Introduction

Global warming, the Russia-Ukraine war, the Israeli–Palestinian conflict, and the trade war between China and the United States have led to a surge in global food prices. Additionally, the urgent search for sustainable protein sources for livestock feed to address the challenge of feeding, approximately 9.5 billion people globally by 2050 has led humans to explore the potential of the BSF, as one of the most promising insect species for industrial protein production to replace protein feeds such as soybean and fish meal1. This is because they are capable of efficiently upgrading organic waste into high-value protein sources, thereby enhancing the productivity and efficiency of the food chain2. The evidence of the high content of protein, fat, and minerals in BSF has been previously reported in our earlier studies3. Nylon bag technique is simple to operate, cost-effective, and yields accurate results4, making it crucial for assessing the degradation rate and quantity of nutrients in the rumen of ruminant animals4,5.

Nitrogen-containing compounds entering the small intestine of ruminants include feed proteins that escape rumen degradation, microbial proteins synthesized in the reticulorumen, and endogenous proteins6. The rumen-undegradable protein (RUP) entering the small intestine depends on the degradability and passage rate of feed proteins in the reticulorumen7. Because indigestible feed in the rumen is closely related to microbial biomass and endogenous secretions, it is challenging to accurately assess the true small intestinal digestion rate of nutrients in ruminants8. One method to overcome these challenges is the use of the nylon bag technique.

Enhancing the passage rate of RUP and reducing the rumen-degradable protein (RDP) is beneficial for improving the production performance of animals9,10. Methods to improve RUP mainly include formaldehyde treatment, tannin treatment, heat treatment, and physical coating11,12. Due to the toxic side effects of formaldehyde and the anti-nutritional properties of tannins, their use is restricted. The physical coating method is currently less utilized and requires further validation. In comparison, heat treatment is a relatively ideal approach as it is neither toxic nor exhibits anti-nutritional properties. The structures of proteins and starch undergo reorganization in the presence of water. This is due to the interaction between water molecules and protein and starch molecules13. Hydrophobic side chains tend to be buried within the protein, forming a hydrophobic core that stabilizes the protein structure. On the other hand, hydrophilic side chains are more exposed to the solvent14. Under high-temperature conditions, hydroxy compounds and amino groups undergo the Maillard reaction, and the resulting complex can reduce degradation in the rumen15.

Many studies have reported that heat treatment of feed could increase RUP. Karlsson et al.16 reported that, after heat treatment of hempseed cake, both RUP and the intestinal digestion rate of RUP showed a linear increase with the temperature rise. The research discovered by Aufrère et al.17, Subuh et al.18, Ljøkjel et al.19, and Xuewei et al.20 indicates that high-temperature treatment could reduce the degradation of crude protein in the rumen without lowering rumen digestibility. Building upon the aforementioned studies, we employed, for the first time, the in situ nylon bag method and an improved three-step in vitro method to assess the impact of different temperature treatments on the ruminal degradation and intestinal protein digestibility of BSF.

Materials and methods

Ethics approval and consent to participate

This experiment was approved by the Animal Welfare Committee of Suranaree University of Technology (korat, Thailand), No. SUT-IACUC-023/2021. The experiment was carried out at the SUT Goat Farm , Nakhon Ratchasima, Thailand (14°53′37.9′′ N, 102°01′22.0′′ E) in accordance with guidelines and regulations and the animal experiments complied with the ARRIVE guidelines.

Experimental design

This experiment will adopt a single-factor completely randomized experimental design. This study comprises 8 treatments, with 6 replications per treatment, each replication consisting of 1 sample. The supplement level of cassava was determined regarding the quantitative relationship between CP rumen degradability and intestinal digestibility after high-pressure treatment of reducing sugars and proteins21. The treatment group mixes 95% BSF or 95% full-fat soybeans (FFS) with 5% cassava. Subsequently, 75% water was added, and the mixture was vigorously kneaded to achieve uniform blending. The resulting mixture is then pressed to a thickness of approximately 5 cm, and will be placed in an oven for drying, with a time setting of 120 min and temperatures of 120 °C and 140 °C, respectively22. There are 8 groups, with FF and BS as the negative control groups, FFC and BSC as the positive control groups, and 12FFC to 14BSC as the treatment groups. Group details are provided in Table 1.

Table 1 Design factor level.
Full size table

Animals and diets

Three Nubian rams (29.1 ± 2.16 kg) each were equipped with permanent cannulas in the rumen and proximal duodenum. They were housed individually in pens (3 × 2.5 m2), and each ram was fed twice a day at 08:00 and 17:00. The diet was formulated to feed 3% of the body weight, and goats had free access to fresh water and mineral salt blocks. The BSF larvae were purchased from a farmer in Henan Province, China. The basic daily feed for BSF consisted of kitchen waste. At 12 days old, the larvae were immersed in water at 90 °C for 90 s, following a ratio of BSF to water was 1:8. Subsequently, the larvae were immersed in cold water and drained, then dried in a 65 °C oven for 48 h. The formulated diet, by the nutritional requirements for goats (NRC 2007) (crude protein: 14.0%, Ca: 0.6%, P: 0.4% of DM), is detailed in Table 2, specifying its composition and nutritional levels.

Table 2 Feed ingredients and nutrient components.
Full size table

Rumen degradability

The nylon bag method, as determined by DONG et al.4, was employed for the rumen culture to assess the degradation kinetics of DM and CP. In short, approximately 3 g of sample (on an air-dry matter basis) were weighed and placed into nylon bags (4.5 × 10 cm2; 40 µm pore size) labeled with numerical codes. Considering that goats are small ruminants, to facilitate the normal degradation of nutrients in the nylon bags in this experiment, the nylon bags were placed into the rumen twice in chronological order. The bags were then incubated in the rumen for 0, 2, 4, 8, 12, 24, and 48 h before being immediately removed. Subsequently, all bags were placed in a bucket filled with cold tap water to halt microbial fermentation. After manual cleaning in cold tap water, the bags were dried in an oven at 65 °C for 48 h. Weigh the extracted material using an analytical balance. Subsequently, grind the residual material from the nylon bag, pass it through a 1 mm sieve, and prepare it for measurement. The 0 h (2 bags for each sample) incubation samples were not incubated in the rumen, but they were washed as described above. Residues from the bags were pooled within the incubation time and treatment.

In vitro three-step procedure

Drawing inspiration from the approach of Boucher et al.23, improvements were made to the in vitro three-step technique. In summary, at the 16th hour, the nylon bag (5 × 12 cm2; 40 µm pore size) inside the rumen was retrieved, suspended in a 0.1% concentration solution of methylcellulose, and incubated at 37 °C for 30 min (Inhibition of microbial activity). Subsequently, it was removed and stored in a − 20 °C freezer. Before gastric protease treatment, the nylon bag was thawed, and subjected to three 5-min wash cycles in a washing machine to eliminate bacteria from rumen residues. Subsequently, the bag was dried in a 65 °C oven and sieved through a 1 mm mesh.

The P-7000 gastric protease and P-7545 pancreatic protease from Sigma Corporation in the United States were employed. One gram of undegraded rumen feed residue is weighed and placed into a nylon bag (5 × 12 cm2; 40 µm pore size). The sample is then immersed in a solution consisting of 10 mL of pH 1.9 and 0.1 N HCl, with 1 g/L gastric protease (Sigma P-7000, Sigma), followed by vortexing and incubation for 1 h at 39 °C in a shaking water bath. After incubation, 0.5 mL of 1 N NaOH solution and 13.5 mL of pancreatic enzyme solution (0.5 M KH2PO4 buffer, standardized at pH 7.75, containing 3 g/L pancreatic enzyme, Sigma P-7545, Sigma) were added. The sample was vortexed and incubated for 24 h at 39 °C in a shaking water bath, with vortexing every 4 h. After the 24-h incubation, the bags were rinsed with tap water and then dried at 65 °C until a constant weight was achieved (approximately 48 h). Weighing was carried out using an analytical balance, and the data were recorded. The collected samples in the sample bags were subjected to laboratory analysis to determine the CP content in the samples.

Chemical analysis

Conventional nutrients

All feed samples were dried at 65 °C in a vacuum oven for 72 h, ground, and sieved through a 1-mm sieve for further analysis. Following the standard procedures outlined by AOAC (2012) 934.01, 927.02, and 976.05, dry matter (DM), crude ash, ether extract (EE), and Kjeldahl nitrogen (N) analyses were conducted on the original feed samples, rumen undegraded residues, and gastric protease/pancreatic enzyme-digested residues. In which the conversion factor of protein was 6.25 that of N. Neutral detergent fiber (NDF) and acid detergent fiber (ADF) analyses were performed using the method of Van et al.24 without the inclusion of sodium sulfite.

Minerals analysis

Mineral content is determined following the method outlined by Pieterse et al.25. In brief, 5 mL of 6 mol L−1 hydrochloric acid is added to 0.5 g of the sample. The mixture was placed in an oven at 50 °C for 30 min, removed, and then 35 mL of distilled water was added. The solution was filtered and adjusted to a final volume of 50 mL. Mineral concentrations were determined using an iCAP 6000 series inductively coupled plasma (ICP) spectrophotometer (Thermo Electron Corporation, Strada Rivoltana, 20090 Rodana, Milan, Italy), which was equipped with a vertical quartz torch and a Cetac ASX-520 autosampler. Mineral concentrations were calculated using TEVA Analyst software.

Fatty acids (FA) analysis

FA were extracted using a chloroform–methanol solution following the method outlined by Tian et al.26. In brief, utilizing n-hexane as the internal standard, a gas chromatograph-mass spectrometer (GC–MS; Thermo Fisher Scientific) was employed under the following conditions: Thermo TG-FAME capillary column (50 m × 0.25 mm × 0.20 µm), 1 µL injection volume, 8:1 split ratio; inlet temperature 250 °C, ionization temperature 230 °C, transmission line temperature 250 °C, quadrupole temperature 150 °C. Helium served as the carrier gas with a flow rate of 0.63 mL/min, and the ionization energy was set at 70 eV.

Amino acid analysis

The preprocessing of BSF amino acid analysis followed the method outlined by Tian et al.27. The UPLC conditions were as follows: individual amino acids were separated on an ACQUITY UPLC BEH C18 column (2.1 × 100 mm × 1.7 µm, Waters, Milford, USA) with a column temperature of 40 °C; the injection volume was 5 µL. The mobile phase consisted of A = 10% methanol (containing 0.1% formic acid) and B = 50% methanol (containing 0.1% formic acid). The gradient elution conditions were as follows: 0∼6.5 min, 10∼30% B; 6.5∼7 min, 30∼100% B; 7∼8 min, 100% B; 8∼8.5 min, 10∼100% B; 8.5∼12.5 min, 10% B. The flow rate was as follows: 0∼8.5 min, 0.3 mL/min; 8.5∼12.5 min, 0.3∼0.4 mL/min. The mass spectrometry (MS) conditions were as follows: electrospray ionization source, positive ion ionization mode; ion power temperature was 500 °C, ion source voltage was 5500 V; collision gas pressure of 6 psi, curtain gas pressure of 30 psi; nebulization gas pressure and aux gas pressure were both 50 psi; and multiple-reaction monitoring scan mode.

Chitin analysis

The chitin content of the BSF meal was analyzed following the method outlined by Liu et al.28 with minor modifications. In brief, an aliquot of the prepupae meal (90–100 mg) was enclosed in an ANKOM filter bag (ANKOM Technology, Macedon, NY, USA) shaped to fit a 15 mL screw cap centrifuge tube. This aliquot underwent demineralization for 30 min in 5 mL of 1 M HCl at 100 °C. The demineralization process was followed by five washing steps in ASTM Type I water, ensuring neutrality. Subsequently, a deproteinization step was carried out in 5 mL of 1 M NaOH at 80 °C for 24 h. Finally, the sample was washed five times in ASTM Type I water until neutrality was achieved. After drying at 105 °C in an air-forced oven for 2 h, the chitin content (CT, g/kg DM) was calculated using the following formula:

$$ CT = 1000 \times \frac{{Fw - \left( {Bw \times C} \right)}}{Sw} $$

where Fw = weight after demineralization, deproteinization, and drying (g), Bw = weight of the modified ANKOM extraction bag (g), C = dimensionless factor taking in account the MEAN weight loss of extraction bags (0.999, n = 6) treated according to the same procedure used for the samples, and Sw = exact amount of sample processed (g).

Kinetic modeling and statistical analysis

The results of DM and CP disappearance from nylon bags were fitted into the following exponential equation of Ørskov and McDonald29 using non-linear regression (SAS 2003):

$${\text{P}} = { }a + b{ }(1 - e^{ - ct}$$

).

where P is the disappearance of nutrients during time t, a is the soluble nutrient fraction rapidly washed out of the bags and assumed to be completely degradable, b is the proportion of insoluble nutrients potentially degradable by microorganisms, e is the natural logarithm, c is the degradation rate of fraction b per hour (i.e. k) and t is a time of incubation.

The effective degradability (ED) of DM and CP in situ of each feed sample for each of the three lambs was calculated as follows:

$$ {\text{ED}} = a + b \times c/\left( {k + c} \right) $$

where ED denotes the effective degradation rate (%); k signifies the ruminal outflow velocity of the feed. For reference, Castro et al.30 recommend using k = 0.08%/h.

The effective degradability of CP is also referred to as the rumen-degradable protein (RDP) content of CP in the feed. The calculation is as follows:

$$ \begin{aligned} {\text{RDP}}\;\left( {{\text{g}}/{\text{kg}}} \right)\; = & {\text{CP}}\left( {\text{\% }} \right) \times {\text{ED}}/10 \\ {\text{RUP}}\;({\text{g}}/{\text{kg}})\; = & {\text{CP}}\left( {\text{\% }} \right) \times 10 - {\text{RDP}} \\ \end{aligned} $$

The content of rumen undegradable protein (RUP), and the digestible crude protein (IDCP) in the small intestine are as follows:

$$ \begin{aligned} {\text{Idg}}\left( {\text{\% }} \right) = & 100 \times \left( {{\text{CP}}_{{16{\text{h}}}} - {\text{CP}}_{{\text{N}}} } \right)/{\text{CP}}_{{16{\text{h}}}} \\ {\text{IDCP }}\left( {{\text{g}}/{\text{kg}}} \right) = & {\text{RDP}} \times 0.85 \times 0.775 + {\text{RUP}} \times {\text{Idg}} \\ \end{aligned} $$

where, Idg: Small intestinal digestibility of protein, CP16 h: protein in rumen-degraded residues of 16 h (g/kg), CPN: protein in residues after small intestinal digestion (g/kg), the rumen-degraded protein coefficient of MCP is 0.85, and the coefficient of small intestinal digestibility of MCP is 0.775 (NRC, 2001).

Data for a, b, and c values and ED and results from the three-step in vitro procedure were analyzed using the GLM option of the software program of SAS (1987) according to the following model:

$$ {\text{Yij }} = {\text{ m }} + {\text{ dij }} + {\text{ eij}} $$

where Yij = the criteria under study, m = overall mean, dij = feed source effect, and eij = residual error.

Data were analyzed by ANOVA with SAS 9.1.3. The statistically significant differences were determined by Duncan’s multiple-range tests. Data were presented as the MEAN and SEM. The significance level was indicated at P < 0.05.

Results and discussion

Proximate composition of FF and BSF

The proximate composition of FF and BSF is shown in Table 3. The CP and P content of FF were significantly higher (P < 0.05) than that of BSF, while other nutritional components and mineral content were significantly higher (P < 0.05) in BSF than in FF. The chitin content was 7.1% in BSF.

Table 3 The proximate composition of BSF (% of DM).
Full size table

Amino acid content of FF and BSF

The amino acid composition of FF and BSF is shown in Table 4. In indispensable amino acids, the levels of arginine, histidine, isoleucine, leucine, lysine, and threonine in FF were significantly higher (P < 0.05) than those in BSF, whereas methionine and phenylalanine in BSF were significantly higher (P < 0.05) than in FF. There was no statistically significant difference (P > 0.05) in valine between FF and BSF. Overall, the content of indispensable amino acids in FF was higher than that in BSF.

Table 4 Amino acid content of BSF (g/100 g).
Full size table

Fatty acids content FF and BSF

The fatty acid composition of FF and BSF is shown in Table 5. FF lacked fatty acids ranging from C6:0 to C12:0, while BSF was abundant in C12:0, C14:0, and C16:0. The content of C18:1 n − 9t in BSF was significantly higher (P < 0.05) than in FF, whereas FF had higher (P < 0.05) levels of C18:2 n − 6c and C18:3 n − 3 compared to BSF. However, BSF had higher (P < 0.05) levels of C18:3 n − 6. Saturated fatty acids (SFA) were significantly higher in BSF compared to FF, while unsaturated fatty acids (UFA) were significantly lower (P < 0.05). The content of n − 3 PUFA and n − 6 PUFA was significantly higher (P < 0.05) in FF than in BSF. There was no statistically significant difference in the n − 3/n − 6 ratio between FF and BSF.

Table 5 Fatty acids content of FF and BSF (g/100 g).
Full size table

Effect of different treatments on DM degradability in the rumen

The ruminal degradability of DM for each treatment is presented in Table 6. Throughout the experimental period, the ruminal degradation rate of DM in the 14BSC group consistently remained the highest (P < 0.05), while that of the BS group consistently remained the lowest (P < 0.05), there was no difference (P > 0.05) between the BSC group and the 12BSC and 14BSC groups between 2 and 24 h. Heat treatment did not affect the group containing full-fat soybeans (P > 0.05). The BSF without heat treatment (BS and BSC groups) showed a significant reduction (P < 0.05) at 48 h.

Table 6 Effect of different treatments on dry matter (DM) degradability in the rumen (%).
Full size table

There were no significant differences (P > 0.05) in a* and c* among all treatments; however, the b* in the FF group was significantly higher (P < 0.05) than in all treatments containing BSF. The a + b* in the FF group was significantly higher (P < 0.05) than in the BS, BSC, and 12BSC groups. There was only the BS group had a significantly lower (P < 0.05) effective degradability (ED) compared to all other groups.

Effect of different treatments on CP degradability in the rumen

The ruminal degradation rates of CP for each treatment are presented in Table 7. The ruminal CP degradation rate was highest (P < 0.05) in the 14BSC group, followed by BSC, BS, and the 12BSC group (P < 0.05). There was no statistical difference (P > 0.05) between the 12FFC and 14FFC groups. The ruminal protein degradation rate was lowest (P < 0.05) in the FF and FFC groups, with no statistical difference (P > 0.05) between these two groups.

Table 7 Effect of different treatments on crude protein (CP) degradability in the rumen (%).
Full size table

For degradation kinetic parameters, the order of values from high to low was as follows: 14BSC, BSC, 12BSC, BS, FFC, 14FFC, 12FFC, and FF (P < 0.05). All a* values containing BSF were higher (P < 0.05) than those containing full-fat soybeans, and there was no statistical difference (P > 0.05) among the soybean groups. The order of b* values from high to low was as follows: 14BSC, FF, BS, BSC, 12BSC, 12FFC, 14FFC, and FFC (P < 0.05). The sum of a and b was highest (s < 0.05) in 14BSC, and lowest (P < 0.05) in FFC, 12FFC, and 14FFC, there was no difference (P > 0.05) between FF, BS, FFC, and 12BSC. The effective degradability (ED) in 14BSC was significantly higher (P < 0.05) than in all groups, and BS and BSC were significantly higher (P < 0.05) than in FF, FFC, 12FFC, 14FFC, and 12BSC.

Effects of different temperature treatments on Idg and IDCP of RUP

The effects of different temperature treatments on Idg and IDCP of RUP are presented in Table 8. The Idg of full-fat soybeans after heat treatment (in the 12FFC and 14FFC groups) was significantly higher (P < 0.05) than in other groups. There was no significant difference (P > 0.05) in Idg among FF, FFC, and the 14BSC group. The Idg of BSF without heat treatment (in the BS and BSC groups) was the lowest (P < 0.05). The IDCP of full-fat soybeans after heat treatment (in the 12FFC and 14FFC groups) was significantly higher (P < 0.05) than in other groups, followed by the FF (P < 0.05) and FFC groups (P < 0.05). All IDCP values containing BSF were significantly lower (P < 0.05) than those containing full-fat soybeans, in the following order from high to low: 14BSC (P < 0.05), 12BSC (P < 0.05), BSC (P < 0.05), and BS groups (P < 0.05).

Table 8 Effects of different temperature treatments on Idg and IDCP of RUP.
Full size table

Effects of different temperature treatments on total digestibility

The effects of different temperature treatments on total digestibility are presented in Fig. 1. Heat treatment did not seem to affect the total digestibility of full-fat soybeans, and treatment at 120 °C did not affect the total digestibility of BSF, but when the treatment temperature reached 140 °C, the abundance of total digestibility was the highest.

Figure 1
figure 1

Effects of different temperature treatments on total digestibility. Idg = small intestinal digestibility of protein, RDP = rumen-degradable protein.

Full size image

Discussion

The degradation rate of DM in the rumen of BSF supplemented with cassava and heat-treated was higher than that of untreated BSF, while the DM degradation rate of full-fat soybeans falls between these two, furthermore, there was no difference between heat-treated and untreated full-fat soybeans. There were considerable differences in the research findings regarding the effect of heat treatment on the degradation rate of DM in the rumen. A study reported by Molosse et al.31 demonstrated that different times and temperatures of treatment reduced the degradation rate of DM in the rumen of cottonseed meal. However, previous research has found that the degradation rate of DM in green tea waste treated at 121 °C increased. Additionally, the concentrations of volatile fatty acids (VFA), acetate-to-propionate ratio, ammonia nitrogen (NH3-N) concentration, and cumulative gas production during in vitro fermentation significantly increased32. Interestingly, a study reported by Campbell et al.33 found that there were no changes in the in vitro digestibility rates of monogastric and ruminant animals after BSF was treated at 90 °C. The factors leading to these results may be related to starch content because the degradation rate of DM in the cassava-containing BSF group in this experiment did not also significantly differ from that of the heat-treated group. There was only the effective degradability (ED) of the BS group was significantly lower than that of the other groups. This may be related to the presence of chitin, which has been shown to have anti-nutritional properties34, heat treatment and the presence of starch may slightly alter its anti-nutritional properties. Heat treatment did not affect the rapidly washable fraction (a*) but reduced the potentially degradable fraction (b*). This finding contrasts with the results of Lund et al.35, who suggested that heat treatment increased the potentially degradable fraction, allowing rumen microbes to enter and degrade. The decrease in the potentially degradable fraction in this study may be related to the protective layer formed by the Maillard reaction between starch and protein36. In summary, heat treatment of BSF was advantageous for increasing the degradation rate of DM, did not affect the rapidly washable fraction (a*), and reduced the potentially degradable fraction (b*).

Heat treatment can protect feed proteins from microbial degradation in the rumen, thereby increasing the amount of digestible amino acids available in the small intestine. The results of this study revealed that the degradation rate and degradation parameters of proteins in the 14BSC group were the highest, while those of the FF control group, which underwent no treatment, were the lowest. Among the treatments containing BSF, the 12BSC group showed the lowest degradation rate. This indicates that the protective effect of the 12BSC group is slightly better. A study reported by Karlsson et al.37 found that the rumen undegradable protein (RUP) of hempseed cake treated at 110 °C, 120 °C, and 130 °C increased linearly with temperature. Rigon et al.38 also found that the rapidly washable fraction (a*) of CP in heat-treated peanut meal was higher, similar to the 14BSC group, with a decrease in effective degradability (ED) and an increase in RUP, which was similar to the 12BSF group. A result of Stern et al.39, concluded that heat treatment led to a decrease in the apparent degradation rate of proteins. The above research findings are consistent with those of the 12BSC group. However, it is disappointing that the results of the 14BSC group did not meet expectations. At present, we are unable to explain the reason for this outcome. The CP degradation rate of BSF is higher than that of full-fat soybeans, the factors leading to this result may be related to the protein source. Animal-derived proteins have a higher nutritional quality, implying that their digestibility is significantly greater than that of plant-derived proteins40. Furthermore, the degradation rate of soybean protein in this study appears to be quite low, possibly due to the high-fat content in both soybeans and BSF. The elevated fat concentration reduces the pH of the rumen, inhibiting the microbial activity associated with the degradation of plant-based protein sources41. In summary, the treatment of BSF at 120 °C increased the RUP, while treatment at 140 °C enhanced the degradation kinetic parameters.

The small intestinal protein digestibility of full-fat soybeans after heat treatment is significantly higher than that of all other treatments, and the small intestinal protein digestibility of full-fat soybeans is higher than that of treatments containing BSF. The factors that led to these results may be related to anti-nutritional factors. In whole-fat soybeans, the main anti-nutritional factors are pancreatic protease inhibitors and soy agglutinins, which are more easily destroyed at high temperatures42. In contrast, the anti-nutritional factor in BSF is mainly chitin43. Chitin is less damaged under heat treatment compared to anti-nutritional factors in soybeans44,45. Therefore, the digestibility of soybean proteins in the small intestine was higher. However, heat treatment also benefits the digestion of BSF proteins in the small intestine, as the digestibility of untreated BSF small intestine proteins was the lowest. This study found that heat treatment had only a slight effect on improving the total digestibility of soybean protein. For BSF, the total digestibility abundance was the highest when treated at 140 °C, while treatment at 120 °C had no effect on the total protein digestibility of BSF. A study by Shimelis and Rakshit46 found that heat treatment led to an increase in the in vitro protein digestibility of beans, which can be explained by the reduction of anti-nutritional factors such as tannins and trypsin inhibitor activity. However, it was difficult to explain why there was a large difference between 120 and 140 °C. It could be because this temperature range is the critical point where some anti-nutrients are destroyed. In summary, the treatment of BSF at 140 °C was conducive to enhancing the digestibility of its proteins in the small intestine.

Conclusion

Compared to the BS group, heat-treated BSF showed increased rumen DM degradability and effective degradability. The 14BSC group increased rumen CP degradability and degradation kinetic parameters, while the 12BSC group decreased these parameters. The CP degradability of BSF was significantly higher than that of full-fat soybeans. The Idg and IDCP of heat-treated full-fat soybeans were significantly higher than those of other treatment groups. At the same time, heat treatment was beneficial to increasing the Idg and IDCP of BSF, and the 14BSC treatment effect was significantly better than that of the 12BSC group. Therefore, based on the results of this experiment, it is recommended to supplement BSF with cassava and subject them to heat treatment at 140 °C.