Determination and prediction of the available energy and amino acids digestibility of full-fat soybean fed to growing pigs

Abstract

Two experiments were conducted to determine the digestible energy and metabolizable energy contents, as well as the apparent ileal digestibility and standardized ileal digestibility of amino acids in full-fat soybean fed to growing pigs. Ten full-fat soybean samples were collected from different areas in China and used in two experiments in this study. In Exp. 1, 66 growing pigs (initial body weight = 18.48 ± 1.2 kg) were randomly allotted to 1 of 11 diets (n = 6) including a corn basal diet and 10 experimental diets formulated by replacing the corn with 30% full-fat soybean. In Exp. 2, 11 growing pigs (initial body weight = 50.45 ± 3.2 kg) were surgically equipped with a T-cannula in the distal ileum and arranged in a 6 × 11 Youden square design with 11 diets and 6 periods. The diets included an N-free diet based on cornstarch and sucrose and 10 experimental diets formulated with full-fat soybeans as the sole source of amino acids. Chromic oxide was added into the diets as an indigestible maker to calculate the digestibility of the amino acids. Results showed that there was considerable variation in neutral detergent fiber, acid detergent fiber, and trypsin inhibitor contents in the 10 full-fat soybean samples with a coefficient of variation greater than 10%. On a dry matter basis, the averaged digestible energy and metabolizable energy values in the 10 full-fat soybean samples were 4,855 and 4,555 kcal/kg, respectively, both were positively correlated with the ether extract content. The best-fitted prediction equations for digestible energy and metabolizable energy of full-fat soybean were: digestible energy, kcal/kg = 3,472 + 94.87 × ether extract − 97.63 × ash (R² = 0.91); metabolizable energy, kcal/kg = 3,443 + 65.11 × ether extract − 36.84 × trypsin inhibitor (R² = 0.91). In addition, all full-fat soybean samples showed high apparent ileal digestibility and standardized ileal digestibility values in amino acids and were all within the range of previously published values. Those values significantly varied among different samples (P < 0.05) for most amino acids, except for glycine and proline. In conclusion, full-fat soybean is a high-quality protein ingredient with high ileal digestibility of amino acids when fed to growing pigs, and the metabolizable energy value of full-fat soybean could be predicted based on its ether extract and trypsin inhibitor contents.

This study determined the digestible energy, metabolizable energy, and digestible amino acids contents of full-fat soybean fed to growing pigs, and established the corresponding prediction equations. The results showed that full-fat soybean is a high-quality protein ingredient.

Introduction

Soybean is a widely used protein ingredient in pig and poultry feed considering its relatively high and balanced indispensable amino acids (AA) contents that were close to the AA requirements of animals (except for the low sulfur-containing amino acids contents) (Woyengo et al., 2014). Consequently, soybean is grown in large quantities and has become one of the most produced crops with an annual yield of over 300 million tons worldwide, and the major producing countries of soybean include Brazil, the United States, Argentina, and China (FAO, 2020). However, the antinutritional factors contained in raw soybeans, especially urease and trypsin inhibitor (TI), can depress the growth rate of pigs and poultry (Grant, 1989). Heating has been proven as one of the most effective approaches to denature the antinutritional factors in soybeans to result in better growth performance of animals (Faber and Zimmerman, 1973).

The full-fat soybean (FFSB) is one of the main products after heating soybean and is usually used as a protein ingredient in diets for weaned and growing pigs (Kim and Kim, 1997). There are different technologies to produce FFSB: extrusion, roasting, jet sploding, and micronization, among which extrusion could yield significantly higher nutrient digestibility compared with the other three technologies (Marty and Chavez, 1993). After such heating procedures, the antinutritional factors were inactivated and the nutrients were easier to be absorbed. Meanwhile, FFSB contained abundant AA compositions and nearly 20% crude fat, which could produce a volatile aroma after heating processing, resulting in a stimulated feed intake of animals. All these advantages make FFSB an ideal protein source for pigs.

However, the relatively high price and lacks of available nutrition information limit the wide utilization of FFSB in animal feed. Previous studies related to FFSB evaluation only focused on the comparison with soybean meal, and only 1 sample was determined in 1 trial (Baker et al., 2010; Park et al., 2017; Kim et al., 2022). As a result, it is vital to conduct comprehensive evaluation trials to acquire accurate information on the available energy and digestible AA contents in FFSB to better formulate a least-cost diet (Noblet, 2007). Therefore, the objectives of this study were to determine the digestible energy (DE) and metabolizable energy (ME) contents, as well as the apparent ileal digestibility (AID) and standardized ileal digestibility (SID) of AA in extrusion FFSB fed to growing pigs, and to develop prediction equations for DE and ME of FFSB.

Materials and Methods

The experiment protocols used in this study were reviewed and approved by the Institutional Animal Care and Use Committee of China Agricultural University (Beijing, China).

General

This study consisted of two experiments and both two were conducted in the Metabolism Laboratory of the Ministry of Agriculture Feed Industry Center at China Agricultural University. Ten FFSB samples were collected from the major produced provinces of China, including Fujian (sample 1), Jiangsu (sample 2, 9), Shandong (sample 3, 5), Henan (sample 4), Liaoning (sample 6, 7) and Hebei (sample 8, 10) provinces. These FFSB samples were all produced using the extrusion method. More specifically, the extrusion method used friction as the heat source, accompanied by high pressure and shear force. Heat and pressure were generated by using a screw with increased resistance to pass the material through a barrel. On the one hand, the shear force destroyed the spatial structure of proteins and saccharides of FFSB, making them more digestible, and on the other hand, the high temperature and pressure inactivate most of the antinutritional factors in FFSB. The chemical compositions and AA contents of the 10 FFSB samples were presented in Table 1. Pigs were housed individually in stainless-steel metabolism crates (length:1.4 m; height: 0.7 m; width: 0.6 m). Each metabolism crate provided a feeder and a water nipple. All pigs in the metabolism crates had free access to water. Room temperature was controlled at 22 °C ± 2 °C.

Table 1.

Analyzed chemical compositions and amino acids of full-fat soybean (DM basis)

Items	Full-fat soybean1
	1	2	3	4	5	6	7	8	9	10	Mean	CV
Dry matter	91.03	90.32	91.59	91.25	91.62	93.31	90.11	90.41	91.56	91.07	91.23	0.95
Gross energy, kcal/kg	5,673	5,680	5,651	5,765	5,702	5,575	5,613	5,638	5,710	5,531	5,654	1.15
Crude protein	41.40	39.67	39.56	38.94	39.27	38.63	39.08	40.62	39.12	38.81	39.51	2.09
Ether extract	21.50	21.43	20.43	20.30	19.58	16.90	21.98	20.63	20.98	19.78	20.35	6.68
Neutral detergent fiber	12.21	12.25	13.95	11.86	13.81	16.35	11.38	12.34	12.20	15.50	13.18	11.98
Acid detergent fiber	6.39	5.29	5.55	5.19	7.94	7.07	4.83	5.72	5.39	8.00	6.14	17.88
Ash	5.87	5.55	5.68	5.68	5.38	5.41	5.61	5.38	5.38	6.23	5.62	4.58
Calcium	0.28	0.26	0.23	0.25	0.20	0.25	0.24	0.25	0.24	0.29	0.25	9.52
Phosphorus	0.42	0.45	0.40	0.48	0.44	0.40	0.44	0.45	0.41	0.48	0.44	6.22
Trypsin inhibitor unit/mg	6.35	6.55	6.21	6.84	5.96	6.75	5.39	4.80	4.05	4.96	5.79	15.44
Indispensable amino acids
Arg	2.51	2.44	2.47	2.45	2.61	2.47	2.55	2.47	2.43	2.58	2.50	2.41
His	0.93	0.91	0.91	0.94	0.99	0.92	0.96	0.92	0.92	0.95	0.94	2.50
Leu	2.67	2.57	2.68	2.65	2.77	2.68	2.80	2.70	2.70	2.84	2.70	2.72
Ile	1.66	1.58	1.66	1.62	1.74	1.69	1.81	1.72	1.68	1.79	1.70	3.89
Lys	2.34	2.34	2.37	2.42	2.48	2.38	2.49	2.41	2.41	2.46	2.41	2.11
Met	0.51	0.48	0.43	0.53	0.52	0.49	0.56	0.51	0.50	0.46	0.50	7.00
Phe	1.84	1.79	1.82	1.79	1.90	1.81	1.92	1.83	1.84	1.97	1.85	3.07
Thr	1.40	1.35	1.41	1.42	1.45	1.40	1.45	1.41	1.40	1.46	1.41	2.19
Trp	0.49	0.47	0.47	0.47	0.49	0.48	0.51	0.50	0.48	0.48	0.48	2.41
Val	1.79	1.72	1.76	1.75	1.80	1.76	1.85	1.77	1.76	1.84	1.78	2.11
Dispensable amino acids
Ala	1.33	1.29	1.31	1.31	1.35	1.31	1.36	1.31	1.30	1.37	1.32	2.04
Asp	4.09	3.97	4.06	4.01	4.21	4.03	4.23	4.06	4.07	4.25	4.10	2.27
Cys	0.51	0.46	0.51	0.52	0.56	0.50	0.53	0.49	0.50	0.48	0.50	5.07
Glu	6.44	6.23	6.46	6.36	6.62	6.41	6.74	6.46	6.46	6.77	6.49	2.46
Gly	1.86	1.80	1.82	1.83	1.89	1.82	1.91	1.83	1.83	1.91	1.85	2.07
Pro	1.94	1.87	2.08	1.87	2.03	1.91	2.01	2.04	1.87	1.98	1.96	3.77
Ser	1.80	1.77	1.81	1.82	1.90	1.81	1.84	1.80	1.80	1.90	1.82	2.23
Tyr	1.01	1.01	1.14	1.12	1.17	1.12	1.10	1.14	1.17	1.22	1.12	5.67

¹Sources of FFSB are: Fujian (sample 1), Jiangsu (sample 2, 9), Shandong (sample 3, 5), Henan (sample 4), Liaoning (sample 6, 7), and Hebei (sample 8, 10).

CV, coefficient of variation.

Experiment 1: energy measurements

Animals, diets, and experimental design

Experiment 1 was conducted to determine the DE and ME of the 10 FFSB samples fed to growing pigs. Sixty-six crossbred barrows (Duroc × Landrace × Yorkshire) with an initial body weight (BW) of 18.48 ± 1.2 kg were allotted to 1 of 11 diets in a completely randomized design with 6 barrows per treatment. The 11 experimental diets contained a corn basal diet and 10 FFSB diets (Table 2). The FFSB diets were formulated by replacing 30% of the energy-supplying ingredients in the basal diet with FFSB. The amounts of dicalcium phosphate, limestone, vitamins, salt, and premix were kept constant in all diets. The chemical compositions of the diets used in Exp. 1 were analyzed and shown in Table 3. All the diets were fed in mash form. Daily diet allowance (4% of each pig BW) was given in two equal meals supplied at 0830 h and 1730 h daily.

Table 2.

Ingredient compositions of the diets used in Exp. 1 and Exp. 2 (as-fed basis)

Ingredients, %	Exp. 1		Exp. 2
	Basal	Full-fat soybean	N-free	Full-fat soybean
Corn	96.77	67.74	77.43	41.83
Full-fat soybean	–	29.03	–	40
Cornstarch	–	–	–
Sucrose	–	–	15	15
Acetate cellulose	–	–	4	–
Dicalcium phosphate	1.26	1.26	1.6	1.6
Limestone	1.17	1.17	0.52	0.52
Chromic oxide	–	–	0.25	0.25
Salt	0.3	0.3	0.3	0.3
Potassium carbonate	–	–	0.3	–
Magnesium oxide	–	–	0.1	–
Premix	0.5	0.5	0.5	0.5
Total	100	100	100	100

¹Made by Chemical Reagents Company (Beijing, China).

²Premix compositions: 5512 IU vitamin A as retinyl acetate, 2200 IU vitamin D₃ as cholecalciferol, 30 IU vitamin E as dl-alpha-tocopheryl acetate, 2.2 mg vitamin K₃ as menadione nicotinamide bisulfite, 4 mg riboflavin, 14 mg pantothenic acid as dl-calcium pantothenate, 27.6 μg vitamin B₁₂, 30 mg niacin, 400 mg choline chloride, 0.7 mg folacin, 3 mg pyridoxine as pyridoxine hydrochloride, 1.5 mg thiamin as thiamine mononitrate, 44 μg biotin, 40 mg Mn as MnO, 75 mg Fe as FeSO₄·H₂O, 75 mg Zn as ZnO, 100 mg Cu as CuSO₄·5H₂O, 0.3 mg I as KI, and 0.3 mg Se as Na₂SeO₃.

Table 3.

Analyzed chemical compositions of the 10 full-fat soybean diets used in Exp. 1 (as fed-basis)

Items, %	Basal diet	Full-fat soybean diets
		1	2	3	4	5	6	7	8	9	10
Dry matter	86.43	88.13	88.16	88.95	88.41	87.36	88.65	88.64	88.35	88.13	88.57
Gross energy, kcal/kg	3,714	4,142	4,119	4,145	4,093	4,126	4,082	4,077	4,086	4,118	4,074
Crude protein	7.14	15.92	15.57	16.10	15.99	16.35	15.18	15.59	15.79	16.05	15.51
Ether extract	2.29	6.37	5.72	5.62	5.99	5.44	5.35	6.63	5.78	6.38	5.70
Neutral detergent fiber	11.28	11.26	11.21	12.80	11.11	12.33	13.77	11.95	11.78	12.80	13.96
Acid detergent fiber	2.34	3.68	3.88	3.70	3.77	4.20	4.48	3.91	3.71	3.93	4.45
Crude fiber
Ash	3.83	4.83	4.75	4.94	4.99	4.74	4.67	4.81	4.72	4.77	4.94
Calcium	0.73	0.77	0.76	0.73	0.78	0.79	0.76	0.79	0.81	0.88	0.82
Phosphorus	0.44	0.49	0.51	0.55	0.51	0.49	0.47	0.47	0.48	0.47	0.52

Sample collection

The animal trial lasted 12 d, including 7 d for adaption of the cage and diets, and 5 d for total collection of the feces and urine. During the collection period, feed refusals and spillage were collected twice daily and then dried and weighed. Feces from each pig were collected quickly and then immediately stored in plastic bags at −20 °C. A bucket containing 50 mL of 6 N HCl was placed under the metabolism crates to collect urine.

The volume of urine collected each day was recorded and 10% of the daily collected urine was stored at −20 °C. At the end of the animal trial, the collected feces and urine samples of each pig were thawed, pooled, homogenized, and weighed. Then fecal subsample and urine subsample were collected. Approximately 600 g of fecal subsamples were dried at 65 °C in a drying oven for 72 h. After drying, the fecal subsamples were ground through a 1-mm screen and stored at −20 °C for further analysis. Approximately 45 mL of urine subsamples were filtered and stored at −20 °C for further analysis.

Chemical analyses

The FFSB samples and 11 experimental diets used in Exp. 1 were analyzed for dry matter (DM, procedure 930.15; AOAC, 2016), crude protein (CP = N × 6.25, procedure 984.13; AOAC, 2016); ether extract (EE; Lyu et al., 2019), calcium (Ca, procedure 968.08; AOAC, 2016), and phosphorus (P, procedure 946.06; AOAC, 2016). The neutral detergent fiber (NDF) and acid detergent fiber (ADF) were determined using a fiber analyzer (ANKOM Technology, Macedon, NY) according to a modification of the procedures of Van Soest et al. (1991). Heat stable α-amylase and sodium sulfite with correction for insoluble ash were used in the determination of NDF. The concentration of TI was also analyzed (method Ba 12–75; AOCS, 2011) as TI units per milligram of the FFSB sample. The gross energy (GE) of diets, FFSB, fecal, and urine samples were determined using an adiabatic oxygen bomb calorimeter (Parr Instruments Co., Moline, IL, USA).

Experiment 2: amino acid digestibility

Experimental design and dietary treatments

Experiment 2 was conducted to determine the apparent ileal digestibility (AID) and standardized ileal digestibility (SID) of CP and AA of the 10 FFSB samples fed to growing pigs. Eleven crossed barrows (Duroc × Landrace × Yorkshire, initial BW: 50.45 ± 3.2 kg) were surgically fitted with T-cannula in the distal ileum following the procedure of Stein et al. (1998). After the recovery period, these barrows were allotted to 1 of 11 experimental diets in a 6-period Youden square design (6 observations per dietary treatment). Each period lasted 7 d, including a diet adaption period (5 d) and an ileal digesta collection period (2 d). During the collection period, the ileal digesta was collected from 0830 h to 1730 h every day. The 11 experimental diets contained 10 FFSB diets and 1 N-free diet (Table 2). The N-free diet was formulated to estimate the basal endogenous losses of CP and AA, and then the SID of CP and AA could be calculated (Chen et al., 2016). Considering the great digestibility of AA and CP in FFSB, 40% FFSB was added to the test diet as the sole source of protein. The 0.3% chromic dioxide was also contained in the test diets as the inert marker. The chemical compositions and AA of the experimental diets were analyzed and presented in Table 4. At the beginning of each period, each pig was weighed, and the daily feed allowance was adjusted to 4% of their BW.

Table 4.

Analyzed chemical compositions of the 10 full-fat soybean diets used in Exp. 2 (as fed-basis)

Items, %	N-free diet	Full-fat soybean
		1	2	3	4	5	6	7	8	9	10
Crude protein	1.11	14.75	15.22	14.14	15.40	15.37	15.17	14.95	14.28	14.49	14.47
Indispensable amino acids
Arg	0.04	1.28	1.25	1.34	1.34	1.06	1.24	1.20	1.22	1.25	1.31
His	0.02	0.36	0.36	0.37	0.37	0.35	0.37	0.37	0.36	0.38	0.32
Leu	0.03	0.95	1.05	0.97	0.97	0.88	1.01	1.01	0.96	0.99	0.98
Ile	0.01	0.57	0.62	0.58	0.58	0.53	0.60	0.60	0.57	0.61	0.59
Lys	0.03	0.89	0.93	0.93	0.93	0.87	0.92	0.90	0.88	0.89	0.83
Met	0.04	0.20	0.20	0.18	0.20	0.19	0.20	0.19	0.16	0.20	0.17
Phe	0.05	0.72	0.79	0.70	0.70	0.74	0.75	0.76	0.75	0.68	0.70
Thr	0.02	0.52	0.55	0.54	0.54	0.50	0.53	0.54	0.52	0.55	0.52
Trp	0.21	0.39	0.39	0.40	0.38	0.39	0.39	0.38	0.39	0.37	0.37
Val	0.05	0.66	0.71	0.68	0.68	0.62	0.68	0.67	0.64	0.68	0.66
Dispensable amino acids
Ala	0.05	0.66	0.71	0.68	0.68	0.62	0.68	0.67	0.64	0.68	0.66
Asp	0.05	1.24	1.57	1.50	1.50	1.31	1.53	1.50	1.44	1.57	1.43
Cys	0.04	0.24	0.24	0.22	0.25	0.23	0.22	0.23	0.21	0.23	0.20
Glu	0.09	2.13	2.52	2.44	2.44	2.21	2.45	2.44	2.36	2.52	2.33
Gly	0.03	0.68	0.71	0.70	0.70	0.66	0.69	0.69	0.66	0.70	0.64
Pro	0.07	0.71	0.77	0.76	0.76	0.70	0.70	0.74	0.73	0.82	0.73
Ser	0.03	0.67	0.72	0.73	0.73	0.66	0.68	0.70	0.67	0.72	0.65
Tyr	0.00	0.40	0.43	0.33	0.33	0.33	0.42	0.44	0.37	0.38	0.40

Sample collection

During the collection period, plastic bags were attached to the open cannula to collect ileal digesta samples. When the bag was half full, the bag was replaced and frozen quickly at −20 °C. At the end of the animal trial, the ileal digesta samples from each pig at each period were thawed, mixed, and subsampled. Approximately 600 g digesta subsamples were lyophilized using a vacuum-freeze dryer (Tofflon Freeze Drying Systems, Minhang District, Shanghai, China). Then the vacuum-freeze dried samples were grounded through a 1-mm screen for further analysis.

Chemical analyses

The CP and AA of FFSB samples, diets, and lyophilized ileal digesta were analyzed. The determination of CP followed the same method described in Exp. 1. The AA contents were analyzed using the standard method (AOAC, 2016). After a cold performic acid oxidation overnight and hydrolyzing using 7.5 N HCl at 110 °C for 24 h, Met and Cys were measured as methionine sulfone and cysteic acid using an amino acid analyzer (model L-8800; Hitachi High-Technologies Corp.). After LiOH hydrolysis for 22 h at a constant temperature of 110 °C, Try was measured using High-Performance Liquid Chromatography (Agilent 1200 Series; Agilent Technologies Inc., Santa Clara, CA, USA). The remaining 15 AA were analyzed using an Amino Acid Analyzer (Hitachi L-8900; Hitachi Ltd., Tokyo, Japan) after being hydrolyzed with 6 N HCl at 110 °C for 24 h.

Calculations

In Exp. 1, the DE, ME, and the apparent total digestibility (ATTD) of GE of the diets and FFSB were calculated following the methods described by Adeola (2001):

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{D}{\mathrm{E}}_{\mathrm{d}\mathrm{i}\mathrm{e}\mathrm{t}}=(\mathrm{G}{\mathrm{E}}_{\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{d}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{t}}-\mathrm{G}{\mathrm{E}}_{\mathrm{f}\mathrm{e}\mathrm{c}\mathrm{e}\mathrm{s}})/\mathrm{F}\mathrm{e}\mathrm{e}{\mathrm{d}}_{\mathrm{i}\mathrm{n}\mathrm{t}\mathrm{a}\mathrm{k}\mathrm{e}},}\end{array}}$$

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{D}{\mathrm{E}}_{\mathrm{a}\mathrm{d}\mathrm{j}\mathrm{u}\mathrm{s}\mathrm{t}}=\mathrm{D}{\mathrm{E}}_{\mathrm{d}\mathrm{i}\mathrm{e}\mathrm{t}}/0.9677,}\end{array}}$$

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{D}{\mathrm{E}}_{\mathrm{F}\mathrm{F}\mathrm{S}\mathrm{B}}=[\mathrm{D}{\mathrm{E}}_{\mathrm{d}\mathrm{i}\mathrm{e}\mathrm{t}}(100\mathrm{\%}-\mathrm{X}\mathrm{\%})\times \mathrm{D}{\mathrm{E}}_{\mathrm{a}\mathrm{d}\mathrm{j}\mathrm{u}\mathrm{s}\mathrm{t}}]/\mathrm{X}\mathrm{\%},}\end{array}}$$

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{M}{\mathrm{E}}_{\mathrm{d}\mathrm{i}\mathrm{e}\mathrm{t}}=(\mathrm{G}{\mathrm{E}}_{\mathrm{i}\mathrm{n}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{d}\mathrm{i}\mathrm{e}\mathrm{n}\mathrm{t}}-\mathrm{G}{\mathrm{E}}_{\mathrm{f}\mathrm{e}\mathrm{c}\mathrm{e}\mathrm{s}}-\mathrm{G}{\mathrm{E}}_{\mathrm{u}\mathrm{r}\mathrm{i}\mathrm{n}\mathrm{e}})/\mathrm{F}\mathrm{e}\mathrm{e}{\mathrm{d}}_{\mathrm{i}\mathrm{n}\mathrm{t}\mathrm{a}\mathrm{k}\mathrm{e}},}\end{array}}$$

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{M}{\mathrm{E}}_{\mathrm{a}\mathrm{d}\mathrm{j}\mathrm{u}\mathrm{s}\mathrm{t}}=\mathrm{M}{\mathrm{E}}_{\mathrm{d}\mathrm{i}\mathrm{e}\mathrm{t}}/0.9677,}\end{array}}$$

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{M}{\mathrm{E}}_{\mathrm{F}\mathrm{F}\mathrm{S}\mathrm{B}}=[\mathrm{M}{\mathrm{E}}_{\mathrm{d}\mathrm{i}\mathrm{e}\mathrm{t}}(100\mathrm{\%}-\mathrm{X}\mathrm{\%})\times \mathrm{M}{\mathrm{E}}_{\mathrm{a}\mathrm{d}\mathrm{j}\mathrm{u}\mathrm{s}\mathrm{t}}]/\mathrm{X}\mathrm{\%},}\end{array}}$$

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{A}\mathrm{T}\mathrm{T}\mathrm{D}\mathrm{o}\mathrm{f}\mathrm{G}\mathrm{E}=\left(\mathrm{G}{\mathrm{E}}_{\mathrm{i}\mathrm{n}\mathrm{t}\mathrm{a}\mathrm{k}\mathrm{e}}\mathrm{G}{\mathrm{E}}_{\mathrm{f}\mathrm{e}\mathrm{c}\mathrm{e}\mathrm{s}}\right)/\mathrm{G}{\mathrm{E}}_{\mathrm{i}\mathrm{n}\mathrm{t}\mathrm{a}\mathrm{k}\mathrm{e}}.}\end{array}}$$

in which all the values were calculated on a DM basis. Feed_intake, GE_intake, GE_feces, and GE_urine were feed intake or GE values measured from each pig during the 5-d collection period. DE_adjust and ME_adjust were adjusted DE and ME values of the basal diet considering the non-energy supply part, and 0.9677 was the percentage of the energy-supplied ingredients in the formula.

In Exp. 2, the AID and SID of CP and AA were calculated following the methods described by Stein et al. (2007):

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{A}\mathrm{I}\mathrm{D}=[1\left(\mathrm{A}{\mathrm{A}}_{\mathrm{i}\mathrm{l}\mathrm{e}\mathrm{a}\mathrm{l}}/\mathrm{A}{\mathrm{A}}_{\mathrm{d}\mathrm{i}\mathrm{e}\mathrm{t}}\right)\times \left(\mathrm{C}{\mathrm{r}}_{\mathrm{d}\mathrm{i}\mathrm{e}\mathrm{t}}/\mathrm{C}{\mathrm{r}}_{\mathrm{i}\mathrm{l}\mathrm{e}\mathrm{a}\mathrm{l}}\right)]\times 100\mathrm{\%},}\end{array}}$$

in which all the values were calculated on a DM basis. AA_diet, AA_ileal, Cr_ileal, and Cr_diet were AA or chromic levels measured from the diets and ileal digesta of each pig during the 2-d collection period. The AID of CP can also be calculated using this equation.

The endogenous loss of CP and AA was measured from the pigs fed the N-free diet, and the calculated equations were shown as follows:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{I}\mathrm{l}\mathrm{e}\mathrm{a}\mathrm{l}\mathrm{A}{\mathrm{A}}_{\mathrm{e}\mathrm{n}\mathrm{d}\mathrm{o}\mathrm{u}\mathrm{g}\mathrm{e}\mathrm{n}\mathrm{o}\mathrm{u}\mathrm{s}}=[\mathrm{A}{\mathrm{A}}_{\mathrm{N}-\mathrm{f}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{i}\mathrm{l}\mathrm{e}\mathrm{a}\mathrm{l}}\times \left(\mathrm{C}{\mathrm{r}}_{\mathrm{N}-\mathrm{f}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{d}\mathrm{i}\mathrm{e}\mathrm{t}}/\mathrm{C}{\mathrm{r}}_{\mathrm{N}-\mathrm{f}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{i}\mathrm{l}\mathrm{e}\mathrm{a}\mathrm{l}}\right)]}\end{array}}$$

The average ileal AA_endougenous was used to calculate the SID of AA in all diets as follows:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{S}\mathrm{I}\mathrm{D}=[\mathrm{A}\mathrm{I}\mathrm{D}+\left(\mathrm{I}\mathrm{l}\mathrm{e}\mathrm{a}\mathrm{l}\mathrm{A}{\mathrm{A}}_{\mathrm{e}\mathrm{n}\mathrm{d}\mathrm{o}\mathrm{u}\mathrm{g}\mathrm{e}\mathrm{n}\mathrm{o}\mathrm{u}\mathrm{s}}/\mathrm{A}{\mathrm{A}}_{\mathrm{d}\mathrm{i}\mathrm{e}\mathrm{t}}\right)\times 100\mathrm{\%}].}\end{array}}$$

Statistical analyses

All the data of Exp. 1 and Exp. 2 were checked for normality and homogeneous variance using the UNIVARIATE procedure of SAS 9.4 (SAS Inst. Inc., Cary, NC, USA). The LSMEANS statement of SAS was used to calculate the means of each treatment group. In Exp. 1, the DE, ME, ATTD of GE, GE intake, and ratio of ME to DE in 11 diets (fed basis) and 11 ingredients (DM basis), as well as the fecal energy (FE) output, urine energy (UE) output in the 11 diets (fed basis) were analyzed using the MIXED procedure of SAS. The significance of differences among the dietary treatments was separated using Tukey’s multiple range test. Significant differences were declared at P < 0.05. Moreover, the PROC CORR procedure of SAS was used to determine the correlation between energy content and chemical composition. The PROC REG procedure of SAS was used to develop the prediction equations for DE and ME of FFSB based on chemical composition parameters. The R², root mean square error (RMSE), Mallows’ statics (C(p)), Akaike Information Criterion (AIC), and Bayesian Information Criterion (BIC) were used to select the best-fitted prediction equations. The equation with the greatest R², the smallest RMSE, AIC, and BIC was considered to be the best-fitted equation (Dong et al., 2014). In the Exp. 2 the AID and SID of CP and AA were also analyzed using the MIXED procedure of SAS. The statistical model included dietary treatment as a fixed effect and pig and period as random effects with individual pig as the experimental unit. The significance of differences among the dietary treatments was separated using Tukey’s multiple range test. Significant differences were declared at P < 0.05.

Results

Chemical compositions and AA contents of full-fat soybeans

As shown in Table 1, there was considerable variation in NDF, ADF, and TI levels among the 10 FFSB samples, with the coefficient of variation (CV) greater than 10%. On a DM basis, the averaged CP, EE, NDF, ADF, Ash, Ca, and P concentrations in FFSB were 39.51%, 20.35%, 13.18%, 6.14%, 5.62%, 0.25%, and 0.44%, with ranges of 38.63 to 41.40%, 16.90% to 21.98%, 11.38% to 16.35%, 4.83% to 8.00%, 5.38% to 6.23%, 0.20% to 0.29%, and 0.40% to 0.48%, respectively. The average GE value of FFSB was 5,654 kcal/kg, with a range of 5,531 to 5,765 kcal/kg.

The AA concentrations of the 10 FFSB samples were also presented in Table 1. The concentrations of all AA were relatively stable with CV less than 10%. The averaged Lys, Met, Thr, and Trp concentrations in FFSB were 2.41%, 0.50%, 1.41%, and 0.48%, with ranges of 2.34% to 2.49%, 0.43% to 0.56%, 1.35% to 1.46%, and 0.47% to 0.51%, respectively.

Energy balance, energy concentrations in diets, and ingredients

No differences were observed in the energy balance and energy concentrations in the 10 FFSB diets (P > 0.05, Table 5). The averaged GE intake, FE output, and UE output in the 10 FFSB diets were 2,673, 309, and 80 kcal/d, respectively. Especially, the UE output in the FFSB diets was higher than the basal diet (80 kcal/d vs. 39 kcal/d). The averaged DE and ME contents in the 10 FFSB diets were 3,629 and 3,506 kcal/kg (fed basis). The averaged ATTD of GE and the ratio of ME to DE in the 10 FFSB diets were 88.39% and 96.59%, respectively (fed basis).

Table 5.

Energy balance, energy content and apparent total tract digestibility of energy of basal diet and 10 full-fat soybean diets (Exp. 1, fed basis)^{1, 2}

Items	Basal diet	Full-fat soybean diets										Mean	SEM	P-value
		1	2	3	4	5	6	7	8	9	10
GE intake, kcal/d	2,461.6	2,711.9	2,686.3	2,719.3	2,678.5	2,698.9	2,664.5	2,660.2	2,681.9	2,693.1	2,535.5	2,673.0	–	–
FE output, kcal/d	281.4	323.1	299.6	342.0	309.6	343.4	326.8	269.4	292.2	294.3	289.4	309.0	21.1	0.23
UE output, kcal/d	38.7	82.0	94.1	85.8	80.7	73.8	75.6	92.6	78.3	74.6	66.7	80.4	10.5	0.67
Diet DE, kcal/kg	3,287	3,642	3,662	3,624	3,618	3,602	3,583	3,660	3,637	3,662	3,603	3,629	21.76	0.11
Diet ME, kcal/kg	3,227	3,517	3,520	3,494	3,497	3,486	3,468	3,517	3,515	3,548	3,494	3,506	25.05	0.62
ATTD of GE, %	88.50	87.92	88.9	87.42	88.4	87.3	87.78	89.77	89.01	88.95	88.44	88.39	0.529	0.07
ME/DE, %	98.17	96.57	96.11	96.41	96.66	96.78	96.79	96.10	96.64	96.87	96.97	96.59	0.397	0.78

¹The data on the basal diet was not involved in the statistical analysis.

²Values are means of 6 observations per sample.

GE, gross energy; FE, fecal energy; UE, urine energy; DE, digestible energy; ME, metabolizable energy; ATTD, apparent total tract digestibility.

The DE and ME contents of corn were 3,931 and 3,860 kcal/kg (DM basis, Table 6). The DE contents in FFSB samples 2, 7, and 9 were greater than that in sample 6 (P < 0.01, Table 6), but not different from that in the other FFSB samples. No significant differences were observed in ME content among all the FFSB samples. On a DM basis, the average DE and ME contents of the 10 FFSB samples were 4,855 kcal/kg (range: 4,580–5,027 kcal/kg) and 4,555 kcal/kg (range: 4,318–4,690 kcal/kg), respectively. The ATTD of GE in FFSB sample 7 was greater than that in sample 6 (P < 0.01, Table 6). No differences were detected in the ME/DE ratio in 10 FFSB samples with an average ratio of 93.83%, respectively.

Table 6.

Energy content and apparent total tract digestibility of energy of the corn and 10 full-fat soybean samples (Exp. 1, DM basis)^{1, 2}

Items	Corn	Full-fat soybean										Mean	SEM	P-value
		1	2	3	4	5	6	7	8	9	10
DE, kcal/kg	3,931	4,909ab	5,023a	4,815ab	4,810ab	4,734ab	4,580b	5,027a	4,926ab	4,956a	4,767ab	4,855	79.60	<0.01
ME, kcal/kg	3,860	4,604	4,651	4,494	4,521	4,464	4,318	4,654	4,630	4,690	4,521	4,555	91.67	0.15
ATTD of GE, %	88.49	86.52ab	88.43ab	85.21ab	83.44ab	83.02ab	82.14b	89.56a	87.36ab	86.80ab	86.19ab	85.87	0.53	< 0.01
ME/DE, %	98.19	93.79	92.61	93.31	94.00	94.31	94.28	92.58	93.99	94.65	94.81	93.83	1.08	0.83

¹The data on the corn was not involved in the statistical analysis.

²Values are means of 6 observations per sample.

^a–cMeans within a row with different superscripts differ (P < 0.05).

GE, gross energy; DE, digestible energy; ME, metabolizable energy; ATTD, apparent total tract digestibility.

Correlation analysis and prediction equations of DE and ME

As presented in Figure 1, the EE content was positively correlated with DE (r = 0.94, P < 0.01) and ME (r = 0.91, P < 0.01) contents of FFSB. In addition, the NDF content was negatively correlated with the GE (r = −0.78, P < 0.01), DE (r = −0.86, P < 0.01), and ME (r = −0.83, P < 0.01) contents, while the ADF content was negatively correlated with the DE (r = −0.70, P < 0.05) content of FFSB.

Figure 1.

Correlation coefficients between chemical compositions and energy values of the 10 full-fat soybean samples (Exp. 1). The darker color indicates a greater correlation and the lighter color indicates a lower correlation. * represents 2 factors that were significantly correlated with P < 0.05. ** represents 2 factors that were significantly correlated with P < 0.01.

The prediction equation with the greatest R² and least RMSE, AIC, and BIC was considered the optimal prediction equation (Table 7). Accordingly, the best prediction equations for DE and ME of FFSB were: DE, kcal/kg = 3,472 + (94.87 × EE) − (97.63 × Ash) (R² = 0.91, Eq. (2) in Table 7) and ME, kcal/kg = 3,443 + (65.11 × EE) − (36.84 × TI) (R² = 0.91; Eq. (5) in Table 7), respectively.

Table 7.

Prediction equations to estimate digestible energy (DE) and metabolizable energy (ME) values of full-fat soybean (Exp. 1, DM basis)

Number	Regression equation1	Statistics					P-value
		RMSE	R 2	C(p)	AIC	BIC
(1)	DE = 2,982 + 91.99 × EE	51.58	0.88	26.15	115.01	111.92	< 0.01
(2)	DE = 3,472 + 94.87 × EE − 97.63 × Ash	46.48	0.91	18.84	117.59	110.8	<0.01
(3)	ME = 3,085 + 72.19 × EE	49.77	0.83	45.75	114.29	111.21	<0.01
(4)	ME = 3,435 + 61.73 × EE − 10.41 × NDF	52.29	0.84	45.97	119.95	113.15	<0.01
(5)	ME = 3,443 + 65.11 × EE − 36.84 × TI	37.67	0.91	21.89	113.37	106.58	<0.01

¹Prediction equations were developed using stepwise regression analyses. EE, ether extract; NDF, neutral detergent fiber; TI, trypsin inhibitor.

RMSE, root mean square error; C(p), Mallows’ statistic; AIC, Akaike information criterion; BIC, Bayesian information criterion.

Digestibility of crude protein and amino acids

The AID and SID of CP and AA in 10 FFSB samples were shown in Tables 8 and 9, respectively. The AID and SID of CP averaged 74.05% and 78.17%, with ranges of 70.00% to 75.78% and 74.02% to 79.86%, respectively. The AID and SID of most AA in all FFSB samples were high and significantly varied among the 10 samples (P < 0.01), except for Gly and Pro. Arginine and Met illustrated the greatest AID and SID values among all the indispensable AA in FFSB, with AID values of 87.92% and 84.73%, and SID values of 92.32% and 90.06%, respectively, while Thr and Trp represented the smallest AID and SID values among all the indispensable AA in FFSB, with AID values of 70.72% and 73.21%, and SID values of 80.23% and 77.03%, respectively. The AID and SID of Lys in FFSB averaged 81.95% and 85.53%, with ranges of 77.53% to 86.37%, and 81.15% to 89.83%, respectively.

Table 8.

Apparent ileal digestibility of amino acids in full-fat soybean fed to growing pigs (%, Exp. 2)¹

Items	Full-fat soybean										Mean	SEM	P-value
	1	2	3	4	5	6	7	8	9	10
CP	75.52a	75.74a	74.62a	74.64a	73.34ab	70.00b	75.78a	73.35ab	73.34ab	74.18ab	74.05	1.03	<0.01
Indispensable AA
Arg	89.01abc	89.76ab	85.13bc	92.05a	82.62c	88.89abc	89.70ab	88.43abc	84.63bc	88.97abc	87.92	1.63	<0.01
His	82.16abc	82.62abc	78.62bc	84.92a	78.71bc	80.69abc	83.63ab	80.54abc	79.13bc	77.51c	80.85	1.25	<0.01
Ile	77.90abc	81.97ab	73.56bc	82.64a	72.48c	77.67abc	81.15ab	74.83abc	72.34c	74.37abc	76.89	2.01	<0.01
Leu	78.01abc	81.63ab	72.95c	82.36a	72.67c	77.03abc	81.21ab	75.00bc	71.21c	73.67c	76.57	1.74	<0.01
Lys	82.62abc	84.63ab	80.87abc	86.37a	80.09abc	82.27abc	85.08a	81.88abc	77.53c	78.19bc	81.95	1.49	<0.01
Met	87.94ab	85.77abcd	80.55d	89.35a	82.40bcd	86.08abcd	86.97abc	81.46cd	81.84bcd	84.92abcd	84.73	1.47	<0.01
Phe	80.73abc	84.53a	74.85cd	85.46a	79.45abc	79.91abc	83.86a	80.83ab	73.18d	77.67bcd	80.05	1.42	<0.01
Thr	73.08abc	73.76abc	69.84abc	74.79ab	68.85abc	68.28c	75.19a	68.47bc	67.92c	67.07c	70.72	1.58	<0.01
Trp	77.27a	75.03ab	73.48ab	73.74ab	73.32ab	68.31c	73.28ab	73.89ab	73.36ab	70.42ab	73.21	1.11	<0.01
Val	76.66abcd	79.31abc	72.97bcd	80.82a	72.39cd	75.70abcd	79.89ab	73.13bcd	69.94d	71.18d	75.2	1.77	<0.01
Dispensable AA
Ala	72.08a	74.63ab	70.42ab	76.22ab	68.28ab	69.63ab	74.03ab	69.77ab	65.94b	67.31ab	70.83	2.24	<0.01
Asp	75.67bcd	82.26a	75.95bcd	83.3a	72.67d	78.95ab	82.28a	77.53bc	74.04cd	76.3bcd	77.9	1.16	<0.01
Cys	76.10ab	77.83a	72.30ab	77.20ab	74.71ab	72.97ab	79.87a	68.28b	69.36b	67.95b	73.66	2.08	<0.01
Glu	82.00abc	85.19a	79.13bc	87.71a	78.87c	84.28ab	85.94a	82.45abc	78.36c	82.36abc	82.63	1.33	<0.01
Gly	58.5	65.53	59.68	64.01	56.39	55.99	67.2	57.71	59.39	58.4	60.28	3.79	0.2
Pro	58.68	59.43	59.49	64.08	60.22	54.26	65.14	58.06	58.75	62.48	60.06	5.24	0.86
Ser	78.49abc	79.46ab	75.40abc	80.89a	74.45bc	75.97abc	80.84a	74.90bc	73.12c	73.73bc	76.73	1.42	<0.01
Tyr	81.20ab	85.11a	69.49c	80.06ab	71.61bc	78.42bc	82.91abc	74.83abc	70.57c	74.73abc	76.89	2.32	<0.01

¹Values are means of 6 observations per sample.

^a–dMeans within a row with different superscripts differ (P < 0.05).

Table 9.

Standardized ileal digestibility of amino acids in full-fat soybean fed to growing pigs (%, Exp. 2)¹

Items	Full-fat soybean										Mean	SEM	P-value
	1	2	3	4	5	6	7	8	9	10
CP	79.65a	79.75a	78.93a	78.59a	77.31ab	74.02b	79.86a	77.61ab	77.54ab	78.39ab	78.17	1.04	<0.01
Indispensable AA
Arg	92.43abc	93.27abc	88.51abc	95.32a	86.75c	92.42abc	93.34ab	92.01abc	88.13bc	92.32abc	91.45	1.63	<0.01
His	86.49abc	86.94abc	82.73c	89.14a	83.13bc	84.83abc	87.83ab	84.89abc	83.18bc	82.28c	85.14	1.25	<0.01
Ile	83.93abcd	87.51ab	79.50bcd	88.56a	79.03cd	83.41abcd	86.88abc	80.87abcd	77.98d	80.20abcd	82.79	2.01	<0.01
Leu	82.82abc	85.97ab	77.71c	87.05a	77.87c	81.56abc	85.73ab	79.75bc	75.83c	78.33bc	81.26	1.74	<0.01
Lys	86.22abc	88.09ab	84.19abc	89.83a	83.78abc	85.75abc	88.66a	85.53abc	81.15c	82.04bc	85.53	1.49	<0.01
Met	92.91ab	90.87abc	86.08c	94.37a	87.58bc	91.15abc	92.24ab	87.64bc	87.00bc	90.75abc	90.06	1.47	<0.01
Phe	84.66abcd	88.14ab	79.00de	89.53a	83.31bcd	83.68bcd	87.60abc	84.61abcd	77.34e	81.70abc	83.96	1.42	<0.01
Thr	82.87abc	82.92abc	79.02abc	84.09ab	78.87abc	77.77bc	84.57a	78.20bc	77.15c	76.83c	80.23	1.58	<0.01
Trp	81.06a	78.8ab	77.18ab	77.63ab	77.08ab	72.10c	77.16ab	77.61ab	77.32ab	74.37bc	77.03	1.11	<0.01
Val	82.58abcd	84.74abc	78.61bcd	86.54a	78.62bcd	81.39abcd	85.68ab	79.18bcd	75.64d	77.11cd	81.01	1.78	<0.01
Dispensable AA
Ala	80.68ab	82.69ab	78.54ab	84.43a	77.22ab	77.97ab	82.46ab	78.57ab	74.16b	75.93ab	79.27	2.24	<0.01
Asp	80.45cd	86.05ab	79.95cd	87.28a	77.21d	82.85abc	86.25a	81.66bcd	77.82d	80.46cd	82	1.16	<0.01
Cys	82.22ab	83.89ab	78.85ab	82.99ab	80.84ab	79.38ab	86.2a	75.27b	75.55b	75.18b	80.04	2.08	<0.01
Glu	85.66abc	88.27a	82.31bc	90.90a	82.39bc	87.45ab	89.13a	85.75abc	81.45c	85.69abc	85.9	1.32	<0.01
Gly	75.31	81.47	75.7	80.32	73.71	72.48	83.65	74.88	75.61	76.1	76.92	3.79	0.28
Pro	93.61	91.63	92.48	96.72	96.03	89.79	98.89	92.32	89.19	96.88	93.75	5.24	0.84
Ser	84.49abcd	85.05abc	80.86bcd	86.43ab	80.51bcd	81.86abcd	86.62a	80.87bcd	78.70d	79.90cd	82.53	1.41	<0.01
Tyr	86.27ab	89.81a	75.69c	86.24ab	77.72bc	83.28abc	87.52a	80.32abc	75.94c	79.78abc	82.25	2.32	<0.01

¹Values are means of 6 observations per sample.

^a–dMeans within a row with different superscripts differ (P < 0.05).

Discussion

Chemical compositions and amino acids of full-fat soybeans

Ten FFSB samples were collected from the major production areas in China, and their chemical compositions were analyzed in the current study. The averaged GE value of those samples (5,654 kcal/kg, DM basis) was close to the values reported by Baker et al. (2010, 5,770 kcal/kg) and Park et al. (2017, 5,719 kcal/kg), and the CP contents of FFSB were also comparable among these three studies (39.51%, 38.98%, and 39.08%, respectively). The FFSB is sourced from the whole soybean, and the processing technology does not contain an oil extraction procedure, leading to the high crude fat content in FFSB. The averaged EE content of FFSB was 20.35% in the current study, which was also approaching the value reported by Kim et al. (2022, 20.56% on a DM basis).

In addition, several chemical composition parameters determined in this study showed different values from previous reports but were within a reasonable variable range. Marty and Chavez (1993) reported 12.7% NDF and 5.7% ADF of FFSB, which were slightly lower than our values (13.18% and 6.14%, respectively), but the 12.7% NDF were greater than those reported in NRC (2012, NDF: 10.86%) and Park et al. (2017, NDF: 11.34%). Those could be attributed to the various planting environments of soybeans used in different studies, for example, soil, rain, and sunlight, which could directly affect the physical characteristics and chemical compositions of crops (Marty and Chavez, 1993; Li et al., 2014a). In addition, fiber content also varied among the 10 FFSB samples collected in this study. It is well known that soybean hulls contain a high amount of fiber, thus, whether soybeans were peeled before processing could greatly influence the fiber content in FFSB (Marty and Chavez, 1993). However, information on the sample peeling procedure is not available, so we can only speculate that the greater fiber content in FFSB samples 6 and 10 may be caused by the remaining soybean hulls during processing. As mentioned before, FFSB could be processed based on 4 major technologies, and even under the same extrusion technology, different processing parameters (e.g., temperature, heating time) could affect the chemical compositions of FFSB. The TI is very sensitive to heating and is considered an indicator of excessive heating (Agrahar-Murugkar and Jha, 2010). Thus, different TI units in the 10 FFSB samples may indicate a discrepancy in the processing conditions of these samples.

The AA contents measured in this study were stable among the 10 FFSB samples, and the results were consistent with those reported by NRC (2012) and Park et al. (2017). The AA compositions in FFSB were similar to those in soybean meal, but the total amount of AA in FFSB was 20% to 30% less, which was mainly due to the oil extraction procedure in soybean meal production, leading to approximately 20% crude fat loss. However, the contents of indispensable AA in FFSB were still greater than those in cereal ingredients (e.g., corn, rice bran) and in most protein ingredients (e.g., flaxseed expellers, rapeseed meal) (Li et al., 2015; Chen et al., 2016; Huang et al., 2018; Lyu et al., 2019). Indispensable AA cannot be synthesized by pigs themselves, so enough intake through the feed is crucial for achieving good growth performance of pigs (Hou et al., 2015), indicating FFSB is a high-quality protein ingredient, especially for pigs at the weaned and growing stages.

Energy concentrations and energy digestibility

The averaged DE and ME contents of FFSB determined in the current study were 4,855 and 4,555 kcal/kg on DM basis, respectively, approximately 300 kcal/kg greater than those reported by NRC (2012, 4,539 and 4,264 kcal/kg for DE and ME, respectively). Nevertheless, Kim et al. (2022) reported greater available energy values of FFSB (5,399 and 5,184 kcal/kg for DE and ME, respectively). Both Kim et al. (2022) and our study used animal trials and the difference method to determine the available energy concentration of FFSB, but with different substitution ratios (40% vs. 30%) on the basal diet, resulting in discrepancies in diet compositions, especially in protein (or AA) levels. Various nutrient levels would lead to different protein catabolism and feeding behavior, as well as the difference in determined DE and ME values of the same ingredient (Noblet et al., 2022).

When the difference method was adopted, the determination of the energy value of the basal diet (or the corn) was very important due to that the available energy contents of the test ingredient were calculated by the difference between the basal diet and the test diet. The ME content of the corn was 3,860 kcal/kg (DM basis), which was close to NRC (2012, 3,844 kcal/kg) but lower than the 3,982 kcal/kg reported by Lyu et al. (2019). The difference was mainly produced by the various EE content in the corn (3.80% vs. our 2.29%) while Lyu et al. (2019) reported that EE was the first key factor influencing the ME content of corn.

The ATTD of GE in FFSB was also greater than most protein ingredients (e.g., cottonseed meal, rapeseed meal, flaxseed meal) according to our results, which could also be attributed to the greater crude fat content in FFSB compared to the other oil-extracted protein ingredients. Compared with carbohydrates and protein, lipid was more easily to be digested and absorbed after being emulsified and broken down into fatty acids combined with lipase (Chilliard, 1993), and lipid could delay gastric emptying, thus resulting in greater digestibility of other nutrients (Low, 1990).

Correlation analysis and prediction equations of DE and ME

The current results showed that EE and NDF were the best predictors for DE and ME prediction in FFSB, with EE positively correlated with the available energy and NDF negatively correlated with the available energy. This was consistent with the results of the previous studies (Li et al., 2014b; Chen et al., 2016; Huang et al., 2018), and was further proved by the fact that FFSB sample 6 had the lowest DE and ME values whereas contained the greatest NDF level and the lowest EE level. Lipid is an important energetic substance with approximately 2.25 times greater energy values compared with carbohydrates and protein (Noblet and van Milgen, 2004), while the increased dietary fiber content could stimulate gut movement, resulting in increased fecal output and decreased energy digestibility (Wilfart et al., 2007), leading to lower available energy values.

Ash was also included in the best DE prediction equation developed in the current study, and similar results were also observed in the study of Huang et al. (2018), indicating that the side effect of ash in diets on energy digestibility was more than the simple dilution effect (Noblet and Perez, 1993). The antinutritional factor TI was also included as a negative factor in the ME prediction equation we developed. The intake of TI could cause pancreatic hypertrophy, leading to a disorder in the secretory activity of the pancreas of animals (Embaby, 2010). Consequently, TI showed a negative effect on the DE and ME values of FFSB. Even though FFSB has undergone heat treatment, it is still difficult to deactivate TI completely.

In the current study, all the developed prediction equations [Eqs. (1–5)] exhibited R² > 0.80, indicating that parameters of chemical compositions were good predictors to predict the DE and ME values in FFSB with high accuracy.

Digestibility of crude protein and amino acids

Similar to the results of energy digestibility, the CP digestibility of FFSB was also greater than other protein ingredients (e.g., flaxseed meal, canola meal, and rapeseed meal; NRC, 2012), which could be attributed to the excellent protein compositions of soybean. Moreover, after extrusion and heating, the antinutritional factors in soybean were inactivated, making FFSB easier to contact with proteases and be digested. The digestibility of AA in FFSB shows differences among different research. The SID of AA reported by Baker et al. (2010) was nearly 5%–10% greater than our results. Two reasons can explain this phenomenon. Firstly, the higher fiber content in our research would hinder protease binding to proteins, thus leading to lower AA digestibility. Secondly, as mentioned before, the higher substitution ratio (51.6% vs. ours 40%) in the research of Baker et al. (2010) would result in a higher diet CP concentration, and then activate relevant metabolism pathways and enhance AA metabolism. However, Park et al (2017) reported similar AID results to ours (e.g., Arg: 86.2% vs. 87.92%) while using a similar substitution ratio (41% vs. ours 40%). This supported the influence of substitution ratio on the digestibility of AA. Interestingly, Marty and Chavez (1993) reported lower AID but a higher SID of AA than our FFSB samples. The different endogenous losses of AA measured by different methods would be responsible for this discrepancy. In detail, Marty and Chavez (1993) used a homoarginine technique while we used an N-free diet to determine endogenous AA excretion. The latter approach always yields low values because the lack of dietary peptides in the gut lumen weakens the endogenous excretion of AA (Butts et al., 1993). However, there are also some limitations in the homoarginine approach, for example, it assumed that the lysine in the protein was homogeneously guanidinated and it ignored the hydrolysis of homoarginine. So, it’s hard to judge which approach is better. In general, the AID and SID of AA in FFSB were within the range of previously published values (Marty and Chavez, 1993; Baker et al., 2010; NRC, 2012; Park et al., 2017).

The results of the current study also showed that the AID and SID of AA varied among the 10 FFSB samples, and the differences in chemical compositions and processing technologies may explain it. The lowest CP and AA digestibility appeared in samples 6 and 10 may result from their great fiber content, which could influence the amount of endogenous AA recovered at the distal ileum (Taverner et al., 1981), and may induce increased digesta flow, shortened digestion time, and obstructed combination of digestive enzymes and substrates (Sauer and Ozimek, 1986). Sample 9 exhibited the lowest AID or SID of Lys, but the fiber content was not very high, which then may be caused by processing. Heat treatment could deactivate TI, while the Maillard reaction would occur under overheating conditions, which then denatured the protein and thus reduced the digestibility of CP and amino acids, especially Lys (González-Vega et al., 2011). Because both Lys and TI were particularly vulnerable to overheating (Vagadia et al., 2017) and a very low concentration of TI was observed in sample 9, the lowest digestibility of Lys may be attributed to an overheating condition. The AA digestibilities in samples 1, 2, 4, and 7 were above the average value, which contained high concentrations of EE than the other samples, indicating the important role of EE in AA digestibility. The relationship between increased dietary EE content and increased AA digestibility was also observed in nursery and growing pigs reported previously (Imbeah and Sauer, 1991; Li and Sauer, 1994) since the increased fat level could delay gastric emptying and slow digesta passage rate, resulting in a longer time of AA exposure to digestive enzymes (Hunt and Knox, 1968). Soybeans also contain nearly 10% oligosaccharides (5% sucrose, 1% raffinose, and 4% stachyose), which were reported could proliferate Bifidobacterium in the hindgut and enhances intestinal functions (Pan et al. 2002). However, Leske et al. (1993) found that soybean oligosaccharides depress 14% CP digestibility in poultry, indicating oligosaccharides in each FFSB sample may also influence AA digestibility. The oligosaccharides in the FFSB samples were unavailable in this study, thus further research on the oligosaccharides in the FFSB is still necessary.

Finally, some AA in FFSB illustrated interesting digestive characteristics. For example, Thr had the lowest AID value, but not the lowest SID value because of the high concentration of endogenous Thr in the distal ileum of growing pigs when fed an N-free diet (Holmes et al., 1974; Sauer et al., 1991). Arginine showed the greatest AID and SID values in all indispensable AA in FFSB, which was the first AA that appeared in isolated intestinal loops after enzymatic hydrolysis, and the specificity of proteases and peptidases in the central region of the small intestine may promote the digestibility of Arg (Low et al., 1980). There was a great gap between the AID and SID values of Pro, which was also observed in the research of Low (1979), Moughan and Schuttert (1991), and Furuya and Kaji (1992). Except for FFSB, the SID of Pro in many other ingredients would also exceed 100% (NRC, 2012) which indicated a high concentration of Pro in the endogenous secretions. The reason for this phenomenon is that endogenous losses are estimated using N-free diets. When pigs were fed N-free diets, they were in a negative nitrogen balance and will break down muscle to produce endogenous AA, and 50% of the AA released are Gln. Then, Gln is broken down into Glu, Cit, and Pro after metabolism (De Lange et al., 1989). Therefore the use of N-free diets here overestimates the endogenous loss of Pro. Overall, the SID values of most AA in FFSB could reach 80% to 90%, indicating that FFSB was a high-quality protein ingredient.

Conclusions

On a DM basis, the average DE and ME values in the 10 FFSB samples were 4,855 and 4,555 kcal/kg, respectively, both were positively correlated with the EE content. The best-fitted prediction equations for DE and ME of FFSB were: DE, kcal/kg = 3,472 + 94.87 × EE − 97.63 × ash (R² = 0.91); ME, kcal/kg = 3,443 + 65.11 × EE − 36.84 × TI (R² = 0.91). In addition, FFSB is a high-quality protein ingredient with high AID and SID values when fed to growing pigs.

Glossary

Abbreviations

AA

amino acids

ADF

acid detergent fiber

AIC

Akaike’s information criterion

AID

apparent ileal digestibility

ATTD

apparent total tract digestibility

BIC

bayesian information criterion

BW

body weight

Ca

calcium

CP

crude protein

C(p)

Mallows’ statics

CV

coefficient of variation

DE

digestible energy

DM

dry matter

EE

ether extract

FE

fecal energy

FFSB

full-fat soybean

ME

metabolizable energy

NDF

neutral detergent fiber

P

phosphorus

RMSE

root mean square error

SID

standardized ileal digestibility

TI

trypsin inhibitor

UE

urine energy

Funding

This work was supported by the National Natural Science Foundation of China (NSFC, 32072764), the fund from the Ministry of Agriculture and Rural Affairs (MOA, 16210081), and the National Key Research and Development Program of China funded by the Ministry of Science and Technology of the People’s Republic of China (MOST, 2021YFD1300205-8 and 2021YFD1300205-9).

Ethics Statement

All experimental protocols including animal care and use were approved by the Institutional Animal Care and Use Committee of China Agricultural University.

Conflict of Interest Statement

All authors declare no conflict of interest.