Crude Protein, Amino Acid, and Iron Content of Tenebrio molitor (Coleoptera, Tenebrionidae) Reared on an Agricultural Byproduct from Maize Production: An Exploratory Study

Abstract

Edible insects offer environmental and nutritional benefits, as they are characteristically nutrient-dense, are efficient biotransformers of organic material, and emit fewer greenhouse gasses than traditional livestock. Cultivating Tenebrio molitor (yellow mealworm) as ‘minilivestock’ is one possible means of increasing access to insect protein for food insecure populations. Tenebrio molitor growth and nutrient content varies with diet and rearing conditions, but little is known about the precise impact of poor quality feedstocks, such as maize crop residue (stover). Stover is widely available across sub-Saharan Africa where maize is a common dietary staple. Early instar larvae were reared under controlled conditions on three feed substrates: a standard control; a mixed soy, maize grain, and stover diet; and a 100% stover diet. Larvae reared for 32 d were analyzed for total amino acid profile, crude protein, and iron content. Larvae fed the three diets contained all essential amino acids for human nutrition and compared favorably to other traditional protein sources. The mixed diet contained 40% stover by weight and yielded amino acid values similar to the control diet, suggesting that some grain feedstock could be replaced with stover without hampering nutrient content. A second experiment demonstrated that T. molitor were able to complete metamorphosis and survive on a 100% stover diet for multiple generations. These results suggest that stover could be a suitable dietary component for T. molitor, which could facilitate the development of low-cost insect farming systems in low-resource settings that stand to benefit from increased access nutrient-dense edible insects.

Edible insects have been identified as one strategy to alleviate global hunger, malnutrition, and food insecurity. They provide a high-quality, digestible animal protein source (Verkerk et al. 2007, Belluco et al. 2013, Finke 2015), with fewer environmental ramifications than conventional livestock (Collavo et al. 2005, Oonincx et al. 2010, van Huis et al. 2013b). Insects have a low feed-conversion ratio (Collavo et al. 2005), large edible body mass percentage (Nakagaki and DeFoliart 1991), and most do not produce methane (Hackstein and Stumm 1994). Historically, edible insects have been an important component of the human diet across the globe (Bodenheimer 1951, Defoliart 1995). They are typically rich in fat (Belluco et al. 2013), with numerous species supplying ample polyunsaturated fatty acids including essential linoleic and linolenic acids (Womeni et al. 2009). Insects are also generally good sources of minerals including potassium, calcium, copper, magnesium (Schabel 2010), manganese, phosphorous, and selenium (Finke 2002, Rumpold and Schlüter 2013a), as well as iron and zinc (Christensen et al. 2006), which are crucial for growth and development. Iron and zinc are also relevant for health and well-being in food insecure populations that may not meet dietary requirements for these micronutrients.

Today, entomophagy (the practice of eating insects) is commonplace for an estimated 2 billion people (van Huis et al. 2013b) from 3,000 ethnic groups in 130 countries (Ramos-Elorduy 2009), who select from more than 1900 known edible species (van Huis et al. 2013a). In Africa alone, at least 500 species are consumed across 40 nations (Jongema 2014). Some regions boast abundant, readily available, affordable supply of edible insects (Illgner and Nel 2000), often a cheaper protein alternative to meat (Raksakantong et al. 2010, van Huis et al. 2013a). However, insects are not always accessible in other areas. The vast majority (~92%) of edible insects consumed globally are collected from the wild (Rumpold and Schlüter 2013b, van Huis 2013, Jongema 2014), and many are seasonal (van Huis et al. 2013a); not surprisingly, consumption also tends to be seasonal (Takeda and Sato 1993, van Huis et al. 2013a).

While in season, wild-harvested insects provide crucial and free food resource, but long-term sustainability and safety require further examination. Entomophagy may be of particular importance for food insecure populations, as it is at times associated with the inaccessibility of staple foods (van Huis et al. 2013a), being a crucial ‘safety net’ (Clarke and Grundy 2004), and providing both food and income that rural communities depend on (Makhado et al. 2009, 2014; Lucas 2010). The future of entomophagy in these contexts is uncertain, however. Growing demand and higher prices could incentivize overexploitation of wild insects (Ramos-Elorduy 2006), and reports suggest that habitat mismanagement and overharvesting are already potentially damaging insect populations and ecosystems in Southern Africa (Faure 1944, Roberts 1998, Illgner and Nel 2000). Moreover, wild-harvested insects may not always be safe to eat. There are a few examples of heavy metal toxicity and pesticide contamination in edible insects (Saeed et al. 1993, Handley et al. 2007). Since insects can bioaccumulate some metals (Diener et al. 2015, van der Fels-Klerx et al. 2016) and the diet of wild-harvested insects is unknown, safety cannot be guaranteed.

One means to boost access to safe insect foods year-round without risking ecological damage is strategic insect cultivation (also called ‘minilivestock’ farming). Minilivestock farming for human consumption is popular in parts of Asia (Yen 2015) and is also gaining momentum in Europe and North America. Some edible insects can be reared on organic side-streams, such as agricultural byproducts or organic waste (van Huis et al. 2013a). This may add value to and serve as a recycling mechanism for biomass not suitable as food for humans or other livestock. It may also be feasible to farm insects indoors, and/or vertically, thus separating production from the need for arable land (Dunkel and Payne 2016).

Tenebrio molitor Linneas, 1758, also called the yellow mealworm, may be especially suitable as minilivestock because they are easy to farm and nutritionally dense. Generally considered a pest of stored grain, T. molitor is found across the globe and is frequently reared at a commercial scale as feed for reptiles, fish, and birds. T. molitor offers a promising alternative to soybean meal for poultry and livestock (Ramos-Elorduy et al. 2002, van Huis et al. 2013a). Humans consume T. molitor larvae in many contexts, particularly across Southeast Asia. The insects breed easily, have a relatively short lifespan, and are typically protein-rich with reported crude protein values ranging between 44 and 68% by dry weight (Finke 2002, Ghaly and Alkoaik 2009, Rumpold and Schlüter 2013c, Oonincx et al. 2015, Alves et al. 2016, Zhao et al. 2016). This protein is high quality, as T. molitor reared on an ideal diet contain all essential amino acids for human nutrition (Jones et al. 1972, Aguilar-Miranda et al. 2002, Ghaly and Alkoaik 2009). Additionally, T. molitor larvae are a good source of iron. Mwangi et al. (2018) calculated the average iron content of T. molitor larvae from seven studies. At 5.3 mg/100 g dry matter, T. molitor larvae contain more iron by weight than pork or chicken, but less than beef on average (Mwangi et al. 2018).

Researchers have established the nutritional requirements of T. molitor for optimal growth (Fraenkel 1950), and the ideal insect diet is grain-based, usually containing mixture of grains, and raw fresh vegetables (Davis 1975, Lyons 1991). T. molitor larvae require the same 10 amino acids essential for growth and development in other animals, while amino acids including serine, tyrosine, glutamic acid, and glycine can be deleted from the diet without having a negative impact on growth (Davis 1975). Methods for rearing T. molitor in captivity vary, but most use wheat brain or wheat flour, nutritional or brewer’s yeast, and a vegetable water source to create a control or optimal diet. It should be noted that T. molitor are considered scavengers (Rees 2004), however, capable of consuming a wide variety of organic materials and wastes (Ramos-Elorduy et al. 2002); thus, investigations into other possible feedstocks for T. molitor cultivation are warranted. Successful minilivestock farming in low-resource settings will require access to a low-cost, consistently available insect feedstock.

While T. molitor has been studied extensively, the nutritional value of larvae reared on variable diets and under variable conditions is less well understood. Oonincx et al. 2015 analyzed the chemical composition of T. molitor reared on four different diets composed of food manufacturing byproducts. The researchers concluded that it is possible to produce T. molitor on food byproducts, but survival, development time, and feed-conversion efficiency are strongly influenced by diet composition. Another study investigated the food value of T. molitor reared on pulp flour from palm trees (Acroncomia aculeate), finding that larvae reared on a mixture of pulp flour, wheat flour, and soybean flour remain a good source of protein and lipids (Alves et al. 2016). Elsewhere, supplementing T. molitor diets with calcium, has been shown to increase the calcium content of the larvae (Klasing et al. 2000). Other researchers reared T. molitor larvae on five experimental diets containing a wide range of organic waste and then fed them in equal protein ratios to broiler chickens. No significant difference between treatments in broiler chickens after 15 d were observed, though the amino acid profiles of the larvae did differ by diet (Ramos-Elorduy et al. 2002). As others have found (Rumpold and Schlüter 2013b), low-nutrient diets and the quality of insect feed impacts the nutritional composition of cultivated insects (Oonincx et al. 2015). More information about the impact of insect diet on nutrient content, growth, and progeny is needed.

The feasibility of rearing T. molitor as minilivestock for human consumption in development contexts may hinge on the availability of an appropriate feedstock. Ideal grain sources may not be readily available or affordable in all regions. Likewise, diverting human food to feed insects could be problematic in situations of poverty and food insecurity. Agricultural byproducts that are inedible for humans may offer one tenable, widely available, and relatively inexpensive feedstock for T. molitor. In Zambia, for example, maize is the most abundant food crop, as approximately 1,253,664 hectares were planted in 2011/12 (Tembo and Sitko 2013). Maize crop residue (henceforth referred to as ‘stover’).the resulting leaves, stalk, husks, and tassels that remain after harvesting maize grain, is one possible insect feed option that has not been assessed. Stover is carbohydrate rich, but notoriously low in protein and nitrogen because at the time of harvest, the maize grain retains more than 60% of total nitrogen in the plant, leaving the rest divided among the stalk, cob, and leaves (Hoeft et al. 2000). Stover is nutritionally inferior to wheat bran and oats, containing about 5% protein (Eastridge 2007) versus 16.9% in oats and 15.5% in crude wheat bran (USDA 2018). If T. molitor can survive on this crop residue, farming T. molitor could be plausible for smallholder maize farmers who otherwise have limited resources for rearing minilivestock.

The purpose of this exploratory pilot study was to measure the nutritional content of T. molitor reared exclusively on stover, which is widely available in Southern Africa. To do this, we conducted a feed study to assess and compare protein, total amino acid profile, and iron content of larvae reared on three diets for 32 d. Study conditions (temperature and humidity) were selected to emulate possible average real-world minilivestock farming circumstances in tropical and humid subtropical regions of Southern Africa, such as Zambia, where agricultural byproducts are an available feedstock.

Materials and Methods

Insect Origin and Diets

Stock culture of early instar, mini-mealworms (~0.635 cm) were obtained from a U.S.-based commercial supplier (Rainbow Mealworms, Compton, CA). Larvae were reared in a laboratory using a Percival Scientific Chamber (Percival Scientific Inc., Perry, IA) to control temperature, humidity, and light.

Three feed mixtures were tested: a control diet based on Weaver and McFarlane (Weaver and McFarlane 1990), a ‘mixed’ diet modified from ZhiXin et al. 2011, and a 100% stover diet (Table 1). Both soy and maize are grown throughout the region and could be utilized to feed insects. Combining these feedstocks with stover allowed us to assess a feedstock option for contexts where wheat bran and oats are not readily available or affordable. USDA certified organic carrots were purchased at local supermarkets for use as a moisture source. USDA certified organic grains (wheat bran, oats, corn meal, soy flour) for the feed substrate mixtures were obtained from a local grocery store, and laboratory grade brewer’s yeast was purchased from a chemical supplier. Certified organic stover was obtained from the Arlington Agricultural Research Station of the University of Wisconsin–Madison, USA. This high moisture (72–75%) maize grown for silage with immature cob formation was harvested and dried in paper bags at 48.89°C (130°F) for one week and then chopped into small pieces (~0.25 cm) prior to use. The dried stover was predominately stalk.

Feed substrate diets (A-C) were created by mixing specific grains, stover, and brewer’s yeast at a percentage by dry weight basis. Four slices of carrot (~0.635 cm thick) were added to the top of the substrate twice per week (Ravzanaadii et al. 2012) to provide moisture. Any remaining old carrot was removed when new carrot was added. Additional feed substrate mixture was added as needed to retain a 2.54 cm depth of the whole substrate bed.

Experiment 1: Nutritional Composition of Larvae Reared on Three Diets

Larvae were reared based on a modified version of the methods described in Ghaly and Alkoiak (Ghaly and Alkoaik 2009). Upon arrival, larvae were fasted for 24 h to clear the gastrointestinal tract of residual food. Next, 500 T. molitor larvae of similar size were weighed and placed in a 22.2 × 22.2 × 6.8 cm square plastic container with an unsealed lid to allow airflow (Ghaly and Alkoaik 2009). Each container was filled with one of the three feed mixtures to a depth of approximately 2.54 cm. Each test was replicated three times in separate containers. T. molitor were reared at 25 ± 1°C and 50 ± 5% relative humidity with a 12-h light:12-h dark photoregime 32 d on one of three feed mixtures (A-C) outlined below (Table 1). Larvae were monitored three to four times per week, and three random samples of 20 T. molitor larvae from each bin were weighed every 3–4 d to calculate average period growth per feed group. Larvae were randomly selected by slightly mixing the entirety of feed substrate and larvae and then scooping approximately ½ cup out. Next, 20 T. molitor larvae were selected from the cup, carefully separated from the substrate using forceps, and weighed collectively using an analytical balance (AG285, Mettler Toledo, Columbus, OH).

When monitoring larvae, the feed substrate was inspected for the presence of dead larvae or pupae. Dead larvae and pupae were identified visually by abnormal color (i.e., black or dark brown color), dehydrated body mass, lack of movement, or changes in body shape. They were removed from the container. At 32 d, T. molitor larvae were harvested, separated from the feed substrate, and weighed before analyzing for nutritional content including crude protein, iron, and amino acid concentration. Whole insects were frozen and held at −20°C until analyzed.

Total crude protein content was determined at Covance Laboratories (Madison, WI) using the Kjeldahl method (AOCS 2011). First, the protein and other organic nitrogen in the sample were converted to ammonia by digesting the sample with sulfuric acid containing a catalyst mixture. Next, the acid digest was made alkaline, and the ammonia was distilled and titrated with standardized acid. The percent nitrogen was determined and converted to protein using the factor 6.25, which is the recommended nitrogen conversion factor of the American Oil Chemists’ Society (AOCS) Ac 4–91 (AOCS 2011).

TAAP were determined using standard automated precolumn derivatization and high-performance liquid chromatography (HPLC) (Schuster 1988, Barkholt and Jensen 1989, Henderson et al. 2000, Henderson and Brooks 2010) at Covance Laboratories. Tryptophan was analyzed separately (AOAC 2012) again using automated precolumn derivatization and HPLC (Schuster 1988, Henderson et al. 2000, Henderson and Brooks 2010). Iron content was determined using inductively coupled plasma emission spectrometry (AOAC 2012) at the same facility.

Experiment 2: Observation of Feasibility and Reproduction on Stover Diet

To further investigate the initial feasibility of rearing T. molitor through the complete life cycle using stover as a feedstock in a low-resource setting, we conducted a second experiment whereby 200 T. molitor larvae of similar size were weighed and placed in plastic containers (22.2 × 22.2 × 6.8 cm) with an unsealed lid to allow airflow. Each container was filled with either Diet A (control) or Diet C (stover) to a depth of approximately 2.54 cm. Each test was replicated three times in separate containers. T. molitor were reared at 25 ± 1°C and 50 ± 5% relative humidity with a 12-h light:12-h dark photoregime and observed through development for 7 mo, until second-generation larvae, pupae, and beetles emerged. Pupae were placed in separate bins with the same feed substrates and allowed to complete metamorphosis. Beetles were observed and cared for in the same way as larvae, and allowed to complete their full life cycle.

Statistical Analysis

Statistical results for both experiments were analyzed using R version 3.3.3 statistical programming language and included descriptive statistics, one-way ANOVA with Tukey’s honest significant difference (HSD) post-hoc test, and growth rate. Growth rate was calculated by linearizing changes in weight for larvae reared on each diet during the experimental period and reporting the slope. Results are expressed as a mean ± SEM (standard error of the mean) with probabilities of P < 0.05 considered significant.

Results

Experiment 1: Growth Rate

There was no significant difference between the average starting weights for larvae in the three diet treatment groups at experiment initiation determined by one-way ANOVA (F = 1.5865; df = 2,6; P = 0.28), but average final weights were significantly different (F = 220.11599; df = 2,6; P < 0.001; Table 2). Post-hoc comparisons using the Tukey HSD test indicated that larvae reared control diet (M = 101.66, SD = 4.6) were significantly heavier than both the mixed diet (M = 70.12; SD = 1.74) and the stover diet (M = 46.75; SD = 2.72) at time of harvest. While all three diets were accepted by T. molitor larvae, growth rate differed across groups. Changes in T. molitor larval weight over time are depicted in Fig. 1. When comparing percent of weight gained calculated as (Final Weight – Initial Weight) / (Initial Weight) × 100%, larvae reared on Diet A (Control) gained 36.2% more weight on average than those reared on Diet B (mixed) and 62% more weight than those reared on Diet C (stover), respectively, reaching a maximum average weight of 101.66 mg per larva during the experimental period. Those reared on Diet B (mixed) gained less weight than the control (reaching an average weight of 70.12 mg) but more than those feeding on Diet C (stover), the lowest-nutrient diet (46.75 mg).

As can be seen in Table 2, Diet A (control) yielded the highest growth rate, followed by Diet B (mixed) and Diet C (stover). Growth rate of Diet A (control) was 3.65 mg per day, about 2.39 times that of Diet C (stover), and 1.56 times that of Diet B (mixed).

Experiment 1: Crude Protein Content

T. molitor larvae reared in this study contained variable amounts of crude protein after 32 d. Those grown on Diet A (control) had the highest concentration of protein per gram bodyweight and those grown on Diet C (stover) had the lowest (Table 3). Crude protein levels in T. molitor larvae ranged between 15 and 19.93 g per 100 g.

Experiment 1: TAAP

The TAAP of larvae fed diets A, B, and C are shown in Fig. 2. T. molitor larvae reared on all three diets contained all dietary amino acids and all nine essential amino acids needed for human nutrition (histidine, isoleucine, leucine, lysine, methionine, phenylalanine, threonine, tryptophan, and valine [NRC 1989]) in varying concentrations.

Larvae fed both Diet A and Diet B differed from those fed Diet C (stover) for all amino acids based on results of one-way ANOVA comparing the mean amino acid concentration in larvae across treatment groups and post-hoc comparisons using the Tukey HSD test. Diet C (stover) yielded larvae with significantly lower content of all amino acids (mg/g) in the whole insect. However, larvae fed Diet A (control) and Diet B (mixed) differed only in tyrosine content, determined by one-way ANOVA (F = 88.04298; df = 2,6; P = 0.0004). Post-hoc comparisons using the Tukey HSD test indicated that larvae fed the control diet had significantly more tyrosine with an average 11.84 mg/g (SD = 0.42) versus 10.24 mg/g (SD = 0.29) for the mixed diet.

Experiment 1: Iron Content

The average iron content in T. molitor larvae grown on Diets A-C ranged from 0.0143 mg/g whole insect (SD = 0.0002) in Diet B (mixed) to 0.0169 mg/g (SD = 0.0002) in Diet C (stover), as shown in Fig. 3. The average iron content of larvae reared on Diet C (stover) was significantly higher than that of insects reared on either Diet A (control) or Diet B (mixed) feed (F = 15.21951; df = 2,6; P = 0.00446).

Experiment 2: Ability to Complete Metamorphosis

To determine if pure stover, as a simple and free feedstock, is a viable option for mealworm production long-term in low-resource settings, we assessed the potential for T. molitor to survive beyond the experimental period (32 d) and undergo full metamorphosis into reproductive adults by repeating the experiment using the same growing conditions from experiment 1 for both Diet A (control) and Diet C (stover). There was no significant difference between the average starting weights for larvae in the two diet treatment groups at experiment initiation determined by an independent-samples two-tailed t-test (P = 0.9132). This experiment continued until second-generation beetles emerged to confirm the production of viable offspring. We observed that T. molitor larvae reared on both Diet A (control) and Diet C (stover) became reproductive adults, and second-generation larvae also successfully hatched, grew, pupated, and became adult beetles.

Discussion

In this study, we assessed the overall effect of low-nutrient stover on larval growth, nutritional composition (crude protein, TAAP, and iron content), and basic ability to reproduce. We evaluated the nutrient content of T. molitor reared under controlled temperature, humidity, and light conditions on three diets: 1) Diet A (a control diet); 2) Diet B (a mixed diet), which was specifically selected to reflect alternative, nonwheat bran, high-nutrient feed substrates that might be available in agrarian Southern Africa; and 3) Diet C (a stover diet) generated from 100% agricultural byproduct from maize production. Our feasibility experiment for low-resource settings assessed the possibility of rearing T. molitor through its full life cycle on the most meager feedstock, and thus compared only Diet C to Diet A. Given that the purpose of this screening study was to evaluate the nutrient content of larvae and basic production feasibility on a meager feedstock, we did not assess the nutrient composition of the insect diets themselves. Nutrient composition of crop residue changes during the process of drying as it turns from green to brown; so, future studies should be designed to measure the nutrient composition of input feedstocks at various times.

In experiment 1, T. molitor developed normally and survived on all three diets for the study duration (32 d), demonstrating the flexibility of T. molitor to eat variable feed substrates. The control diet facilitated T. molitor larval growth and development better than the mixed and stover diets. None of the trials yielded weight losses however, and all feed substrates fostered larval growth. Protein content and individual amino acid concentrations of the larvae reared on Diet C were lower than the other two diets, but remarkably high considering the low-nutrient quality of stover. Compared with maize grain, stover contains significantly less protein and energy (Hoskinson et al. 2007, Karlen et al. 2015). Larvae in this experiment performed well on Diet B, a mixed diet that contained 40% stover, suggesting that insect farmers may be able to supplement some high-quality feed with stover and still yield an edible product with high protein and iron content, provided that farmers have access to such a higher-quality supplement. Diet B (mixed) yielded larvae with comparable amino acid content to the control diet at harvest.

T. molitor, like other edible insects, are a good source of high-quality protein. T. molitor larvae in this experiment had crude protein levels between 15 and 19.93g per 100g, yielding between approximately 37 and 49% protein by dry weight. These values are lower than those previously reported, as Ghaly and Alkoaik (2009) calculated protein levels between 63.31 and 68.86% by dry weight in T. molitor larvae reared on a commercial wheat germ and oat feed (though exact rearing conditions in that study were not explained). Results from our study are consistent, however, with previous work that has established that edible insects contain high crude protein levels (Melo et al. 2011, Belluco et al. 2013), with values between 40 and 75% of dry weight (Schabel 2010). This protein quality of edible insects is typically excellent (Collavo et al. 2005), with some insects containing more protein than ground beef on a dry matter basis (Onyeike et al. 2005, Ramos-Elorduy 2005, Banjo et al. 2006, Gahukar 2011). While crude protein is a good approximate of true protein for most insects (Finke 2007), chitin does contribute to nonprotein nitrogen present in the insect body and crude protein cannot provide detail on the content of specific amino acids. Such information is useful in determining the utility of insects to meet dietary needs and the health of the insect reared on low-nutrient feed substrates.

Additionally, the growth trajectory for T. molitor in this experiment was higher and faster in larvae fed the control and mixed diets, compared to those reared on stover alone. T. molitor grow faster and gain more weight when they consume a diet that contains optimal dietary protein, including the same 10 amino acids essential for rats and other vertebrates (Davis 1975). A future experiment should measure the nutrient composition of feed substrates. Despite being of lower nutritional quality, larvae reared on stover alone still had adequate concentrations of essential amino acids for human nutrition and iron. In experiment 2, T. molitor reared on both Diet A and Diet C completed metamorphosis and reproduced. More research is needed to assess impact of diet on reproductive capacity and survival, but this study finds that T. molitor can survive and reproduce without an ideal diet. This finding is of significance for the potential to farm this insect in a low-resource context.

TAAP is a good proxy for protein quality. The TAAP of T. molitor larvae observed in this trial compares favorably to that of other meat products such as beef and chicken (Table 4). Note that TAAP is a good proxy for protein quality, and it is more accurate than total nitrogen content for certain foods (Hall and Schönfeldt 2013). While T. molitor reared on Diet C (stover) had lower concentrations of all amino acids, their values are still higher than cow’s milk, which is a key protein source for many populations as it is generally more affordable and widely available than meat. The amino acid content and composition of most edible insects is comparable to the World Health Organization (WHO) recommendations for amino acids required for human nutrition (Rumpold and Schlüter 2013b), but it is known that feed substrate quality, including nitrogen content, can impact insect nutritional quality. Grasshoppers reared on bran, for example, which is high in essential fatty acids, contain almost double the protein content of those fed on maize (van Huis et al. 2013a). Protein content and digestibility of edible insects can vary drastically across a single species (Ramos-Elorduy 1997, Bukkens 2005) and differ by insect life-stage, feedstock, and processing methods.

In addition to containing overall essential amino acid values comparable to traditional meat and dairy products, the lysine content of T. molitor larvae reared in this experiment is relevant. Lysine is the most limiting amino acid in maize grain for human nutrition (Alan 2009). Populations that consume more than 50% of their daily calories from maize, such as many residing in Southern Africa, are at risk of protein energy malnutrition (PEM) and with severe PEM, individuals face greater susceptibility to life-threatening diseases including tuberculosis and gastroenteritis (Nuss and Tanumihardjo 2011). Severe cases of PEM, such as kwashiorkor, can lead to muscle atrophy, wasting syndrome, and even death. All-cause undernutrition is responsible for more than 45% (>3 million) of deaths among children under-five in developing countries (Black et al. 2013). In Zambia, for example, the traditional staple of the diet is nshima, a stiff, maize-based porridge. Nshima is typically consumed with a relish of vegetables and sometimes meat, sour milk, eggs, or edible insects. Malnutrition remains a major problem in Zambia today, where chronic undernutrition plagues about 45% of the population (WFP 2015) and contributes to more than half (52%) of deaths in children under 5 yr old (CSO et al. 2009). There is a need for better access to essential amino acids, including lysine, in diets where the majority of calories come from maize.

Lysine requirements for children younger than three are greater than that of adults (Institute of Medicine 2006a, WHO 2007). Children of 1–3-yr old need 45 mg kg⁻¹ day⁻¹, but adults need just 30 mg kg⁻¹ day⁻¹ (Institute of Medicine 2006b, WHO 2007). On a per gram basis, T. molitor larvae from this experiment (fresh weight) and farmed T. molitor larvae reported elsewhere (dry weight) (Ghaly and Alkoaik 2009) contain significantly more lysine than maize meal or whole milk (Table 5). As a diet supplement, edible mealworms could increase lysine intake for both children and adults.

Larvae from this experiment contained slightly less iron than beef (Table 6), averaging about 1.7% compared to 2.4% (USDA 2016). Interestingly, the lowest-nutrient diet (Diet C) yielded the highest iron content in the larvae. Higher iron content in the stover-fed larvae could be due to iron content levels in the stover or represent a change in insect biology. High iron levels in the insects also offers the potential for a boost in much needed iron intake among populations that are iron deficient. The bioavailability of insect iron is not fully understood, but one study using in vitro simulated peptic-pancreatic digestion found that several edible insects had higher iron bioavailability (uptake) than whole-wheat flour, and buffalo worm iron was more bioavailable than sirloin steak (Latunde-Dada et al. 2016). More research and testing are needed to understand the implications of insect iron consumption and the impact of feedstock on iron content in minilivestock.

Iron content of T. molitor is important given that iron deficiency anemia is a primary public health concern in many developing countries including Zambia, as the mineral is crucial for growth and cognitive development during childhood. The prevalence in Zambia is 53% in children under-five and approximately 22.5% in pregnant women (NFNC 2004). Iron deficiency is the result of iron stores depletion over time, as dietary absorption does not meet metabolic demands to replenish losses and sustain growth (Wood and Ronnenberg 2005). Deficiency can stem from low dietary intake or inadequate intake of bioavailable iron forms, increased iron requirements due to pregnancy, menstruation, rapid growth, excess blood loss caused by pathogenic infections, or impaired iron absorption. Bioavailability of iron is low for populations that consume homogenous plant-based diets with very little meat (Zimmermann and Hurrell 2007), including many in Zambia. Plant-based diets often contain high levels of phytates, which are a primary inhibitor of iron absorption (Shlemmer et al. 2009).

There are several limitations of this study and remaining questions that require further investigation. This exploratory pilot study involved only three replicates for each feed type. Larger experiments with more repetitions are needed to increase the power of these findings. Early instar mini-mealworms from a commercial supplier were utilized in this study, so larvae in experiment 1 were not exposed to experimental diets for their entire lives. Future studies that control insect diet from hatching are needed, and a clear measurement of insect survival will be critical. Now that we have observed insect growth and development using a 100% stover diet, a full nutrient profile of all the feeds should be calculated to determine direct impact of feedstocks on insect growth, development, and nutrient content. Additional research into optimal rearing conditions for T. molitor in variable contexts is warranted. For example, investigations into larval density may be important. Weaver and McFarlane (1990) observed improved growth with T. molitor larvae kept at greater larval density, with no change in survival. However, they noted that cannibalism and incomplete larval-pupal development was also more prevalent in higher density populations (Weaver and McFarlane 1990). More research is needed to assess how larval density might impact growth, fecundity, and nutrient content in T. molitor minilivestock systems. Lastly, from a human health perspective, measurement of insect nutrient bioavailability in vitro and in vivo will provide insight into potential nutritional impacts of consumption.

This study has implications for future efforts to evaluate and potentially instigate minilivestock cultivation to promote food security. In parts of Southern Africa, entomophagy is common. In Zambia, for example, while T. molitor are not currently consumed, other beetle larvae such as Acanthophorus capensis White and Pachylomera femoralis Kirby (dung beetles) are eaten by some (Mbata 1995). Understanding social perceptions of edible insects will be crucial to determining the cultural acceptability of minilivestock farming and introducing new edible insect species into the diet in Zambia (Stull et al. 2018). However, there may be multiple ways to use T. molitor larvae in contexts like Zambia. The larvae have a benign flavor, are easy to farm, have a relatively short lifespan and flexible diet, can be fed to humans, poultry, or fish, and can be farmed year-round. The waste produced by T. molitor could also potentially serve as a fertilizer for farmers in southern Africa. Annelid and insect frass (also called castings) have both been used as fertilizers, especially earthworms whose castings are linked to plant growth improvements (Hidalgo et al. 2006), higher nutrient uptake (Kang and Ojo 1996), and increased flower number (Hidalgo et al. 2006) in certain species. Anecdotal evidence suggests that T. molitor frass is a useful fertilizer, but studies on frass quality given variable feed substrates are lacking.

These benefits, along with their nutritional quality, make T. molitor farming in Zambia appealing, either as a new human food source or as an animal feed. Small-scale farming in Zambia is overwhelmingly dominated by maize, with 86% of smallholders cultivating it (Tembo and Sitko 2013). Many smallholders have limited access to cattle and other animals that might eat significant quantities of residual stover. Strategic and limited removal of stover as an insect feed source may be agroecologically tenable. Moreover, there may be alternative diets generated from crop byproduct or byproducts that could yield nutritious larvae and healthy T. molitor populations. Rural agrarian communities in Zambia are also rich in other natural resources that could be added to and complement stover as a useful feed substrate. For example, deciduous fruit trees that retain green leaves year-round may provide ample nitrogen and supplement feed for T. molitor or other edible insects.

Conclusions

This study provides preliminary evidence that T. molitor larvae can consume a diet of 100% agricultural byproduct (stover) from organic maize production with a carrot water source, suggesting that this insect species may be a suitable candidate for farming in low-resource contexts. This feedstock (stover) slows insect growth compared to the control diet, but it still yielded on average a high protein end product (15.17%) with ample essential amino acids and more slightly more iron (mg/g) than larvae reared on higher protein diets. T. molitor larvae reared on Diet C (stover) contained significantly lower essential amino acid concentrations relative to the control and mixed diets, but all essential amino acids were present. Substitution of 40% of a mixed dry feed by weight with a low-quality feedstock (stover) yielded amino acid values consistent with the control, suggesting that up to 40% of T. molitor feedstock could be substituted with stover in future minilivestock farming initiatives. Additionally, we observed that T. molitor are able to reproduce on diet of up to 100% stover. Feedstock for minilivestock is an important consideration given the resources available to smallholders in low-resource contexts like rural Zambia, where diverting human food to insect production would be counterintuitive. Additionally, if insects can be cultivated using locally available materials, this gives participants autonomy over their food production, as opposed to other agricultural activities that require agribusiness to supply inputs or seeds.

While T. molitor larvae are not currently consumed in Southern Africa, the nutritional value, ease of farming, and flexible diet of the insects make them a reasonable candidate to supplement the maize-based diet or to use as animal feed. An important next step is to develop a suitable mixed feedstock for T. molitor consisting of stover, along with other available biomass that does not divert food from human consumption. Exploring ways to optimize T. molitor growth, survival, and reproduction without relying on grain will be important. Options may be found in other agricultural byproducts, locally available plants, or tree leaves. The mango tree, for example, is evergreen. Future studies should investigate whether mango leaves might be suitable as a supplemental feedstock for edible insects like T. molitor. To facilitate adoption of T. molitor minilivestock cultivation by farmers in food insecure regions, more research is needed concerning optimal strategies for reducing agricultural and economic risks. Finally, cultural factors affecting perceptions of entomophagy, minilivestock, and edible insects must also be considered before embarking on T. molitor cultivation in variable contexts.

Acknowledgments

Special thanks to Jack Cook for help with data collection and Anders Gurda for facilitating access to stover.

References Cited