Crude protein (CP) is a vital nutrient in poultry nutrition, essential for maintaining bird welfare, supporting optimal growth, and facilitating key physiological functions (Dilelis et al., 2021; Lalev et al., 2023). Modern broilers require diets containing more than 20% high-quality crude protein (CP), well-balanced in amino acids (AA), during the starter and grower phases (Aviagen, 2022). Soybean meal (SBM) is usually incorporated into poultry diets, but is expensive for those who import it, which impacts the cost of broiler production (Pirgozliev et al., 2024). While numerous research efforts are focused on replacing SBM with more accessible and cost-effective locally produced alternative protein sources, their use is often limited by suboptimal amino acid profiles, reduced nutrient availability, and lower nutrient yield (Whiting et al., 2019; Watts et al., 2021). However, recent research has demonstrated that insects are a promising high-quality sustainable source of dietary CP and fat for poultry and other nonruminant animals (Al-Qazzaz et al. 2016; Bovera et al., 2018; Lalev et al., 2023; Yordanova et al., 2024). The meal produced from insects is rich in CP and AA, but contains chitin, which is considered as an antinutrient for poultry, reducing dietary energy and CP availability (Lubitz et al., 1943; Kobayashi and Itoh, 1991; Lokman et al., 2019).

The use of dietary enzymes such as phytase, protease, and carbohydrase is a well-established strategy to enhance nutrient and energy utilization, thereby improving the overall quality and efficiency of poultry diets (Šimić et al., 2023). Protease is a less extensively studied enzyme, but experiments with broilers fed SBM-based diets have shown that protease supplementation can improve growth performance (Angel et al., 2011; Cowieson et al., 2016; Xu et al., 2017), as well as dietary metabolizable energy, apparent ileal AA and CP digestibility (Adebiyi and Olukosi, 2015; Stefanello et al., 2016; Watts et al., 2020). However, the results of supplementing protease in insect meal-containing diets for fish (Bolton et al., 2021), pigs (Go et al., 2022) and laying hens (Facey et al., 2024) have not demonstrated efficacy expected for commercial animal production.

Updated information on the effects of supplementary protease in insect meal–containing diets for broilers is required to inform modern diets that include insect-derived ingredients. Accordingly, the objective of this study was to investigate the effects of exogenous protease in isocaloric and isonitrogenic diets formulated with and without insect meal on the dietary apparent metabolizable energy (AME), N-corrected AME (AMEn), dry matter (DMR), nitrogen (NR) and fat (FR) retention coefficients. Daily feed intake (FI), weight gain (WG) and feed conversion ratio (FCR) were also determined.

All animal procedures were carried out in accordance with the ethical guidelines for animal experimentation outlined in European Directive 2010/63/EU and its implementation in Bulgaria under Regulation 20/2012. The study was conducted at the poultry research facility of Trakia University (Stara Zagora 6000, Bulgaria; 42°23'54" N, 25°34'17" E) and received approval from the University Research Ethics Committee.

Crude protein was the main constituent of the BSFLM meal, followed by crude fat (Table 1). Among the essential AAs, leucine, lysine, and valine were present in the highest concentrations, while tryptophan was found at the lowest level (Table 1). Aspartic and glutamic acids were the main dispensable AAs, with cystine having the lowest content (Table 1).

Compared to the calculated values, determined dietary CP was 7.0 and 18.7% higher in CD and ED, respectively than formulated (Table 2). Conversely, the determined values of oil in CD and ED were respectively 4.3 and 6.3% lower compared with calculated values (Table 2). It can be assumed that the observed differences were due to the composition of the actual ingredients used in the present study and the values given by the diet formulation software for the same ingredients. The diets contained similar GE, but the ED had approximately 28% more NDF than the CD (Table 2).

The mean body weight of the birds was 1493 g (SD ± 153.2) at start (at 28 d old) and 2096 g (SD ± 207.4) at the end of the study (at 38 d old) (data not shown). Overall, birds fed ED containing BSFLM, tended to have lower feed efficiency (higher FCR; P = 0.061) compared to birds fed CD (Table 3). Protease fed birds consumed 8.4% more feed than birds fed unsupplemented diets (P = 0.036; Table 3). There was a BSFLM by protease interaction regarding daily WG, as birds fed protease supplemented SBM-based diet reached 28.2% greater WG (P = 0.021) than birds fed the unsupplemented SBM-based diet (Table 3). However, in the BSFLM diet there was no response to protease for WG (P>0.05; Table 3). The AME and AMEn of the CD were more than 1.3 MJ greater than those of the BSFLM based ED (P<0.05), and there was no response to protease (P>0.05) and no response to protease was observed (P>0.05; Table 3). Birds fed protease supplemented diets consumed 9.6% more AMEn than those fed the unsupplemented feed (P = 0.036). There was no response (P>0.05) of DMR and NR to dietary type or protease, although the CD had a greater FR coefficient (P = 0.001; Table 3).

Table 3

Treatment	FI (g/b/d DM)	WG (g/b/d)	FCR (g:g)	AME (MJ/kg DM)	AMEn (MJ/kg DM)	AMEn intake (MJ/d)	DMR	NR	FR
BSFLM
no	109.2	61.2	1.812	13.66	13.03	1.42	0.684	0.571	0.839
yes	113.6	58.3	1.960	12.19	11.51	1.30	0.646	0.550	0.730
SEM	2.78	2.08	0.0521	0.379	0.355	0.046	0.0204	0.0220	0.0192
Protease
no	106.9	55.9	1.934	12.82	12.18	1.30	0.661	0.551	0.795
yes	115.9	63.6	1.838	13.03	12.36	1.43	0.670	0.570	0.774
SEM	2.78	2.08	0.0521	0.379	0.355	0.046	0.0204	0.0220	0.0192
BSLFM × Protease
no no	101.8	53.6a	1.921	13.73	13.10	1.33	0.689	0.571	0.869
no yes	116.6	68.8b	1.702	13.59	12.96	1.51	0.680	0.571	0.809
yes no	112.1	58.2a	1.947	11.92	11.26	1.26	0.632	0.530	0.720
yes yes	115.2	58.4a	1.974	12.46	11.75	1.35	0.660	0.569	0.739
SEM	3.93	2.94	0.0737	0.536	0.502	0.065	0.0289	0.0311	0.0272
P-value
BSFLM	0.278	0.384	0.061	0.014	0.008	0.085	0.206	0.505	0.001
Protease	0.036	0.019	0.211	0.709	0.724	0.058	0.757	0.549	0.464
BSFLM × Protease	0.153	0.021	0.114	0.537	0.539	0.483	0.530	0.542	0.166

BSFLM - Black Soldier Fly larvae meal; SEM - pooled standard errors of mean.

The analyzed composition of BSFLM fell within the expected ranges reported in previous studies (Cullere et al., 2016; Bovera et al., 2018; Mahmoud et al., 2023; Chobanova et al., 2023). However, variations in fat, protein, mineral, and chitin contents are well recognized and primarily influenced by factors such as the larvae’s age and the rearing substrate used (Makkar et al., 2014; Kieronczyk et al., 2022). The Ca content, which forms part of the total ash, corresponds with the elevated ash levels observed in the BSFLM sample of this study. The ADF content of BSFLM is commonly linked to the chitin present in insects. To improve the accuracy of chitin quantification, ADF values can be adjusted to account for amino acids (Finke, 2007) or acid detergent lignin (ADL) content (Hahn et al., 2018). The chitin levels observed here were consistent with previous findings (Schiavone et al., 2017; Chobanova et al., 2023), although variability is expected depending on the specific chitin determination method employed.

The discrepancies observed between the analyzed and calculated CP and fat contents of the diets are likely due to variations in the nutrient composition of the actual ingredients used compared to the values employed in the formulation software. The determined AME in CD was 13.24 MJ/kg on an as-fed basis, in agreement with the calculated values, but the AME of ED was 12.15 MJ/kg, or 1.0 MJ lower than the expected. Although BSLFM has a relatively high ME value (Bovera et al., 2018; Cutrignelli et al., 2018; Chobanova et al., 2024), dietary fiber influences AME in poultry diets, as high fiber content is generally associated with for lower AME (Pirgozliev et al., 2016). Dietary fiber fractions can correlate with chitin content in insect meals, a nitrogen-containing carbohydrate that is nutritionally unavailable to birds (Lubitz et al., 1943). In this study, the ED contained 28% more NDF compared with the CD, which may explain the lower AME and AMEn values in ED. Kobayashi and Itoh (1991) and Lokman et al. (2019) suggested that chitin itself may contribute to the low production and metabolizable energy. Previous research also found a reduced dietary nutrient and energy availability in diets containing over 10% insect meal (Bovera et al., 2018; Cutrignelli et al., 2018; Chobanova et al., 2023). Yang et al. (2025) verified that chitinase is effective to improve the nutrient digestibility in growing pigs. Therefore, the use of chitinase in diets with BSFLM can be an alternative to mitigate the negative effects of chitin on digestibility.

The birds remained healthy throughout the study, with no mortalities recorded. However, their body weights at both the start and end of the trial were lower than the Ross 308 female broiler target weights, which may be attributed to the use of mash feed instead of pelleted feed and housing in small groups (Pirgozliev et al., 2016; Yang et al., 2020). Despite this, these factors were not considered to compromise the study’s objectives. Although mash diets can potentially lead to selective feeding of certain ingredients, this effect is expected to be minimal in practice, and no data on selective feeding were collected in the current study.

Dietary CP content is an important factor in elucidating the effectiveness of the protease supplementation. Proteases typically improve performance in birds fed low CP diets by releasing additional available amino acids and fulfilling the needs of the animals. In this study, despite the intention to produce low CP diets, the ED contained 203 g/kg vs 195 g/kg required CP. Thus, the birds had enough dietary CP for growth and maintenance, such that additional release of amino acids from the protease was not essential. In the current study, it was not possible to measure ileal AA digestibility, however this may be explored in future work. In the absence of ileal AA digestibility data, the positive response of the CD to protease may be due to the greater suitability of the enzyme to the plant-based diet and to the 6% reduction of CP in CD compared with the ED.

Most of the published literature on the feeding value of protease when supplemented to diets is based on vegetarian protein sources. Although in some reports supplementary protease did not improve performance or nutrient digestibility in broilers and turkeys (Freitas et al., 2011; Walk et al., 2018), in other reports the enzyme improved growth performance (Angel et al., 2011; Cowieson et al., 2016; Xu et al., 2017), dietary AMEn, and apparent ileal AA and CP digestibility (Adebiyi and Olukosi, 2015; Stefanello et al., 2016; Watts et al., 2020). However, the lack of response of the ED to the enzyme regarding WG requires further explanation. Proteases are categorized into six main classes: aspartate, cysteine, glutamate, metallo, serine, and threonine proteases. Among these, serine proteases account for over one-third and are characterized by endoproteolytic activity that typically relies on a catalytic triad composed of aspartate, histidine, and serine residues (Walk et al., 2018). The protease used in this study is from the serine group and was developed and very successfully used in vegetable/SBM-based diets but may not be so efficient when the diet contains a high amount of animal CP, such as insect meal. Lu et al. (2025) concluded that the cecal microbiome diversity is enhanced in laying hens fed high levels of BSFLM, linking it to suboptimal protein digestion and an undesired increase in protein fermentation, which can be partially addressed by protease supplementation. Bolton et al. (2021) did not find changes in the performance of fish fed 150 to 300 g/kg insect meal with and without exogenous protease (produced by thermophile bacteria) in an eight-week feeding study. Additionally, Go et al. (2022) did not find differences in growth performance of pigs when fed diets containing insect meal with and without supplementary alkaline serine endopeptidase for four weeks. Also, after feeding 130 g/kg insect meal to laying hens for 20 weeks, Facey et al. (2024) concluded that the supplementary protease, possessing acid, neutral, and alkaline activity and produced by Bacillus licheniformis, did not improve growth or performance in hens. Walk et al. (2018) reported that novel protease supplementation improved AA digestibility in vegetable diets, but no changes in growth performance of birds were observed. Since the high CP dietary content in our study may compromise the efficiency of protease, it is not possible to draw firm conclusions regarding protease effects in insect meal–based diets.

This study investigated a commercially available protease and insect meal for broiler production using the widely available Ross 308 broiler model. The research indicates that there is a need for further research in this area to make insect meal a consistent and viable alternative protein to soybean meal.

Feeding 10% BSFLM reduced dietary metabolizable energy for broiler chickens but did not affect other variables, except fat retention. An interaction was observed between dietary type and protease supplementation whereby protease was effective in SBM diets but not in BSFLM diets. Research on the impact of protease low-protein BSFLM diets, rather than protein sufficient diets, is recommended.