Skip to main content
European Commission logo
español español
CORDIS - Resultados de investigaciones de la UE
CORDIS
CORDIS Web 30th anniversary CORDIS Web 30th anniversary
Contenido archivado el 2024-06-18

Enabling the exploitation of Insects as a Sustainable Source of Protein for Animal Feed and Human Nutrition

Final Report Summary - PROTEINSECT (Enabling the exploitation of Insects as a Sustainable Source of Protein for Animal Feed and Human Nutrition.)

Executive Summary:
The overarching goal of the PROteINSECT project was to evaluate the potential use of insects as a novel source of protein for inclusion in animal feed. Focus on organic manures as rearing substrates for fly larvae enabled us to evaluate the possibility of deriving safe, high quality and sustainable feed protein whilst at the same time reducing volumes of low value wastes.
New rearing systems have been established in Ghana, China and the UK and improvements have been made to those already established in Mali and China. Systems ranged from semi-commercial scale production to those designed for use by small-scale livestock farmers. Whilst overall emphasis was placed on the rearing of houseflies (Musca domestica), production systems were also developed for black soldier fly (Hermetia illuscens) and blowfly (Chrysomya megacephala). Considerable improvements to the efficiency and productivity of the rearing systems were made through, for example, the development of separation and drying techniques. Data was used as the basis for economic, social and environmental impact assessments allowing recommendations for the future development of insect rearing stations at different geographical locations to be defined.
The nutritional quality of dried dipteran larvae was found to be highly compatible with use in feed, having both high total protein content and levels of key amino acids comparable with those found in fishmeal. The value of dipteran meal is difficult to predict given fluctuations in feed prices but was estimated to be at least double that of soymeal, but lower than fishmeal. Solvent extraction was found to be an efficient and scaleable processing method yielding protein enriched samples suitable for formulation. The production of more than 2.3 tonnes of dried larvae enabled fish, poultry and pig feeding trials to be conducted in Europe using both crude and protein enriched material. Fish and poultry trials were also conducted in Mali, Africa and China using both dried and live insects. Animals reared on insect containing diets performed as well as those reared on commercial feeds with no taints detected in fish and meat derived from the feeding trials. These results provide powerful evidence for the suitability of fly larvae as a partial replacement for soy or fishmeal in fish and monogastric feeds.
A state of the art screening programme assessed the safety of dried larvae supplied by the various rearing systems. Levels of more than 500 potentially toxic chemical contaminants were all below recommended maximum amounts suggested by bodies such as the European Commission, World Health Organisation and Codex. However, elevated levels of the toxic heavy metal cadmium in some samples indicated that pre-screening of rearing substrates would be necessary to mitigate risk. Tests found no evidence for the presence of viable biological contaminants such as Salmonella and Campylobacter. This data made a valuable contribution to the preparation of a risk profile related to production and consumption of insects as food and feed published by EFSA in October 2015.
A positive and receptive platform in Europe is vital for the future utilisation of insect protein in animal feed. Key stakeholders from industry, regulatory bodies, academia and retailers were engaged throughout the project facilitating the creation of a consensus business case and a subsequent White Paper presented at a European Parliamentary Reception in April 2016. Two information gathering exercises involving > 2000 consumers found a high level of support for the use of insects in feed together with an expressed desire for more information. Determination to communicate the findings of the project is evidenced by participation of PROteINSECT partners in 54 dissemination events over the course of the project. More than 53 000 unique visitors have visited the project website reflecting a high level of public interest in the topic, with more than 300 articles and films on insects in animal feed tracked in the print and broadcast media. The final project Conference attracted a global audience of 130 interested stakeholders. Eleven peer reviewed publications have been produced with a further 14 anticipated in 2016.
The work of PROteINSECT has established that it is possible to obtain high quality insect protein from fly larvae reared on low value wastes that is safe for use in animal feed. Further research and legislative changes are required to realise the potential of this approach and thereby reduce Europe’s heavy reliance upon imported feed protein.

Project Context and Objectives:
The European Union protein deficit
The global population is currently approximately 7 billion and it has been estimated by the United Nations that it will reach over 9 billion by 2050. In addition there has been an approximate fivefold increase in global meat consumption since the 1940s as a result of income growth, increasing urbanisation and changes in lifestyles and food preferences. This has resulted in increasing pressures on the production of protein crops for animal feed.
The principal sources of protein for animal feed are currently soya and fishmeal. However, the total EU protein crop production currently supplies only 30% of the protein crops consumed as animal feed in the EU, with a trend over the past decade towards an increase in this deficit. Therefore, the EU imports millions of tonnes of plant protein, primarily in the form of soymeal, mainly from Brazil, Argentina and the USA. New customers for South American suppliers, notably China, may weaken the stability of the markets and the EU supply chain. The dependency on protein imports for animal feed also makes the EU livestock sector vulnerable to price volatility and trade distortions. This situation is further exacerbated by the rise in the consumption of fish with aquaculture now the fastest growing food production system in the world. Aquaculture relies heavily upon fishmeal as a protein source but fish farms are increasingly looking to crop proteins as a lower cost replacement for fishmeal. Fishmeal is not only used in aquaculture but is also an excellent source of highly digestible protein ideal for poultry and pig diets. Thus fish and animal farming are together placing growing demands upon protein crops and fishmeal supply. Further issues in relation to land and water availability will place additional pressures on the world’s ability to meet growing demands for animal feed protein production.
In 2011 the European Parliament adopted a resolution to address the EU’s protein deficit, stating that urgent action is needed to replace imported protein crops with alternative European sources. The European Parliament highlighted a number of factors affecting the EUs protein deficit including protein supply, demand and international trade
Insects offer a promising alternative to conventional protein sources for animal feed for fish and monogastric animals (poultry and pigs) as they are rich in protein. Depending on the insect species, protein levels can be up to 63% on a dry matter basis. The use of insects as a source of protein for animal feed has been the subject of evaluations for several decades (e.g. Bondari and Sheppard, 1987; Newton et al., 2005; Hem et al., 2008), and although the concept has been established and some animal feeding trials undertaken, there have been few studies that have examined processing of the produced insect material or that have assessed the safety, economic, environmental, social and acceptability issues.

Choice of insect species
Fly species are the most extensively researched insect order for mass production and use as animal feed. Furthermore they can utilise a wide range of waste substrates offering potential for low economic and environmental costs.
The larvae of fly species such as those from black soldier fly (Hermetia illucens) and house fly (Musca domestica) contain up to around 63% protein and 36% fat (d.wt) (Makkar et al, 2014). They possess high levels of key amino acids (e.g. lysine, tryptophan) when compared to most crop plants. Fly larvae exhibit rapid growth and short life cycles. For example, the house fly, depending on rearing substrate and temperature, can develop from egg to a harvestable larval stage in as little as 4-5 days. Black soldier fly and house fly larvae can utilise a range of low value waste materials for development and insect producers in countries such as China, South Africa and the United States are already rearing large quantities of fly larvae on organic wastes for aquaculture and poultry feed (FAO 2013). Smaller scale local farming of fly larvae in rural Africa is helping to reduce reliance on manufactured feeds.
The European Union generates 88 million tonnes of biodegradable organic waste and as much as 1.4 billion tonnes of manure each year. With growth in the consumption of animal products this is set to increase and, as such, practical solutions are needed to deal with this high volume low value waste stream. Insects are not only able to provide the potential to extract protein from waste material but also facilitate significant reductions in waste volume. Dipteran larvae can in fact reduce the mass of organic waste by up to 60% in 10 days (Miller et al., 1975, Sheppard, 1983). There is evidence that the remaining digestate can be exploited for added value in a number of ways. This is based on the fact that fly digestion only reduced the values of key elements (N, P, K, C) within the substrate by 40-60% (Newton et al., 2005). Such potential uses for residual material include compost, fertilizer, soil remediation material, and as a substrate for biogas generation (anaerobic digestion).

Processing of insects for incorporation in animal feed
In addition to protein insects may also be a source of other high value products. Lipids are a large component of fly larvae with crude fat content (dry wt) in meal produced from house fly larvae reported to range from 14 to 27 % (Fasakin et al., 2003; Aniebo et al., 2008, Pretorius, 2011) with variations likely to be due to the differences in composition of feed substrates (Pretorius, 2011). The proportion of fat in M. domestica larvae can vary depending on the age of the larva and the processing method used. In a study by Aniebo and Owen (2010) fat content was shown to increase from 22.4% to 27.3% on a dry matter basis as the age of larvae increased from two to four days. The principal fatty acids found in M. domestica larvae and pupae are palmitic, palmitoleic, oleic and linoleic (St-Hilaire et al., 2007; Hwangbo et al., 2009). The potential to convert oils derived from the black soldier fly to biodiesel has been demonstrated (Li et al., 2011a,b). Both M. domestica and H. illucens are a source of lipids that have a range of uses for nutrition and potential for exploitation as an industrial feedstock or a sustainable energy source. Possibilities for obtaining other valuable by-products from the insects such as lipids, vitamins, minerals and chitin as well as fertilisers from the digested substrates should also be considered.
The value of insects as complete feeds or feed ingredients has been established for a number of fish species in both temperate and tropical climes. The use of black soldier fly larvae has been demonstrated in diets for channel catfish (Ictalurus punctatus), and blue tilapia (Oreochromis aureus), as well as rainbow trout (Oncorhynchus mykiss) (Bondari & Sheppard, 1987; St Hilaire et al., 2007). A trial using Atlantic salmon (Salmo salar) demonstrated that fish fed on a diet containing black soldier fly meal performed as well as those fed on the control (fish meal) diet and an increase in feed conversion efficiency was observed (Lock et al., 2014).
A feeding study for pigs using a dried meal prepared from a wild population of H. illucens larvae was conducted by Newton et al. as early as 1977. The nutritional composition (proximate analysis & amino acid content) of the meal was compared to a soybean meal, and protein digestibility studies were carried out. The authors concluded that the larval meal would be a suitable ingredient in swine diets but more work was required to develop efficient larval production methods and to establish optimum inclusion levels for pig diets. No evaluation was made of the safety of the product by assessing the potential presence of pathogens or undesirable substances such as contaminants or residues
Insects are a natural component of the diet of chickens and other poultry. Although there is only a relatively small quantity of experimental data available to date, it would appear that replacing a proportion of the plant-derived protein within a given diet with fly larvae can produce birds of equal or higher quality (Ocio et al., 1979, Hwango et al., 2009) to those reared on conventional feed. Trials have assessed different inclusion rates of the larval meal and some studies have shown that high inclusion rates may result in lower feed intake and reduced performance (Bamgbose, 1999). Studies have also been conducted on the use of live fly larvae to supplement the diet of rural chickens. In a study in Togo this resulted in a higher growth rate of the chickens and a higher clutch size and egg weight (Ekoue and Hadzi, 2000).

Quality and safety of insect protein
The quality of the protein is important in animal feed and essential amino acids need to be present for a balanced diet. In addition to protein, total fat and its constituent fatty acids are also an essential component of the diet. The nutritional composition of insects varies depending on the species and the feedstock used to grow them. For example, feeding black soldier fly larvae on fish offal increased their lipid content, including omega-3 fatty acids (St. Hilaire et al., 2007). A recent study comparing nutritive characteristics of a range of insects has shown that the amino acid profile of dipteran insects is superior to soybean meal and more similar to fishmeal (Barroso et al., 2014).
A major consideration in the use or applicability of any novel feed product is to demonstrate its safety, in particular if the initial substrate used for its production is a waste product. Information on the safety of the use of insect protein is very scarce in the literature. Research to assess the potential effects of the presence of some metals (cadmium, lead and zinc) and to determine possible bioaccumulation revealed accumulation patterns according to metal type and concentration; cadmium was accumulated, lead suppressed and zinc remained constant (Diener et al., 2009, 2011). In addition it was observed during field experiments that high concentrations of zinc in the growth substrate led to problems with the fly population. The authors recommended developing a process that allows separation of heavy metals from prepupae and residue.
Studies on the microbiological and chemical safety of insects reared for feed are also limited. Recently, review papers have been published on the microbiological and chemical safety of insects used for feed and food (Belluco et al. 2013; Van der Spiegel et al. 2013). Food safety authorities in the Netherlands (NVWA 2014) and Belgium (FAVV 2014) have published their opinions with an opinion from the European Food Safety Authority also published in September 2015.
Information about the possible transfer of substrate specific hazards such as prions from waste streams containing specified risk materials (e.g. abattoir, supermarket, restaurant or household food waste) and mycotoxins (e.g. from contaminated or degraded cereal products) are missing at present. Allergenic risk in relation to animal welfare and occupational exposure during feed manufacturing is also of concern.
Hazards associated with insects for use in feed depend on; feedstock, species, and production/ processing conditions. To date, little information is publically available about the hazards associated with insects for use in feed and this should be rectified through the adoption and sharing of best practise supported by robust analytical data for R&D and regulatory purposes. Efforts to establish guidance in relation to standardisation of insect rearing and processing practises will also help to ameliorate risks, acknowledging that production scale and local requirements will have a significant influence on the approach taken.

Sustainability of insect protein
The reliance on plant based protein and the use of fishmeal in animal feed is not sustainable given the increasing global population and the increased demand for land and water resources to provide sources of human nutrition. The environmental, social and economic impacts of insect production have only been studied to a limited degree. The extent to which a life cycle assessment of the production process can be made will depend on the availability of evidence from insect rearing systems, quality and safety analysis and animal feeding trials. Once a baseline for these factors is established, the environmental, social and economic life cycle impacts of insect based animal feed production systems can be assessed. This assessment can then be used to facilitate the design of optimised and sustainable production systems suitable for adoption by both small and large-scale operations in different geographical locations. Ultimately life cycle assessment will enable policy and technical recommendations to be provided for the establishment of economically viable and efficient fly rearing systems in different locations.

Regulatory considerations
The use of insects in animal feed is not currently recognised or specified in European legislation or regulation and this presents a major barrier to the development of industrial insect-rearing facilities. Key areas that need to be addressed include the authorisation of insect processed animal protein (PAP) for use in non-ruminant feed, the need for solid scientific evidence of the safety of products to permit the rearing of insects on animal manures and legislation to address novel issues associated with mass production. The regulations currently prohibit the use of processed insects as a feed material, as the TSE regulation prohibits the feeding of all farmed animals with PAP, which includes that derived from insects, except for PAP derived from non-ruminants (including insects) that may be fed to aquaculture species. However, the requirement for PAP to be processed in a registered slaughterhouse means that this is not technically possible. The substrate on which the insects are reared is also a consideration. Insects produced for feed would be classed as “farmed animals” for which only category 3 material is allowed as a feed source. Manure is a category 2 material and therefore is not permitted to be fed to “farmed animals”. Other issues such as the welfare of insects raised for use in animal feed also need to be considered.

Consumer acceptance
Consumer acceptance is key to the successful adoption of insects as a source of protein for livestock. Key stakeholder groups (eg. producers of feed and feed ingredients, consumers, livestock farmers, government) need to engage with the concept and to be kept informed of developments in the scientific and technological developments in this area. This will enable them to make informed decisions on the potential use of insect protein in animal feed and will encourage the adoption of sustainable protein production technologies in order to alleviate the reliance of the feed industry on plant/fish derived proteins.

Objectives of the PROteINSECT project
The principal objective of the PROteINSECT project is to facilitate the use of insects as an alternative source of protein for animal feed. To achieve this objective the project has focussed on five key areas:
1. The co-ordinated development and optimisation of fly production methods for animal feed production in EU and International Cooperation Partner Countries (ICPC)
2. Determination of safety and quality criteria for substrate and insect protein products
3. The evaluation of crude and refined insect protein extracts in fish and monogastric (poultry and pig) animal feeding trials.
4. The determination of optimal design(s) of insect-based animal feed production systems for both EU and ICPC countries through comprehensive ILCD assessments
5. Building a Pro-insect platform in Europe to encourage a holistic adoption of sustainable protein production technologies in order to reduce the reliance of the feed industry on plant/fish derived proteins in the short term, and to encourage the acceptance of insect protein as a direct component of human food in the longer term.

Project Results:
Work Package 1: Sustainable insect production

1.1. Introduction
The objectives of Work Package (WP) 1 were as follows:
• Develop and maintain a database on fly rearing methods and fly production for animal feed
• Define the characteristics of the existing fly rearing systems presently used by the partners
• Set up new fly rearing systems at partner organisations using other partners expertise
• Optimize/improve existing and newly set up rearing systems
• Provide fly larvae for WP2 (protein processing technologies and animal feeding trials) and WP3 (quality and safety) and data for WP4 (life cycle assessment) that will contribute to the design of efficient insect rearing stations to be integrated into local farms in Europe, China and Sub-Saharan Africa

The main outcome from this work package was the development of fly larvae production systems in Europe, China and West Africa and recommendations for improvements that were guided by the results of the socio-economic and life-cycle assessments in Work Package (WP) 4. Additional outcomes were the development of a database of scientific information on the use of flies for animal feed and the production of fly larvae for the feeding trials, analyses of quality and safety, and the testing of protein processing technologies (WP 2 and WP 3).

1.2. Development and optimization of fly larvae production systems

The following fly larvae production units were developed within the framework of the project:
a. Musca domestica (house fly) system in UK (Grantbait and Fera)
b. Hermetia illucens (black soldier fly) system in Ghana (FfA and Stirling)
c. M. domestica system in Ghana (FfA and CABI)
d. M. domestica system in Mali (IER and CABI)
e. M. domestica system for individual farmers in China (HZAU)
f. M. domestica system for semi-commercial scale production in China (GEI)

A brief description of each of these systems is provided below, with a summary of the major research carried out during the project. More detailed descriptions and recommendations are found in in the annexes of Deliverable 4.5 for the systems b, d, e and f, which can already be recommended for implementation in Africa and Asia. System c (M. domestica in Ghana) cannot yet be recommended because it still shows disadvantages compared to systems b and d that are applicable under similar climatic and economic conditions. A M. domestica system in Europe would need substantial improvements, in particular in automatization, before becoming economically viable and environmentally sustainable. This was neither possible nor intended in the framework of PROteINSECT. Therefore, only general recommendations are provided in D.4.5.

1.2.1. Musca domestica system in UK (Grantbait and Fera)
A new system for the rearing of M. domestica on poultry manure was established at Grantbait at the beginning of the project. This system was based on the expertise of Grantbait in the rearing of Calliphora vomitoria on abattoir waste for fishing bait. The production method has been refined and adapted in conjunction with research carried out at Fera. Initial studies to develop a system for M. domestica used pig manure as a substrate for rearing of the larvae. As pig manure gave highly variable yields of M. domestica larvae a system using poultry manure as the main substrate was subsequently developed. A number of improvements have been tested on-site for all stages of the system throughout the project.
In addition, research in Fera focused specifically on:
• Testing the suitability of different pig and poultry manures, as well as the effect of different manure qualities on yields.
• Assessing the potential for mass rearing of two blow-fly species, Calliphora vomitoria and Lucilia sericata
• Assessing an alternative to fly species, i.e. the lesser mealworm beetle Alphitobius diaperinus.

1.2.2. Hermetia illucens system in Ghana (FfA and Stirling)
The University of Stirling and Fish for Africa worked together to develop a system to mass produce H. illucens (Black Soldier fly - BSF) in the rural conditions of coastal Ghana, in the suburbs of Accra. This was the only BSF system developed in PROteINSECT, with the exception of rearing trials at IER (Mali) that collapsed after a few generations. The BSF system in Ghana was developed from scratch, but based on previous experience of the University of Stirling in Indonesia. It also encountered difficulties in initial establishment (e.g. capture and domestication of H. illucens, breeding in captivity), but determination and a flexible approach facilitated the successful development of a BSF rearing system. It was developed for small scale production for the local market or to feed a fish or poultry production unit. The system was developed in the conditions of the “tropical wet and dry” climate of the Gulf of Guinea. As such, it is not clear whether a similar BSF production system would be effective in dryer areas.
The BSF rearing system is organised around several production steps and separated into two operating units: the insectarium where adults are reared for egg production and the larvarium, where larvae are produced. Various substrates were tested and used during the project, but the main ones were mixtures of chicken manure, brewery waste and fish feed waste.
Research on the system included:
• Identification and testing of rearing substrates in Ghana
• Improvements to the insectarium and larvarium buildings
• Improvements to the oviposition support systems
• Testing of various rearing trays
• Improvements in techniques for harvesting larvae
• Control of a pupal parasitoid, the main natural enemy of BSF in Ghana
• Up-scaling the production system

1.2.3. Musca domestica system in Ghana (FfA and CABI)
While the priority of FfA and Stirling in Ghana was the establishment of a BSF production unit (see 1.2.2. above) for the provision of larvae for fish trials, a production system for house fly larvae was also developed at the same site, based on the expertise of Chinese and Malian partners and supported by CABI. In contrast to the production system in Mali that is based on natural oviposition, the system developed in Ghana was based on adult rearing and inoculation of eggs on to the substrate, as used in China and the UK. The main substrates used in this system were chicken manure, brewery waste and mixtures of these. The production system was set up in late 2013 and improved in 2014, in part as an MSc thesis. The system was evaluated in WP4 using life cycle analysis and although the performance was acceptable, it was considered that under the basic conditions of rural Africa, M. domestica was more easily produced using the natural oviposition system developed in Mali (see 1.2.4 below). Maintaining a sizable and constant egg production was nearly as complicated as for BSF, but the latter species had the advantages of higher yields and a larger spectrum of possible substrates for development. Research on the system included the following topics:
• Effect of adult population on egg production
• Effect of adult diet on egg production
• Effect of oviposition site on egg production
• Testing of various attractants for oviposition
• Development of a pupae production system
• Larval production: search for the best inoculum size and rearing substrate
• Larvae harvesting methods
• Larvae drying methods

1.2.4. Musca domestica system in Mali (IER and CABI)
The M. domestica larval production system in Bamako, Mali, was the only system that existed at the beginning of the project. It was developed in the 1990s by Mr N’Golopé Koné (IER), but had to be discontinued for various reasons before being restarted with the inception of PROteINSECT at the IER campus in Bamako, in collaboration with CABI. The system was unique in PROteINSECT because fly larvae were obtained simply by exposing variable quantities of suitable substrates, in rearing beds, to naturally occurring house flies. The main substrates tested in the system were chicken manure alone or with blood, and sheep manure with blood or fish offal.
This system is suitable wherever house flies occur in abundance the whole year round. It is thus most suited for tropical regions. In PROteINSECT, the system was developed in a “semi-arid” climate with a rainy season of 5 months. The system was primarily developed for use for a micro-enterprise that would sell dried house fly larvae as an animal feed complement, for a poultry producer that would use the manure produced on farm to produce its own protein feed, or for a fish farmer to feed its own fish. It can also be potentially down-scaled to a smallholder farm system. Up-scaling is more difficult because of the need for a large surface area at ground level.
Research on the system included the following topics:
• Testing and improvement of substrates, including investigation of the most efficient mixtures
• Temporal variations of yields
• Influence of the rearing bed position and neighbouring substrate exposures
• Replacing the ground rearing beds by trays on shelves

We also developed an adult rearing and egg production system, similar to the M. domestica system in Ghana. However, it was found that the system based on adult rearing performed poorly in the very basic facilities available in Mali, whereas, in contrast, the system based on natural oviposition became increasingly efficient. Therefore, in the last year of the project, it was decided to focus on, and promote exclusively the natural oviposition system.

1.2.5. Musca domestica system for individual farmers in China (HZAU)
An experimental M. domestica larvae production system was set up by HZAU to represent an average single family farm in the region of Liling (Hunan Province). The system was designed such that the volume of maggot production was sufficient to feed up to 1000 hens and chickens. In this system, the house flies were reared in cages to produce eggs and small larvae were then used to inoculate pig manure. The system has already been promoted in the region and many variants have been adopted by farmers.

Research on the M. domestica system at HZAU covered various aspects of the following topics:
• Enhancement of egg production and egg survival
• Optimisation of larval production in the rearing beds
• Adult production for rearing

A very similar system was also used to rear the blow fly Chrysomya megacephala at the same farm, with positive results, especially in the summer, when the temperature is too high for M. domestica. Chrysomya megacephala has been a major target of the research carried out at HZAU during PROteINSECT since their expertise and knowledge in the biology and ecology of this fly was very limited compared to that for M. domestica. Research on C. megacephala included various aspects such as:

• Body weight and length of larvae at different day-ages
• Effect of constant temperatures on larval and pupal development
• Effect of thermoperiod temperatures on larval and pupal development
• Effect of the duration of thermoperiod temperatures on development
• Ultrastructure of sensilla on larvae and adults of Chrysomya megacephala

1.2.6. Industrial Musca domestica system in China (GEI)
A new house fly larval production system was set up by GEI at the beginning of the project, attached to a large poultry production company in Huizhou, Guangdong Province. The goal was to use the chicken manure produced by the farm to rear larvae that would complement the diet of the poultry produced by the company. Alternatively, larvae could also be sold to other animal production units. This system followed the same principles as the HZAU system, with a separate adult rearing for egg production, but it was characterised by a higher level of automatization that allowed a higher production level. It also included the fermentation of substrates using microbes and the automatic extraction of larvae from the substrate by inducing anoxia.

Research, partly carried out on-site and partly in the laboratories of GEI, covered the following topics:

• Influence of different yeasts on the development of larvae as additives to wheat bran
• Influence of adult rearing density on egg laying
• Effect of hen’s egg as an additive to the housefly feed on the amount of fly eggs laid
• Determination of the optimal housefly egg quantity to be added to the chicken manure
• Determination of the optimal thickness of chicken manure for larvae rearing
• Influence of silicon as an additive to chicken manure on the development of larvae
• Influence of microorganisms as an additive to chicken manure on the development of larvae
• Influence of different Bacillus concentrations for the treatment of chicken manure on the development of larvae
• Influence of the timing of microbe treatment in chicken manure on the development of larvae

Research was also performed in the laboratory of CABI China in Beijing to identify chemical compounds as oviposition attractants for M. domestica. Volatiles from fermented wheat bran were identified as strong attractants that could be used to enhance oviposition and, thus, egg production.

1.4. Production of fly larvae for protein processing and animal feeding trials
All fly rearing systems have been used to produce larvae for nutritional and safety analyses (WP2 and WP3) and for the animal feeding trials (WP2). More than 2 tonnes of dried larvae were produced during the project for the feeding trials.

1.5. Database on fly rearing methods and fly production for animal feed
A database on flies and their use in animal feed was built using the free access software Jabref. It started in the first month of the project and has been regularly updated. The latest version contains nearly 400 references, mainly journal articles, but also unpublished theses, reports and patents. Most include an abstract in the database. An important effort was made to translate abstracts of key publications, in particular in the Chinese literature. However, it must be noted that the number of papers on fly rearing for feed, particularly BSF is expanding rapidly. Thus, this database will quickly become obsolete and it was decided not to maintain this database after the end of the project.

Work Package 2: Protein processing technologies and animal feeding trials

2.1 Introduction
The objectives of Work Package 2 were as follows:
• To review the most suitable processing technologies for protein extraction/conversion
• To test different processing technologies for protein extraction/conversion at laboratory scale
• Scale-up (to pilot scale) of retained processing technologies
• Comparison of crude and processed insect protein extracts with conventional feed stuff in aquaculture and livestock (poultry and pigs) feeding trials
• Defining leading prototype processing technologies suitable for insect biomass protein fractions
The major outcomes of this work package were the development of a processing method for the extraction of insect protein and the completion of animal feeding trials for fish, poultry and pigs in different geographical locations, which examined zootechnical performance and animal health.

2.2 Processing technologies for protein extraction/conversion
During the first 18 months of the PROteINSECT project, a strengths, weaknesses, opportunities and threats (SWOT) analysis was performed on potential protein extraction methodologies for M. domestica (house fly) and H. illucens (black soldier fly) and for their residual substrates (biomass), for their inclusion as a protein source in animal feed applications. The purpose of the SWOT analysis was to define at least two methodologies for protein extraction and processing within each of the following categories: physical, chemical and biotechnological. Based on the outcomes, general approaches for protein extraction and processing were derived, but these were always based on existing literature, patents and practices in the participating countries.
Identified physical, chemical and biotechnological approaches were scaled-up and evaluated in order to convert insect larval protein into different protein concepts suitable for animal nutrition. As a result, we have provided three major technologies at the pilot-scale. Table 1 summarizes the technologies, indicating the maximum protein yield obtained at the pilot scale as well as the maximum relative protein increase reached for each of the technologies.
Although all three insect larvae protein processing technologies are suitable for scale-up, one major disadvantage is the presence of an enriched (polymeric) fibre fraction, representing chitin, from those processes. Fibre contains mainly chitin, which is not as yet proven to be safe for animal nutrition. Although the presence of chitin in insect larvae protein can pose a potential risk for processing into animal feed applications, it is anticipated that the insect protein is likely to be used at relatively low levels in diets and that this may actually be beneficial for animals as it promotes an immune response.
In addition, the application of chitin products has received considerable attention during the past few years in different projects. Specific chitins and their derivatives – such as chitosan - have great economic value because of their application in food, cosmetics, pharmaceuticals, textile industries and agriculture.
From the pilot-scale processing technologies evaluated, solvent extraction yielded samples with the highest protein content. As such, the consortium decided to evaluate crude ground larvae and solvent extracted proteins as animal feed ingredients in pig, poultry and fish diets. These two sources of insect protein are considered to be the most close to market assuming a positive evaluation in animal feeding trials.

2.3 Animal feeding trials
Currently, aquaculture, pig and poultry feeds remain reliant upon fishmeal or soybean meal as a major source of high quality protein. Insects may offer a valuable, and more ecological alternative. Larvae of the housefly (M. domestica) and black soldier fly (H. illucens) were shown during through the PROteINSECT project to be of high nutritional value. The effect of the inclusion of crude insect meal, live larvae and extracted insect proteins was assessed in aquaculture, pig and poultry species in both EU and ICPC countries, taking into consideration both zootechnical performance and health issues.
Aquaculture. Crude housefly (M. domestica) larval meal was compared with enriched (de-fatted) insect protein extracts in an Atlantic Salmon (Salmo salar) parr trial conducted in the UK. Six iso-nitrogenous and iso-energetic experimental diets were formulated such that the main protein source fishmeal (FM) was gradually replaced with insect meal. Growth performance (weight gain, specific growth rate and FCR) of 3600 tank reared fish fed for 8 weeks on diets containing either crude meal at 25 and 50% FM replacement levels or de-fatted meal at 50% FM replacement was comparable to the control (commercial diet) treatment. However, diets with a higher inclusion of crude meal (75% and 100% FM replacement) appeared to have a negative impact on growth. Results also indicated that the lipid composition of crude meal may induce a reduction in dietary lipid digestibility, although this was improved when 50% of the FM was substituted with de-fatted meal. Overall the results indicated that maggot meal could successfully replace up to 50% FM and that the use of protein enriched fractions may be advantageous for lipid digestibility.
No significant difference in feed intake and apparent digestibility coefficient were detected when insect meal prepared from larvae fed on wheat bran were fed to Nile Tilapia in a trial in China. Growth performance and nutrient utilization were similar to the control treatment at up to a 75% replacement level of fishmeal by insect meal, whereas significantly lower survival rate and zootechnical performance was observed when fishmeal was completely replaced. A Tilapia trial in Ghana validated these results where a 75% substitution of fishmeal with insect meal also led to a significantly increased survival rate.
In Mali, feeding dry larvae to Clarias resulted in fish weighing far more than the market average with a largely reduced feed conversion ratio, especially when 70% of the protein source was insect larvae. In line with the trial on Tilapia in China, using 100% insect larvae was suboptimal, although in this trial it gave rise to better performance compared to the control. It is likely that the combination of insect meal with conventional protein sources provides a more balanced diet for the animals, resulting in better protein retention efficiency and a more balanced amino acid profile.
Overall, there are some indications, proven by trials on different continents, that fishmeal as a protein source can be partially replaced by insect meal, without harming animal performances. It might even lead to some synergistic effects resulting in enhanced zootechnical performance.
Poultry. In Mali, a trial was performed feeding dried insect larvae to Cobb broilers. Results showed that animals receiving a mixed protein source performed significantly better than those on 100% fish or insect meal. These data are probably due to differences in crude protein and energy content which was highest in the “mixture” group, making it difficult to draw final conclusions.
The results from the broiler trial in Mali were not validated in a trial in China, where housefly larvae raised on pig manure were fed to Shankengfengxiang broiler chickens. In this trial, no significant effects on zootechnical performance were observed. A broiler trial was also conducted in Europe and no significant differences were seen between the control group and broilers receiving 2.0% crude insect meal or 1.25% extracted insect protein as a partial replacement for soy meal. In the European trial, it was also demonstrated that the gastro-intestinal animal health was not negatively impacted by feeding insects to the animals.
In China, feeding 5% (w/w) live insect larvae reared on chicken manure to Huxu broiler breeder chickens improved their production performance (i.e. laying rate and feed-to-egg ratio) and egg characteristics, such as egg fertility and hatchability. Also it was found that feeding higher amounts of insect larvae (10% and 15%) resulted in an increased laying rate, egg fertility and hatchability. These results were validated in the trial in Mali. Layer chickens fed on insect larvae in the trial in Mali also had a better FCR compared to the control chicks. In contrast, a slightly lower performance in egg production was seen in layers fed on insects. The exact composition of the commercial feed was not available and, therefore, we cannot conclude whether this was due to the insect larvae or to other compounds. Overall, it seems that insect larvae-based diets can replace traditional poultry diets based on other protein sources, with similar performances. Further trials, with more precise data, are however needed to further investigate this hypothesis.
Overall, it seems that partially replacing standard protein sources by insect protein, will at least lead to the same zootechnical performance in poultry as that observed with commercial diets.
Pigs. In a trial conducted in Belgium, no significant differences were observed in body weight, daily gain, feed intake and feed conversion ratio, indicating that including 2% crude insect meal or 1.25% extracted insect protein in the diet of the piglets (in exchange for fish meal) had no negative effect on animal performance. The gastro-intestinal health of the animals fed insects was not negatively affected. This was further confirmed by the gastro-intestinal simulation trials, where including insect meal or insect extract in the piglet feed matrix reduced the growth of the Escherichia coli (negative microflora), whereas the Lactobacillus amylovorus / Lactococcus lactis (positive microflora) remained at the same level.
In conclusion, this trial showed no negative effects of including crude insect meal or extracted insect protein in the diet of weaned piglets on zootechnical performance and animal health. In addition, some beneficial effects on digestibility and gastro-intestinal health were observed.
Conclusion. Taking together the results from all animal species (aquaculture, pigs and poultry) across all regions worldwide (Europe, Africa and Asia), it seems that partial replacement of fish- or soybean meal by live larvae, crude insect meal or defatted insect protein does not lead to a negatively impacted zootechnical performance or animal health. More (and larger) trials are needed to confirm this.

Work Package 3: Quality and Safety of Insect Protein

3.1 Introduction
The objectives of Work Package 3 were as follows:
• Undertake risk and opportunity assessments in relation to the increased use of insect protein in the human food chain
• Provide nutritional data to guide the optimisation processes undertaken in WP1 and 2
• Establish the safety of insects for use in feed and food. Assessing insect derived products according to EU regulations specifying maximum permissible levels of contaminants (e.g. dioxins, PCBs, heavy metals etc.) microbiological safety and allergenicity
• Determine the safety and organoleptic properties of meat and fish derived from insect based diets
• Assess potential to exploit other high value by-products from insects e.g. extracted chitin, fats, residual waste e.g. for fertiliser
The major outcomes of this work package were the identification of the potential chemical and biological risks of protein derived from insects and the analysis of four insect species to obtain data on the nutritional profile and safety of insect larvae. In addition the meat and fish derived from the animal feeding trials were analysed for chemical and biological safety and taint. The data obtained within this work package has been used by EFSA in developing the scientific opinion on the safety of insects as food and feed.
3.1 Nutritional profile and safety of insect larvae
Insect larvae were analysed for a range of chemical/microbiological hazards, allergens and for nutritional composition. Safety data were obtained from the larvae of the four fly species that have perhaps the greatest economic relevance in relation to their use as animal feed being; house fly (M. domestica), blue bottle (Calliphora vomitoria), blow fly (Chrysomya spp.) and black solder fly (H. illucens).
Diverse rearing methods were used to produce larvae fed on a range of waste substrates and in four geographically dispersed locations being; UK, China, Mali and Ghana.

3.1.1 Chemical safety
Chemical safety data were collected by a fully accredited laboratory in the UK. The levels of the main subclasses of chemical contaminants considered for animal feed were determined being; veterinary medicines, pesticides, heavy metals, dioxins and polychlorinated biphenyls, polyaromatic hydrocarbons and mycotoxins.
The larvae analysed generally possessed levels of chemical contaminants which were below recommended maximum concentrations suggested by bodies such as; the European Commission, the World Health Organisation and Codex. However, the toxic heavy metal cadmium was found to be of concern in three of the M. domestica samples analysed.
3.1.2 Metal accumulation studies
Concentrations of lead (Pb), cadmium (Cd), arsenic (As), chromium (Cr), copper (Cu), mercury (Hg) and nickel (Ni) were measured in pig and chicken litter substrate pre and post Musca domestica larvae production. The larvae were raised in the substrate for 9 days and were also analysed. Only cadmium from pig litter showed any potential bioaccumulation with a mean fold increase of 2.7. No other metal bioaccumulation was observed.

3.1.3 Microbiological safety
The absence of viable Salmonella, Campylobacter and Listeria monocytogenes was confirmed for all samples tested. The samples were those listed in 3.2.1 with rearing and processing as described in Charlton et al. 2015. Two of the M. domestica samples were however found to contain Enterobacteriaceae with levels of 6900 and 15,100 colony forming units (cfu) Enterobacteriaceae /g.

3.1.4 Nutritional analysis
Analytical data have been generated to show that the nutritional profiles of fly larvae are highly compatible with use in feed, having both high total protein content and levels of the key amino acids that are comparable with those found in fishmeal. Using a linear programming software tool for broiler chicken and piglet feed compilation, M. domestica meal is estimated to be worth a minimum of 700 € / tonne. The value of the fat profile for M. domestica is in many ways similar to that of palm oil containing comparable levels of palmitic acid.
Concentrations of protein, amino acids, fibre, fat, fatty acids and trace mineral profiles all show differences within the M. domestica samples and between insect species. This indicates that the country of origin and ultimately the larval production method (i.e. rearing substrates used) has a key impact on final nutritional content.

3.1.5 Allergens
Analysis of proteomics data obtained for larvae of the four different species of fly were assessed to determine the homology of tropomyosin, arginine kinase and myosin light chain with the crustacean orthologous proteins and other known allergenic proteins. The results indicate that the three proteins share homology with known allergens and therefore it is likely that they are also allergens in insects (Romero et al, 2016). It is therefore essential that allergenicity is considered in the design of insect rearing facilities and the need for personal protective equipment (PPE) is emphasised.

3.2 Insect reared meat
Salmon fingerlings, broiler chickens and piglets were used for animal feeding trials which provided meat samples for analysis. Details of these trials including inclusion rates of insect meal are described in WP2.

3.2.1 Samples
Fera received a total of 18 salmon fingerling samples ranging between 80 – 140 g per sample, 9 chicken breast samples ranging between 230 – 390 g per sample and 9 pork chop samples ranging between 160 – 250 g.

3.2.2 Chemical safety
1. Veterinary medicines: No residues were found in any of the fish, chicken or pork samples.
2. Pesticides: No residues were detected in all fish and chicken samples. The chlorinated disinfectants benzyldimethyldodecylammonium chloride (BAC12) and N-Decyl-N,N-dimethyl-1-decanaminium chloride (DDAC) were detected in a number of the pork samples. The most recent legislation (EU) No 1119/2014 of 16 October 2014 amending Annex III to Regulation (EC) No 396/2005 details maximum residue levels in all foodstuffs listed as 0.1 mg/kg for both DDAC and BAC12. It is unlikely that insect feed replacement has contributed to the presence of these residues as there are similar or greater concentrations in the controls. These persistent chemicals are likely derived from disinfectant practices in the pig farm.
3. Metals: Heavy metal concentrations in all samples did not exceed regulatory limits (where limits exist). The relevant regulations are summarised below:
i) EU Commission Regulation 2015/1005 of 25 June 2015 indicates maximum levels of lead in fish and meat should not exceed 0.3 mg/kg and 0.1 mg/kg respectively.
ii) EU Commission regulation 488/2014 indicates maximum levels of cadmium in fish and meat should both not exceed 0.05 mg/kg.
iii) EU Commission regulation 1881/2006 indicates maximum levels of mercury in fish should not exceed 0.5 mg/kg.
4. Dioxins, polychlorinated biphenyls (PCBs) and polyaromatic hydrocarbons (PAHs): Dioxins, polychlorinated biphenyls (PCBs) and polyaromatic hydrocarbons (PAHs) concentrations in all samples do not exceed regulatory limits or action limits (where limits exist) as summarised below:
i) EU Commission Regulation 2011/1259 indicates maximum levels for the sum of ICES 6 as 125 µg/kg in wild fresh water fish (or 75 µg/kg in farmed fish) and 40 µg/kg in both poultry and pig meat (equivalent concentration in fat).
ii) EU Commission Regulation 2011/1259 indicates maximum levels (WHO-TEQ) for the sum of dioxins as 3.5 ng/kg in all fish, 1.75 ng/kg in poultry meat and 1 ng/kg in pig meat (equivalent concentrations in fat for meat products).
iii) EU Commission Regulation 2013/711 indicates action levels (WHO-TEQ) for the sum of dioxins as 1.5 ng/kg in all farmed fish, 1.25 ng/kg in poultry meat and 0.75 ng/kg in pig meat (equivalent concentrations in fat for meat products).
iv) EU Commission Regulation 2011/1259 indicates maximum levels (WHO-TEQ) for the sum of dioxins, non ortho and ortho PCBs as 6.5 ng/kg in wild and farmed fresh water fish, 3 ng/kg in poultry meat and 1.25 ng/kg in pig meat (equivalent concentrations in fat for meat products).
v) EU Commission Regulation 2013/711 indicates maximum levels (WHO-TEQ) for the sum of dioxin like PCBs as 2.5 ng/kg farmed fish, 0.75 ng/kg in poultry meat and 0.5 ng/kg in pig meat (equivalent concentrations in fat for meat products).
There are generally lower concentrations of dioxin, PCB and PAH in the insect fed salmon samples as fishmeal is being replaced with insect meal, providing tentative evidence that insect meal is safer than fishmeal in regards to these compounds.
5. Mycotoxins: No mycotoxin residues were found in any of the fish, chicken or pork samples.

3.2.3 Microbiological safety
A screen for Salmonella spp, Escherichia coli 0157 and Listeria monocytogenes was applied to all samples. Enumeration for total Escherichia coli was also undertaken for all samples. Methodology for all analyses are described in ISO 6579:2002 +A1:2007 (Salmonella), BS EN ISO 16654:2001 / BS EN ISO 16649-3:2015 (E.coli) and BS EN ISO 11290-1:1996 / BS 5763-18:1997 (Listeria).
Salmonella spp, Escherichia coli 0157 and Listeria monocytogenes were not found in any of the fish, chicken or pork samples. A very small amount of total E. coli was detected in two of the salmon samples but at less than 100 cfu/g. This is unlikely to be a problem as the meat industry guide published by the UK FSA (August 2015) states that E. coli levels should be < 500 cfu/g.

3.2.4 Nutritional analysis
Total protein, amino acid profiles, total fat, fatty acid profiles and nutritional trace elements analyses were undertaken on all salmon, chicken breast and pork chop samples.
All samples tested from animals fed an insect fortified diet had nutritional parameters similar to the controls. It is possible but not yet conclusive due to the small number of samples analysed (n=3 for each group) that chicken breastss have a greater fat content in animals fed non “de-fatted” insect meal replacement at 2% inclusion rate with the mean concentrations differing with a p value of 0.01.

3.2.5 Allergenicity
Potential symptoms of allergic response were monitored in piglets and chickens as part of the animal feeding trials in WP2. IgE concentrations were also measured in blood taken from all chickens and piglets, as food allergy reactions are mediated by IgE. IgE concentrations were very low being < 0.01 µg/ml for all piglet blood samples. IgE concentrations for chicken blood did not increase in samples from animals fed an insect fortified diet.

3.2.6 Taint analysis
All fish, poultry and pork samples were assessed for taints, including hexane derivatives which may be present due to the insect larvae fat wash procedure (for selected samples derived from a “fat removed” insect fortified feed). Samples were screened for 3,4 Hexanedione, 2,5 Hexanedione, 1- Hexanol, Hexanal and 2,3 Hexanedione. The samples were also analysed to detect any compounds present in the insect fed animals that were not present in the controls for each matrix type. Any unknowns found were searched against the MS library NIST which contains approximately 40,000 compounds with a particular emphasis on highlighting potential taints. For all 3 matrix types no increased amounts of hexane derivatives were found in “fat washed” insect fed samples. From the unknown analysis, no taints were found in any of the samples fed an insect fortified diet for the 3 matrix types.

3.3 NPK analysis
Nitrogen (N), phosphorous (P) and potassium (K) concentrations were measured in chicken and pig litter substrate pre and post a 9 day Musca domestica larval digestion. Three replicate litter trays of 2.4 Kg for each substrate type pre and post larvae were prepared and analysed. N concentration decreased significantly and P remained the same in both post larvae substrates. K decreased in the pig litter substrate only.
When calculating NPK on a dry matter basis (i.e. taking a dried 100g portion of the pre and post larvae substrate) P and K show an increased concentration in chicken litter substrate (K only in the pig litter) with N concentration decreasing in both cases. Therefore there is evidence to support the use of insect frass as an enriched source of P and K fertiliser when compared to chicken manure.
Analysis of various insect feeding substrates used in the project has largely highlighted that fly larvae will mainly deplete the N content of their feeding substrate whilst P and K remain largely unchanged.

WP4: Environmental, social and economic life cycle assessment

4.1 Introduction
The objectives of WP4 were the following:
• Ex-ante life cycle sustainability assessment of insect based animal feeds in different geographical regions
• Comparison of the sustainability performance of insect based feeds against conventional protein feeds
• Development of optimization routes towards more sustainable insect production systems

The analysis was performed from a life cycle perspective, which in accordance to the ISO standards, includes four phases: 1) goal and scope definition; 2) inventory analysis; 3) impact assessment; and 4) interpretation. Within the goal and scope phase, the system boundaries were iteratively adjusted based on data availability. The studied system comprises the extraction of raw materials, manufacturing of inputs including rearing substrates, the insect rearing and residue substrate separation, as well as the processing of the final co-products, i.e. from “cradle to gate”. We considered five geographical regions for the production of insect based feeds (Figure 1).
After the examination of production designs either on-site or as a desktop review, the life cycle inventory phase was executed following a generic approach. A generic approach, in contrast to a specific approach, consists of a function-based life cycle inventory that represents the biophysical flows of a larger group of similar production systems, irrespective of the specific features of the original system. In order to retain a maximum of geographical distinction and system characteristics of the original set ups of the PROteINSECT consortium, all generic models have been extrapolated from averages of available inventory data and were further parameterized with site-specific economic and environmental data. Although this generic approach is accompanied with a loss of detail and representativeness, it allows for a more flexible assessment of changes, a much-valued feature when dealing with systems under development.

All biophysical and economic flows were related to 1 kg of insect product (functional unit). To parametrize the modelled life cycle inventories with economic metrics, market prices of inputs and outputs were surveyed either through desktop reviews or in interviews on site with the coordinators of the respective systems. Results were communicated within the PROteINSECT consortium and an iterative process allowed for improvements on the life cycle inventory data and, consequently, on the downstream analysis.

The life cycle inventory analysis was succeeded by the socio-economic & environmental impact assessment phase. For the socio-economic assessment, the potential economic performance of the focal system design was estimated as a total cost assessment, analysing the cost break-down structure of the life cycle from the perspective of the economic actor producing the functional unit, i.e. insect farmer or feed producer (Swarr et al. 2011). Following the basic accounting principle in microeconomics, i.e. revenues must balance with expenditures; the total cost provides an indicator for a hypothetical market price of the focal system design.

In order to derive indicators of relevance for social sustainability from economic inventory data, a set of conditional assumptions were made: (i) production systems perform under conditions of general economic equilibrium; (ii) all economic actors tend to maximize their private welfare; (iii) workers receive salaries that align with skill level; (iv) money spent on local markets, particularly payments for human labour, generates welfare benefits in respective communities. Built on such assumptions, analysis of societal performance followed different analytical steps. First, production expenditures were analysed by their spatial distribution, i.e. local, national, or international markets (import). To estimate welfare gains in a local community expenditures were further grouped to benefiting stakeholder groups, i.e. ‘labourer’, ‘processors, farmers & service providers’ and ‘retailer’, whereby expenditures are ranked in descending order to the conjectured societal benefits. To assess the societal strain of work input, we related information on the economic significance of employed labour with information on the skill-specificity of the needed labour work inputs, i.e. composition of workforce with reference to share of untrained and trained labourers. This enable an estimate of the likeliness of worker exploitation in regions with poorly defined employment practices to be obtained, i.e. cost reduction measures that could compromise the quality of working conditions or payed wages.

The environmental impact assessment was conducted by applying the ReCiPe method (which conforms with designated ISO standards) on background data from ecoinvent® (Goedkoop et al. 2013). Whenever confronted with multifunctional processes and the output of co-products, the environmental loads have been partitioned by means of economic allocation, considering applicable market prices as a measure to capture the complex and varying attributes of products (Guinée et al. 2004, Ardente and Cellura 2012, Pawelzik et al. 2013). Using the example of a residue substrate as a co-product with economic value, then the environmental burden of the production system should be partitioned between the insect based feed and the residue substrate considering the respective market values.

4.2 Optimisation pathways for insect rearing systems
Environmental and economic assessments were presented to the PROteINSECT consortium allowing optimization pathways for each system were developed (see Tables 2 and 3). The definition of the optimization followed a stepwise approach, with a first set of measures (marked as #2) representing an improved set up for larval production, and a second set (marked as #3) included #2 measures, but excluded drying by considering the use of fresh larvae as a feedstuff.

The agreed system improvements were applied to the generic models and the socio-economic and environmental indicators were re-assessed. The performance of baseline and optimized scenarios was further compared with a suitable reference system. Fishmeal was chosen as the reference system, given its nutritional content and position in the trophic network, which are both similar to the insect based feed. For all systems, including fishmeal, the developed indicators for each of sustainability’s three dimensions were normalized by calculating their proportional distance to the best score. Hence, all performance indicators ranged between 0% and 100%, with this last score indicating the best performing system. By computing the non-weighted average across all considered indicators of one sustainability dimension, the results were further aggregated into a single score for the economic, societal and environmental dimension of sustainability (see Table 4 for results). As shown Table 4, the fishmeal system had the best economic performance (100%), but the FfA and IER systems at 93% and 88% support the hypothesis that insect based feeds have commercial potential. As most of the production designs operate in a simple setup, the required labour inputs, especially of trained workers, were in some cases critical to the operational benefits. Though cost-wise a disadvantage, these process characteristics acted as an advantage to their social sustainability performance. This is true due to the high skill-specificity (share of trained workers), which is considered a positive social trait.

Due to a relatively high share of system expenditures allocated to local stakeholders, particularly labourers, processors and farmers, the simple and yet laborious processes, such as in the IER_CM, IER_SMBL and FfA_BSF system, scored highest in skill-specificity and local welfare benefits. The fishmeal system was ranked best for the indicator of economic significance of labour, but the average over all three indicators neutralizes this superior result.

For environmental performance, the different feed production systems show variability in ranking amongst the different impact categories. Fishmeal showed the lowest impact (100%) in climate change potential, terrestrial acidification, human toxicity, particulate matter formation, terrestrial ecotoxicity and marine ecotoxicity. In contrast, representatives of the insect based feed production systems were evaluated best in the impact categories of freshwater eutrophication (IER_CM#3), marine eutrophication (GEI#3), photochemical oxidant formation (GEI#3), freshwater ecotoxicity (GEI#3), natural land transformation (GEI#3) and fossil depletion (GEI#3). As the final score of the environmental performance was calculated as a non-weighted average of the impact categories, these particular variations between environmental categories are lost. The GEI#3 system showed more balanced results across all environmental impact categories, which resulted in a high average of 83% and thus best score in the environmental dimension. With an average of 72%, the fishmeal system ranked second, shortly followed by the HZAU#3 (69%) and IER_CM#3 design (62%). The outstanding results of the GEI#3 system are particularly interesting, as the corresponding system design represents the most sophisticated and up-scaled production design of the six evaluated insect based feed systems. These findings clearly support our projections that further up-scaling measures will be mirrored in substantial impact reductions. Conditional to the geographical conditions on site, such economy of scale effects are likely to arrive at a point where the environmental performances of insect based feed becomes superior to fishmeal.

A holistic analysis of the three sustainability dimensions, relied upon the concepts of weak and strong sustainability and their most common representations. For weak sustainability, two commonly accepted representations are considered, where either the three sustainability dimensions have the same weight (Triple Bottom Line approach) or the economic is given higher relevance, followed by society and environment (commonly called the “Mickey Mouse” scheme). For the strong sustainability concept, the environment is the most relevant and limiting factor to socio-economic development (Nested approach). Here fishmeal performance was set as the default reference in all three dimensions. Correspondingly, the sustainability of insect based feeds was gauged by computing the proportional distance to the performance scores estimated for the fishmeal production. Without any agreed consensus in the literature the proposed methodology departs from the assumption that any improvements to current production and consumption patterns would be a step in the right direction. The proposed assessment was applied to both weak and strong sustainability concepts and is presented in Figure 2. For a given insect based feed system, the distance to the borders of the circle in Figure 2 reflects its performance in comparison to the one of fishmeal i.e. if the insect based feed systems’ performance is found inferior to the one of fishmeal, it will be located outside the circle, if performance is superior to fishmeal, the insect based feed system will be located inside the circle and the distance to the centre of the circle also holds a meaning. For illustrative matters, if the performance score of an insect based feed is located halfway between the circle edge and its centre, it means its production causes half the impact of the fishmeal system.

When evaluated in the light of the current state, i.e. where highest priority is given to the economic performance (scheme A, Figure 2), fishmeal would be the protein feed of choice. However, the performances of FfA_BSF#3 and IER_CM#3 are approaching the performance of fishmeal. In contrast, highest priority to the strong sustainability concept (scheme C, Figure 2) makes the GEI#3 or HZAU#3 system the protein feeds of choice. Since the production process of these systems were assessed to have a lower environmental impact, adherents to the strong sustainability concept would advise abandonment of fishmeal in the applicable geographical conditions. Treating each of the dimensions with equal importance (scheme B, Figure 2), advocates of the weak sustainability would focus on systems that showed superior performance in any of the three sustainability dimensions. With the exception of FERA#3, this elimination process would comprise all considered insect based feed production systems. We suggest that efforts to further develop systems with a superior performance to fishmeal in more than one dimension should be given priority.

These results formed the basis for the provision of technical and policy recommendations and future research and development activities. When assessed in the light of changing geographical settings, the application potential of insect based feed production designs was variable. This research may serve as a basis to assist the strategic and site-specific judgement of prospective insect farmers. We advocate interested stakeholders to take this report as guide assist in the implementation of a strategy that is better geared towards performance-critical socioeconomic and environmental site conditions and whenever possible, valorise the comparative environmental, social and/or economic advantages of a given geographical context.

WP5: Pro-Insect Platform in Europe
5.1 Introduction
The objective of this work package was to support the evolution of a positive and receptive platform in Europe for the utilisation of novel insect based proteins in animal feeds by engaging with and informing key stakeholder groups. This was achieved through the following activities:
• Mapping of the European, African and Chinese regulation, legislation, policies and guidance relevant to use of insects in animal feed and food
• Creation of the consensus ‘business case’ for use of insects in animal feed in Europe
• The development of position papers, tailored for identified key stakeholder groups, to inform and create dialogue with these ‘gatekeepers’
• Mapping and evaluation of current consumer opinion on the use of insects in the feed chain
• The creation of a ‘White Paper’ that was launched at a parliamentary reception

5.2 Consumer perception
Minerva UK Ltd conducted two surveys on consumer acceptance and perceptions to gain a better understanding of consumer views about eating insects and animals fed on insects. The first survey was conducted in 2013/14 with a broad remit of benchmarking attitudes towards eating insects and animal fed on insects. The second survey 2015/16 focused on the use of insect meal in farmed animal feed compared to other current and novel protein sources.
More than 2,400 consumers, from over 70 countries, participated in the multi-language surveys ensuring that a representative sample was taken. This is the largest exercise in consumer opinion on eating animals fed on insects known. The results from the surveys have been presented by Minerva at five scientific conferences, attracted significant media coverage and have been used to supplement project partner interviews on the BBC TV’s Countryfile programme (audience of 7 million) (Figure 3). The data from the consumer surveys has also been made available to the University of Sheffield and the Croatian Food Agency for additional analyses which will help to ensure the project’s legacy.

Overview of findings:
1st Survey Statistics:
Available in English, French and German.
Ran from October 2013 to March 2014
Received 1302 responses
Gender spilt- 55.9% male, 43.7% female
Responses from 71 different countries.

Main findings:
66% of respondents thought that insects are a suitable source of protein for use in animal feed
73% of respondents would be willing to eat fish, chicken or pork from animals fed on a diet containing insect protein
39% of respondents had eated insects directly themselves
57% of respondents thought that there should be appropriate labelling of fish, chicken or pork fed on insect protein
88% said that more information on the use of insects as a feed source should be made available

2nd Survey Statistics:
Available in in English, Spanish, Italian, Croatian and Czechoslovakian
Received 1150 responses
Ran from July 2015 to November 2015
Gender split – 38% male 60% female

Main findings:
70% of respondents considered that it was totally acceptable or acceptable to feed insect protein to farmed anumals including fish
66% of respondents would be very comfortable or comfortable eating meat from a farmed animal fed on insect meal
64% of respondents said there is no risk or low risk to human health in eating farmed animals fed on insct meal
There was a 30% gap in the difference between how knowledgeable respondents considered themselves to be and how knowledgeable they considered they should be
Media coverage from the consumer perception work included:
• All About Feed
• Feed Navigator

Alongside the formal consumer surveys Minerva has, since 2013, been monitoring media coverage (collected from online sources) on the topic of insects in food and feed. The tone, topic, expected audience and source country has been recorded for each media item identified. To date Minerva has tracked over 1,000 media articles providing a useful monitoring tool to track what is being said about PROteINSECT and the tone utilised, whilst also generating a wider picture of what is being written about insects as food and its inclusion in animal feed.

Full results analysis of media tracking is available on the PROteINSECT website.

5.2 Legislative development
5.2.1 Review of Legislation
Minerva UK Ltd undertook a full review of the current European, African and Chinese regulation, legislation, policies and guidance relevant to use of insects in animal feed and food. The European element of this work was developed through core research undertaken and via the development of a positive relationship with DG Sante (formerly DG SANCO). PROteINSECT has maintained these strong relationships to ensure that the project’s state of the art evidence has been, and will continue to be, used to inform the legislative and regulation debate.

Minerva UK Ltd published the PROteINSECT Mapping Exercise Report and an Executive Summary in July 2013 making it publicly and freely available via its website. By working with DG SANCO (now DG Sante), PROteINSECT was able to identify and agree the key areas of legislation that would need to be formally reviewed if insect protein (reared on animal manure) was to be adopted. The documents have remained a key reference point for legislative discussions during dissemination events and closed meetings. PROteINSECT continues to work with DG Sante, sharing and developing the required evidence base.

5.2.2 Scientific Opinion
In May 2014 DG SANCO (now DG Sante) requested ‘an initial scientific opinion on the safety risks arising from the production and consumption of insects as food and feed’ be made by the European Food Safety Authority (EFSA). EFSA accepted the request and set up a scientific committee to review the evidence available for the risk assessment. The work of PROteINSECT partners was vital to satisfy the call for evidence from EFSA with specific reference to the project’s quality and safety data. Project partner Dr Adrian Charlton of Fera accepted an invitation to join the scientific committee, and other project partners contributed by taking part in EFSA meetings when requested by the committee.

The publication of EFSA’s Scientific Opinion on the potential risks associated with using insect protein in food and feed in October 2015 provided PROteINSECT with a further opportunity to highlight its role, not only in the EFSA committee, but also the wider state of the art scientific work being undertaken. In October and November 2015 Minerva tracked 55 media articles covering insects as food and feed with specific reference of the PROteINSECT project highlighted in 11 of these articles.

The EFSA Opinion highlighted the need for additional scientifically-published nutritional and safety data for insects reared on organic wastes such as manure. Relevant research has been undertaken by the PROteINSECT project with further peer-reviewed publications due during 2016 and beyond.

5.2.3 Continued Engagement
In April 2016 Dr Wolfgang Trunk confirmed that the EC is committed to internal discussions around the barriers to adoption of insect as feed in the EU. Dr Trunk highlighted the critical importance of DG Sante’s responsibility for safety in feed and food as this remains their priority focus. Dr Adrian Charlton, the PROteINSECT partner responsible for quality and safety, is working closely with Dr Trunk to share state-of-the-art research findings in support of these ongoing discussions.
PROteINSECT continues to proactively work with DG Sante and EFSA to ensure that the state of the art evidence from the project is available in the development of evidenced based policy making.

5.3 Stakeholder groups
Minerva has led the specific development and engagement with European Stakeholder Groups in WP5. This has been achieved through a range of formal and informal activities. Each key engagement activity has been sequential in its approach, resulting in a clear line of sight to achieving a receptive platform for the discussion of the use of insects in Europe, primarily in animal feed (Figure 4).

Stakeholder engagement and work flow
The event drew together key European stakeholder groups, with representation from the PROteINSECT project team, to facilitate a discussion on state-of-the-art insect protein production and utilisation for animal feed. This meeting was designed to bring a variety of ‘voices’ together in order to develop a receptive platform for the utilisation of novel insect based proteins in animal feeds in Europe by engaging with, informing and understanding the views of these voices, establishing a two-way dialogue.
PROteINSECT Key Opinion Leaders Group members contributing their voice and perspectives:
• Hanover University of Veterinary Medicine
• European Reference Laboratory for Animal Proteins (EURL-AP)
• European Feed Manufacturers' Federation (FEFAC)
• European Aquaculture Technology Innovation Platform (EATiP)
• Waste & Resources Action Programme (Wrap)
• International Platform of Insects for Food and Feed (IPIFF)
• Metro Group (European Supermarket)
• World Wildlife Fund (WWF)
• European Commission - Directorate General for Health and Food Safety (DG SANCO)
• Association of Poultry Processors and Poultry Trade in the EU (a.v.e.c.)
• European Aquaculture Society
• European Rural Poultry Association
• European Food Safety Authority (EFSA)
• Food Standards Agency (UK)
• Nutreco Skretting

Invited stakeholders were provided detailed briefings on the current evidence base for Insect Production, Processing, Quality and Safety and the Life Cycle Assessment (LCA) work undertaken within the PROteINSECT project. Through a series of managed discussions, co-chaired by an independent chair Emile Frisson with Elaine Fitches, coordinator of PROteINSECT, attending stakeholders debated six key themes regarding the development of insect protein in Europe:
1. The need for additional safety data from across the regulatory spectrum.
2. The need for additional quality data to show the potential of the use of insect protein.
3. Animal welfare - two key areas; insect processing techniques and approaches; and welfare of the animals being fed on the insect protein.
4. The importance of clear scientifically based and neutral communications.
5. The separation of insects for food and insects as feed as topics.
6. Life Cycle Assessment is a key element of work for the development of insect derived protein and its contribution to providing a sustainable novel and additional source of quality protein for animal feed
These themes were shared widely across the project team to ensure, where appropriate, they could be addressed within existing work packages. The results provided valuable input for both the internal project deliverable report on the meeting as well as the Consensus Business Case.

5.3.1 Stakeholder Briefing Example – Consensus Business Case (July 2015)
The Business Case Report provided state-of-the-art evidence and holistic perspectives (February 2015) to support the evolution of a positive and receptive platform in Europe for the utilisation of novel insect based proteins in animal feeds to a broad range of stakeholders across policy and political circles, feed industry, farmers, retailers, researchers, consumer groups. The Business Case will be of interest to all parties considering the issue of sustainability of animal feed for livestock and fish production now and in the future. The publication was constructed utilising the knowledge and materials provided by PROteINSECT partners on their research undertakings and reports, supplemented by the discussions, perspectives and material brought to the Round Table meeting by our Key Opinion Leaders Group.
The Business Case Report was published having achieved consensus from our broad stakeholder group and specific endorsement from the stakeholder organisations listed below:
• Hanover University of Veterinary Medicine
• European Reference Laboratory for Animal Proteins (EURL-AP)
• Waste & Resources Action Programme (Wrap)
• Metro Group (European Supermarket)
• Food Standards Agency (UK)
• PROteINSECT partners

To support the launch of the Business Case Minerva issued a press release and undertook targeted social media activity and made the report freely available on the PROteINSECT website. This activity led to significant media coverage, examples of news articles include:
• Farming Futures
• Fefac
• All About Feed

In addition to media coverage the Consensus Business Case has been downloaded over 3,400 times from the PROteINSECT website (as at 1 May 2016) confirming the reach of the report and the project.

5.3.2 Stakeholder Messages Example - Position Papers (September 2015):
Position papers were developed and tailored to individual stakeholder groups by utilizing the updated knowledge and evidence from PROteINSECT’s research team, and from the information gleaned from the project’s established relationships with a broad range of European stakeholder organisations and key individuals.
Thirteen key stakeholder groups were identified to enable PROteINSECT to support the evolution of a positive and receptive platform in Europe for the utilisation of novel insect based proteins in animal feeds. These 13 key stakeholder groups are:

1. Animal Nutritionists
2. Aquaculture Producers
3. EC Policy Makers
4. Feed Producers
5. Pig Farmers
6. Poultry Farmers
7. Consumers
8. Environmentalists
9. Members of the European Parliament (MEPs)
10. Distribution chain – wholesalers to retailers
11. Food Certification Bodies
12. Human Nutritionist
13. Waste Recycling - Commercial and Research

Tailored position papers were developed for each of the 13 identified key stakeholder groups and shared with project partners.
It should be noted that the Key Opinion Leader Round Table, Consensus Business Case and Position Papers were utilised to develop the PROteINSECT White Paper ‘Insect Protein – Feed for the Future’ which is covered within another section of this report.

Potential Impact:
Potential impact
The rising global population and trend for the increased consumption of meat, milk and eggs, particularly in developing countries such as China, threatens the sustainability of protein production for animal feed. Not only is there a greater demand for livestock products, but animal feed suppliers also have to manage increasing safety and welfare concerns. The use of insects as a source of protein offers potential to help alleviate pressures upon the supply of protein for animal feed but its use is currently prohibited in Europe for animals raised for human consumption. Whilst there are a number of alternative protein sources that are being investigated for potential inclusion in animal feed, the production of insects as a source of protein is already at a stage where commercial operations exist. The PROteINSECT project has advanced the state-of-the-art in the five key areas identified at the start of the project:
• Development and optimisation of fly production methods for animal feed production in EU and International Cooperation Partner Countries (ICPC)
• Determination of optimal design(s) of insect-based animal feed production systems for both EU and ICPC countries through comprehensive life cycle impact assessments
• Determination of safety and quality criteria for substrate and insect protein products
• Evaluation of crude and refined insect protein extracts in fish and monogastric (poultry and pig) animal feeding trials
• Building a Pro-insect platform in Europe to encourage a holistic adoption of sustainable protein production technologies

The project has developed rearing systems suitable for the production of different fly species in several geographical locations. In some cases the production systems have been developed to a stage where they can be recommended for use by smallholders in Africa and China to feed livestock on small family farms. Training in the use of the systems developed in Mali, together with the production of a short film produced by the PROteINSECT project, has promoted the uptake of insect production in the region. In addition, a house fly rearing system developed by HZAU in China has been promoted and used as the basis for the establishment of similar systems by farmers in the region. As such the project has delivered positive short-term impacts by providing a means to enable small-scale farmers to produce a new feed for their livestock. In the longer term it is anticipated that this will improve the sustainability and economic viability of local livestock production by reducing reliance upon costly imported feeds.

The knowledge and expertise gained during the PROteINSECT project has been used to obtain funding for further research in West Africa. The black soldier fly (BSF) system established by PROteINSECT in Ghana was used as the basis for the development of another BSF production unit in the region of Accra. This work is being carried out by the University of Stirling and another Ghanaian partner, Animal Research Institute (ARI), in the framework of a sister project, Ento-Prise funded by the UK Department for International Development. The system is being further improved in another new project involving CABI, Fish for Africa, and ARI, Insects as Feed in West Africa (IFWA), funded by the Swiss Agency for Development and Cooperation and the Swiss National Science Foundation. It is envisaged that these research projects will have further positive impacts upon the future sustainability of small-scale livestock production in Africa.

Work to establish and improve insect rearing systems located in Africa, China and the UK went hand–in-hand with life cycle analysis. The assessed rearing systems covered a wide diversity of scenarios as different species of flies, rearing substrate, technological setups, production scale and geographical conditions were considered. Assessment of the potential environmental, social and economic performances of each insect production system allowed strengths and weaknesses to be highlighted. PROteINSECT prepared a publicly available toolkit and document providing recommendations to prospective insect farmers, which has the potential to influence the future design and implementation of insect-to-feed farming projects. Impacts in the longer term may be evidenced by the generation of insect farms with higher environmental and socio-economic performances, particularly in areas dominated by small-scale farmers.

The rearing system set up in the UK established proof of concept for the rearing of fly larvae on manure in Northern Europe providing an enabling platform for further research. We were fully aware of the need to develop large-scale automated production processes to achieve economically viability in a European setting, but this was well beyond the scope of the PROteINSECT project. In fact, a growing number of entrepreneurs across Europe have established insect production facilities during the course of the project, being collectively represented by the International Platform of Insects for Food and Feed. Such activities demonstrate growing awareness of the huge potential that exists in Europe for the use of insects in food and feed.

In the European Union, the use of insects as a source of protein for animal feed is currently prohibited for animals raised for human consumption. An understanding of the potential risks associated with the use of insects in animal feed is vital for the development of an appropriate regulatory framework that permits the use of insects in animal diets. PROteINSECT conducted a state of the art screening programme to assess the safety of dried fly larvae supplied by the different rearing systems established by the project partners. Given that the analysis was conducted on insects that had been reared on animal manures this screening presented what might be envisaged as a “high risk” scenario for feed safety. Levels of more than 500 potentially toxic chemical contaminants were all found to be below recommended maximum amounts for feed suggested by bodies such as the European Commission, World Health Organisation and Codex. Similarly tests found no evidence for the presence of viable biological contaminants such as Salmonella and Campylobacter. The importance of substrate quality was highlighted by elevated levels of the toxic heavy metal cadmium that were identified in some samples indicated that pre-screening of rearing substrates would be necessary to mitigate risk. PROteINSECT also highlighted the potential for allergenic risks in relation to both animal welfare and worker exposure during feed manufacturing. This data made a valuable contribution to the preparation of a risk profile related to production and consumption of insects as food and feed published by EFSA in October 2015. This document highlights future research that needs to be conducted to ensure that an appropriate regulatory framework can be established in Europe to allow the future adoption of insects in animal feed.

The quality of insect material must be clearly defined to ensure that feed formulations are optimal for animal performance and well-being. PROteINSECT found the nutritional quality of fly larvae to be highly compatible with use in feed, having both high total protein content and high levels of key amino acids that were comparable with those found in fishmeal and higher than most crop plants. This data was in agreement with an increasing body of evidence in the literature suggesting that insects are a suitable alternative source of protein to both animal (e.g. fish meal, meat and bone meal) and plant (e.g. soy and maize) derived sources. PROteINSECT also evaluated extraction methods and found solvent extraction to be an efficient and scalable processing method yielding protein enriched material suitable for formulation. Fish, poultry and pig feeding trials were conducted in Europe using both crude and refined insect material. Fish and poultry trials were also conducted in Africa and China using both dried and live insects. All animals reared on insect containing diets performed as well as those reared on commercial feeds with no taints detected in fish and meat derived from the feeding trials. These results provided a powerful body of evidence for the suitability of fly larvae as a partial replacement for soy or fishmeal in fish and monogastric feeds. This data will help to inform regulators as consideration for the use of insects in feed continues. Assuming that future regulatory changes will permit the inclusion of insects in animal feeds, this work will have a direct impact upon feed manufacturers by providing a basis from which new feeds can be formulated.

The adoption of insects into the feed chain is ultimately dependent upon consumer attitudes towards the consumption of meat and dairy products derived from animals fed on insect containing diets. PROteINSECT has made significant efforts to raise public awareness and to gauge the likely level of consumer acceptance of the use of insects in feed. Two surveys completed by more than 2000 consumers found a high level of support for the use of insects in feed together with an expressed desire for more information. Media communications and publications generated by the project have started to fill this knowledge gap. Careful consideration of project messages to the media have ensured that claims about the potential benefits or risks of using insects in feed have not been sensationalized. At all times PROteINSECT has promoted the need to identify and address the knowledge gaps that must be filled to ensure the safe use of insects in the food chain. We believe this approach has laid an appropriate foundation from which the public can make an informed decision about the acceptability of insects in the feed chain. The engagement of partners in a broad range of dissemination activities across the world has successfully raised awareness of the potential use of insects in feed to a wide range of key stakeholders including, regulatory bodies, feed manufacturers, farmers, retailers, research funding bodies, and academia.

A substantial rise in interest as to the potential use of insects in feed and as food has occurred, particularly in Europe, over the past 4-5 years. PROteINSECT research and communications have helped to drive progression of the concept of insects in feed towards their safe, acceptable and approved use as a feed ingredient in the future. The adoption of insects in feed has potential for positive global societal impacts by helping to reduce current reliance upon the land-limited supply of protein for meat production.

Communication and Dissemination Activities
An integrated dissemination strategy has been fundamental to communicating the benefits and positive impact of the uptake of knowledge generated by PROteINSECT to stakeholders. This aimed to drive a supportive regulatory framework to enable adoption of the technology by demonstrating its feasibility, safety and acceptability. In parallel PROteINSECT developed a pro-insect platform across Europe to increase acceptance among consumers and other key stakeholders, and to develop the market for food (pork, chicken and fish raised on insect containing feeds) for human consumption.

The specific objectives of the dissemination activities were to build awareness for the project and its results. The dissemination plan laid the foundations for an effective and diverse communication of potential benefits to interested stakeholders beyond the project’s partners.
The overall dissemination strategy reached a broad range of activities and goals:
• Raise awareness of the project itself through developing its ‘brand’ via a logo and project website
• Produce appropriate materials for use in dissemination activities (leaflet, bookmark)
• Identify all appropriate external stakeholders (national, EU and global) and the channels through which those stakeholders can be reached effectively
• Promote and disseminate all the project materials appropriate for public dissemination to the public and to key stakeholders (since the objective of PROteINSECT is to develop novel technologies for feed and feed additive producers, some results will remain confidential)
• Inform decision-makers at policy and legislative levels and regulatory bodies of the research and development work of the PROteINSECT consortium

PROteINSECT has actively engaged through the project lifecycle in a significant range and number of communication and dissemination activities including managing social media, television/radio interviews and hosting a scientific conference. Extensive participation in conferences, workshops and meetings across Europe, China and Africa has ensured that a wide audience has been reached during the course of the project. This continuous activity has focused on providing high quality scientific evidence in audience appropriate formats, allowing a variety of perspectives to be heard. This has led to a strong network of followers allowing the project to ensure their message is heard in a variety of environments from specialist to general audiences. The breadth of dissemination activities is illustrated in Figure 5 and particular highlights are shown below:
• More than 1.3 million visitors to the PROteINSECT website since March 2013 with visits peaking in November 2015 with 68,800 page views
• More than 1200 followers on Twitter
• More than 980 Facebook likes
• There have been over 3400 downloads of the Consensus Business Case
• The BBC Countryfile programme featuring the research of PROteINSECT attracted 7 million live viewers
• PROteINSECT researchers participated in two online expert blogs with 26,000 views
• PROteINSECT was recognised for its ambitious innovation potential and received an Innovation Award under the EC funded CommBeBiz Project. The award was presented at the EBN Congress on 29th October 2015.

Insects to Feed the World Conference – Ede, The Netherlands (May 2014)
A dissemination highlight was the participation of PROteINSECT at the first ever ‘Insects to Feed the World’ international conference in Ede, the Netherlands in May 2014. The conference was jointly organised by the FAO and Wageningen University to promote the use of insects as human food and as an ingredient in animal feed as a way of assuring food security. Over 450 individuals from 45 countries participated in the conference, representing diverse areas including science, policy and industry.

The conference provided the ideal platform to share the latest results from the project with a wide range of interested stakeholders – breeders, producers, researchers, regulators and environmentalists. It also enabled us to increase awareness of PROteINSECT and explain how we are contributing to the effort to add insect protein to the permitted list of ingredients for animal feed. In addition to engaging directly with those at the conference, we were also able to promote PROteINSECT more widely via the press and social media outreach. Furthermore, it was an excellent opportunity to meet and exchange ideas with people who are working in a similar area.

It was important to ensure that PROteINSECT had a strong, visible presence at the conference and this was achieved via oral presentations, posters and the PROteINSECT stand.

Article in Science magazine
A special highlight of the PROteINSECT dissemination activities was the article in Science (http://www.sciencemag.org/news/2015/10/feature-why-insects-could-be-ideal-animal-feed)

Parliamentary Reception and Final Conference (April 2016)
Parliamentary Reception
The PROteINSECT White Paper ‘Insect Protein – Feed for the Future. Addressing the need for feeds of the future today’ was launched at and presented to a Parliamentary Reception sponsored by Jan Huitema MEP and PROteINSECT partners on Tuesday 26th April 2016 and made publically available on Wednesday 27th April 2016. Working with Jan Huitema MEP, a targeted invite list was developed to ensure a targeted and relevant reception within the European Parliament. The reception was attended by 5 MEPs or their representatives as well as invited industry experts. To date the White Paper has been provided in hard copy to over 130 conference attendees and 30 Parliamentary Reception guests and downloaded over 200 times from the Project website (as at 1 May 2016).

Final Conference
The PROteINSECT Final Conference ‘Insect Protein – Feed for the Future’ demonstrated and disseminated the project’s research protocols, methods and results and provided an opportunity to discuss our work with scientific and other interested and relevant communities. The scope of the Final Conference was agreed with the project’s Work Package leads to ensure that key project outcomes were provided with the appropriate platform to share results within the global context. The conference attracted a trans-national audience of 130 interested parties from research, policy, industry and the media on Wednesday 27th April 2017 in Brussels.

PROteINSECT also used the conference as an opportunity to highlight its Student Engineering Competition winner Simon Schantl from the University of Applied Sciences JOANNEUM and his concept ‘ENTODRYA’.

Media Activity
Minerva UK Ltd created a significant level of media engagement for the Parliamentary Reception and Final Conference with project partners engaging in this activity through offering interviews and statements to attending journalists, via pre-agreed statements together with a BBC radio interview.
As a result the project received significant media attention.

Media coverage examples focused on the Parliamentary Reception, White Paper & Final Conference:
• New Scientist “Animals may be fed manure-bred maggots to make meat substitute”
• International Business Times “Maggots and manure: How to meet the world’s meat demand with insect food”
• Feed Navigator “PROteINSECT: ‘The process towards adoption of insect protein in feed (in the EU) is already underway’”
• Pig World “Insect protein feeding trials deliver positive results”
• The Aquaculturalists “White Paper published today on insect protein for feed”
• Farming UK “’Huge potential’ for using insect protein as a source of animal feed”
• All About Feed “New whitepaper on insect protein for feed”
• Pig World “Feed industry is ‘ready’ for insect protein says trial leader”
• Fera “Insects no more risky than other proteins as animal feed”
• All About Feed “Best practices in the insect protein sector”
• All About Feed “A glimpse at the recent insect meal trials”
• Feed Navigator “EU: Insect protein use in aqua feed likely within 12 months, says industry insider”
• Phys Org “Insect larvae as an additional source of protein for Europe’s animal feed”

Social Media
Minerva UK Ltd led social media activity throughout the life of the project, boosted under the hashtag ‘#insect4feed’ to engage conference attendees and a wide variety of non-attendees. Below is a snapshot overview of the impact of this activity:

Key stats from PROteINSECT social media:
• Received >17,300 tweet impressions
• Top tweet received 1,671 impressions
• 48 tweets posted
• 829 profile views
• 60 mentions
• 54 new followers

Website Activity
In addition to the mainstream media articles and social media activity the level of online engagement with the project increased during the White Paper Launch and Final Conference, building on its already strong online presence (Figure 6). Secondary benefits were achieved by drawing an increasing number of unique and return visitors to the PROteINSECT website leading to existing project material also receiving additional spotlight. One example is the Consensus Business Case which has now been downloaded >3,400 times from the PROteINSECT website.
Key Website Stats:
• More than 53 000 unique visitors to www.proteinsect.eu
• The highest month for visits to the website was November 2015 coinciding with the launch of the EFSA opinion (3,965 visits)
• There have been 1.3 million website views of PROteINSECT to date

The consensus business case for use of insects in animal feed in Europe
The Consensus Business Case (CBC) Report was developed utilising the knowledge and materials provided by PROteINSECT partners based on their research undertakings and reports, supplemented by the discussions and material brought to the Round Table meeting by our Key Opinion Leaders (KOL) Group. The CBC Report was finalised and submitted in February 2015 with content agreed by all stakeholders. The CBC was launched into the public domain on 29 July 2015 via media communications and targeted social media activity, with the report freely available to view and download from the project’s website. As at 1 May 2016 the CBC has been downloaded over 3,200 times.

EFSA opinion
Members of the PROteINSECT consortium were involved with the production and providing evidence for the EFSA Scientific Opinion ‘Risk profile related to production and consumption of insects as food and feed’ (doi:10.2903/j.efsa.2015.4257) which was published in October 2015. In particular, data on the safety of insect protein was provided for consideration by the EFSA panel.

Future dissemination and exploitation
Peer reviewed publications
A variety of publications have been produced during the time frame of PROteINSECT (listed below in the publications section) Three publications could not be added to this list as the journal title was not available as an option to upload. These were as follows:

Adrian J. Charlton, Michael Dickinson, Maureen Wakefield, Elaine Fitches, Marc Kenis, Richou Han, Fen Zhu, N’Golope Kone, Michael Grant, Gabriel K.D. Koko, Emilie Devich & Geert Bruggeman (2015). Assessing the chemical safety of fly larvae as a source of protein for animal feed. Journal for Insects as Food and Feed http://www.ipiff.org/public/studies/charlton-et-al-2015---chemical-safety.pdf

Martin Roffeis, Bart Muys, Joana Almeida, Wouter M.J. Achten, Santos Rojo, Anabel Isabel Martinez-Sanchez, Berta Pastor, Yelitza Velasquez (2015). Pig manure treatment with house fly [Musca domestica] rearing - An environmental Life Cycle Assessment. Journal of Insects as Food and Feed http://rua.ua.es/dspace/bitstream/10045/53087/1/2015_Roffeis_etal_JIFF.pdf

Romero, R., Claydon, A., Fitches, E., Wakefield, M. Charlton, A. (2015) Sequence homology of the fly proteins tropomyosin, arginine kinase and myosin light chain with known allergens in invertebrates. Journal of Insects as Food and Feed http://www.wageningenacademic.com/doi/abs/10.3920/JIFF2015.0067

It is anticipated that further publications will include data from PROteINSECT. These will be listed on the PROteINSECT website when they are published and made available if open access.

Future funding opportunities
Opportunities for further project funding were discussed by partners at the final project meeting. It was agreed that the PROteINSECT project partners will create a project network for information exchange and to plan next steps will be formed.

PROteINSECT website
The PROteINSECT website will remain operational until December 2017. Key documents, for example the Consensus Business Report, White Paper and final conference proceedings will be available to download from the website.

List of Websites:
www.proteinsect.eu