Protein feed ingredients are characterized by a relatively high cost, with soybean meal (SBM) being the predominant protein source in widespread utilization. The incorporation of SBM into animal feed typically constitutes a proportion of 23% to 30% of the overall feed composition [1]. Indonesia’s livestock feed industry relies mainly on imported SBM (over 5.3 million metric tons) to meet this demand [2]. The condition is challenging in managing feed prices, which have shown a tendency to steadily rise. As a result, there has been a focus on enhancing the efficiency of utilizing local feed as viable substitutes for conventional feed ingredients.

Considering this circumstance, contemporary research is focused on exploring alternative protein sources that are both sustainable and do not pose a threat to human food production. The application of black soldier fly larvae meal (BSFLM) has been recognized as a potential alternative protein, energy source in broiler chicken diets, more sustainable than soybean and does not compete with human food production [3,4]. They possess a high content of crude protein (CP) (55.42% to 65.50%) [5,6] and energy (7% to 39%) [7], along with a favorable amino acid (AA) composition, rich in calcium (1.21% to 4.39% dry matter [DM]), and phosphorus (0.74% to 0.95% DM) [8]. According to Gariglio et al [9] partially defatted BSFLM contains lauric acid (C12:0) 49.7% of the total fatty acids followed by palmitic acid (C16:0; 13.3%) and myristic acid (C14:0; 10.1%). Meanwhile, according to Rostagno et al [10], the standardized ileal digestible (SID) value of SBM is 44%, which is higher than BSFLM. According to Tansil et al [11], the SID value of partially defatted black soldier fly larvae (BSFL) was found to be 2.7%, slightly lower than SBM but 5% higher than fish meal. This is related to the chitin content in BSFLM, influenced by the age of the larvae. The BSFLM is composed of chitin and medium-chain fatty acids, including lauric and myristic acid. These components are believed to enhance intestinal and immunological health in broiler chickens due to their prebiotic and antibacterial effects [12,13].

On the other hand, insects such as BSFL not only have a high nutrient content, but also produce low greenhouse gas and ammonia emissions, have a good feed conversion ratio (FCR) as cold-blooded animals, and require little water and land to grow [14–16]. These conditions accord with a concerted initiative to optimize the utilization of CP in feed rations by implementing N-balanced diets based on easily digestible amino acids, multi-phase feeding, addition of essential amino acids and other feed additives [17]. Reducing feed protein by an average of 20 to 30 g/kg is an effective recommendation for sustaining animal performance and production. However, lowering CP by more than 30 g/kg has a detrimental effect on performance and leads to an increase adipose fat [18,19]. On the contrary, the reduction in dietary CP levels from 17.22% to 13.54% did not result in a decline in body weight gain (BWG), productivity, or feed consumption among ducks [20]. In another study, it was shown that the inclusion of crystalline amino acids with low dietary CP levels of 16% to 12% did not provide a significant impact on final body weight (BW), BWG, and feed intake (FI). However, a notable enhancement in FCR was seen [21].

Moreover, there is a limited amount of information regarding the application of BSFLM in low-protein diets in duck production. In Indonesia, ducks have emerged as a viable alternative for animal protein, as evidenced by a growing demand over time. Consequently, enhancing production efficiency becomes a crucial factor to consider. Hence, the aim of this research was to investigate the potential of BSFLM as a source of protein feed and determine the efficiency of CP utilization in duck diets. The findings will offer theoretical and technical support for using a low-protein diet in the duck industry.

The advantages of lowering protein levels in poultry diets while supplying AA requirements have been reported in several studies of poultry performance [25,26]. However, in the present study, meeting AA requirements such as methionine, lysine, threonine, arginine, valine, isoleucine, and tryptophan on a low-protein diet resulted in linearly depressed growth performance (p<0.05), regardless of various protein sources. The lowering CP levels affected to reduce the growth performance, as also reported by Wang et al [27], the result was validated by the data, where reduction CP levels suppressed the FI, BW, weight gain, and ADG and increased FCR. Wang et al [28] in their research stated that a decrease in feed protein levels (13.5% and 15.5%) reduced FI and BW in ducks aged 15 to 35 days. The suppression of growth performance could be associated with lower levels of non-essential amino acids such as glycine, serine, or glutamine [29] and feed ingredient digestibility [28]. This was reported by Widyaratne and Drew [30] that applying a low-protein diet to poultry could affect growth performance at similar levels to high-protein diets when using highly digestible feed ingredients. In other research by Wang et al [28] stated that a high protein level decreased growth performance due to increased levels of lower digestible ingredients. In the present study, the interaction between the main factors showed BSFLM-14% as the lowest growth performance. This related to the low passage rate and length of time of feed in the gastrointestinal tract, which were associated with nutrient content and feed particles [31]. It is interesting to note that a reduction of 2% in CP levels in SBM as a protein source did not have a significant impact on the growth performance of ducks. This finding was in line with the results of studies by Xie et al [20] and Sigolo et al [25], which found that adding crystalline amino acids decreased CP levels by 2.5% to 3.68% without affecting chicken productivity. On the other hand, the treatment of BSFLM protein sources affected growth performance and FI. Thus, the decrease in productivity could be associated with the chitin content of BSFL (6.67%) which influenced digestibility value and feed transit duration in the digestive tract [31,32].

The data showed that descending CP levels in a diet increased the concentration of triglycerides, lipids, albumin, and phosphate. This was confirmed by Law et al [33] in their study, which indicated a rise in triglycerides, albumin, free fatty acids [33,34], cholesterol, LDL, and HDL in the blood [31]. The significant rise in these parameters was due to an increase in the proportion of energy sources in the feed, which was related to the process of lipogenesis in the liver. Kamran et al [35] also confirmed that dietary CP levels and in vitro lipogenesis were inversely linked. The higher dietary CP content decreased glucose utilization and increased its synthesis from substrates that were used for fat synthesis previously. Kamran et al [35] stated that birds with low protein diets would use carbohydrates as an energy source rather than free fatty acids, showing greater triglycerides in their blood plasma. This would also affect the level of LDL that is a lipoprotein responsible for transporting triglycerides from liver via the blood to extrahepatic tissue. In the present study, the dietary BSFLM on ducks increased the concentration of lipid and decreased albumin and phosphate levels. This related to BSFL consisting of 52% saturated fatty acid in the form of lauric acid, which stimulates cholesterol synthesis in the blood by raising the activity of enzymes involved [36]. In our research, the treatment of CP levels and protein sources resulted in a total protein in serum between 2.76 to 3.38 g/dL, with the lowest shown on BSFL-14%. According to Behera et al [37], the total blood protein of 8-week-old ducks is between 6.3 to 6.4 g/dL in the treatment of reduced protein levels up to 4%.

The longer the villi, the greater the value of nutritional absorption. Decreasing villus height leads to a lower intestinal absorption area, therefore, gut morphology is crucial in determining the efficiency of dietary nutrients [31]. In this present study, the interaction between dietary CP levels and protein sources markedly on VH. Lowering CP on BSFLM dietary levels had higher VH compared with higher CP levels, this result agreed with Shazali et al [38] but this contradicted with using SBM on low CP. Gu and Li [39] reported that increasing CP levels affected a higher number of goblet cells in the distal jejunum that was associated with VH, CD, and epithelium cell size. Interestingly, decreasing 2% CP levels (SBM-18% to SBM-16%) did not show a significant effect on jejunum morphology, thus indicating that decreasing CP levels could be carried out by supplying synthetic amino acids. In the current study, applying BSFL showed lower VH and CD than SBM as a protein source (p<0.05). Jiang et al [31] stated that the height of the villi and CD were mostly determined by diet. This is related to the digestible value of each protein source, SBM has a higher digestibility value (SID 44%) than BSFL [10]. According to Tansil et al [11], SID-partially defatted BSFL was 2.7% lower than SBM but 5% higher than fish meal. The variation in digestibility influenced the availability of amino acids, which are essential for villus development. The supply of non-essential amino acids such as glycine, glutamine, and proline contributed to the development of intestinal villi [33]. In their study, Wang et al [27] discovered that a decrease in essential amino acids, particularly free branched chain amino acids, damaged intestinal morphology. This was supported by the findings of Facey et al [32], who discovered that low concentrations and digestibility of proteins had a negative impact on development and function of the intestine. Depending on the protein source, the percentage of undigested CP in various poultry feedstuffs ranges from 8% to 35% [40]. Interestingly, the dietary CP levels markedly enhanced the relative weight of the gastrointestinal tract of ducks (p<0.05) when the CP levels were reduced. In contrast, the length of the proventriculus, gizzard, and large intestine significantly decreased as a result of reducing CP levels. The development of gastrointestinal tract (GIT) was related to amino acids supplied in the starter phase as energy sources and building blocks, especially for the small intestine. It has been shown that quality and AA levels can affect gene expression of transporter proteins in the small intestine. Thus, suboptimal AA (protein quality) suppressed gut function. The optimal AA supplied in the starter diet will improve GIT development and its capacity for digestion and nutrient absorption [41]. In contrast, in the present study, lowering CP levels with AA supplementation showed improved weight of small intestinal organs. On the other hand, this result related to the improvement of feedstuffs proportion and particle formation. The higher the reduction of CP levels, the more corn is added to achieve the energy requirement. Furthermore, decreasing 2% CP on the SBM diet had no marked effect on gastrointestinal tract development except for jejunum and ileum weight. According to Kleyn and Chrystal [42], increasing the feed particle size had a positive impact on gizzard development (as much as 5%) by optimizing size, function, and nutrient absorption. In addition, large particles, such as in whole grains, stimulated 44% of relative gizzard weight [43], decreased pH, improved proventriculus dilatation, digestibility, and starch retention [44]. The particle structure and fiber components of feed played a function in modulating the passage rate of digesta through the GIT [31]. Coarse feed particles will remain in the gizzard for a longer period, resulting in greater muscle activity and gizzard weight [45]. According to Auza et al [46], providing 11.25% BSFL flour had a substantial effect on expanding the length of the ileum, this was related to chitin component [47]. Furthermore, the presence of antimicrobial peptide in BSFL contributed to the development of gastrointestinal tract and inhibited pathogenic microbes in the gut, hence enhancing gut balance [48].

Decreasing the CP levels in the diet reduced the percentage of nitrogen, ammonia (p = 0.088) and pH of litter (p<0.05). It is well-documented that decreasing dietary CP levels (1%) reduced nitrogen excretion (about 10%) [49] and litter moisture [50]. Collin et al [51] reported that most N losses through excreta were caused by a dietary CP inability to meet AA requirements, notably imbalances between AA. This had a significant impact on poultry metabolism and endocrine functioning. In addition, reducing dietary CP levels had a correlation with enhancement of animal proportion without lesions [52], decreased water consumption and ammonia concentration (acidic fluids) [53] as well as heat increment [54]. This condition affected the pH level, the lower the pH, the less unionized ammonia is available for volatilization. Ammonia concentration tended to rise with increasing moisture content of the litter due to the majority of microbe species relying on water for proliferation [55]. Normally, the pH of litter ranges between 6.5 and 8.5 [55], a pH below 7.0 (neutral) declined the uricolytic bacterial population is responsible for the production of ammonia and the resulting growth of other bacteria that absorb ammonia. This process leads to a reduction in the volatilization of ammonia in the environment around it [26].

In conclusion, supplementation with BSFLM in low-protein diets declined duck performance, reduced nitrogen, and ammonia in the litter, increased the morphological size of the jejunum, digestive tract, and blood lipid concentration. In contrast, the reduction of 2% CP in SBM-treated diets did not affect growth performance but decreased nitrogen and ammonia concentrations. Thus, the supplementation of BSFLM on low-protein duck diets must consider the stage of larvae related to the digestibility value.