Novel foods represent sustainable alternatives to traditional farming and conventional foodstuffs. The house cricket (Acheta domesticus) is considered as one of the most promising reared insects due to their attractive nutritional profile and lower feed conversion ratio compared to other animals. However, putative health hazards associated with consumption of crickets have previously not been investigated. The present study assesses the risks of A. domesticus reared in closed systems controlled by the implementation of hazard analysis and critical control points and good farming practices. Due to the novelty of the topic, data scarcity has been a limiting factor, hence comparative evidence from closely related species belonging to the order Orthoptera (e.g. grasshoppers, locusts, and other cricket species) have been included. The present risk profile identified as main hazards: (1) high total counts of aerobic bacteria; (2) presence of spore-forming bacteria post thermal processing; (3) accumulation of cadmium and other heavy metals; and (4) a possible increase of allergenic reactions due to exposure to insects and insect derived products. Important data gaps regarding edible crickets and their safety as novel foods have been highlighted in the future perspective section, representing aims for future research. Identified data gaps include: (1) farming conditions of the insects being studied; (2) data on the impact of thermal processing of the products prior to consumption; (3) fungal communities and mycotoxins-producing fungi in reared crickets; and (4) heavy metals not fully assessed (chromium, aluminium and arsenic) and other chemical hazards produced during processing (i.e. heterocyclic aromatic amines, acrylamide). The present risk profile explores food safety risks related to consumption of A. domesticus, thereby constituting an example of chemical and microbial hazards risk profiling on edible insects, covering rearing to consumption.

The effect of heat treatment on the protein value of field beans, lupins and peas was studied using an in vitro method. Protein feeds were subjected to heat treatment for 30, 60 and 90 min in forced air oven at 120, 140 and 160 °C and in autoclave at 105, 120 and 135 °C. The heat-treated protein feeds were incubated in buffered rumen fluid together with grass silage and barley in complete isonitrogenous diets. The gas production (GP) was recorded continuously, and ammonia-N (NH -N) concentrations were determined during the in vitro incubation at 8, 24 and 48 h and used to determine the utilisable crude protein (uCP) at the duodenum at 16 h of incubation (uCP ). Heat treatments decreased the concentration of soluble crude protein and increased neutral detergent insoluble CP (NDICP) in all protein feeds compared to untreated. Inclusion of protein feeds to basal diet showed no increase in the uCP in untreated field bean diet and only a small numerical increase in the uCP concentrations from 160 g/kg dry matter (DM) to 166 and 172 g/kg DM in untreated lupine and pea diets, respectively, indicating high degradability of untreated feeds. Increasing the time and temperature of the heat treatment linearly increased the uCP concentrations in field bean and pea diets, but not in lupin diets. Autoclave treatment was more effective in decreasing uCP than oven treatment despite the lower temperatures used. However, the combination of highest temperatures and treatment time in autoclave increased acid detergent insoluble CP (ADICP) concentrations in protein feeds, indicating protein damage and decreased intestinal digestibility. Determining in vitro uCP and ADICP shows to be a promising method for evaluating protein value in heat-treated animal feeds.

In an approach combining microbiological culture methods with high-throughput sequencing, this study investigated the microbial communities (bacteria, moulds and yeasts) in Swedish-produced edible crickets (Acheta domesticus) reared in a controlled environment. The effects of different feeds on microbial loads and populations in crickets were also studied. The crickets used were third-generation offspring from wild-caught individuals from Sweden, which are adapted to grow in a laboratory environment. The efficiency of rinsing to decrease microbial load was evaluated not obtaining a significant decrease of plating counts for total aerobic counts (TAC) and Enterobacteriaceae. Crickets were divided into three batches and fed different diets (control feed, early-cut red clover hay (ECH), late-cut fresh red clover (LCF)) for 62 days. Bacterial numbers (TAC and Enterobacteriaceae) on whole raw crickets ranged between 7 and 8 log cfu/g. Pre-rinsing in water did not reduce these levels (P=0.19). All batches tested negative for the food-borne bacteria Salmonella, Listeria monocytogenes, Bacillus cereus and Clostridium perfringens. The mean mould count for crickets fed control feed was 2.8 log cfu/g, while the values for crickets fed ECH and LCF were 4.2 and 4.5 log cfu/g, respectively. The dominant bacterial communities were Proteobacteria, Bacteroidetes and Firmicutes, with Firmicutes and Proteobacteria dominating in crickets fed control feed, Firmicutes dominating in crickets fed LCF and Proteobacteria dominating in crickets fed ECH. Aspergillus flavus, a fungus that is capable of producing mycotoxins, was detected in control feed and ECH reared crickets. More work is needed to identify specific food-borne pathogens in edible crickets and establish possible bacterial quality reference values, as an important step in developing microbial quality and safety parameters to ensure consumer safety.

This study evaluated the potential of red clover as a sole diet for house crickets (Acheta domesticus, AD) and the effect of ensiling or drying red clover biomass on growth, survival and water consumption of AD. Wild AD were caught near Uppsala, Sweden, and reared in a climate-controlled room under a 12-h light regime. One day-old third-generation cricket nymphs (n=2,880) were used in a 56-day feeding trial. The experimental diets (n=8) were early-cut (pre-bloom) and late-cut (late-bloom) red clover, preserved as frozen-fresh, dry-silage, haylage and hay, and a control diet. All clover diets were fed as sole diets with salt block available ad libitum in every treatment. Feed dry matter intake, feed conversion ratio (FCR, kg feed dry matter per kg weight gain), number and weight of crickets and water consumption were recorded every five days. FCR was higher for late-cut than early cut-red clover, but overall cricket weight and survival rate did not differ between cutting times (P=0.939). FCR, weight and survival were not affected by forage conservation method. Total feed consumption was highest for red clover hay. Crickets fed red clover diets had lower (P<0.01) weight and higher FCR than crickets fed the control diet, but their survival rate was not different from that of control crickets during the first 25 days of life. Water consumption of AD fed the control diet was about half that reported for pigs and poultry. Crickets fed on fresh red cover had lower (P=0.04) water consumption compared with crickets fed dried or ensiled red clover, but ensiling did not reduce water consumption compared with hay. Red clover cannot be recommended as a sole feedstuff for AD, but early and late-cut red clover had similar effects. The possibility to partly include late cut red clover in cricket diets is interesting from an ecosystem service perspective since the flowering crop will provide feed for declining populations of bees and other pollinators.

House crickets (Acheta domesticus) are increasingly being used as food. This has the potential for a more efficient food production that also may benefit agroecosystems. As the first study to compare feeds based on European wild flowering plants as feed for house crickets we examined five common plants known to support biodiversity of wild insects (white clover (Trifolium repens), white nettle (Lamium album), common nettle (Urtica dioica), rough comfrey (Symphytum asperum) and common gypsophila (Gypsophila paniculata). These plants and a control diet were fed as dried and sole feeds ad libitum to one-day-old house crickets for 62 days in a climate-controlled laboratory. Cricket weight, mortality and feed intake were recorded every 7 days. An additional test examined crickets preferences for forages and the effect on maturation and weight, by providing a free choice of rapeseed and wheat meal and either an early- or late-cut red clover (Trifolium pratense) or white nettle for 62 days. Mortality of crickets fed common nettle, rough comfrey and common gypsophila exceeded 80% in first 7 days, so these plants were removed from the trial after 14 days. Survival of crickets on white nettle and control feed was 59% after 28 days and average cricket weight was similar. By day 62, the weight of crickets fed white nettle and white clover was on average 32 mg, compared with 201 mg on control feed. In the free choice test, crickets consumed 15-30% red clover, 31-37% wheat meal and 39-64% rapeseed. Crickets with access to red clover showed a higher (P<0.001) proportion of adults (28%) compared with the control (5%). We conclude that white nettle has potential as feed for house crickets during the early growth stages, and that red clover supplementation increase cricket maturation. Using feeds including white nettle and red clover in cricket rearing additionally benefits to support wild biodiversity and lower feed costs.

The present study was conducted to investigate ruminal N metabolism in dairy cows using N-labeled N sources and dynamic models. The data summarized in this study were obtained from 2 of 4 treatments whose effects were determined in a 4 × 4 Latin square design. Soluble N (SN) isolated from timothy grass silage labeled withN and ammonia N (AN) labeled with N were administered into the rumen contents of 4 ruminally cannulated dairy cows. Ruminal N pool sizes were determined by manual evacuation of rumen contents. The excessN-atom% was determined in N-fractions of rumen digesta grab samples that were collected frequently between 0 to 72 h and used to determine N metabolism in the rumen. Calculations of area under the curve ratios ofN were used to estimate proportions of N fractions originating from precursor N pools. A model including soluble nonammonia N (SNAN), AN, bacterial N, and protozoal N pools was developed to predict observed values of N atomic excess pool sizes. The model described the pool sizes accurately based on small residuals between observed and predicted values. An immediate increase inN enrichment of protozoal N suggests physical attachment of bacteria pool to protozoa pool. The mean proportions of bacterial N, protozoal N, and feed N in rumen solid phase were 0.59, 0.20, and 0.21, respectively. These observations suggest that protozoal N accounted for 0.25 of rumen microbial N. About 0.90 of the initial dose of AN was absorbed or taken up by microbes within 2 h. Faster N enrichment of bacterial N with SN than with AN treatment indicates a rapid adsorption of SNAN to microbial cells. Additionally, the recovery ofN as microbial and feed N flow from the rumen was approximately 0.36 greater for SN than for the AN treatment, indicating that SNAN was more efficiently used for microbial growth than AN. The present study indicated that about 0.15 of microbial N flowing to the duodenum was of protozoal origin and that 0.95 of the protozoal N originated from engulfed bacterial N. The kinetic variables indicated that 0.125 of SNAN escaped ruminal degradation, which calls into question the use of in situ estimations of protein degradation to predict the flow of rumen undegradable protein.

This study evaluated the relationship between utilizable crude protein (uCP) at the duodenum estimated in vitro and omasal flow of crude protein (CP; omasal flow of nonammonia N × 6.25) measured in lactating dairy cows. In vivo data were obtained from previous studies estimating omasal digesta flow using a triple-marker method and N as microbial marker. A total of 34 different diets based on grass and red clover silages were incubated with buffered rumen fluid previously preincubated with carbohydrates for 3 h. The buffer solution was modified to contain 38 g of NaHCO and 1 g of (NH)HCO in 1,000 mL of distilled water. Continuous sampling of the liquid phase for determination of ammonia-N was performed at 0.5, 4, 8, 12, 24, and 30 h after the start of incubation. The ammonia N concentrations after incubation were used to calculate uCP. The natural logarithm of uCP [g/kg of dry matter (DM)] at time points 0.5, 4, 8, 12, 24, and 30 h of incubation was plotted against time to estimate the concentration of uCP (g/kg of DM) at time points 16, 20, and 24 h using an exponential function. Fixed model regression analysis and mixed model regression analysis with random study effect were used to evaluate the relationships between predicted uCP (supply and concentration) and observed omasal CP flow and milk protein yield. Residual analysis was also conducted to evaluate whether any dietary factors influenced the relationships. The in vitro uCP method ranked the diets accurately in terms of total omasal CP flow (kg/d) or omasal CP flow per kilogram of DM intake. We also noted a close relationship between estimated uCP supply and adjusted omasal CP flow, as demonstrated by a coefficient of determination of 0.87, although the slope of 0.77 indicated that estimated uCP supply (kg/d) was greater than the value determined in vivo. The linear bias with mixed model analysis indicated that uCP supply overestimated the difference in omasal CP flow between the diets within a study, an error most likely related to study differences in feed intake, animals, and methodology. Predicting milk protein yield from uCP supply showed a positive relationship using a mixed model (coefficient of determination = 0.79), and we observed no difference in model fit between the time points of incubation (16, 20, or 24 h). The results of this study indicate that the in vitro method can be a useful tool in evaluating protein value of ruminant diets.

An in vitro method based on 15N-labelled forage nitrogen (N) was developed to study ruminal N metabolism of soluble N (SN), insoluble N (ISN) and neutral detergent insoluble N (NDIN) fractions of timothy forage. Timothy grass was grown on replicated experimental plots with one plot receiving 15N-labelled and the other unlabelled N fertilizer. Harvested grass was preserved as dried grass or as formic acid treated or untreated silage. The intact forages and their corresponding N fractions were incubated in buffered rumen fluid in vitro to determine degradation parameters based on the 15N fluxes between labelled feed N and ammonia N pools. A high percentage (25–38%) of 15N-labelled ammonia disappeared from ammonia N pool during the first 15 min of incubation. Microbial uptake of dried grass SN fraction was higher than of silage SN fractions. Fractional degradation rates of SN from formic acid treated silage, untreated silage and dried grass during the first 6 hours of incubation were 0.145, 0.125 and 0.115 /h, respectively. By the end of the incubation period (28 h), 69, 66 and 43%, of the SN fraction of formic acid treated silage, untreated silage and dried grass, respectively were recovered as ammonia. The percentage of ISN fractions degraded to ammonia N were 9, 34 and 27%, respectively. Based on the changes in 15N-labelled ammonia N pool in blank incubation and appearance of 15N to ammonia N pool from 15N-labelled NDIN fractions, it was estimated that a significant portion of microbial lysis occurred when incubations were carried out for longer than 20 hours. With dried grass the contribution of ammonia N for microbial N synthesis was greater than with silages. Use of 15N-labelled forages together with this in vitro method is a promising technique for determining soluble N degradation parameters, but it requires further development to be used for determining degradation parameters of insoluble N fractions and work with whole feeds.