Introduction

Food waste refers to garbage generated in daily life, including leftovers, meat residue, etc. The main sources of food waste are household kitchens, public canteens, and catering industries (Baiano 2014; Chang and Hsu 2008; Grizzetti et al. 2013). Polysaccharides (starch), protein, lipid, and cellulose are the main ingredients in food waste (Zhang et al. 2013). In addition, due to the difference in customs, culture, geography, and eating habits, the composition of food waste also has obvious regional characteristics (Chang and Hsu 2008). Non-standard stacking treatment can lead to secondary pollution (e.g., landfill leachate), polluting groundwater or surface water (Li et al. 2014; Pham et al. 2015). The long-term stacking is easy to breed germs and cause disease transmission, and then pose the environment and human health risk (Chang and Hsu 2008; Lebersorger and Schneider 2011; Pham et al. 2015). Food waste has been a global problem and causes economic, environmental, and social consequences (Betz et al. 2015; Brautigam et al. 2014; Grizzetti et al. 2013). The Food and Agriculture Organization of the United Nations (FAO) reported that globally about one-third of food produced is lost or wasted, the amounts to approximately 1.30 billion ton per year (Gustavsson et al. 2011). In South Africa, the output of food waste was estimated at 9 million tons per year (Oelofse and Nahman 2013). In North America, the per capita food waste was 95.0~115 kg every year (Brautigam et al. 2014). In the European Union, food waste was expected to increase from 89.0 million tons in 2006 to 126 million tons in 2020 (Pham et al. 2015). The urban food waste generated at least 60 million tons each year in China, especially in cities with developed catering industry such as Beijing, Guangzhou, and Chengdu (Li et al. 2014; Wen et al. 2016). Therefore, more and more countries are concerned about food waste and advocate recycling, recycling of food waste (Lebersorger and Schneider 2011).

In general, the disposal of food waste is the same as municipal solid waste, which mainly includes filling and incineration (Gustavsson et al. 2011). Landfills take up large amounts of land resources and release greenhouse gases such as methane and carbon dioxide (Kwon et al. 2010; Lo and Woon 2016). In addition, incomplete combustion can cause toxic gases such as dioxin to pose a threat to humans (Li et al. 2014). Food waste is rich in nutrients such as crude protein and fat, and its the most potential resources utilization method was protein feed (Baiano 2014; Pham et al. 2015). The methods of using food waste to produce feed include silaging, low-pressure frying, boiling drying, and high-temperature dry (Sugiura et al. 2009). Due to the high moisture and nutrient contained in food waste, it causes rapid decays and breeds pathogenic bacteria, which can lead to diseases such as foot and mouth disease, toxoplasmosis, and swine fever virus (Chen et al. 2015). Moreover, homologous animal protein feeds have certain risks such as mad cow disease (Gustavsson et al. 2011).

Fly larvae can rapidly digest organic waste such as food waste and livestock manure waste, reducing its volume while absorbing nutrients from organic waste (Cickova et al. 2015). Yang and liu (2014) used Chrysomya megacephala convert about 700 kg of fresh pig manure (73% moisture) within one week, to gain 18.2 kg dried larvae biomass containing. The average yield of larvae Lucilia sericata and Sarcophaga carnaria larvae bred on fish and poultry waste was approximately 304 g/kg (Ogunji et al. 2008). A protein meal and oil derived from the dehydrated fly larvae may become a valuable fodder for livestock and aquaculture (Ogunji et al. 2008). Previous studies showed the larvae meals of the house fly, yellow mealworm, and black soldier fly are excellent sources of the nutrients for broilers chickens and a valuable source of the digestible amino acid (De Marco et al. 2015; Milutin et al. 2008). In addition, fly larvae growing on food waste was also a raw material for biodiesel production (Cickova et al. 2015).

The housefly (Musca domestica L.) is a worldwide health pest, widely distributed in all parts of China, and is the dominant species in most places (Zhu et al. 2015). The larvae, pupae, and adults of houseflies contain about 16 kinds of amino acids with high protein and high purity chitin on their body surface. After being extracted, they are widely used in industry, medicine, agriculture, and food (Cickova et al. 2015). Previous studies showed that using housefly larvae to conduct bioconversion with food waste is a universal, efficient, and environmentally friendly method for dealing with food waste (Cickova et al. 2015; Niu et al. 2017). However, some factors may affect the growth and development of fly larvae in the process of biotransformation of food waste, such as the temperature of the substrate rises, pH changes from neutral to alkaline, ammonia release increases, and moisture decreases, thereby reducing the conversion efficiency of food waste (Cickova et al. 2015).

Therefore, the main objectives of this investigation were to (1) a mixed-level orthogonal array design was employed for the optimum conditions of breeding housefly larvae by food waste, (2) assess the quality and safety of housefly larvae products for feeds, and (3) evaluate the effects on use of fly larvae as a feed for tilapia and the residual manure as organic fertilizer.

Materials and methods

Housefly

To avoid the biological invasion, that housefly (Musca domestica L.) was collected by flies trap cage from a suburb of Chengdu city, Sichuan, China. Adult flies were fed with a milk/sugar mixture (1:1) and water, and cultured in mesh-covered plastic cages (10,000~20,000 adults in 0.50 m3) at 50.0~60.0% relative humidity, 26.0~30.0 °C, 12 h light/12 h darkness (06:00–18:00). The larval oviposition substrates were a mixture of raw pork, dry milk, wheat bran, and water (2:2:5:11 by mass) placed in a tray (20 × 12 × 5 cm). The housefly was cultured for more than 10 generations.

Food waste

In this study, food waste was from local restaurants and hotels, which was mainly post-consumption waste. The staple food waste (cereals) and dish waste (a mixture of meat and vegetables) was separated from the food waste and rinsed four times with tap water to remove most of the lipids and salt, then drained the water and cut into small pieces (about 1.5 cm). The pretreated food waste was stored at room temperature. The pretreatment can effectively reduce the sodium content in food waste; moreover, the bones and other debris are reduced and the nutritional components in the food waste were improved, which is conducive to the growth of housefly larvae in the food waste culture substrate.

Experimental design

A mixed-level orthogonal array (OA) design was employed for the optimum conditions of breeding housefly larvae by food waste. Three variables that may affect the larvae weight, larval yield, crude protein, and food waste mass reduction rate (FMRR) and their possible interactions were studied. These variables selected are as follows: (1) substrate ratio (factor A): The culture substrates consist of staple food waste and dish waste in different proportion at weight, and the ratio was 1:0, 3:1, 1:1, 1:3, 0:1, respectively; (2) Breeding density (factor B): the substrate (1000 g) were placed flat to about 3.00~5.00 cm thickness in plastic bowl (30 × 20 × 12 cm). The housefly eggs were inoculated into the substrate at five types of breeding densities (2.50 g/kg, 5.00 g/kg, 7.50 g/kg, 10.0 g/kg, and 12.5 g/kg). The breeding density had approximately 2000 neonate housefly larvae per gram (< 24 h after hatch); (3) Feeding mode (factor C): the larval stage of housefly is 4~6 days, so this study set three feeding modes of substrate in the larval stage: one-time feeding, daily feeding and feed on demand. Because three-levels and two five-levels variables are to be optimized, the OA25 (31×52) matrix is used to assign the variables considered. The assignment of factors and levels of the experiment using an OA25 (31×52) matrix is shown in Table 1, one column in the three-level design provides only two degrees of freedom, whereas the five-level column has four degrees of freedom (Webster 1990). Under ambient conditions, housefly larvae (1 g) at 2, 3, 4, and 5 days of age was collected from each bowl, and all late instar housefly larvae (near pupation) were separated from residues by sieving (20 mesh). The separated larvae and residues were stored at − 20 °C for subsequent analysis.

Table 1 Experiment results according to the OA25 (31×52) matrix
Full size table

The tilapia (Oreochromis spp) were obtained from Guangxi Fisheries Research Institute, Nanning, China. The initial length and weight of the tilapia fry were 11.0 ± 1.23 cm and 25.0 ± 1.52 g, respectively. There were four diet groups in the feeding experiment. The commercial fish feed (Kangdaer® 825 formulated feed) was used as a control feed. The commercial dried housefly larvae were the product of housefly biotransformation of pig manure, purchased from Dayuan Biotechnology Co., Ltd., China. The fresh housefly larvae were reared under the optimal condition based on the orthogonal test. The larvae were dried at 80 °C for 6 h to make dry larvae. The feeding experiment was carried out in 200-L fish tanks (each tank filled about 150 L of tap water) with continuous aeration (pH, 7.00~8.00; water temperature, 22~25 °C; dissolved oxygen, 6.00~9.00 mg/L; ammonia nitrogen content ≤ 0.02 mg/L; sunshine time, 12 h). Each group has three replicates and 60 fish per fish tank. The fish were fed two times per day (10:00 and 18:00) at 5.00% of body weight for 45 days. In addition, the water of fish tank was refreshed once every three 3 days for maintaining water quality. The fish were randomly sampled after all the fish stop feeding for 24 h and the lengths and weights of the fish were measured.

Chemical analyses

Experimental housefly larvae, fish feeds, and fish were analyzed for crude protein, crude fat, ash, amino acids, potassium (K), sodium (Na), phosphorus (P), and calcium (Ca) according to the standard method by China National Standards Management Department (GB13078 2017; GB19164 2003). For the organic fertilizer, the contents of moisture, organic matter (OM), total nitrogen (TN), total phosphorus (TP), and total potassium (TK) were determined following the national standard of China (NY525 2012).

0.5 g of freeze-dried fish samples was weighed to determine the trace element concentrations. Concentrations of cadmium (Cd), lead (Pb), chromium (Cr), copper (Cu), nickel (Ni), and zinc (Zn) in the samples were determined by ICP-OES (Perkin-Elmer, Elan 8300, Norwalk, CT), followed the procedures described previously (USEPA 1996). All of the samples were tested three times. Two standard reference materials (SRMs) were obtained from the National Institute of Standards and Technology, USA, (NIST 1566b, oyster tissue) and the National Research Council of Canada (TORT-2, lobster hepatopancreas). The recoveries for 7 elements ranged from 87 to 106%.

Statistical analyses

Each statistical test was performed using SPSS 20.0 for Windows. The arithmetic mean for each kind of index was used to compare other studies. Normality of the data was checked by Shapiro-Wilk test. Means of different groups were compared using one-way ANOVA test. Data of mixed-level OA (31×52) were analyzed using correlation analysis. The probability value of p < 0.05 was set as the level for statistical significance.

Result and discussion

Food waste

Due to differences in living areas, living standards, eating habits, and other factors, the properties of food waste were also different. Therefore, this study divides food waste into staple food waste and dishes waste and treats them separately. After pretreatment, the content of crude protein, crude fat, and sodium in dishes waste was about 25.4 ± 2.99%, 35.7 ± 1.99%, and 9.24 ± 0.18% respectively, while the crude protein content of the dish food waste increased by about 5.00% (p < 0.05), and the sodium content decreased by about 0.86% (p < 0.05). It may be due to the reduction of the influence of bones, paper towels, and other debris after pretreatment of food waste. Salinity in the culture substrate is an important factor affecting the growth of insects (Cickova et al. 2015). There was no significant difference in crude protein (9.01 to 12.0%), crude fat (1.06 to 1.41%), sodium (0.10 to 0.18%), potassium (0.23 to 0.43%), and phosphorus (0.80 to 0.09%) contents between pre-treated staple and raw staple food waste (p > 0.05). In most methods of using fly larvae to convert food waste, the food waste was homogenized and fed directly, but the excessive salt and water in the food waste slurry are not conducive to the growth and conversion of fly larvae (Cickova et al. 2015; De Marco et al. 2015). Therefore, it is necessary to add supplementary materials (such as wheat bran and sawdust) to adjust the properties of food waste culture substrate to maintain the conversion efficiency of food waste and larval production, but a large number of supplementary materials will inevitably increase the production costs. Classifying food waste (staple food and dishes) and applying it proportionally can not only avoid excessive moisture and oil in the culture substrate but also increase the contact area between food waste and air, which is conducive to the biodegradation of food waste by fly larvae (Wang et al. 2017). In addition, because of the low content of crude fat and sodium in staple food waste, it can replace the supplementary materials to regulate the properties of food waste culture substrate, thereby reducing the input of supplementary materials in the process of larvae converting food waste and saving costs.

Feeding housefly larvae with food waste

Under the conditions of temperature 25.0 to 35.0 °C and relative humidity 50.0 to 60.0%, the adult life span of the house fly larvae fed with food waste was 30–50 days, and the oviposition period is 10 to 25 days. The adult feedstuff, oviposition bait, and the density of adult flies could affect the oviposition rate of adult flies (Cickova et al. 2015). In this study, the density of housefly adults is 30.0 to 50.0 thousand, and the daily egg production could reach 2.00 to 6.00 g/day, which can meet the follow-up experiment. The housefly eggs can be hatched 6 to 10 h later. The larval stage of a housefly is about 4 to 6 days, which is the key period in the life of the housefly (Zhu et al. 2015). The individual size, reproductive efficiency, and even life span of houseflies are affected by the development degree of the larval stage.

The mixed-level OA test program and results are listed in Table 1. The results showed that the effects of these factors (culture substrate ratio, A; breeding density, B; feeding mode, C) on the weight of 50 larvae (R1: A, 0.42; B, 0.40; C, 0.13), larvae yield (R2: A, 100; B, 54.6; C, 11.2), and crude protein content (R3: A, 13.7; B, 7.97; C, 6.06) were A > B > C. In addition, the culture substrate ratio had a positive effect on the weight of 50 larvae and larvae yield (p < 0.05). The animal and plant protein content in dishes waste is rich and higher than that in staple foods, and these proteins can promote the growth and development of animals (Titi et al. 2000; Wen et al. 2016). With the increase of the proportion of dish waste, the crude protein and crude fat content in the culture substrate also increased; thus, the energy and utilization efficiency in the substrate was improved, which was conducive to the growth and nutrient absorption of housefly larvae. The breeding density was positively correlated with larvae yield and negatively correlated with 50 larvae weight (p < 0.05). Under the same nutritional conditions, the yield of housefly larvae increased with the increase of breeding density (p < 0.05). When the breeding density was 7.5 to 10.0 g/kg, the maximum larval yield (133 to 176 g/kg) could be obtained. However, when the breeding density continued to increase, the yield tended to be stable. At relatively high breeding density, the larvae cannot take enough nutrients for growth and development, resulting in small, incomplete development, or even death of housefly larvae, which leads to the stability or even decline in yield (Green et al. 2003). Similar results were also found in the study of Manduca sexta caterpillars (Petersen et al. 2000). The culture substrate ratio had a positive effect on protein content (p < 0.05). This is because the housefly larvae feed on high-protein feeds, and they obtain more nutrients, which in turn increases the crude protein content of the larvae, and the result is similar to the study on bee larvae feeding different proportions of feeds (Fang et al. 2010). When the culture substrate ratio (staple food waste: dishes waste) was 1:3, the crude protein content of larvae reached the maximum value (56.6–58.2%). However, as the proportion of dishes waste increased, the amount of crude fat in the culture substrate also increased (p < 0.05). Hu (2012) observed that the fly larvae would use the crude fat in the culture substrate to convert into their own fat, thus reducing the proportion of protein in the body.

The food waste mass reduction rate (FMRR) increased with the increase of breeding density (p < 0.05), and its peak value appeared at 10.0 g/kg (83.6%). The structure of culture substrate changed in the process of housefly larvae converting food waste. Due to the fly larvae are photophobic, most larvae biodegrade food waste by drilling and peristalsis under the surface of the culture substrate, resulting in increased the temperature and pH of the substrate, increased release of ammonia and moisture, and promoting the growth of aerobic microorganisms (Beard and Sands 1973; Zhu et al. 2012). On the other hand, the number of larvae converting a certain amount of food waste increased with the increase of breeding density, leading to the relative lack of food in the housefly larvae, forced to feed on foods with low nutritional value such as vegetable protein and crude fiber to maintain their survival (Cickova et al. 2015). Therefore, breeding density plays an important role in the conversion rate of food waste. But, when the breeding density increased to 12.5 g/kg, the average FMRR decreased to 70.7%. This may be due to the excessive feeding density which leads to the weight reduction and development imperfection of housefly larvae, resulting in the decrease of feeding capacity and activity of the larvae (Wang et al. 2013), thus affecting the larval yield and the FMRR. As a result, a moderate increase in breeding density could promote the conversion efficiency of housefly larvae to food waste.

There was no significant correlation between the feeding mode and the growth of housefly larvae or the FMRR (p > 0.05) (Table 1). Due to the fecundity of female flies decreases after 30 days in the adult stage, a sufficient number of strong and healthy individuals are required to maintain the reproductive capacity of the population (Zhu et al. 2015). Therefore, the optimum conditions for the housefly larvae to convert food waste were as follows: culture substrates ratio 1:3, breeding density 10.0 g/kg, and all substance added in the first day, and the verification test results showed that the larvae yield, protein content, and FMRR were 153 g/kg, 56.1%, and 77.3%, respectively.

Quality of larvae products

The composition of the housefly larvae fed with food waste was shown in Fig. 1. The crude protein content of housefly larvae (56.1 ± 2.12%) was similar to that of the housefly larvae after bioconversion of pig manure (56.9 ± 2.49%) (Wang et al. 2013; Zhu et al. 2015), and reached the second grade of Chinese fish meal standard (crude protein content ≥ 55%) (GB19164 2003). In addition, the essential amino acids (Eaa) (Thr, Val, Met, Leu, Phe, Ile, Cys, and Lys) accounted for approximately 45.1% of the total amino acids (Taa) and the contents of nonessential amino acid (Naa) Glu (7.42%) and Asp (5.21%) were higher than other amino acids (Fig. 2). The value of Eaa/Taa (45.1%) and Eaa/Naa values (0.83%) were in line with the Food and Agricultural Organization (FAO) and World Health Organization (WHO) requirements for feed protein (Eaa/Taa≈40% and Eaa/Naa > 0.6) (WHO 1979, 1985). Therefore, larvae products have the potential to be a good source of protein feed.

Fig. 1
figure 1

The composition of housefly larvae fed with food waste

Full size image
Fig. 2
figure 2

The composition and content of larvae amino acid

Full size image

However, the crude fat content of housefly larvae fed with food waste (30.1 ± 1.18%) was higher than of the housefly larvae after bioconversion of pig manure (22.0%) and the fish meal standard of China (Fig. 1). Due to the higher crude fat content in food waste, housefly larvae can convert various fats (animal, vegetable, or mixed) into self-fat, and then relatively reduce their own protein content, thus affecting the quality of housefly larvae as a feed protein source (Cickova et al. 2015; Hu 2012). Previous studies have found that about 87.8% of fatty acid methyl esters can be extracted from the larvae of the Chrysomyia megacephala larvae with the oil content of about 24.4 to 25.3%, and properties of the obtained methyl esters were within the specifications of the EU biodiesel standard (Li et al. 2012). Therefore, the potential of the housefly larvae fat component after the conversion of food waste and its raw material for biodiesel production deserves further study. The housefly larvae fed with food waste also represented a potential alternative feedstock for biodiesel production.

Residue

Physicochemical parameters of food waste residues after bioconversion by housefly larvae were shown in Table 2. The moisture content in the residue was 25.4 ± 5.02%, which is about 59.5% lower than that in the food waste, and the dry matter mass of food waste was reduced by 75% through bioconversion of housefly larvae. The similar results were also found in other studies of insect bioconversion organic waste, such as earthworm (Libyodrilus violaceus), black soldier fly (Hermetia illucens L.) Larvae, and Chrysomyia megacephala (Li et al. 2012; Loh et al. 2009). The residues in this study contained higher TP (0.93 ± 0.23%) and TK (1.20 ± 0.11%), and lower moisture than that in the residues of housefly larvae conversed pig manure (Zhu et al. 2012). In addition, the contents of total nutrients (N+P+K ≥ 5.0%) and heavy metals (As ≤ 15 mg/kg, Hg ≤ 2 mg/kg, Pb ≤ 50 mg/kg, Cr ≤ 150 mg/kg, Cd ≤ 3 mg/kg) in the residues of this study met the Chinese agriculture industry standard for organic fertilizer (NY525 2012). This indicates that the residue converted from food waste could be used as organic fertilizer.

Table 2 Physicochemical parameters of food waste residues after bioconversion by housefly larvae
Full size table

Tilapia

Effects of four feeds on physiological and biochemical indexes of tilapia for 45 days are shown in Table S2. Tilapia fed with commercial and experimental dried housefly larvae presented the best growth performance and nutrient content in the experiment group (p < 0.05). Therefore, the housefly larvae could be used as fish feed or instead of fish meal to produce fish feed, which achieves a similar or better yield and quality than commercial feed. Although the tilapia feed with fresh larvae had the worst growth performance and nutritional concentrations, its survival rate (90%) was the highest (p < 0.05). The relatively high crude protein in the fish feeds can promote the growth of cultured fish, but excessive intake of crude protein by farmed fish may cause its metabolism to produce excessive ammonia-nitrogen-containing excreta, and lead to the decline of water quality (Chen et al. 2002; Wong et al. 2004). Furthermore, too much protein would have toxic effects on the fish body, affecting the growth and development of the fish (Ogunji et al. 2008). As a live bait, the fresh housefly larvae were rich in vitamins A and D, and a variety of bioactive substances (such as antibacterial peptides, antibacterial proteins, and chitin), which could improve the immune capacity of the fish fed with fresh housefly larvae, and then improve its survival rate (De Marco et al. 2015; Lee et al. 2008). In addition, the trace elements concentration in tilapia raised with the four kinds of feeds complied with the maximum levels of contaminants in foods in both China and WHO (FAO/WHO 1995; GB2762 2017). These findings show that the housefly larvae products that converted food waste are suitable for use in the production of fish feed.

Conclusion

After pretreatment, the protein content of dish food waste increased by 5%, and the sodium content decreased by 0.86%. The mixed-level OAD test results showed the culture substrate ratio had a positive effect on the weight of 50 larvae, protein content. and larvae yield. Under the same nutritional conditions, the yield of housefly larvae increased with the increase of breeding density, and when the breeding density was 10.0 g/kg, the maximum larval yield and food waste mass reduction rate could be obtained. The value of E/T and E/N values in the housefly larvae fed with food waste were in line with the FAO and WHO requirements for feed protein. However, the crude fat content of housefly larvae fed with food waste was higher than the fish meal standard of China. The contents of total nutrients and heavy metals in the residues of this study met the Chinese agriculture industry standard for organic fertilizer. Tilapia fed with commercial and experimental dried housefly larvae presented the best final weight, weight gain ratio, and protein content. In addition, the trace elements concentration in the cultured tilapia were all met the maximum levels of contaminants in foods in both China and WHO. These findings show that food waste feeds are suitable for use in the production of fish feed.