There has been a major shift to diets with increased consumption of animal products, including fish. Indeed, people have never consumed so much fish or depended so greatly on the sector for their well-being as today: in 2012, fish provided 17% of the world population’s intake of animal protein. As capture fishery production has been relatively stable at about 90 million tonnes since the 1990s, the rising demand for fishery products has been met by a fast-growing aquaculture industry, which set an all-time high record at 67 million tonnes in 2012, providing 50% of the fish used for human consumption (FAO, 2014). Fish feeds, notably those of salmonids and marine fish, are usually based on fish meal and fish oil obtained from pelagic species captured for this purpose (Médale et al., 2013). Fish meal is a highly regarded source of protein with an excellent composition of essential amino acids, while fish oil provides long-chain omega-3 fatty acids favored for their health benefits (Olsen and Hasan, 2012). However, this reliance on wild fish capture for fish farming is under question. Not only fish meal and fish oil may contain contaminants such as polychlorinated biphenyls and dioxins, but consumers are now interested in sustainability metrics such as the ratio of wild fishery inputs to farmed fish outputs (Naylor et al., 2009). Also, the volatility and rise of fish meal prices is a matter of concern for fish farmers (Olsen and Hasan, 2012). Furthermore, while aquaculture’s share of fish meal and fish oil consumption has been increasing, reaching 88% by 2007 (Tacon and Metian, 2008), the production of fish meal decreased between 1994 and 2012 and is now about 5 to 6 million tonnes (Médale et al., 2013; FAO, 2014). As a consequence, there has been an ongoing search for alternative sources of protein that would allow aquaculture to remain economically and environmentally sustainable (Barroso et al., 2014). Non-animal proteins derived from legume and/or oil seeds or cereal gluten are now introduced in fish diets (Médale et al., 2013), but plant sources have limitations, such as palatability issues, presence of anti-nutritional substances, low concentrations of sulfur amino acids, and high proportions of fiber and non-starch polysaccharides (Sanchez-Muros et al., 2014).
In the recent years, insects have received wide attention as a potential source of protein both for humans and livestock. Insects grow and reproduce easily, have high feed conversion efficiency, and can be reared on biowastes (van Huis et al., 2013; Makkar et al., 2014). One kilogram of insect biomass can be produced from on average 2 kg of feed biomass (Collavo et al., 2005). This article presents the current status on the insects that are the best candidates as fish feed ingredients in partial or complete substitution for fish meal, with regard to their nutritional attributes, ease of rearing, and biomass production: larvae or pupae of Diptera black soldier fly (Hermetia illucens) and house fly (Musca domestica); larvae of mealworm [Tenebrio molitor (Coleoptera)]; adult Orthoptera from the Acrididae (locusts and grasshoppers), Gryllidae (crickets), and Tettigoniidae (katydids) families; and pupae of silkworm [Bombyx mori (Lepidoptera)]. Many fish species consume insects in the wild: omnivorous species prey on insects found on the bottom of water bodies whereas juvenile stages of carnivorous species eat insects before switching to fish-based diets (Riddick et al., 2013).
Protein and lipids
The main chemical constituents of insects are presented in Table 1. The crude protein (CP) content of insects is high and varies from 42 to 63%, a range comparable to that of soybean meal but slightly less than that of fish meal. Diptera larvae (black soldier fly and housefly) and mealworm larvae contain less protein than adult Orthoptera (locusts and crickets) and silkworm pupae.
|Constituents||Black soldier fly larvae||Housefly maggot meal||Mealworm||Locust meal||House cricket||Mormon cricket||Silkworm pupae meal||Silkworm pupae meal (defatted)||Fishmeal||Soymeal|
|Crude protein||42.1 (56.9)*||50.4 (62.1)||52.8 (82.6)||57.3 (62.6)||63.3 (76.5)||59.8 (69.0)||60.7 (81.7)||75.6||70.6||51.8|
Insects often accumulate fat, especially during their immature stages (Manzano-Agugliaro et al., 2012). The lipid content of non-defatted insects is highly variable and varies from 8.5 (adult locust) to 36% (mealworm larvae). However, variability in lipid concentration is high even within the same species; for instance, oil values as high as 30% have been reported for locusts because it is influenced by the stage of development and by the diet (Barroso et al., 2014). The defatted meal, being richer in CP than soybean meal and fish meal, could find a place as a protein-rich resource in fish diets.
Insects contain relatively low levels of carbohydrates compared with plants, typically less than 20% (Barroso et al., 2014). The carbohydrate most commonly encountered by fish in the wild is probably chitin, a polymer of glucosamine found in the exoskeleton of arthropods (Lindsay et al., 1984). However, the amount of chitin in insects is variable because it depends on the species and development stage and also on the method of analysis. Very high [>10% of the dry matter (DM)] as well as very low values (<100 mg/kg DM) have been reported (Finke, 2007). The ability of fish to digest chitin is also a matter of debate. Chitinase activity has been observed in several fish species, and benefits of incorporating chitin into marine fish diets have been reported, but it is generally agreed that chitin is one of the factors limiting the use of insects in fish feeds (Ng et al., 2001; Sanchez-Muros et al., 2014).
The amino acid profiles of various insects are given in Table 2. Compared with fish meal, the CP of Orthoptera and mealworms tend to contain less lysine while Diptera and silkworms are relatively rich in lysine. Sulfur amino acids (in percent CP) tend to be less in insects than in fish meal, except for silkworms. Threonine levels are roughly comparable but are greater for silkworms. Tryptophan levels are generally less, except for silkworms and housefly maggot meal. For optimum growth, and depending on the specific requirement of the fish species, supplementation with synthetic amino acids could therefore be recommended. Compared with soybean meal, silkworms and Diptera have a globally better amino acid profile and could be better substitutes of fish meal than soybean meal.
|Amino acids||Black soldier fly larvae||Housefly maggot meal||Mealworm||Locust meal||House cricket||Mormon cricket||Silkworm pupae meal||Silkworm pupae meal (defatted)||Fishmeal||Soymeal||FAO Reference protein|
Ash contents of insects are generally low, except for black soldier fly larvae, for which values greater than 15% have been reported. Black soldier fly larvae are rich in calcium (7.6% DM), but other insects have very low calcium levels, and calcium supplementation would be required. Calcium fortification of the rearing substrate can increase the calcium level in larvae meals (Table 1). Calcium:phosphorus ratios in insects vary from 0.2 to 1.2 (except for black soldier fly larvae, which have a ratio of 8.4) and are thus less than the optimal values recommended for fish (1.1–1.4) (Chavez-Sanchez et al., 2000; Kumar et al., 2012). In some insects (e.g., housefly maggot meal and Mormon cricket), phosphorus levels are particularly high (1.0 to 1.6%).
Fatty acid composition
The fatty acid profiles of various insects are given in Table 3. Concentrations of unsaturated fatty acids are high in mealworm, house cricket, and housefly maggot meals (60–70%), and lowest in black soldier fly larvae (19–37%) due to high levels of saturated fatty acids. Linoleic acid (18:2n-6) concentration is much greater than that of α-linolenic acid (APA, 18:3n-3), as in many plant oils (including soybean and sunflower). Compared with fish oil, terrestrial insects contain greater quantities of n-6 polyunsaturated fatty acids and negligible amounts of eicosapentaenoic acid (EPA, 20:5n-3) and docosahexaenoic acid (DHA, 22:6n-3). This lack of EPA and DHA is a limiting factor to the use of terrestrial insects in marine fish, which require these fatty acids but have limited abilities to synthetize them. Salmonids can synthetize EPA and DHA from APA, but dietary supply is more efficient (Médale et al., 2013; Sanchez-Muros et al., 2014). Aquatic insects, on the other hand, contain significant amounts of EPA and have been proposed as source of feed for freshwater fish (Sanchez-Muros et al., 2014). For instance, the lipids of freshwater insects that are part of the natural diet of the Atlantic salmon (Salmo salar) contain more than 15% EPA (Bell et al., 1994). It has been shown that the lipid concentration and the lipid profile of insects are highly dependent on the diet and that they can be modified by changing the composition of the substrate (Sanchez-Muros et al., 2014). For instance, changing the substrate from cow manure to a 50:50 mix of cow manure and fish offal increased the level of omega-3 fatty acids in the black soldier fly larvae from 0.2% to 2% (total fatty acids basis) and total lipid concentration from 20 to 31% (DM basis) (St-Hilaire et al., 2007b).
|Constituents in (% fatty acids)||Black soldier fly larvae||Housefly maggot meal||Mealworm||House cricket||Fish oil|
|Saturated fatty acids (%)|
|Lauric, 12:0||21.4 [49.3] (42.6)||-||0.5||-|
|Myristic, 14:0||2.9 [6.8] (6.9)||5.5||4.0||0.7||3.7-7.6|
|Palmitic, 16:0||16.1 [10.5] (11.1)||31.1||21.1||23.4||10.2-20.9|
|Stearic, 18:0||5.7 [2.78] (1.3)||3.4||2.7||9.8||1.1-4.7|
|Monosaturated fatty acids (%)|
|Oleic, 18: 1n-9||32.1 [11.8] (12.3)||24.8||37.7||23.8||11.4-18.6|
|Polyunsaturated fatty acids (%)|
|Linoleic, 18:2n-6||4.5 [3.7] (3.6)||19.8||27.4||38.0||1.1-1.3|
|Linolenic, 18:3n-3||0.19 [0.08] (0.74)||2.0||1.2||1.2||0.3-0.8|
|Eicosapentaenoic (EPA), 20:5n-3||0.03  (1.66)||-||-||-||3.7-16.9|
|Docosahexaenoic (DHA), 22:6n-3||0.006  (0.59)||-||-||-||2-21.9|
Utilization of Insects in Fish Feeding
Black soldier fly larvae (Hermetia illucens)
Several experiments have shown that black soldier fly larvae could partially or fully substitute for fish meal in fish diets. However, additional trials as well as economic analysis are necessary because reduced performance has been observed in some cases and the type of rearing substrate and the processing method affect their utilization by fish.
Channel catfish (Ictalurus punctatus).
Chopped soldier fly larvae grown on hen manure fed to channel catfish alone or in combination with commercial diets resulted in similar performance (body weight and total length) as with the control diets. The fish aroma and texture were acceptable to the consumer. Young catfish refused whole larvae but consumed chopped ones (Bondari and Sheppard, 1981). Replacement of 10% fish meal with 10% dried soldier fly larvae resulted in slower growth over a 15-wk period for subadult channel catfish grown in cages but not in fish grown in tanks. In tank-grown fish, feeding 100% larvae did not provide sufficient DM or CP intake for good growth. Chopping of the larvae was not recommended, as it improved weight gain and increased feed consumption but resulted in reduced feed efficiency and greater feed waste (Bondari and Sheppard, 1987). A comparison between menhaden fish meal and black soldier fly prepupae meal showed that the latter could be advantageous up to an inclusion rate of 7.5% as a replacement for fish meal provided it was also supplemented with soybean meal to obtain isoproteic diets (Newton et al., 2005).
Yellow catfish (Pelteobagrus fulvidraco).
In yellow catfish, 25% replacement of fish meal by black soldier fly larvae meal produced no significant difference in the growth index and immunity index compared with the control group (Zhang et al., 2014).
Blue tilapia (Oreochromis aureus).
Chopped soldier fly larvae grown on hen manure fed to blue tilapia catfish alone or in combination with commercial diets resulted in similar performance (body weight and total length) as with the control diets and in fish aroma and texture acceptable to the consumer (Bondari and Sheppard, 1981). In a later experiment, feeding dry black soldier fly larvae as the sole component of the diet did not provide sufficient DM or CP intake for good growth for tilapia grown in tanks. However, chopping improved weight gain by 140% and feed efficiency by 28% (Bondari and Sheppard, 1987).
Rainbow trout (Oncorhynchus mykiss).
Black soldier fly prepupae meal reared on dairy cattle manure enriched with 25 to 50% trout offal could be used to replace up to 50% of fish meal protein in trout diets for 8 wk without significantly affecting fish growth or the sensory quality of trout fillets although a slight (but nonsignificant) reduction in growth was observed (Sealey et al., 2011). In a 9-wk study, replacing 25% of the fish meal protein in rainbow trout diets with black soldier fly prepupae meal reared on pig manure did not affect the weight gain and feed conversion ratio (St-Hilaire et al., 2007a).
Atlantic salmon (Salmo salar).
A control diet containing 20% fish meal was replaced by black soldier fly larvae meal at 25, 50, or 100% fish meal replacement, resulting in similar growth and sensory testing of fillets, greater feed conversion efficiency, and an absence of histological differences (Lock et al., 2014). However, these authors did caution that the method of preparation of insect could impact performance.
Turbot (Psetta maxima).
Juvenile turbots accepted diets containing 33% defatted black fly soldier larvae meal (as a replacement of fish meal) without significantly affecting feed intake and feed conversion. However, specific growth rate was less at all of the inclusion rates. Greater inclusion rates decreased the acceptance of the diet, resulting in reduced feed intake and growth performance. The presence of chitin might have reduced feed intake and nutrient availability and therefore reduced growth performance and nutrient utilization (Kroeckel et al., 2012).
Housefly maggot meal and housefly pupae meal (Musca domestica).
The use of housefly maggots as supplements in fish diets has been mostly studied in Nigeria for tilapia and catfish species.
African catfish (Clarias gariepinus, Heterobranchus longifilis, and hybrids).
There have been numerous experiments in Nigeria on the use of housefly maggots in the diets of African catfish, mostly Clarias gariepinus, Heterobranchus longifilis, and hybrids. The results are generally positive, but the inclusion of maggot meal should be limited to 25 to 30% because performance tends to decrease when greater inclusion rates are used (Fasakin et al., 2003; Idowu et al., 2003; Madu and Ufodike, 2003; Sogbesan et al., 2006; Aniebo et al., 2009; Adewolu et al., 2010; Ossey et al., 2012).
Nile tilapia (Oreochromis niloticus).
Nile tilapia fed a 4:1 mixture of wheat bran and live maggots had a better growth performance, specific growth rate, feed conversion ratio, and survival than fish fed only wheat bran (Ebenso and Udo, 2003). When maggot meal was included at 15 to 68% in the diet replacing fish meal, best performance and survival were obtained at 25% inclusion (34% substitution of fish meal), with no adverse effects on the hematology and homeostasis. However, sources of n-6 and n-3 fatty acids should be included in the diet to enhance the fatty acid profile in fish (Ogunji et al., 2007; Ogunji et al., 2008a,b).
Mealworm (Tenebrio molitor)
African catfish (Clarias gariepinus).
Fresh and dried mealworms have been found to be an acceptable alternate protein source for the African catfish. Replacing 40% of fish meal with mealworm meal in isoproteic diets resulted in growth performance and feed utilization efficiency similar to that obtained with the control diet, and performance was still similar at 80% substitution. Catfish fed solely on live mealworms had a slight depression in growth performance, but fish fed live mealworms in the morning and commercial catfish pellets in the afternoon grew as good as or better than fish fed the commercial diet. Live and dried mealworms were found to be highly palatable. Catfish fed mealworm-based diets had significantly more lipids in their carcass (Ng et al., 2001).
Gilthead sea bream (Sparus aurata).
In gilthead sea bream juveniles fed diets containing mealworm meal replacing 25 or 50% of fish meal protein, 25% substitution did not affect weight gain and final weight negatively, while 50% substitution induced growth reduction and less specific growth rate, feed conversion efficiency, and protein efficiency ratio. The whole body proximate composition was unchanged (Piccolo et al., 2014).
Rainbow trout (Oncorhynchus mykiss).
Mealworm added to a diet (containing 45% CP) at levels of 25 and 50% by weight (as a replacement of fish meal) showed that it could be included at up to 50% without reducing growth performance (Gasco et al., 2014a).
European sea bass (Dicentrarchus labrax).
In European sea bass, including up to 25% of mealworm meal in isoproteic diets as a replacement of fish meal had no adverse effects on weight gain. Inclusion at 50% reduced growth, specific growth rate, and feed consumption ratio slightly but not protein efficiency ratio, feed consumption, and body composition. Mealworm inclusion influenced the fatty acid composition of body lipids (Gasco et al., 2014b).
Locust Meal, Locusts, Grasshoppers, and Crickets
African catfish (Clarias gariepinus).
Desert locust meal (Schistocerca gregaria) could replace up to 25% dietary protein in C. gariepinus juveniles without significant reduction in growth. Chitin may have contributed to reduced performance when greater rates were used (Balogun, 2011). Meal of adult variegated grasshopper (Zonocerus variegatus) could replace up to 25% fish meal in the diets of C. gariepinus fingerlings without any adverse effect on growth and nutrient utilization at the same protein level in the diet. Greater inclusion rates decreased digestibility and performance (Alegbeleye et al., 2012).
Walking catfish (Clarias batrachus).
Several studies have investigated the effects of feeding dried Indian grasshoppers (Poekilocerus pictus) on the histological and physiological parameters of walking catfish. A 91-d feeding of dried grasshoppers had no adverse effect on hematological parameters but resulted in a little shrinkage in the gills as well as a reduction in ovarian steroidogenesis, which may reduce fertility (Johri et al., 2010; Johri et al., 2011a,b).
Nile tilapia (Oreochromis niloticus).
Migratory locust meal (Locusta migratoria) could replace fish meal up to 25% in isoproteic diets of Nile tilapia fingerlings without an adverse effect on the nutrient digestibility, growth performance, and hematological parameters (Abanikannda, 2012; Emehinaiye, 2012).
Silkworm Pupae Meal (Bombyx mori)
In the common carp (Cyprinus carpio), it was possible to replace 100% of fish meal protein with non-defatted silkworm pupae meal with no adverse effect on growth and feed conversion (Rahman et al., 1996; Nandeesha et al., 1990). Silkworm pupae meal could be safely used up to 50% in the diet without adversely affecting growth and flesh quality (Nandeesha et al., 2000). In a comparison between silkworm pupae meal and alfalfa or mulberry leaf meals, feed conversion efficiency, nutrient digestibility, and nutrient retention were better for diets based on silkworm meal than for diets based on plant leaf meals (Swamy and Devaraj, 1994).
In a polyculture system based on Indian carp (Catla catla), mrigal carp (Cirrhinus mrigala), rohu (Labeo rohita), and silver carp (Hypophthalmychthys molitrix), fermented silkworm pupae silage (replacing fish meal) included in formulated diets gave better survival rate, feed conversion ratio, and specific growth rate than untreated fresh silkworm pupae paste or fish meal (Rangacharyulu et al., 2003). In rohu, non-defatted silkworm pupae and defatted silkworm pupae resulted in significantly greater protein digestibility values than fish meal (Hossain et al., 1997).
Silver barb (Barbonymus gonionotus).
In silver barb fingerlings, highest growth performance was observed with a diet where silkworm pupae meal replaced 38% of total dietary protein (Mahata et al., 1994).
Mahseer (Tor khudree).
Mahseer fingerlings fed a diet containing 50% defatted silkworm pupae at 5% of body weight had a better growth and survival than fingerlings fed no or reduced amounts of silkworm pupae (Shyama and Keshavanath, 1993).
Mozambique tilapia (Oreochromis mossambicus).
Mozambique tilapias could utilize the protein of both defatted and non-defatted silkworm meal with a high apparent protein digestibility of 85 to 86% (Hossain et al., 1992).
Asian stinging catfish (Heteropneustes fossilis).
Silkworm pupae meal could replace fish meal at up to 75% protein substitution in Asian stinging catfish diets without adverse effect on growth (Hossain et al., 1993).
Walking catfish (Clarias batrachus).
Non-defatted silkworm pupae meal was found to be a suitable fish meal substitute in diets for walking catfish. Digestibility of the CP in silkworm meal was found to be similar to that in fish meal (Borthakur and Sarma, 1998a). Walking catfish fingerlings fed silkworm meal had slightly lower specific growth rate and poorer feed conversion ratio (2.81 vs. 2.45) than fingerlings fed on fish meal (Borthakur and Sarma, 1998b).
Chum salmon (Oncorhynchus keta).
Chum salmon fry fed over 6-wk diets supplemented with 5% silkworm pupae meal at the expense of fish meal did not show improvement in growth rate and protein content although silkworm supplementation enhanced feed efficiency (Akiyama et al., 1984).
Japanese sea bass (Lateolabrax japonicus).
In Japanese sea bass, the energy digestibility (73%) of non-defatted silkworm pupae meal was less than that of poultry by-product meal, feather meal, blood meal, and soybean meal but comparable to that of meat and bone meal. Crude protein digestibility (85%) was also less than that of poultry by-product meal, blood meal, and soybean meal but was comparable with that of feather meal and greater than that of meat and bone meal (Ji et al., 2010).
The insect species presented in this review have potential for use as a source of protein in the diets of farmed fish. Insects are valuable ingredients rich in protein, lipids, and energy. Numerous trials with carnivorous, omnivorous, and herbivorous fish have demonstrated that insects can be successfully included in fish diets as a substitute for fish meal although there have been more studies on omnivorous species than on carnivorous ones. Most trials recommend replacement rates less than 25 to 30%. In some cases, greater rates and even total substitution have been found technically or economically feasible.
Use of insects for the feeding of farmed fish faces several challenges from a nutritional perspective. One is the composition of insects and thus their nutritional value, which is highly dependent on the species, stage of development, and substrate used to feed the insects. Protein, lipid, and mineral composition are all highly variable, even within a taxon at the same development stage. For instance, the lipid concentration reported in the literature ranges from 15 to 35% for black soldier fly larvae and from 9 to 26% for housefly maggots (DM basis). Such a wide variation is a challenge when formulating feeds at an industrial scale although recent developments in on-line estimation of chemical composition using near infrared spectroscopy (NIRS) could theoretically assist the industry in addressing this challenge. Another caveat is that none of the species reviewed here can be considered as a perfect substitute to fish meal. Diptera larvae are most similar to fish meal in terms of amino acid composition and protein digestibility, but all insects reviewed in this paper except silkworm pupae have lesser concentrations of sulfur amino acids than fish meal. The absence of EPA and DHA in the fatty acid profile of insects is also a limitation to their inclusion in marine fish diets. Depending on the insect and fish species, supplementation with other sources of amino acids or fatty acids will therefore be required for optimal growth and fish quality. It is also possible to change insect composition through manipulation of their diets.
Before insects can be used for the industrial production of fish feed, research and development are needed in the following areas.
The feasibility of scaling up insect production into an economically viable business able to provide insects in industrial quantities needs to be investigated beyond experimental or pilot units. This includes the development of cost-effective insect diets and the engineering of specific infrastructures, including the automation of rearing to reduce labor costs. For insects to be competitive with the traditional protein sources, they must have distinctive advantages in terms of nutritional value and price and should be available year-round in well-defined and consistent qualities.
Further work is required on the nutritional value of insects for fish feeding, and particularly for carnivorous fish: factors influencing the chemical composition as well as nutrient and energy bioavailability; dietary manipulation of the profiles of amino acids, fatty acids, and minerals; processes (such as defatting and pelleting); palatability and feeding preferences of fish; and adaptation of fish to insect-based diets.
Because one of the main benefits of insects is their ability to turn biowastes into valuable organic matter, sanitation procedures need to be defined for the safe use of substrate to obtain insects that are free of diseases and undesirable substances.
There is a need to develop a regulatory framework and legislations for use of insects as animal feed and to improve risk assessment methodologies.
Studies on the impact of feeding insects on the safety, quality, and social acceptance of fishery products obtained on feeding insects should be conducted.
Life cycle assessments of insect production compared with that of other feed protein production such as fish meal and oilseed meals should be conducted.