Digestible and metabolizable energy concentrations and amino acid digestibility of dried yeast and soybean meal for growing pigs

Abstract

Energy values and amino acid (AA) digestibility of dried yeast (DY) and soybean meal (SBM) were determined in 2 experiments with growing pigs. Experiment 1 was conducted to determine the digestible energy (DE) and metabolizable energy (ME) in DY and SBM. Thirty barrows with a mean initial body weight (BW) of 20.6 kg (SD = 1.04) were assigned to 5 dietary treatments in a randomized complete block design with period and BW as blocking factors. A reference diet was prepared with corn, canola meal, and soybean oil as energy-contributing ingredients. Four additional diets were prepared by adding 5% and 10% DY or SBM at the expense of energy-contributing ingredients in the reference diet. The ratio of corn, canola meal, and soybean oil was kept consistent across the experimental diets. Each experimental period consisted of 5-d adaptation and 5-d quantitative collection of feces and urine. Test ingredient-associated DE or ME intake (kcal/d) was regressed against test ingredient intake [kg dry matter (DM)/d] to estimate the DE or ME in test ingredients as the slope of linear regression model. The DE in DY was estimated at 3,933 kcal/kg DM, which was not different from the estimated DE in SBM at 4,020 kcal/kg DM. Similarly, there was no difference between DY and SBM in the estimated ME (3,431 and 3,756 kcal/kg DM, respectively). Experiment 2 was conducted to determine the standardized ileal digestibility (SID) of AA in DY and SBM. Twenty-one barrows with a mean initial BW of 20.0 kg (SD = 1.31) were surgically fitted with T-cannulas at the distal ileum and assigned to 3 dietary treatments in a randomized complete block design with BW as a blocking factor. Two semi-purified diets containing DY or SBM as the sole nitrogen source and one nitrogen-free diet (NFD) were prepared. The NFD was used to estimate the basal ileal endogenous losses of CP and AA. Pigs were fed the 3 diets for 5 d as adaptation, followed by 2 d of feeding with ileal digesta collection. The SID of AA, except Gly and Pro, in DY was less (P < 0.05) than in SBM. The SID of indispensable AA in DY ranged from 64.1% for Thr to 85.2% for Arg, and those in SBM ranged from 83.9% for Thr to 91.8% for Arg. In conclusion, energy values of DY are not different from those of SBM, whereas AA in DY is less digestible than in SBM. The estimated DE and ME as well as the SID of AA in DY and SBM can be used in diet formulation for growing pigs using these ingredients.

Introduction

Swine diets in the United States have been generally formulated with corn and soybean meal (SBM) as major feed ingredients because of its nutritional quality and economic advantages (Shannon and Allee, 2010). However, due to the unstable price of corn and SBM for the last decade, researchers have focused on alternative feed ingredients which can partly replace conventional feed ingredients in swine diets (Woyengo et al., 2014). In order to use alternative feed ingredients with the least cost formulation, an accurate determination of energy values and amino acid (AA) digestibility in the potential alternative feed ingredient is necessary.

Fermentation techniques using yeast have been widely used to produce various products such as pharmaceuticals, ethanol, or beverages (Shurson, 2018). Following fermentation, the remaining yeast can be separated from the fermentation medium and dried to produce dried yeast (DY). Due to the high concentration of crude protein (CP), results from several studies have suggested that DY may be used as a major protein supplement in diets for weanling and growing pigs (Cruz et al., 2019, 2020). Energy values including digestible energy (DE) and metabolizable energy (ME) as well as digestibility of AA in DY have been reported in previous studies (Kim et al., 2014; Lagos and Stein, 2020). However, there are considerable variations in nutritional values among products of DY due to the differences in condition of fermentation such as strains of yeast or substrates. Therefore, two experiments were conducted to determine the DE, ME, and standardized ileal digestibility (SID) of AA in DY derived from the pharmaceutical industry and SBM fed to growing pigs. The null hypothesis of this study was that the DE, ME, and SID of AA in DY are not different from those in SBM.

Materials and Methods

Protocols of the animal experiments were reviewed and approved by the Purdue University Animal Care and Use Committee (West Lafayette, IN). Crossbred barrows (Duroc × Yorkshire × Landrace) were used in both Exp. 1 and 2.

Experiment 1: DE and ME

Experiment 1 was conducted for 2 experimental periods with 15 barrows per period. Each experimental period lasted for 10 d. At the beginning of each experimental period, pigs were selected based on body weight (BW) and individually housed in metabolism crates (1.22 × 1.22 m) which enables restricted individual feeding and quantitative collection of feces and urine. Mean initial BW of pigs in the first and second periods were 20.6 kg (SD = 0.87) and 20.7 kg (SD = 1.21), respectively. Within experimental periods, pigs were divided into 3 blocks based on BW and allocated to 5 diets in a randomized complete block design, resulting in 6 replicates per diet.

A reference diet was prepared with corn, canola meal, and soybean oil as energy-contributing ingredients (Table 1). Crystalline or synthetic AA supplements were added to provide limiting AA in the reference diet and to meet the ratio of indispensable AA with Lys suggested in National Research Council (NRC, 2012). The reference diet was prepared to meet or exceed the mineral and vitamin requirement estimates (NRC, 2012). Four additional diets were prepared by adding 5% and 10% DY or SBM at the expense of energy-contributing ingredients in the reference diet. The source of yeast to produce DY used in the current study originated from fermentation plants producing pharmaceuticals. The ratio among corn, canola meal, and soybean oil was maintained at 22.3:9:1 across the experimental diets.

Table 1.

Ingredient and analyzed chemical composition of experimental diets used in Exp. 1, as-fed basis

	Diet
		Dried yeast, %		Soybean meal, %
Item	Reference diet	5	10	5	10
Ingredient, %
Ground corn	66.85	63.40	59.94	63.40	59.94
Canola meal	26.90	25.51	24.12	25.51	24.12
Dried yeast	0.00	5.00	10.00	0.00	0.00
Soybean meal	0.00	0.00	0.00	5.00	10.00
Soybean oil	3.00	2.84	2.69	2.84	2.69
Ground limestone	1.00	1.00	1.00	1.00	1.00
Monocalcium phosphate	1.00	1.00	1.00	1.00	1.00
Salt	0.40	0.40	0.40	0.40	0.40
L-Lysine·HCl	0.47	0.47	0.47	0.47	0.47
DL-Methionine	0.02	0.02	0.02	0.02	0.02
L-Threonine	0.06	0.06	0.06	0.06	0.06
L-Tryptophan	0.03	0.03	0.03	0.03	0.03
Vitamin premix1	0.15	0.15	0.15	0.15	0.15
Mineral premix2	0.07	0.07	0.07	0.07	0.07
Selenium premix3	0.05	0.05	0.05	0.05	0.05
Total	100.00	100.00	100.00	100.00	100.00
Analyzed chemical
DM, %	88.3	88.1	88.5	88.3	88.4
GE, kcal/kg	4,075	4,094	4,164	4,086	4,081

¹Provided the following quantities per kilogram of complete diet: vitamin A, 3,960 IU; vitamin D₃, 396 IU; vitamin E, 26.4 IU; menadione, 1.30 mg; riboflavin, 5.30 mg; _D-pantothenic acid, 13.2 mg; niacin, 19.8 mg; and vitamin B₁₂, 0.02 mg.

²Provided the following quantities per kilogram of complete diet: I, 0.258 mg; Mn, 12.0 mg; Cu, 6.33 mg; Fe, 136 mg; and Zn, 104 mg.

³Provided 0.3 mg Se/kg of complete diet.

Daily feed allowance was set at 4% of mean BW of pigs within blocks which was equally divided into 2 meals and offered to pigs at 0800 and 1700 hours. Water was available at all times via nipple drinkers. Each experimental period consisted of 5-d adaptation and 5-d quantitative collection of feces and urine. During the 5-d adaptation, pigs were adapted to the metabolism crates and experimental diets. Quantitative collection of feces and urine was conducted based on the procedure described by Kong and Adeola (2014). On days 6 and 11, chromic oxide was added at 0.5% of the first meal of the day as a marker. The collection of feces initiated at the first appearance of the first marker in the feces and terminated when the next marker appeared in the feces. From days 6 to 11, urine was quantitatively collected in plastic buckets containing 10 mL of 10% formic acid. Collected urine was weighed and subsampled daily. Collected feces and subsamples of urine were immediately stored at –20°C. Daily feed intake of pigs was precisely recorded by collecting any feed refusal or waste.

All the feces collected from pigs were dried in a forced-air drying oven at 55°C to constant weight and finely ground (<1 mm) using a hammer mill. Subsamples of urine were filtered through glass wool and dried in a forced-air drying oven at 55°C to constant weight. Samples of feed ingredients and experimental diets were finely ground (<0.75 mm) using a centrifugal grinder (ZM 200; Retsch GmbH, Haan, Germany) and analyzed for dry matter (DM) by drying at 105°C for 24 h in a forced-air drying oven [Precision Scientific Co., Chicago, IL; method 934.01; Association of Official Analytical Chemists (AOAC), 2006]. The concentrations of gross energy (GE) in feed ingredients, experimental diets, feces, and urine were analyzed using an isoperibol bomb calorimeter (Parr 6200; Parr Instrument Co., Moline, IL). Ground feed ingredients were analyzed for ether extract [method 920.39 (A); AOAC, 2006] and ash (method 942.05; AOAC, 2006). A fiber analyzer (Ankom 2000 Fiber Analyzer; Ankom Technology, Macedon, NY) was used to analyze the concentration of neutral detergent fiber (Van Soest et al., 1991) and acid detergent fiber [method 973.18 (AD); AOAC, 2006] in feed ingredients.

The apparent total tract digestibility (ATTD) and metabolizability of GE (%) in experimental diets were calculated by the difference between the intake of GE and fecal and urinary GE output (Kong and Adeola, 2014):

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{A}\mathrm{T}\mathrm{T}\mathrm{D}\mathrm{o}\mathrm{f}\mathrm{G}\mathrm{E}=100\times \left[\left(\mathrm{G}{\mathrm{E}}_{\mathrm{i}\mathrm{n}}-\mathrm{G}{\mathrm{E}}_{\mathrm{f}}\right)/\mathrm{G}{\mathrm{E}}_{\mathrm{i}\mathrm{n}}\right];}\end{array}}$$

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{m}\mathrm{e}\mathrm{t}\mathrm{a}\mathrm{b}\mathrm{o}\mathrm{l}\mathrm{i}\mathrm{z}\mathrm{a}\mathrm{b}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}\mathrm{o}\mathrm{f}\mathrm{G}\mathrm{E}=100\times \left[(\mathrm{G}{\mathrm{E}}_{\mathrm{i}\mathrm{n}}-\mathrm{G}{\mathrm{E}}_{\mathrm{f}}-\mathrm{G}{\mathrm{E}}_{\mathrm{u}})/\mathrm{G}{\mathrm{E}}_{\mathrm{i}\mathrm{n}}\right],}\end{array}}$$

where GE_in represents the intake of GE (kcal/d) which is the product of GE (kcal/kg) in experimental diets and feed intake of pigs (kg/d); GE_f and GE_u represent the fecal and urinary GE output (kcal/d), respectively, which are the products of GE (kcal/kg) in feces or urine and the weight of feces or urine output (kg/d). The DE and ME (kcal/kg) in experimental diets were calculated by multiplying the GE (kcal/kg) in experimental diets by the ATTD and metabolizability of GE, respectively.

The DE and ME (kcal/kg) in test ingredients (i.e., DY or SBM) were calculated by the difference procedure (Kong and Adeola, 2014):

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{D}{\mathrm{E}}_{\mathrm{t}\mathrm{i}}=[\mathrm{D}{\mathrm{E}}_{\mathrm{t}\mathrm{d}}-(\mathrm{D}{\mathrm{E}}_{\mathrm{r}\mathrm{d}}\times {\mathrm{P}}_{\mathrm{r}\mathrm{d}})]/{\mathrm{P}}_{\mathrm{t}\mathrm{i}};}\end{array}}$$

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{M}{\mathrm{E}}_{\mathrm{t}\mathrm{i}}=[\mathrm{M}{\mathrm{E}}_{\mathrm{t}\mathrm{d}}-(\mathrm{M}{\mathrm{E}}_{\mathrm{r}\mathrm{d}}\times {\mathrm{P}}_{\mathrm{r}\mathrm{d}})]/{\mathrm{P}}_{\mathrm{t}\mathrm{i}},}\end{array}}$$

where DE_ti, DE_td, and DE_rd represent the DE (kcal/kg) in test ingredient, test diet (i.e., experimental diets containing test ingredients), and reference diet, respectively; P_rd and P_ti represent the proportion of reference diet and test ingredient in the test diet, respectively; ME_ti, ME_td, and ME_rd represent the ME (kcal/kg) in test ingredient, test diet, and reference diet, respectively. Multiple regression analysis was used to estimate the DE and ME in DY and SBM as the slope of linear regression model:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{D}\mathrm{E}{\mathrm{I}}_{\mathrm{t}\mathrm{i}}=(\mathrm{D}{\mathrm{E}}_{\mathrm{D}\mathrm{Y}}\times \mathrm{F}{\mathrm{I}}_{\mathrm{t}\mathrm{i}})+(\mathrm{D}{\mathrm{E}}_{\mathrm{S}\mathrm{B}\mathrm{M}}\times \mathrm{F}{\mathrm{I}}_{\mathrm{t}\mathrm{i}});}\end{array}}$$

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{M}\mathrm{E}{\mathrm{I}}_{\mathrm{t}\mathrm{i}}=(\mathrm{M}{\mathrm{E}}_{\mathrm{D}\mathrm{Y}}\times \mathrm{F}{\mathrm{I}}_{\mathrm{t}\mathrm{i}})+(\mathrm{M}{\mathrm{E}}_{\mathrm{S}\mathrm{B}\mathrm{M}}\times \mathrm{F}{\mathrm{I}}_{\mathrm{t}\mathrm{i}}),}\end{array}}$$

where DE_DY and DE_SBM represent the estimated DE in DY and SBM (kcal/kg), respectively; ME_DY and ME_SBM represent the estimated ME in DY and SBM (kcal/kg), respectively; FI_ti represents the test ingredient intake (kg/d) which is the product of feed intake of pigs (kg/d) and P_ti; DEI_ti and MEI_ti represent the test ingredient-associated DE and ME intake (kcal/d) which is the product of FI_ti (kg/d) and DE_ti and ME_ti (kcal/kg), respectively. The intercept of the linear regression model is set at 0 as the FI_ti of pigs fed the reference diet.

Before statistical analysis, normality of residuals and outliers in data were tested using univariate procedure of SAS (SAS Inst. Inc., Cary, NC). Outlier was not detected in the dataset. Thereafter, data were analyzed by ANOVA using general linear model (GLM) procedure of SAS with the model including experimental diet, period, and block within period as independent variables. Contrast option was used to determine the linear and quadratic effects of the test ingredient. Multiple regression analysis and the comparison of estimated DE or ME between DY and SBM were conducted using GLM procedure of SAS as described by Littell et al. (1997). Experimental unit was pig in all statistical analyses. P-value less than 0.05 was considered as statistical significance, and P-value between 0.05 and 0.10 was considered as tendency.

Experiment 2: digestibility of AA

Twenty-one barrows were surgically fitted with T-cannulas at the distal ileum based on the procedure described by Dilger et al. (2004). After the surgery, pigs were individually housed and monitored for 7 d as a recovery period. Then, pigs were individually weighed and moved to metabolism crates. Mean initial BW of pigs was 20.0 kg (SD = 1.31). Pigs were divided into 7 blocks based on BW and allocated to 3 diets in a randomized complete block design. Therefore, there were 7 replicates per diet.

Three experimental diets were prepared based on cornstarch, dextrose, and sucrose (Table 2). Two diets were prepared to contain DY or SBM as the sole source of nitrogen (N). A N-free diet (NFD) was also prepared to determine the basal ileal endogenous losses (BEL) of CP and AA in pigs. All diets were prepared to meet or exceed the mineral and vitamin requirement estimates (NRC, 2012). Chromic oxide was added to all diets at 0.5% as an index marker.

Table 2.

Ingredient and analyzed nutrient composition of experimental diets used in Exp. 2, as-fed basis

	Diet
Item	Dried yeast	Soybean meal	Nitrogen-free
Ingredient, %
Cornstarch	31.13	36.86	65.03
Dried yeast	39.30	0.00	0.00
Soybean meal	0.00	33.30	0.00
Dextrose	10.00	10.00	10.00
Sucrose	10.00	10.00	10.00
Soybean oil	4.00	4.00	4.00
Cellulose1	0.00	0.00	4.00
Ground limestone	2.05	1.27	1.30
Monocalcium phosphate	0.35	1.40	2.00
Salt	0.40	0.40	0.40
Potassium carbonate	0.00	0.00	0.40
Magnesium oxide	0.00	0.00	0.10
Vitamin premix2	0.15	0.15	0.15
Mineral premix3	0.07	0.07	0.07
Selenium premix4	0.05	0.05	0.05
Chromic oxide premix5	2.50	2.50	2.50
Total	100.00	100.00	100.00
Analyzed nutrient, %
DM	90.9	90.6	90.8
CP6	17.5	15.9	3.2
Indispensable AA
Arg	0.55	0.90	0.01
His	0.28	0.34	0.01
Ile	0.70	0.62	0.01
Leu	1.03	1.03	0.03
Lys	0.96	0.82	0.02
Met	0.23	0.18	0.00
Phe	0.60	0.66	0.01
Thr	0.68	0.50	0.01
Trp	0.16	0.20	0.00
Val	0.80	0.62	0.01
Dispensable AA
Ala	0.88	0.59	0.02
Asp	1.44	1.42	0.01
Cys	0.13	0.19	0.00
Glu	2.19	2.37	0.02
Gly	0.57	0.56	0.01
Pro	0.51	0.72	0.03
Ser	0.55	0.56	0.02
Tyr	0.44	0.44	0.01

¹Solka-Floc 40 FCC (International Fiber Corporation, Urbana, OH).

²Provided the following quantities per kilogram of complete diet: vitamin A, 3,960 IU; vitamin D₃, 396 IU; vitamin E, 26.4 IU; menadione, 1.30 mg; riboflavin, 5.30 mg; _D-pantothenic acid, 13.2 mg; niacin, 19.8 mg; and vitamin B₁₂, 0.02 mg.

³Provided the following quantities per kilogram of complete diet: I, 0.258 mg; Mn, 12.0 mg; Cu, 6.33 mg; Fe, 136 mg; and Zn, 104 mg.

⁴Provided 0.3 mg Se/kg of complete diet.

⁵Provided 5 g chromic oxide/kg of complete diet.

⁶Analyzed nitrogen concentration multiplied by 6.25.

Daily feed allowance was set at 4% of mean BW of pigs within blocks which was equally divided into 2 meals and offered to pigs at 0800 and 1700 hours. Water was available at all times via nipple drinkers. Pigs were fed experimental diets for 5 d as adaptation, followed by 2 d of feeding with ileal digesta collection. Plastic sample bags (Whirl-Pak bag; NASCO, Fort Atkinson, WI) were used to collect ileal digesta samples from the T-cannulas for 9 h in each collection day. A 10 mL of 10% formic acid was added in the plastic sample bags. Bags were changed every 30 min, and removed bags containing ileal digesta samples were immediately stored at –20°C. Collected ileal digesta samples were slightly thawed, pooled within pigs, homogenized, and subsampled at the end of experiment.

Ileal digesta samples were lyophilized before chemical analyses. Samples of experimental diets and lyophilized ileal digesta were finely ground (<0.75 mm) using a centrifugal grinder. The concentration of DM in ground samples was analyzed as described in Exp. 1. Experimental diets and ileal digesta samples were analyzed for the concentration of N by a combustion method using TruMac N (LECO Corp., St. Joseph, MI; method 990.03; AOAC, 2000), which was multiplied by 6.25 to estimate the concentration of CP. The concentrations of AA in ground samples of DY, SBM, experimental diets, and ileal digesta were analyzed at University of Missouri Agricultural Experiment Station Chemical Laboratories (Columbia, MO). Samples were hydrolyzed by 6 M HCl (or BaOH for Trp analysis) at 110°C for 24 h under N atmosphere and analyzed for AA concentrations using a high-performance liquid chromatography after postcolumn derivatization [method 982.30 E (a, b, c); AOAC, 2006]. Performic acid hydrolysis was conducted before acid hydrolysis for the analysis of S-containing AA. Ground samples of experimental diets and ileal digesta were digested by nitric acid and 70% perchloric acid and analyzed for the concentration of Cr using a spectrophotometry (Spark 10M; Tecan Group Ltd., Männedorf, Switzerland) at 450 nm of absorption (Fenton and Fenton, 1979).

The apparent ileal digestibility (AID; %) and BEL of CP and AA [g/kg DM intake (DMI)] were calculated by the index method with the concentration of Cr in experimental diets and ileal digesta (Park et al., 2017):

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{A}\mathrm{I}\mathrm{D}=100\times [1-\left(\mathrm{A}{\mathrm{A}}_{\mathrm{o}\mathrm{u}\mathrm{t}}/\mathrm{A}{\mathrm{A}}_{\mathrm{i}\mathrm{n}}\right)\times \left(\mathrm{C}{\mathrm{r}}_{\mathrm{i}\mathrm{n}}/\mathrm{C}{\mathrm{r}}_{\mathrm{o}\mathrm{u}\mathrm{t}}\right)];}\end{array}}$$

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{B}\mathrm{E}\mathrm{L}=\mathrm{A}{\mathrm{A}}_{\mathrm{o}\mathrm{u}\mathrm{t}}\times \left(\mathrm{C}{\mathrm{r}}_{\mathrm{i}\mathrm{n}}/\mathrm{C}{\mathrm{r}}_{\mathrm{o}\mathrm{u}\mathrm{t}}\right),}\end{array}}$$

where AA_in and AA_out represent the concentration of CP or AA (g/kg DM) in experimental diets and ileal digesta, respectively; Cr_in and Cr_out represent the concentration of Cr (g/kg DM) in experimental diets and ileal digesta, respectively. To calculate the SID of CP and AA (%) in DY and SBM, the AID of CP and AA were corrected for the BEL of CP and AA:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{S}\mathrm{I}\mathrm{D}=\mathrm{A}\mathrm{I}\mathrm{D}+100\times \left(\mathrm{B}\mathrm{E}\mathrm{L}/\mathrm{A}{\mathrm{A}}_{\mathrm{i}\mathrm{n}}\right).}\end{array}}$$

Normality of residuals and outliers in data were tested as described in Exp. 1. Outlier was not detected in the dataset. Data were analyzed by ANOVA using GLM procedure of SAS. Independent variables in the model were experimental diets and block. Experimental unit and declaration of significance and tendency were the same as Exp. 1.

Results

Pigs used in both Exp. 1 and 2 maintained good condition throughout the experimental periods.

Chemical analysis

The GE in feed ingredients used in this study ranged from 3,855 kcal/kg in corn to 4,783 kcal/kg in DY (Table 3). Dried yeast contained greater GE but less CP concentration than SBM. The analyzed concentrations of ether extract in DY and SBM were 1.26% and 1.87%, respectively. The concentrations of indispensable AA, except Met and Thr, in SBM were greater than the values in DY.

Table 3.

Analyzed nutrient composition of corn, canola meal, dried yeast, and soybean meal used in Exp. 1 and 2, % as-fed basis

Item	Corn	Canola meal	Dried yeast	Soybean meal
DM	87.2	89.3	87.3	88.3
GE, kcal/kg	3,855	4,346	4,783	4,057
CP1	8.1	38.7	40.6	49.6
Ether extract	2.96	3.69	1.26	1.87
Ash	1.25	6.49	8.07	4.72
Neutral detergent fiber	7.52	21.79	0.62	8.58
Acid detergent fiber	1.91	15.66	0.12	4.38
Indispensable AA
Arg	—	—	1.42	3.23
His	—	—	0.72	1.18
Ile	—	—	1.80	2.22
Leu	—	—	2.58	3.49
Lys	—	—	2.47	2.87
Met	—	—	0.61	0.61
Phe	—	—	1.54	2.32
Thr	—	—	1.72	1.70
Trp	—	—	0.49	0.69
Val	—	—	2.06	2.24
Dispensable AA
Ala	—	—	2.28	1.96
Asp	—	—	3.71	5.03
Cys	—	—	0.35	0.64
Glu	—	—	5.49	8.11
Gly	—	—	1.54	1.84
Pro	—	—	1.37	2.37
Ser	—	—	1.33	1.78
Tyr	—	—	1.27	1.60

¹Analyzed nitrogen concentration multiplied by 6.25.

Experiment 1: DE and ME

Feed intake of pigs was not affected by the inclusion of DY or SBM in the reference diet (Table 4). The GE in feces linearly increased (P < 0.001) with increasing concentration of DY in the diets, which resulted in the tendency on linear increase (P = 0.060) in fecal GE output and on linear reduction (P = 0.065) in ATTD of GE in experimental diets. However, the DE in experimental diets was not affected by the addition of test ingredients in the reference diet. There was a linear increase (P = 0.020) in urinary GE output as the concentration of DY in experimental diets increased. Consequently, the metabolizability of GE in experimental diets linearly decreased (P = 0.003) with increasing concentration of DY. There was a quadratic tendency (P = 0.075) in the ME in experimental diets as the concentration of DY increased. Increasing concentration of SBM did not affect the ATTD and metabolizability of GE in experimental diets.

Table 4.

Apparent total tract digestibility (ATTD) and metabolizability of GE in experimental diets fed to growing pigs in Exp. 1¹

	Diet2						P-value3
		DY, %		SBM, %			DY		SBM
Item	RD	5	10	5	10	SEM	L	Q	L	Q
Feed intake, g/d	772	775	806	824	818	30.3	0.437	0.714	0.292	0.439
GE intake, kcal/d	3,144	3,172	3,354	3,367	3,338	123.9	0.245	0.617	0.282	0.417
Feces output, g/d	123	133	138	134	132	7.0	0.150	0.793	0.382	0.469
GE in feces, kcal/kg	4,434	4,486	4,602	4,461	4,418	20.6	<0.001	0.211	0.588	0.178
Fecal GE output, kcal/d	547	596	635	597	583	31.1	0.060	0.903	0.426	0.409
DE intake, kcal/d	2,597	2,576	2,719	2,769	2,755	100.8	0.402	0.515	0.281	0.457
ATTD of GE, %	82.7	81.2	81.1	82.3	82.5	0.57	0.065	0.334	0.872	0.629
DE in diet, kcal/kg	3,369	3,324	3,377	3,361	3,368	23.6	0.831	0.107	0.977	0.791
DE in diet, kcal/kg DM	3,818	3,773	3,816	3,808	3,811	26.7	0.952	0.199	0.865	0.849
Urine output, g/d	936	1,029	1,287	1,045	739	182.2	0.188	0.718	0.455	0.362
GE in urine, kcal/kg	74.8	98.3	80.1	88.5	116.9	18.79	0.844	0.376	0.129	0.753
Urinary GE output, kcal/d	59.9	77.1	95.7	79.8	76.0	9.97	0.020	0.952	0.270	0.345
ME intake, kcal/d	2,537	2,499	2,623	2,690	2,679	98.7	0.544	0.510	0.322	0.506
Metabolizability of GE, %	80.7	78.8	78.2	79.9	80.3	0.51	0.003	0.275	0.549	0.341
ME in diet, kcal/kg	3,290	3,225	3,257	3,264	3,276	21.1	0.281	0.075	0.650	0.472
ME in diet, kcal/kg DM	3,728	3,661	3,680	3,698	3,707	23.8	0.175	0.153	0.544	0.522

¹Each least squares mean represents 6 observations.

2RD, reference diet; DY, dried yeast; SBM, soybean meal.

3L, linear effect; Q, quadratic effect.

On an as-fed basis, the estimated DE in DY and SBM were 3,565 and 3,629 kcal/kg, respectively, which were not different from each other (Table 5). The ME was estimated at 3,119 kcal/kg in DY and 3,393 kcal/kg in SBM. The DE in DY was estimated at 3,933 kcal/kg DM which was not different from the estimated DE in SBM at 4,020 kcal/kg DM. Similarly, there was no difference in the estimated ME between DY and SBM (3,431 and 3,756 kcal/kg DM, respectively). In all multiple linear regression, the estimates of intercept were not different from 0, and the R² value ranged from 0.863 to 0.865.

Table 5.

DE and ME in dried yeast (DY) and soybean meal (SBM) estimated by regression analysis in Exp. 1¹

		Slope2				P-value
Item	Intercept	DY	SBM	R2	SD	Model	DY vs. SBM3
As-fed basis
DE	–23.9	3,565	3,629	0.865	46.1	<0.001	0.846
SE	15.73	313.9	308.1
P-value	0.140	<0.001	<0.001
ME	–17.2	3,119	3,393	0.863	42.1	<0.001	0.337
SE	14.37	286.9	281.5
P-value	0.242	<0.001	<0.001
Dry matter basis
DE	–21.1	3,933	4,020	0.864	45.3	<0.001	0.811
SE	15.43	348.4	342.2
P-value	0.183	<0.001	<0.001
ME	–14.4	3,431	3,756	0.863	41.2	<0.001	0.315
SE	14.04	317.0	311.4
P-value	0.313	<0.001	<0.001

¹Regression analysis was conducted with test ingredient intake (kg/d or kg DM/d) as independent variable and test ingredient-associated DE or ME intake (kcal/d) as dependent variable (30 observations).

²The DE or ME in DY and SBM were estimated by the slope of regression analysis.

³Comparison of the slope between DY and SBM.

Experiment 2: digestibility of AA

The AID of CP and AA, except Ala and Pro, in DY were less than in SBM (Table 6). The AID of Gly in DY tended to be less (P = 0.061) than in SBM. The AID of indispensable AA in DY ranged from 57.4% for Thr to 77.2% for Lys, and those in SBM ranged from 74.8% for Thr to 86.6% for Arg. Pigs fed NFD had the BEL of CP at 13.6 g/kg DMI (Table 6). The BEL of indispensable AA ranged from 0.083 g/kg DMI for Met to 0.534 g/kg DMI for Leu. Pigs fed the diet containing DY had less (P < 0.05) SID of CP and AA, except Pro, than those fed the diet containing SBM (Table 7). The SID of Gly in DY tended to be less (P = 0.058) than in SBM. The SID of indispensable AA in DY ranged from 64.1% for Thr to 85.2% for Arg, and those in SBM ranged from 83.9% for Thr to 91.8% for Arg. The SID of Lys, Met, Thr, and Trp in DY were 81.4%, 79.5%, 64.1%, and 81.4%, respectively, whereas respective values in SBM were 87.6%, 89.7%, 83.9%, and 90.6%, respectively.

Table 6.

Apparent ileal digestibility (AID; %) of CP and AA in dried yeast and soybean meal fed to growing pigs and basal ileal endogenous losses (BEL; g/kg DMI) of CP and AA in growing pigs fed nitrogen-free diet in Exp. 2¹

	AID, %
	Diet				BEL, g/kg DMI
Item	Dried yeast	Soybean meal	SEM	P-value	Mean	SD
CP	65.4	79.7	1.64	<0.001	13.6	1.44
Indispensable AA
Arg	76.8	86.6	1.70	0.006	0.513	0.0644
His	75.8	84.9	0.87	<0.001	0.171	0.0276
Ile	74.4	83.5	0.91	<0.001	0.316	0.0617
Leu	77.0	83.6	0.82	0.001	0.534	0.0996
Lys	77.2	82.7	0.92	0.006	0.439	0.0762
Met	76.2	85.5	1.04	<0.001	0.083	0.0222
Phe	76.7	84.6	0.87	<0.001	0.306	0.0559
Thr	57.4	74.8	0.89	<0.001	0.498	0.0466
Trp	74.3	85.0	0.71	<0.001	0.124	0.0306
Val	71.5	78.6	0.86	0.001	0.453	0.0642
Dispensable AA
Ala	74.3	77.7	1.39	0.139	0.487	0.0597
Asp	73.8	81.0	0.84	<0.001	0.724	0.1135
Cys	21.0	74.4	1.82	<0.001	0.166	0.0243
Glu	81.0	86.4	1.03	0.010	0.858	0.1434
Gly	49.0	64.4	4.71	0.061	1.085	0.1176
Pro	18.8	34.2	17.43	0.556	3.173	0.8015
Ser	59.5	79.5	1.02	<0.001	0.421	0.0376
Tyr	73.0	82.6	1.02	<0.001	0.258	0.0536

¹Each least squares mean represents 7 observations.

Table 7.

Standardized ileal digestibility (SID; %) of CP and AA in dried yeast and soybean meal fed to growing pigs in Exp. 2¹

	Diet
Item	Dried yeast	Soybean meal	SEM	P-value
CP	72.4	87.5	1.64	<0.001
Indispensable AA
Arg	85.2	91.8	1.70	0.035
His	81.3	89.4	0.87	<0.001
Ile	78.5	88.1	0.91	<0.001
Leu	81.7	88.3	0.82	0.001
Lys	81.4	87.6	0.92	0.003
Met	79.5	89.7	1.04	<0.001
Phe	81.3	88.8	0.87	<0.001
Thr	64.1	83.9	0.89	<0.001
Trp	81.4	90.6	0.71	<0.001
Val	76.6	85.3	0.86	<0.001
Dispensable AA
Ala	79.3	85.1	1.39	0.026
Asp	78.3	85.7	0.84	<0.001
Cys	32.6	82.3	1.82	<0.001
Glu	84.6	89.7	1.03	0.013
Gly	66.3	81.9	4.71	0.058
Pro	75.3	74.1	17.43	0.961
Ser	66.4	86.3	1.02	<0.001
Tyr	78.4	87.9	1.02	<0.001

¹Each least squares mean represents 7 observations.

Discussion

The analyzed GE and nutrient concentrations in corn and canola meal agree with the reference values (NRC, 2012; Maison et al., 2015; Park et al., 2019). The analyzed GE in DY is within the range of values reported in previous studies (NRC, 2012; Kim et al., 2014; Cruz et al., 2020). The concentration of CP in DY is close to the concentration of CP in single cell protein (SCP) reported in NRC (2012) and that in 2 sources of DY reported in Kim et al. (2014), although it was generally lower than the values in brewers’ yeast and torula yeast (NRC, 2012; Kim et al., 2014; Lagos and Stein, 2020). The differences in the concentration of CP in DY among studies may be due to the differences in the strain of yeast or substrates of fermentation. However, it should be noted that the considerable proportion of N in DY originates from nucleotides, and therefore, the estimated concentration of CP may not represent the actual concentration of protein in DY (Shurson, 2018). The concentrations of GE, nutrients, and AA in SBM are consistent with the previously reported values (NRC, 2012; Kim et al., 2014; Park et al., 2017).

Dried yeast contains not only nutrients and energy but also cell wall components. Yeast cell wall mainly consists of β-glucan and mannan as 2 dominant polysaccharides, both of which have been found to carry beneficial effects on gut health of pigs (Kogan and Kocher, 2007; Shurson, 2018). In brief, dietary β-glucan may improve intestinal immune responses by modulating immune cells in the Payer’s patch or intraepithelial lymphocytes (Volman et al., 2008), and dietary mannan may bind to the surface of pathogenic bacteria and prevent their colonization (Spring et al., 2015). Many studies have been conducted to evaluate the effects of feeding β-glucan or mannan as feed additives on growth performance of pigs (Miguel et al., 2004; Vetvicka et al., 2014) in which the concentrations of purified polysaccharides were generally lower than 0.5% in experimental diets. However, in Exp. 1, the concentration of DY gradually increased in experimental diets, and therefore, the concentration of yeast cell wall components also increased which may be responsible for the linear increase of analyzed GE in feces and the tendency of linear reduction of the ATTD of GE in diets. Soluble dietary fiber including β-glucan may increase the viscosity of digesta and hinder the absorption of nutrients through the mucus layer (Johnson and Gee, 1981). In addition to polysaccharides in yeast cell wall, DY also contains nucleotides in considerable amounts, which may be beneficial to intestinal physiology and immunity (Sauer et al., 2011). However, disproportionate amounts of nucleotides absorbed into the body are subject to catabolism, leading to an increased production of urea for excretion of N. This may explain the linearly increased urinary GE output and linearly decreased metabolizability of GE with increasing concentration of DY in diets.

In Exp. 1, regression analysis was used to estimate the DE and ME in test ingredients together with the difference procedure which has been used in previous studies (Zhang and Adeola, 2017; Zhong and Adeola, 2019). Similarly, regression and direct procedures do not give different estimates of DE and ME (Bolarinwa and Adeola, 2016). The reason for using both regression analysis and difference procedure was to reduce the standard errors for determined DE and ME values. Because direct procedure that requires a diet containing test ingredient as a sole source of energy was not an option for 2 test ingredients used in the current study, difference procedure was used which relies on proportional contribution of energy from the reference diet and test ingredients. However, standard errors for values obtained from difference procedure are generally greater than direct procedure, especially when the concentration of test ingredient is low in experimental diets (Oliveira et al., 2020a). In order to reduce the standard error and increase the accuracy of estimation, regression analysis was used which involves greater number of observations with 3 different inclusion rates (i.e., 0%, 5%, or 10%) of test ingredients. It should be noted that the regression analysis used in the current study is different from the regression procedure suggested by Adeola (2001). In Exp. 1, the DE and ME in test ingredients were estimated by the slope of linear regression, whereas in the procedure suggested by Adeola (2001), the energy value (or digestibility of nutrient of interest) is estimated by the extrapolation to 100% test ingredient in experimental diets where the independent variable (i.e., x-axis) of the linear regression model is the concentration of test ingredient in experimental diet.

Regression analysis indicated that the estimated DE and ME in DY were not different from the values in SBM despite the numerical difference at 325 kcal/kg DM in ME values between DY and SBM. The lack of difference in ME values may be partly due to the greater standard errors, which were relatively greater than previous studies conducted with the difference method (Kim et al., 2014; Son and Kim, 2016; Lagos and Stein, 2020). Perhaps the lower concentrations of test ingredients in experimental diets compared with previous studies ranging from 15% to 40% of diets are partly responsible. In the current study, the concentrations of test ingredients were limited to 5% and 10% in order to moderate palatability and reduce urinary N excretion from relatively higher nucleotide concentration in DY. It should be noted that the results of current study do not indicate that the energy values in DY are equivalent to those in SBM. Simple replacement of SBM by DY may result in reduction of growth performance due to the numerical difference in ME values between DY and SBM.

The DE and ME in DY estimated in Exp. 1 are comparable to the values in one SCP reported by Son and Kim (2016) but are somewhat less than previously reported values in SCP (NRC, 2012), turola yeast (NRC, 2012; Lagos and Stein, 2020), and 2 sources of DY reported by Kim et al. (2014). Perhaps the discrepancy is due to the differences in experimental methodologies, the strain of yeast, or substrates of fermentation. Even though methodology in the current experiment was different from previous studies, estimated DE and ME in SBM are within the range of values observed in the previous studies (NRC, 2012; Kim et al., 2014; Oliveira et al., 2020b).

The BEL of CP and AA in pigs observed in Exp. 2 are close to the values reported in previous summaries (NRC, 2012; Park et al., 2013; Adeola et al., 2016). Dried yeast had less AID and SID of CP and all indispensable AA than SBM. Similar to the tendency of linear reduction in ATTD of GE in diets with increasing DY concentrations in Exp. 1, the reason for less digestibility of CP and most AA in DY compared with SBM may be due to polysaccharides which hinder the enzymatic digestion of proteins as well as absorption of AA. In addition, heat processing during the production of DY might further reduce the digestibility of AA. Increasing temperature and time of heat processing has been shown to reduce the SID of AA in plant protein sources including SBM (Oliveira et al., 2020b), canola meal (Almeida et al., 2014), or distillers’ dried grains with solubles (Almeida et al., 2013). Among the indispensable AA, Lys is known to be the most sensitive AA to heat damage due to the Maillard reaction in which Lys binds to reactive sugar and becomes unreactive Lys (Kim et al., 2012). This may partly explain the less SID of Lys in DY than in SBM. In addition, the SID of Thr in DY was particularly less than the other indispensable AA. This observation may be related to the increased secretion of mucin stimulated by yeast cell wall components. Luo et al. (2019) reported that the relative expression of messenger ribonucleic acid for genes related to mucin production (i.e., MUC1 and MUC2) in weanling pigs were increased by dietary supplementation of β-glucan extracted from selected bacterium. Because Thr is the major indispensable AA in mucin, feeding diets containing high concentration of DY may have stimulated the secretion of mucin and increased the specific ileal endogenous loss of Thr, leading to the reduction of SID of Thr in DY. The SID of CP and AA in DY observed in Exp. 2 were greater than in brewers’ yeast and SCP provided by NRC (2012) but less than the values in torula yeast reported by Lagos and Stein (2020). Dried yeast used in the current study also had less SID of CP and AA than in yeast extract reported by Mateo and Stein (2007), probably due to the fact that cell wall components were removed in yeast extract. The SID of CP and AA in SBM are consistent with the values in previous studies (NRC, 2012; Park et al., 2017; Oliveira et al., 2020b).

In conclusion, DE and ME of DY used in the current study were not different from the values in SBM. Dried yeast had less SID of most AA compared with SBM. Energy values as well as the digestibility of AA in DY and SBM determined in the current study can be used when formulating diets for growing pigs with these ingredients. Further research is needed to evaluate the optimal dietary inclusion of DY to maximize the growth performance not only focusing on nutritional values but also considering the beneficial effects on gut health of pigs.

Glossary

Abbreviations

AA

amino acid

AID

apparent ileal digestibility

ATTD

apparent total tract digestibility

BEL

basal ileal endogenous losses

BW

body weight

CP

crude protein

DE

digestible energy

DM

dry matter

DMI

dry matter intake

DY

dried yeast

GE

gross energy

ME

metabolizable energy

NFD

nitrogen-free diet

SBM

soybean meal

SCP

single cell protein

SID

standardized ileal digestibility

Conflict of Interest Statement

The authors declare that there is not conflict of interest in the current study.