With the livestock industry at its limit in terms of sustainable production capacity, and the pet food business in constant growth, new sources of protein are being sought in order to meet the market's demand and the expectations of pet owners (1). Insects may provide a possible solution as an alternative feed, since they can partially replace traditional feed sources, while they also provide a means to bio-converting organic waste (2). Of the various insects being considered, the black soldier fly (Hermetia illucens) is showing particular promise due to its immediate potential for large-scale production (3).

The black soldier fly (BSF) has a balanced protein composition and one of the highest amino acid scores compared with other currently reared insects or traditional protein sources (such as fish meal) (4). Compared with crickets and mealworms, BSF boasts a more stable nitrogen and phosphorus composition and has a more advantageous feed conversion ratio (5). It can also be considered a possible sustainable solution due to the possibility of rearing the insects on materials deemed unsuitable for human nutrition, such as alimentary by-products and organic substrates (6).

As pointed out by Böhm et al. (7), insects may constitute an appropriate novel protein source for dogs, presenting cutaneous adverse food reactions. Nevertheless, societal negative opinions about the use of insect meal in pet nutrition have arisen, especially due to insect phobia and concerns about safety. Security aspects about insect consumption were also discussed critically in EFSA Scientific Opinion (8), where uncertainty regarding the risk of non-processed items, due to the lack of data, has been acknowledged. However, EFSA concluded that microbiological risks are expected to be comparable with other food raw materials, provided that insects are fed with allowed feedstuff. Consumers from Western countries still continue to have prejudices regarding the introduction of insects in their diet (9), and, due to the current “humanization trend” (10), this fact could be also translated to their pets. Notwithstanding, public opinion seems to be less concerned about the use of veterinary-prescribed diets based on insects (11). Indeed, veterinarians have expressed interest in hypoallergenic food alternatives prepared using insects (12). According to the Commission Regulation (EU) 2020/354 (March 4, 2020) (13), a product can be claimed to reduce ingredient and nutrient intolerances if it is composed of hydrolyzed proteins or selected and limited protein sources or selected carbohydrate sources. Therefore, according to the current European Regulations, a product composed only of insects as the main source of protein could be considered with the particular purpose of reduction of food intolerance. Concurrently, and reflecting the growing interest in this field of research (14), various recent studies have investigated the possibility of feeding BSF larvae to poultry (15–18), fish (19–21), and swine (22, 23). Recently, a thorough review from Bosch and Swanson (24) explored in depth the palatability, digestibility, and nutritional aspects of the inclusion of insects in dog and cat diet, showing the potential of insects as future pet food products.

The aim of the present study was to evaluate the inclusion of defatted BSF larvae meal in extruded dog food in terms of its in vivo and in vitro digestibility, in order to assess its suitability for the pet food market. Furthermore, the purpose of the study was to evaluate if the in vivo marker method and the in vitro digestibility method could be comparable to the traditional in vivo total collection method also in these particular diets. The estimated in vivo digestibility of organic matter and crude protein calculated by means of predictive equations utilizing data obtained by in vitro analysis was also assessed.

All the experimental procedures were approved by the Bioethics Committee of the University of Turin (Italy) (prot. n. 336595).

The foods were well-accepted during all the trial lengths, and no episode of nausea or vomiting has been reported. The in vivo ATTDC digestibility results are summarized in Table 2. The two methods used to estimate in vivo digestibility (TFC and Cr₂O₃) showed similar results between the CTRL and BSF groups in relation to DM, OM, EE, ash, and phosphorus. The ATTDC of CF was significantly lower (p < 0.001) in the BSF diet compared with the CTRL diet. On the contrary, the ATTDC of calcium was significantly higher (p < 0.05) in the BSF compared with the CTRL diet. A statistical trend (p = 0.066) was observed for the ATTDC of CP, being higher in the animals fed the BSF compared with the CTRL diet.

No statistical differences were observed between the two ATTDC methods (TFC vs. Cr₂O₃). Furthermore, no statistical interaction between diets and methods was found.

The in vitro digestibility data and estimated in vivo digestibility results, obtained utilizing the regression equations described in Hervera et al. (33, 34), are reported in Table 3. The digestibility values for DM, OM, and CP obtained using the in vitro method were higher for both the CTRL and the BSF diet (by an average of +8.43, +5.25, and +6.08%, respectively) compared with those obtained using in vivo methods. The estimations of in vivo digestibility of OM and CP (based on in vitro data) were consistently higher than the data obtained using in vivo ATTDC methods: in vitro estimation of in vivo digestibility overestimated OM and CP digestibility by up to 4.0% and 9.8%, respectively, compared with the in vivo methods.

This study evaluated the nutritional quality of defatted BSF larvae meal as a potential sustainable novel raw material for pet food, to be integrated into extruded diets as a protein source. In addition, it explored the suitability of the in vivo marker method and the in vitro digestibility method with the traditional in vivo total collection method.

Although the control (containing venison meal) and insect-based diets were formulated to be isonitrogenous, our analysis showed CP content to be almost 4% lower in the former (16.97 vs. 20.70%, respectively); the discrepancy between the diets was nevertheless within the limits stipulated in the EU regulation 2017/2279 regarding “Tolerances for analytical constituents” (36). It is also important to remember that since chitin is a nitrogen-containing polysaccharide, this could also have led to a mild overestimation of the protein content in the BSF diet (6, 37).

We must also acknowledge that the higher crude protein content of the BSF diet compared with the CTRL diet could be an overestimation due to our use of a nitrogen to protein (N:P) conversion factor of 6.25. In fact, several authors recently pointed out that this conventionally used conversion factor may lead to the overestimation of protein content in a variety of feedstuffs (38, 39), including insect meals (40, 41). Furthermore, although Finke et al. (42) estimated that the amount of nitrogen in insect chitin would not significantly affect the total amount of nitrogen, other authors support the hypothesis that the presence of non-protein nitrogen (NPN) in insect CP could cause the overestimation of CP (40, 41).

In our trial, the ATTDC of DM, OM, and EE were similar in both BSF and CTRL groups, whereas the ATTDC of CP were higher in the BSF vs. CTRL group. A similar result was obtained by Lei et al. (43), where increasing levels of BSF meal inclusion (at 0, 1, and 2%) in Beagle dog rations raised nitrogen digestibility, whereas EE digestibility remained similar to that of the control diet. However, Gariglio et al. (18) observed that up to 9% BSF meal inclusion in the diet of growing Muscovy ducks did not change diet digestibility, with the exception of the ATTDC of EE, which was improved in BSF groups. In line with these data, Biasato et al. (23) observed no change in the ATTDC of BSF diets (up to 10% inclusion) in growing piglets. Similarly, Freel et al. (44) did not notice any difference in ATTDC of DM, CP, and EE in a trial involving 56 Beagle dogs fed with diets containing graded levels of BSF meal (5.0, 10.0, and 20.0%) and BSF oil (1.0, 2.5, 5.0%). Furthermore, in a study where BSF meal completely replaced soybean meal in the diet of laying hens, Cutrignelli et al. (45) found BSF to correlate with lower crude protein digestibility, whereas lipid digestibility remained unaffected. Likewise, Kröger et al. (46), in a study involving 12 Beagles, observed a decrease in ATTDC of CP in the BSF group compared to the control group, while the ATTDC of DM was increased when dogs were fed the diet containing the BSF meal (at 20.0% of inclusion). This result could be explained by differing levels of chitin, which can negatively affect protein digestibility (47). Indeed, the reported difference in fiber digestibility between the diets supports this result and explanation, since chitin gets recognized as part of the crude fiber fraction during the analysis (48). Furthermore, the mean values of crude protein ATTDC (for BSF-based diets) observed in our study were in line with those found in Kröger et al. (46) but below those recovered in Freel et al. (44).

Hydroxyproline can be used as an index of protein quality (49), due to its being a marker of collagen content (50). The levels of collagen and of hydroxyproline were higher in the control diet compared with the BSF diet, probably due to the fact that collagen is limited in insect meal compared to that in vertebrate protein meal. This could also explain the higher level of digestibility of the BSF diet compared with the control diet, at least with regard to crude protein digestibility, since the net protein utilization of collagen is zero (51). Collagen content also influences the N:P ratio of protein sources, and consequently the real CP content of the diets, in particular that of the control diet (39). It may also be speculated that the control diet had a decreased crude protein digestibility due to the higher ash content; however, high levels of crude ash did not appear to decrease protein digestibility, as previously reported by Bockskopf and Kamphues (52).

The difference in calcium digestibility could be due to the use of different ingredients to adjust the calcium level of the diets. Indeed, calcium carbonate was added to the BSF diet to obtain the minimum requirements for dogs, whereas in the CTRL diet the calcium requirements were satisfied by the presence of ground bone in the venison meal (thus avoiding the need for any calcium salt addition), and this could have led to the discrepancy. Interestingly, Lei et al. (43) noticed significant increases in the level of calcium in the blood of beagles as the BSF larvae meal content of their food was increased. This result points toward a potential increase in the bioavailability of this macro-element that depends on the inclusion of BSF larvae meal in the diet; however, further investigations are required to confirm and understand the basis of any possible relationship.

It is important to note that no statistical differences were observed between the ATTDC values determined using the marker method and the total collection method for both CTRL and BSF diets, confirming the validity of the marker method as an alternative to the total collection method (28). The values of in vitro DM, OM, and CP digestibility were also similar to the results obtained with the two in vivo methods, despite being, in line with the previous literature (33, 34), slightly overestimated in the former. We also evaluated whether the equations for the estimation of in vivo crude protein and OM digestibility, utilizing in vitro digestibility data, as described in Hervera et al. (33, 34), fitted with the results obtained in this study (shown in Table 3). Since the predictive equations proposed were only used to assess feedstuff based on vertebrates and, to our knowledge, no other study inspected if they could be applicable to invertebrates, we decided to include these findings. For both the venison and insect diet, the predictive equations gave slightly overestimated values compared with the mean of the in vivo digestibility results, even though they were substantially similar from a nutritional perspective. Indeed, the discrepancy between the crude protein digestibility estimated using the equation and the in vivo crude protein digestibility results ranged from 3.2 to 9.8%, whereas the overestimation of the OM digestibility ranged from 0.2 to 4.0%, with lower deviations and a narrower range. According to these results, predictive equations utilizing in vitro digestibility values appear to constitute a valid tool for the analysis of feedstuff digestibility and therefore offer a means to reduce, if not avoid, the use of live animals.

The present study suggests that the inclusion of BSF in extruded diets for dogs (at 36.5%) offers a promising alternative source of dietary protein for this species, in particular in relation to the digestibility profile of crude protein, crude fat, and OM. Our findings also highlight the need for further studies in order to understand the effect of chitin on fiber digestibility and mineral absorption in a BSF-based diet.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The animal study was reviewed and approved by Ethic Committee of Turin University. Written informed consent was obtained from the owners for the participation of their animals in this study.

AS, EP, LPe, and LPr conceived and designed the experiment. EP and NR collected the experimental data. EV, FH, JM, JN, and SM carried out the chemical analyses. AS, LPe, and UA performed the statistical analysis. All the authors interpreted the data. AS, LPe, and LPr wrote the first draft of the manuscript. All the authors reviewed the manuscript for intellectual content and gave approval for the final version to be published.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.