as proposed by DGHM (2010). Certain product categories would fit into the physicochemical and nutritional criteria of foods included in the DGHM criteria. For raw protein powders the microbiological criteria were applied according to milled cereal products. The existing raw product-related microbiological quality specifications of commercial protein powders are defined from the industry and these reference values have no legal character. They include specifications for aerobic mesophilic counts, Enterobacteriaceae, yeasts and moulds, coliforms, E. coli, coagulase-positive staphylococci and Salmonella. For each of the analysed protein powder sample different thresholds were defined by the related manufacturer. For extruded high-moisture products no official microbiological thresholds are available. Therefore the reference values for tofu from soy protein and fresh-packaged pasta were used as references DGHM.
In general, among the raw materials examined, the sample with the highest microbial load was the Lupin protein isolate with average aerobic mesophilic count of 2.4 x104 cfu g-1 followed by Pea protein isolate with a count of 3.4 x103 cfu g-1. Aerobic mesophilic values of the other raw materials were below 1000 cfu g-1 in average. All extruded products had aerobic mesophilic counts below 100 cfu g-1. The contamination with enterobacteriaceae and coliforms could especially be observed for the Pea protein isolate (1.1 x 103 cfu g-1 and 4.6 x 102 cfu g-1, respectively). All extrudates were negative for the presence of enterobacteriaceae and coliforms. The contamination of raw materials with Bacillus spp. has to be outlined for the Pea protein isolate (2.6x102 cfu g-1) and the Lupine protein isolate (9x102cfu g-1). Wheat gluten was also positive for Bacillus species but with lower counts. However, Bacillus cereus could not be detected in this protein powder. The confirmation procedure of obtained Bacillus isolates was done according to the rules of the applied ISO standard method.
Based on the results of the analysis, the following conclusions could be made:
All extruded proteins showed high aw-values and were in the neutral pH range. Further trials will help to describe possible proteolytic spoilage and the expected shelf live. The recontamination of the extruded proteins seems to be very critical. As the initial bacterial counts are very low and no competing microbiota is present, the fast growth of spoiling and/or pathogenic microorganisms during shelf life is possible. Due to the elevated microbial load of Pea protein powder, further products should be analysed carefully, however the extruded proteins tested so far were of good microbial quality.
From the methodological point of view, all methods were applicable for each of the raw materials and the extruded products. Same handling problems of the intermediate products were solved by modification of the homogenisation procedure. The additional application of chromogenic media was indispensable because no experimental data on Bacillus spp. counts, their growth and morphology were available for the specific microbial analysis of protein powders and extruded intermediate products thereof. Remarkable differences in growth could be observed on differential and chromogenic Bacillus media. Further tests will help to formulate recommendations for their application in raw and intermediate meat-like products.

The microbiological quality of extruded products was satisfactory among all samples tested. This leads to the assumption that all microorganisms monitored in the raw material were not able to survive the extrusion cooking process. Therefore, it is necessary to maintain Good Hygiene Practice (GHP) to ensure that re-contamination may not occur.

WP3: Identification of critical microorganisms in protein-rich matrices:

Within this work package the micro-flora of raw materials and extruded products was characterized with respect to its level and composition. Further, inoculation trials to test the inactivation capacity of bacterial endospores by extrusion cooking were performed. The primary microbiological contamination occurs through the raw materials, mainly dry protein powders. The low water activity (aw value) does not promote the growth of microorganisms, but several bacterial species and bacterial endospores are able to survive these conditions and were isolated from dried food ingredients. The aw value, an intrinsic physical parameter of food, indicates the physically available water for microorganisms. It is also well known that heat resistance of certain microorganisms is highly increased by a decreased water content with an optimum aw-value of 0.20 - 0.50. By adding water to the dry protein powder prior extrusion cooking, an initial micro-flora, derived from the raw material and/or the production environment, can emerge and multiply. Additionally, poor sanitation conditions, lack of hygienic design of the equipment and deficient water hygiene pose a contamination risk for the intermediate and final protein matrix. The further product refinement by adding technological ingredients (e.g. hydrocolloids) or food ingredients (e.g. spices, herbs and marinades) pose another potential re-contamination source.
The typical composition of a finished meat analogue is characterized by a high protein content, high water activity (aw 0.94 - 0.99) level, neutral to alkaline pH, low salt content, intermediate carbohydrate content (e.g. sugars, gelatinised starches) and a low fat content. The high aw value is critical because at these levels existing Gram-positive and Gram-negative microorganisms as well as yeasts and molds are able to grow in the matrix. Based on an appropriate temperature profile of extrusion cooking, vegetative microorganisms are probably eliminated (DoW, 2010). Potential heat-stable bacterial endospores (e.g. Clostridium spp. and Bacillus spp.) may survive processing and germinate under storage conditions.
These potentially surviving microorganisms and the post-process contaminations necessitate further processing or preservation steps. In order to provide an acceptable shelf-life period and consumer safety, post-processing treatments, such as pasteurization or sterilization will be necessary. This is the task of Work Package (WP) 7. Soil is known as major habitat for spore-forming bacteria, whereas it is important to note that the diversity of spore-forming bacteria varies to a high extent. The sporulation and germination process of bacterial endospores strongly depends on the environmental conditions (e.g. food composition, food treatment and storage conditions).
Further, sporulation conditions presumably vary according to environmental factors e.g. the culture media applied in the laboratory. This issue is relevant for the inoculation trials applied with spore forming bacteria in WP 3, where the growth and harvesting of spores in different media is of high relevance. Representatives from the genera Clostridium and Bacillus and related Gram-positive endospore-forming bacteria are mainly involved in foodborne poisonings and spoilage.

Task 3.1: Evaluation of suitable routine methods:
The aim was to evaluate and adapt microbiological standard methods for their applicability in routine detection and enumeration of relevant microorganisms in LikeMeat raw materials, intermediates and developed protein-rich end products. The outcomes of task 3.1 have partly been summarized in WP 2 in D 2.2. There, raw materials are mainly focused with respect to the methodology and microbial ecology. The methodologies for the cultivation and recovery of heat-stable microorganisms and endospore forming bacteria in intermediate products are further discussed in this deliverable.

Task 3.2: Degradation kinetics of heat stable microorganisms and spores:
A fast and easy to use protocol for the growth and harvest of spores from facultative-anaerobic and strictly anaerobic bacterial spore-formers was developed based on current literature. The literature was differing regarding different applications for the growth and harvesting of spores. The most important factor was the incubation time. It was described from 3 days up to 10 days of incubation. From practicability aspects a compromise was chosen to earn as much spores as possible in a short time frame. After several pre-examinations (time, temperature and media) one protocol for Bacillus spp. and another for Clostridium sporogenes was defined. Both strains were applied as surrogate strains for the inoculation trials. A surrogate strain should behave similarly to the target pathogen aimed for lethality testing in food products. The before mentioned bacterial endospores were inoculated at two process levels (labscale and pilot plant scale) in order to optimize process conditions regarding hygenic processing. The recipes were chosen upon prior product optimizations. Also non-inoculated extrudates were examined for the whole range of relevant spoilage, indicator and pathogenic microorganisms. Total counts for samples in the lab-scale extrusion format revealed higher CFU levels compared to those processed at pilot plant scale. Results for non-inoculated samples showed a higher microbial load compared tot he inoculated samples. Here, the total microbial load from all trials was generally low (< 100 CFU/g) after extrusion cooking. This may be explained by the applied high temperatures, which had obviously a bactericidal effect and a low post-process contamination level (recontamination). Lab-scale extruded products showed a higher level of contamination, however without spoiling indications. Based on the microbiological methods, a comparable sporicidal and bactericidal effect could be shown, either by pilot plant-scale and lab-scale extrusion. However, this is not necessarily always the case. Therefore, a technological upscale is inevitable to obtain valuable and reproducible results for the risk assessment of food. The inoculated Clostridium sporogenes ATCC 19404 spores were not detectable in any of the samples after extrusion. Bacilli were detectable in low counts (< 1000 CFU/g) in inoculated and not-inoculated samples. This means that possible post-process contamination happened. After the microbiological screening of the inoculated and non-inoculated protein extrudates and the isolation of inoculated spore forming strains, species specific PCR, 16S rRNA sequencing and molecular subtyping have been performed in order to demonstrate the diversity and potential sources of obtained isolates. The non-inoculated samples were mainly dominated by Enterococcus durans and Exiguobacterium acetylicum, followed by Bacillus amyloliquefaciens. Other isolates were identified as Acinetobacter spp. and Staphylococci. Extrudates among all trials mainly harboured Bacillus amyloliquefaciens AB255669 and Enterococcus durans AJ276354. Also the wild-type strain Bacillus amyloliquefaciens LM008A-1 (wheat gluten, extrudate) could be found in different inoculated samples. The micro-flora in the lab-scale extrusion was dominated by Bacillus amyloliquefaciens. Due to the fact that the total micro-flora is very low immediately after extrusion and vacuum-packing, possible post-process contamination needs to be excluded. Within a LikeMeat matrix (high protein, low salt, high aw value, near to neutral pH) re-contaminating/surviving microorganisms are exposed to ideal multiplication conditions, even if considering chilled temperatures. Enterococcus spp., frequently isolated in LikeMeat products is for instance capable to grow under these conditions.

WP4: Development of a vegan and a non-vegan basic recipe

Task 4.1 Development of vegan and non-vegan basic recipes:
Based on the outcomes from WP2 the consortium discussed and decided on three routes for further recipe developments during the 6M meeting. Accordingly recipes based on legume protein, a mixture of legume protein and wheat gluten and a third formulation that contains also milk and egg proteins were developed in lab scale. For the first recipe (vegan and gluten-free) the three potential components soya protein concentrate, pea protein isolate and lupine protein extract were mixed according an experimental design to a total of eleven formulations. The relating test strands from pilot extrusion trials were subsequently tested by sensorial analysis. By correlating the test results with the used formulations using response surface methods and optimising the recipe to set attributes by calculating respective regression equations. For the second recipe (vegan) a pre-selected mixture of soya protein concentrate and pea protein isolate (deduced from recipe 1) was mixed in different ratios with wheat gluten. A total of 5 different formulations were tested by sensorial analysis. Relevant attributes and hedonic preferences were correlated with the recipe composition and optimised by using experimental design software. The third recipe (vegetarian) was dedicated to contain a mix of plant and high-quality animal proteins. Animal proteins such as from milk and egg are highly digestibility and highly nutritional making them an interesting protein source for products aiming for sectors like sport and nutrition. The suitable percentage of each single protein in the recipe was estimated based on the knowledge that was gained from the study on texturisation behaviour of raw materials and sensory as well as physical properties of textured proteins which is reported in D2.1. It turned out that the addition of hen egg albumen should not be too high because otherwise it develops a strong undesirable flavour and taste during the high temperature short time extrusion process. The addition of whey protein was limited because of its very strong gelling properties and cheesy. Milk protein that contains about 90% of casein and 10% of whey protein was varied between 40% and 70%. Among plant proteins each pea, soya and wheat proteins were suggested to a good partner to combine with animal proteins.

Task 4.2 Scale up of basic recipes:
A new, multifunctional cooling die system was built up for a Clextral BC45 -extrusion line. The configuration was able to manage capacities of 60-100 kg/h product in the used configuration. The reached throughput rate was excellent, considering the early stage of the process development. The system allows a good control of process parameters such as temperature and pressure. After the first test runs minor improvements of the design were discussed with partners and subsequently conducted. The products from pilot scale extrusions trials showed a surprising good fibrous texture and taste for all three basic recipes from task 4.1. Compared to lab scale experiments, the products were somewhat more fibrous and less soft. Thus slightly higher water content was used in pilot scale to receive a comparable product. LikeMeat base product was subsequently processed in pilot scale setting a throughput rate of 80kg/h in four different flavours: blank, chicken, beef and curry. A total of about 500 kg product was produced to serve as raw material for production developments and production of foods for the consumer tests.

Task 4.3 Aroma development during extrusion processing:
During the cooking extrusions trials for task 2.1 a manifold of effects on aroma was determined for the different pure raw materials. Most stunning were the loose of negatively associated taste attributes in whey and milk protein isolate, whereas other raw materials developed new aromas. These changes were described in deliverable 2.1. Further, a one day training workshop on sensory analysis was given by ETOL for the project partners. This workshop gave a perfect overview about the great possibilities of sensory analysis but also its limitations.

Since pure protein-rich raw materials showed already big influence on aroma, this was also expected for more complex formulations. This task focused therefore on the aroma development and degradation during extrusion of one of the three evaluated basic recipes. Due to the wide range of effects already noticed in the first trials, the consortium decided to extend and prolong this task to follow aroma developments during the whole recipe development stage.

The formation of aroma and odorants, respectively, was monitored by sensory methods as well as identification and quantification of odour-active compounds. The impact of potential precursors, which are known to generate odorants, was investigated intensively. For this propose, sugars and amino acids were added to the basic formula and the influence of the compounds on aroma was studies after extrusion. Furthermore, the use of spices, aroma and combinations of them as flavouring ingredients was tested successfully in two workshops with the partners dealing with aroma and spices. The workshops took place at Fraunhofer IVV and WIBERG. Finally, the stability of a flavour was characterised by following the concentration changes of key odorants during extrusion process. The results helped to gain a much better and deeper insight in both aroma generation and degradation induced by the extrusion process.

WP5: Improvement of texture and juiciness

Again, the production of juicy and fibrous meat analogue products in a cooking extrusion process relies on the network formation of the protein molecules under very specific process conditions. In WP2 protein-rich raw materials were characterised with respect to techno-functional, sensory and microbial properties. Based on this knowledge and with consideration of the outcomes from WP1 “Consumer preferences, acceptance and needs” three basic formulations, both vegan and non-vegan were developed within WP4. WP5 focused on the improvement of the sensorial aspects of the raw LikeMeat product. Recipe2 based on legume and wheat protein was used as starting material for the further developments.

Task 5.1 Determination of the Influence of hydrocolloids, starches and fibres:
Overall objective of work package WP5 was to further improve the texture of the protein based formulations. Therefore the addition of about twenty different non-protein ingredients both as single component and in combination with other ingredients were investigated in lab scale extrusion experiments. The respective developments were performed using the vegan recipe 2 that was developed within WP4. Products from the final optimised recipe were delivered as basis for the development of refined meat analogue products in WP6. First the influence of single ingredients such as different sugars, starches and starch-rich ingredients, as well as fibres, hydrocolloids and lipids on the texture and sensory characteristics of the extrudates was investigated. Special consideration was given to starch gelatinisation during the cooking extrusion process and its influence on the protein network. Based on the knowledge gained from these results, adapted non-protein compounds were developed for being added to the basic recipe. The new recipes containing adapted compounds were developed with regard to the juicier and more fibrous texture in comparison to the basic recipe. Sensory properties were investigated using trained and untrained panel for descriptive and team liking evaluation. Among physical properties tensile strength of textured strands was characterised using a texture analyser as objective indication of fibrousness.

Task 5.2 Scale up of optimised meat analogue recipes in pilot plant scale:
The optimised formulation was processed in pilot scale setting a throughput rate of 80kg/h. The LikeMeat base product was produced in four different flavours: blank, chicken, beef and curry. A total of about 300 kg product was produced and distributed to all partners to serve as raw material for production developments and in-house testing. Further a bench mark test was performed using a trained sensory panel. The LikeMeat product showed highest meat like appearance among non-meat products with a distinctive, anisotropic structure.

WP6: Development of refined products

The objectives of this work package was to create prototypes of refined products, one for each country, depending on asuccessful match between consumer target groups and intended usage; to gain information about the sensory and processing qualities of the products, e.g. handling and storage properties; and to report on the sensory and microbial changes in the products during storage, as well as comparison of the nutritional qualities with equivalent meat products. Considerable effort was spend to test the safety of the LikeMeat material over (storage) time.

Task 6.1 Formulation of recipes:
The product characteristics that need to be optimised change with each adaption of the products, and changing the ideas about the products in discussion with the partners in the consortium. Therefore, this task was worked on in an integrated approach with task 1.5: Identification of product requirements. The findings were reported in deliverable D6.1. By doing so, the report represent a comprehensive body of knowledge on the next stage of product development as it was presented in the model, the ‘development funnel’. The stage that is refered to is called ‘Idea refinement’. In WP1, the product ideas were conceptual. The tasks reported in these deliverables are all on making these product concepts concrete with respect to aspects that matter directly to consumers, i.e. content of the recipes, safety, nutrition. D6.1 reports on the creation of three recipes based on the concepts of WP1. Over all development steps ‘idea refinement’ puts up the products that reach the final stage before consumer testing. Throughout the stages it becomes obvious that the sequence of selecting is not a linear one. Because of different development processes playing a role at the same time, some products remain in while excluded in the selection step, some go out despite being favourably judged. Reasons to stay in are thus a) favourable evaluation in the selection step, and b) intervention by the industrial partners. We consider this a normal process. It shows that three important selection criteria play a role in selection: i) technical developments: some technical limitations or discoveries led to exclusion of product concepts; ii) consumer research input: consumer and researcher evaluations give indications as to what products have the most potential for further development; iii) business: although not necessarily trained in a systematic selection approach, in the end the gut feeling of the industrial partners is probably the most important selection tool for market success. Using these tools, five final products were selected and ranked (from highest to lowest potential): 1. Mexican minced meat; 2. Baked stripes; 2. Asian Cuisine; 4. Schnitzel; 5. Healthy bites. Four more elaborate recipes were described in this report. The potential of the fifth product (Baked stripes) lies in the shape that makes it fit for multiple use. The exact recipe depends on that use (occasion). Considerable effort was spend to test the safety of the LikeMeat material over (storage) time. This work refers to Task 6.2.

Task 6.2: Hygiene and sensory evaluation of the refined products:
Within task 6.2 the focus has been much more on microbial than sensory stability. The main reason for this is that the hygiene aspects seemed to be more critical both for the tests at FBR and overall for the marketing of LikeMeat. In order to prepare the LikeMeat products for consumer testing on larger scale, a series of studies were performed on the sensory and microbial changes in the products during storage. To ensure the microbiological safety of intermediate and refined LikeMeat products for the consumer tests, it was necessary to install a monitoring system at the pilot plant in Freising. Hence a sampling strategy was jointly prepared (BOKU, FRAUNHOFER, DLO-FBR) on the basis of production capacity and duration. Depending on the developed product types appropriate storage conditions and durations were selected. The sensory evaluation of the base product by BOKU, learned that the dominant sensory characteristics did not change during storage at 6°C, even when the total microbial count rises. Equally, the sensory characteristics of the base product did not change significantly during frozen storage during about 6 months as observed by Fraunhofer. The sensory of the refined products tested at Food & Biobased Research, Wageningen, was observed not to change significantly during 6 or even more storage weeks, when stored at -20°C. These observations together with the microbiological information below reflect the related efforts from the RTDs.
Important conclusion is that the LikeMeat products should not be kept stored longer or in other conditions than fresh meat products. Positive is that the sensory characteristics do not change much remarkably during chilled storage of the intermediate product. However, this also means that a natural spoilage warning signals typical for protein-rich foods is lacking could not be observed for the tested intermediate LikeMeat, requiring the development of other ways to guarantee safety.

Task 6.3 Nutritional quality of the refined products:
The nutritional qualities of the LikeMeat material were compared to that of different commercial meat types in a comparative nutritional analysis. In all cases the LikeMeat product performs well in comparison: more carbohydrates, more protein, less fat and water, more fiber. Energy density is slightly higher. All in all, the conclusion of WP6 was that the product concepts of work package 1 were worked out in acceptable real products that can be transferred to the next stage: consumer testing (WP8).

WP7: Improvement of freshness and shelf-life of packed meat analogues:

Microbiological activity can be seen as the most contributing factor for the spoilage of food products. This process within a food ecosystem can be measured by shelf-life studies promising a safe product for the consumer within a certain storage life span. Intrinsic and extrinsic factors can be applied to describe the influence on food deterioration throughout the supply chain. The before mentioned intrinsic (implicit) factors are caused by biotic changes mainly caused by a great variety of bacteria, moulds and yeasts. However, depending on the process stage (from raw to canned products) certain foods have characteristic spoilage behaviour represented by specific bacterial consortia. In general the storage temperature, packaging type (atmosphere, vacuum-packed) and physico-chemical properties (aw, pH) cause a substantial selection of the spoilage microbiota.
To obtain data about the shelf-life, the intermediate and refined products were stored under two different temperatures (6°C and 10°C) for one, four and if necessary more weeks. Subsequently, product changes during storage were analysed by microbiological and parallel sensory evaluation (optical and textural changes). As shown before, the shelf-life of foods strongly depends on the extrinsic factors, such as processing conditions, transportation and storage. Hence, it was necessary to monitor the hygiene conditions during the whole production of LikeMeat. This was especially important, as these samples were used in the consumer studies (out-of-home tests) in the Netherlands. The microbiological safety of the products had the highest priority and therefore the handling and storage conditions had to be monitored.

Different LikeMeat product categories were tested for relevant microbiological parameters (raw products, sterilized products, products for consumer tests and optimized products). The sampling plan was designed in correspondence with Fraunhofer (Freising, Germany) in order to cover the whole extrusion process duration. Storage tests were performed with the non-sterilized intermediate products, sterilized intermediate products, selected products for the consumer tests (refined LikeMeat products) and optimised LikeMeat products.
The intermediate products were vacuum packaged immediately after processing. The samples were stored at -20°C. Sterilized intermediate products were vacuum-packaged immediately after processing and then sterilized at 121°C for 20 min. The LikeMeat products were further optimised regarding their sensory profile. Especially juiciness and the texture could be improved through the application of certain concentrations of starch. The aroma profiles of beef, chicken, chicken-curry and blank samples were used for testing of shelf life. The samples were stored at -20°C.At sample receipt the lots were checked optically for correct packaging conditions (labelling, bag sealing and temperature) and photographed for documentation purposes. The LikeMeat samples were then stored at -20°C prior analysis. Important physicochemical properties (aw-values, pH and receiving temperature) and thereby the potential spoilage behaviour of extruded proteins were determined. Microbiological routine methods were applied for the screening of the extruded protein-rich recipes. The microbiological examination of most samples (except storage tests) was carried out immediately after delivery. If an immediate analysis was not possible, the products were stored at 4°C (sterilized intermediate products), -20°C (non-sterilized intermediate products) and -80°C (products for the consumer tests and refined products). Non-sterilized intermediate products were analysed to monitor the production process. Hence it was decided to take a sample approximately every 30 min.

The effectiveness of the sterilization process resulted in a microbiologically safe LikeMeat product. These products were further used for consumer tests. All product concepts (concepts decided earlier) were refined by the SME (burger, minced meat, minced meat for sugo, baked stripes, raw LikeMeat) intended for consumer tests and further storage trials. Storage tests showed, that without sterilisation the shelf life of the fresh LikMeat products achieved approx. one week. However, foodborne pathogens could not be detected. Sterilized intermediate products were stored for approx. one month at 4°C. After this period no microbial activity could be detected by the application of culture based methods. Sterilization of the intermediate products achieves a minimum shelf life of one month at chilled storage. Storage tests were only performed with schnitzel, sterilized LikeMeat and burger, as the other products (minced LikeMeat and baked stripes) had already too high microbial loads at the starting point of the storage test. The samples were stored at 6°C. The burger showed a very good shelf life at 6°C. Higher counts could be achieved after 12 weeks of storage and were mainly dominated by bacilli. The optimised products (chicken, beef, raw, chicken-curry flavour) were microbiologically examined immediately after production and further stored at 6°C. After one week of storage at 6°C the raw LikeMeat (blank) and the chicken-curry showed microbiological activity. After two weeks of storage all products showed elevated bacterial counts. In the chicken-curry sample the storage trials for this product type has been stopped after 2 weeks at 6°C. The beef sample showed the lowest bacterial loads at this period. After three weeks of storage the total viable counts of beef, raw LikeMeat and chicken increased from 10 log cfu up to 13 log cfu per gram. After the microbiological screening of the above mentioned products, obtained bacterial isolates were submitted for molecular analysis (PCR-based methods). The aim of these procedures was to identify possible sources as well as possible cross-contamination routes in the production of LikeMeat products. All in all it could be shown, that the bacterial microflora of stored Likemeat products was dominated by Enterococcus species.The origin of the enterococci isolated from the LikeMeat products are believed to be some kind of (re-)contamination. The source could not be identified. The application of molecular methods (Sequencing, PCR-based subtyping) is inevitable for the microbiological characterisation of “novel” foods and their microflora. The shelf-life of intermediate LikeMeat products (non-sterilized) with or without flavouring can be recommended for one week at 6°C. The application of spices and herbs can negatively influence the shelf life. A case to case assessment is necessary. The determined dominating Gram-positive spoilage microflora did not exhibit strong spoilage signals and therefore it seems to be difficult to recognize symptoms, typical for the spoilage of protein-rich foods. As the intermediate Likemeat products in this project did not contain relevant amounts of Gramnegative spoilage microorganisms, no information about typical spoilage symptoms can be provided.

Task 7.2 Sensory evaluation of stored products:
The sensory evaluation of the base product by BOKU, learned that the dominant sensory characteristics did not change during storage at 6°C, even when the total microbial count rises. Equally, the sensory characteristics of the base product did not change significantly during frozen storage during about 6 months as observed by Fraunhofer. The sensory of the refined products tested at Food & Biobased Research, Wageningen, was observed not to change significantly during 6 or even more storage weeks, when stored at -20°C. These observations together with the microbiological information below reflect the related efforts from the RTDs. Task 7.3 Application of different herb extracts and marinades to improve the shelf-life of the products –microbiological analysis: The application of spices and herbs can negatively influence the shelf life. A case to case assessment is necessary. One recipe (chicken-curry) was monitored during production at pilot plant scale. An increase in bacterial numbers could be determined as a function of time. This means, after a few hours of processing the bacterial counts increased due to process conditions during extrusion. Additional examination of spices (curry powder) showed typically high bacterial levels including aerobic spore formers which may survive the processing conditions. Therefore, the application of spices should be tested before applied in novel recipes. The outcomes of the storage trials with chicken-curry recipes reflected the problems with curry powder.

WP8: Overcoming barriers to implementation:

The objectives of this work package was to test the developed products for overall consumer appreciation and the willingness of consumers to buy the products in in-home and out-of-home situations and in complete meal type products. Successful implementation of experience, credence and search attributes had to be assessed in real life situations. Additionally, the aim was to identify hurdles that have negatively impacted market introduction of meat-like products in the past.

Task 8.1: Consumer testing for products in real life setting: Out-of-home, Task8.2: Consumer testing of products in real life settings: In-home.
Tasks 8.1 and 8.2 were performed in an integrated way, as the unit of analysis was the LikeMeat product. The different consumer tests provided information with respect to (partly) the same products, thus providing more information how to improve the particular product. For Task 8.1 over a period of six weeks, 88 consumers came to the Restaurant of the Future (Wageningen UR, the Netherlands) once a week, to have a hot meal that they could choose from a selection of three buffets. Each buffet contained a dish with a LikeMeat meat replacer: Mexican dish, Italian pasta sauce or Baked stripes. Also a central location test was held, where consumers tested the different LikeMeat products in a controlled situation (sensory booths) at the facilities of the Restaurant. For Task 8.2 in parallel to the Restaurant of the Future test (Task 8.1) participants were asked to prepare and evaluate a LikeMeat Burger or piece of unprepared (raw meat) LikeMeat (named ‘chicken steak’) at home. The products were tested with consumers in real life situations, including as many of the aspects that belong to normal consumption patterns and that determine product success, but that are often not tested in regular product testing. Additionally, different LikeMeat base materials were tested by the participants at their own facilities. A standardised questionnaire was prepared on their handling and their experiences at cooking and evaluation of the end-products.

The central location test showed that firmness was only moderately associated to overall liking and liking of the taste of the LikeMeat products. Liking was mostly positively associated to spiciness, but negatively to ‘sponginess‘. In further development of the production technique therefore, the sponginess of the material needs to be reduced. The Mexican dish was the most successful product, regarding the liking ratings, sensory evaluation, choice behaviour, experiences at home and remarks of the participants. Also likelihood of eating the LikeMeat replacer again and willingness to buy the product in a shop were highest for the Mexican dish. The best indicator of potential success may be the fact that the quantity eaten of the Mexican dish was highest. Additionally participants judged the Mexican dish as innovative. Next, Schnitzel might have potential to become a successful product, as it was also rather highly liked by the participants in the central location taste test. The product rated best for appearance (and smell) was the Burger. However, the Burger product scored worst on mouthfeel. The PCA plot showed that the Burger products were evaluated to be too loose (not firm). Generally, the LikeMeat products were evaluated more positively in the Restaurant setting compared to the in-home setting. For Baked strips, Schnitzel and Pasta sauce, the products were liked slightly better when information was provided. Likely, the information helped consumers to understand product deviations from ‚normal‘, which made them more positive. This supports our view that the LikeMeat products not necessarily need to be perfect imitations of meat: they can be different, as long as the expectations of consumers are well managed. We expect that it is possible to build on this positive effect with better information, thus improving the appreciation of the LikeMeat products. Overall conclusion of this evaluation from the SME partners was that the raw material still needs improvement. Some of the flavoured products can actually look nice, but only few manage to be acceptable in taste and texture. Preparations that added flavours to the products seems to improve overall acceptance a lot, which confirms the consumer findings as discussed above. With respect to market segmentation, the appreciation ratings for LikeMeat Italian pasta sauce point to relatively committed and conscious consumers that have an interest in plain and healthy food. Appreciation of the LikeMeat Baked Strips point to consumers that like good ‘value for money’. LikeMeat Burgers seem to appeal to consumers that are not too bothered about price or calorie content, but like a filling product. As the same pattern emerged with the chicken analogue LikeMeat product it was concluded that the LikeMeat chicken analogue was apparently not associated to chicken meat. Liking ratings for the LikeMeat Mexican dish was not associated strongly to any of the segmentation variables. That may be (another) part of the explanation why it generally comes out as the best liked LikeMeat product: it is liked by most consumers. It appears that most LikeMeat product concepts are targeted to different consumer segments. The Mexican dish seems to appeal to most groups, showing it has potential not only from a technical and consumer (sensory) evaluation perspective, but also from a marketing and business perspective.

For the consumer tests in the Restaurant of the Future and in-home we got a clearance that the research fulfills the Wageningen code of conduct.

Task 8.3: Lessons learned from earlier market introductions:
Based on a Innova database and literature search, it became apparent that, for texturized vegetable proteins to be successful on the market the most important aspects seem appearance, taste, texture and also convenience, price and credence quality attributes (e.g. healthiness). Apparently, products that do well in the market already look acceptable, have good sensory properties (taste, texture), are convenient and sustain one or more credence attributes. Meat replacers that can also easily be used as an ingredient for a home dish seems to have an advantage. Most of these ‘must haves’ of meat replacing products can be associated to one of the particular flexitarian target groups as they were defined in work package 1 (see deliverable 1.1.2) thus confirming the value of the work in WP1. We suggest that in order to be a success, future meat replacing products should match characteristics with at least one of these consumer target groups. Selling points should focus on one or preferably more of the aspects ‘health’, ‘convenience’, inexpensive, sustainably and animal friendly produced, and/or associated to ‘fun’ or a particular cultural style for social events. It seems that the LikeMeat material and the variety of (concept) products that were developed in the project have the potential to combine one or more of the ‘must have’ characteristics – either intrinsically and/or in their marketing and communication characteristics. The final conclusion of WP8 was that although we definitely find a lot of promising results and a lot of progress has been made over the project, the absolute level of appreciation for the LikeMeat products still needs improvement. WP8 gives many very specific clues as to what should and can still be improved. It shows there is definitely market potential for LikeMeat, considering the abundance of hopeful results. The next step should be to focus more on the product concepts that already are relatively successful and on the target groups that are willing to buy these products. Not only the motivations and preferences of these target groups, but also the shopping behaviour (locations, frequencies) and media use. A lot of this information is already available in the results of WP1. Direct interaction with consumers from the selected target groups will provide valuable information that, next to improving sensory aspects, can help to manage consumer expectations and direct marketing messages. This report provides a broad overview of methods and analyses that can be used for consumer product testing at the final stages of product development. A more general recommendation is that also in the future a careful selection of appropriate consumer tests will prove its value. We recommend that in similar product development processes the creativity and experience of the companies is used more in the beginning stages of the process than was the case in the current set-up. However, this requires dedicated commitment in time and energy of the project participants. It should be stimulated that partners actually join each other in their kitchens more often and be more active in the development process. Even closer involvement of the SME’s and practical interaction with research experts might improve the speed and focus of the results. Continuation of the development of LikeMeat products should again include a cooperation between technological development, consumer and marketing expertise, and add more business oriented tasks to the process.
Potential Impact:
The results of the LikeMeat Project will enable SMEs to produce premium meat alternative products. The products developed within the project are an excellent basis for a large range of applications in dishes showing the same indulgence properties than conventional meat products would do. However, “LikeMeat” does not include meat and thus can address additionally to conventional consumers also to vegetarians and vegans. A large number of dissemination activities took place in all kind of media but also at scientific and public conferences, round table discussions, trade fairs and exhibitions to transfer this information to consumers, the food industry and to the scientific community. Keeping in touch with the consumer is of outstanding importance for the development of a consumer product. For the new food product “LikeMeat” the transparent communication is a must, in order to avoid any misconception.

The gained knowhow within “LikeMeat” offers several routes to be exploited post-project. An IPR management and watch were installed during the project. For the exploitation post-project a clear regulation was agreed on. This regulation, described in a joint ownership agreement and in the consortium agreement, includes terms for the ownership of know-how, means to handle IPR related aspects and protocols for future decision making. It enables both the use of the foreground generated within “LikeMeat” by the involved SME partners and conditions to licence towards third parties.

Meat production is known to have a strong environmental impact regarding water use and the production of greenhouse gases. Depending on the source, meat production is said to account for about 18% of all greenhouse gases, causing in any case more greenhouse gases than transportation and industry. The optimised meat substitutes in combination with the investigation of consumer needs and the overcoming of market barriers can help to introduce a radical change in the most important part of the human diet: Reduction of the consumption of meat.

The teamwork within the LikeMeat participants improves both the economic and social cohesion of the EU. This co-operation will be kept established after the end of the project, as the members of the consortium work together to further develop and exploit the results of the technology. Finally, in terms of the exploitation of the results of this transnational European research effort, the inclusion of a synergistic group of SME partners and other enterprises from across Europe will serve to ensure that the results of this project are used at European level and that the SMEs, that will make up the supply chain of the LikeMeat technology, can reach pan-European markets (and beyond).

Main Dissemination Activities:

During the project period a large number of dissemination activities took place in order to inform about project objectives and results. Since different groups were addressed such as project partners, the scientific community, the food industry and last not least the public a wide range of dissemination channels was used.

Fraunhofer developed and established with the contribution from all partners a project website with public and internal areas, which is updated continuously. This tool is simple, user friendly and supports collaboration between partners. The internal area is used for communication between project partners. The website acts as a dissemination tool for all meeting reports, which was also distributed to partners and to the European Commission via Email. The website ensures the visibility of the project and disseminates information directed to the public and to industry. The website was transferred to the SME partners post-project and will be kept visible for at least one more year.

Training and transfer of know-how to the SME
Knowledge generated by the project was transferred by the RTDs to the technical and managerial staff of the participating SMEs. Knowledge transfer comprised the use of the new ingredients, their handling and required processing conditions as well as new and/or adapted technologies/methods for the production of improved meat analogue products including legal and safety aspects. Additionally a deep insight into state of the art consumer research was given.
Additionally several partners attended conferences and trade fairs or travelled to other places of interest to learn and bring knowhow into the project.

Dissemination by the SMEs to consumers
The single companies contributed to several dissemination activities that were conducted by the LikeMeat consortium, such as the website, participation at ANUGA FOODTEC trade fair or the successful applications for the Science to Business Award and the Ferchau Innovation Award. Additionally several activities were conducted by the companies such as publishing LikeMeat in newsletter, presenting LikeMeat towards business partners and customers or giving interviews. Refined LikeMeat products were developed by a large number of partners and subsequently tested in-house or with external persons such a consumer. By all these means potential consumers were informed about the new products and their advantages. The information of the consumer and the optimisation of LikeMeat according to the consumers’ feedback give the SME the chance to bring products on the market that contribute to direct and prompt economic benefits by increasing their turnover. A high and prompt return on investment is therefore plausible. Additionally, the knowledge gained in the project can be used to develop further meat-free products for their portfolios.

The consortium actively sought to publish results where this is compatible with IPR policy. In terms of scientific publications, the RTDs performers discussed and will discuss their intentions with the SMEs in order to obtain their consent in advance on publication materials as mentioned above.
In total 17 articles for general project information to the public were launched, such as “Is this the end of meat?” (The Independent 2012). Furthermore 12 contributions were launched in radio and television to make the LikeMeat project public. Also in different lectures and round table discussions the project´s topics were discussed.
Direct Information for the food industry was provided in 8 online and print media articles, for example “Ersatz mit Biß und Aroma” (Food Technologie 2012) or “Die vegetarische Schnitzelfabrik” (Lebensmitteltechnik 2013).
The scientific community was informed about selected project results in 4 scientific papers and poster presentation as well as in 9 contributions and participations in international scientific conferences, trade fairs and innovation contests.

Information sheets and LikeMeat logo
Information sheets were published to inform potential consumers and consumers about the products, their ingredients and production technology. Since meat analog products are relatively unknown to most consumers a transparent way of communication is a must to avoid misleading fears. The information sheets were used and will be used in the future at various presentations and trade fairs.
Further a logo was created to mark LikeMeat activities in the form of a corporate design.

Exploitation activities and results:

Project results: In the course of the project substantial knowledge about consumers’ behaviour, technological aspects and safety with respect to meat analogue products could be gathered. This knowledge is summarised in 7 project results.
These results comprise:
1) Knowledge on consumers’ behaviour with respect to meat analogue products.
2) Knowledge about specific techno-functional properties of raw materials and there interaction.
3) Knowledge about aroma formation and degradation in the extrusion process.
4) Protocols and recipes for the production of two raw meat analogues.
5) Protocols and recipes of the production of three differently refined products.
6) Knowledge about general microbial stability of extruded meat analogues and specific knowledge for two refined products.
7) Specific knowledge about consumers’ acceptance of two refined products.

At the end of the project, the development of a product prototype was completed, so that every company is able to use the project results. Each partner has at least one new product (pre-mix, raw meat analogue, refined product, and aroma) available that differ fromexisting products in his portfolio. The developed products can be launched in the marketplace after minor adaptions to individual requirements, while the SMEs keep on developing further products. Marketing and distribution will follow existing routes, saving time and effort. The exploitation will be done by one or several of the following options: By SMEPs themselves, by the joint founding of a start-up in cooperation with FRAUNHOFER or by a non-exclusive licencing towards third parties from food industry.

The first option address to the individually participating SMEP, whereas the other two options address to the SMEP consortium as such. The results from the project are available to be used for the further exploitation. This will give the SME partners the possibility to develop a knowledge based strategy to develop new products and markets. This increases their turnover and lead to direct and prompt economic benefits. A high and prompt return on invest is therefore likely.

It is therefore realistic that new meat substitutes will be in the marketplace within 12 months after completion of the project.

List of Websites:

Project coordinator:
Dr. Florian Wild
Fraunhofer Institute for Process Engineering and Packaging
Giggenhauser Str. 35
85354 Freising
Phone: +49 8161 491-416

Business contacts:
CyberColloids Ltd
Sarah Hotchkiss
Unit 4A Site 13,
Carrigaline Industrial Estate
Carrigaline, County Cork
Phone: +353 21 4375773