Chemical composition and standardized ileal digestibility of crude protein and amino acid in whole yeast and autolyzed yeast derived from sugarcane ethanol production fed to growing pigs

Objective This research determined the chemical composition and the apparent and standardized ileal digestibility (AID and SID) of crude protein (CP) and amino acids (AA) in whole yeast and autolyzed yeast derived from sugarcane ethanol production fed to growing pigs. Methods Six growing pigs were randomly allocated in a replicated 3×3 Latin square design with 3 diets and 3 periods of 7 days each, resulting in a total of 6 experimental replications. Three assay diets were formulated using whole yeast, autolyzed yeast, or soybean meal as the sole sources of dietary CP and AA. Pigs were allowed to adapt to the assay diets for 5 days. Thereafter, ileal digesta samples were collected continuously for 8 hours on days 6 and 7. Results There was no difference in the chemical composition between whole yeast and autolyzed yeast, but whole yeast had low digestibility of CP and AA due to the presence of a rigid cell wall. As conducting autolysis can induce cell wall damage, the AID and SID of CP and AA were greater in autolyzed yeast than in whole yeast. Conclusion The information obtained on the SID of CP and AA in both yeast products can be used for the accurate estimation of the bioavailability of CP and AA in feed formulations. The yeast products derived from sugarcane ethanol production are an alternative protein source in pig diets.


INTRODUCTION
Thailand is one of the topfive countries for sugarcane production globally with annual exports of approximately 10 million metric tons, ranking second behind Brazil [1]. Sugarcane molasses is a byproduct of sugar production and is widely used as a substrate for ethanol fermentation [2]. This process is commonly performed using Saccharomyces cerevisiae as this yeast strain has high potential for fermenting sugar to produce a high ethanol con centration [3]. As the yeast biomass is a rich source of nutritional values, including protein, peptides, amino acids (AA), carbohydrate, fatty acids, vitamins and minerals [4], the sugar cane yeast derived from ethanol production can be included in animal feed as an alternative protein source [5].
Whole yeast typically consists of the cell wall, periplasm, plasma membrane, organelles and cytoplasm [6]. Nutritionally, the highly digestible compounds (including proteins, AA, polysaccharides and nucleotides) are present in intracellular constituents [7], whereas the major structural components strengthening the yeast cell wall are glucans, mannans and chitin [6]. The rigid yeast cell wall has limited digestibility due to the absence of those endogenous enzymes in the small intestine [8]. Alternatively, the autolysis of yeast cell disrup tion has been reported using mechanical (bead mill, high pressure homogenizer, and ultrasound) and nonmechanical (electrical, enzymatic, physical such as osmotic pressure or heat shock, and chemical) methods to release intracellular bioactive compounds [7].
Numerous yeast products are commercially produced and supplemented in animal feed to improve growth perfor mance and gut health function [4]. As standardized ileal digestibility (SID) of crude protein (CP) and AA has been widely accepted as a more suitable approach for bioavailabil ity of CP and AA in feed ingredients [9], the SID of CP and AA have been reported for brewer's yeast [10], torula yeast [11], yeast extract [12,13] and yeast products produced by the ethanol industry [14]. However, there is no information on the nutrient digestibility of yeast products derived from sugarcane ethanol production. The additional process of au tolysis method may break the rigid yeast cell walls, and then release intracellular nutrients, resulting in an improvement of protein digestion in growing pigs. Therefore, the objective of this research was to determine the chemical composition, apparent ileal digestibility (AID) and SID of CP and AA in whole yeast and autolyzed yeast fed to growing pigs. For com parison, the chemical composition and the AID and SID of CP and AA were determined in soybean meal, as it is a pro teinrich ingredient used extensively in swine diets.

MATERIALS AND METHODS
The research proposal was reviewed and approved by the Animal Care and Use for Scientific Research Committee, Kasetsart University, Bangkok, Thailand (ACKU64AGR021) and care of animals throughout this experiment was in an accordance with the corresponding Ethical Principles for the Use of Animals for Scientific Purposes [15].

Animals, housing, and surgical procedures
Six growing pigs (Duroc×Large white×Landrace) with ini tial mean (±standard deviation) body weight (BW) of 67.60 ±6.15 kg were used to determine AID and SID for CP and AA in whole yeast, autolyzed yeast, and soybean meal. The pigs were housed in individual pens (2.0×2.5 m). An evapo rative cooling system was used to control environmental temperature with an average of 25°C to 26°C. Each pen was equipped with a lowpressure drinking nipple that allowed free access to water. The pigs were surgically fitted with a simple Tcannula at the distal ileum according to the proce dures described by Li et al [16]. Until surgery and during the recovery period, pigs were fed a commercial grower diet with 17.5% CP and 3,200 kcal metabolizable energy (ME)/kg (as fed basis). The pigs were allowed a recuperation period of at least 7 days.

Experimental design, diets, and procedures
Six growing pigs were randomly allocated in a replicated 3×3 Latin square design with 3 diets and 3 periods of 7 days each, resulting in a total of 6 experimental replications. Ex perimental diets were fed to the pigs at a daily level of 2.5 times the estimated maintenance requirement for energy (197 kcal metabolizable per BW 0.60 ; NRC [17]). Body weight was determined at the beginning of each experimental period to keep the daily feed intake constant in relation to the ani mals' BW throughout the experiment. The diets were divided into 2 equal meals and fed at 0800 and 1600 h.
Soybean meal, whole yeast, and autolyzed yeast were used as the sole sources of CP and AA. The soybean meal was purchased from Thai Vegetable Oil Public Company Limited (Thonburi, Bangkok, Thailand) and used as the control group. The commercial yeast products, consisting of whole yeast (KOBOTEN) and autolyzed yeast (KOBOTENnutri30), were supplied by Mitr Phol Bio Fuel Co., Ltd. (Bangkok, Thailand). The processing steps for the yeast products are presented in Figure 1. In brief, Saccharomyces cerevisiae was used as a se lected yeast strain to supplement in sugarcane molasses under fermentation conditions at 30°C to 34°C for 36 to 40 h in a bio reactor tank (Step 1). Thereafter, the yeast biomass was separated from solubles using centrifugation at 13,614 g (Step 2). The supernatant was carefully removed and used to produce ethanol, whereas the remaining sediment (yeast cells) was washed with water (Step 3). After that, the sedi ment was dried at 80°C under the evaporation system (Step 4). Finally, spray drying occurred at 180°C (Step 5) to pre pare the powder form of whole yeast. For autolysis of the yeast product, the sediment from Step 3 was further auto lyzed at pH 5 to 6 at 50°C to 60°C for 24 h, followed by Steps 4 and 5, respectively.
Three assay diets were formulated using whole yeast, auto lyzed yeast, and soybean meal as the sole sources of dietary CP and AA (Table 1) to meet or exceed the dietary threshold levels for CP and AA according to NRC [17] nutrient recom mendations for pigs from 50 to 75 kg BW. All diets contained 0.5% chromic oxide as an indigestible index. Vitamins and minerals were supplemented to the assay diets to meet or exceed the requirement estimate in NRC [17]. An Nfree diet (Table 1) was used to determine basal ileal endogenous losses of CP and AA (IAA end ) in an additional period at the end of the experiment.
In each of the 4 experimental periods, pigs were allowed to adapt to the assay diets for 5 days. Thereafter, ileal digesta samples were collected continuously for 8 h on days 6 and 7. The samples were collected using plastic bags attached to the barrel of the cannula with elastic bands. The bags were changed whenever they were filled with digesta and immediately fro zen at -18°C. During collection, 4 mL of 2.5 M formic acid were added to the sampling bags to minimize further micro bial fermentation. The frozen ileal digesta samples were allowed to thaw at room temperature, and mixed within animal and period. The ileal digesta samples were dried using freeze dryer (FDB5503; Operon Co., Ltd., Gyeonggido, Korea), and were finely ground to 0.5 mm before chemical analysis.

Chemical analyses
The samples (soybean meal, whole yeast, autolyzed yeast, assay diets and ileal digesta) were analyzed in duplicate. Official standard methods [18] were used to determine the contents of dry matter (DM; method 930.15), ether extract (method 920.38), crude fiber (method 978.10), CP (method 984.13), neutral detergent fiber (NDF; method 2002.04), and acid detergent fiber (ADF; method 973.18). The CP contents were calculated by multiplying the content of nitrogen by 6.25. The AA contents were determined using ultra high performance liquid chromatography (UHPLC) (Nexera X2; Shimadzu Scientific Instruments, Kyoto, Japan) after being hydrolyzed with 6 N HCl for 18 h at 115°C. Tryptophan was not determined. The chromium contents in the diet and ileal digesta samples were analyzed using a spectro photometer (UV 1800; Shimadzu Scientific Instruments, Japan) according to the AOAC [18] procedure (method 968.088D).

Calculations
The AID of CP and AA in the assay diets were calculated ac cording to the equation: Where AID D = AID of CP and AA in assay diet (%), I D = marker content in the assay diet (g/kg DM), A I = CP or AA content in ileal digesta (g/kg DM), A D = CP or AA content in the assay diet (g/kg DM), and I I = marker content in ileal digesta (g/kg DM).
The SID of CP and AA in whole yeast, autolyzed yeast and soybean meal were calculated by correcting the AID of CP and AA in the assay diet for IAA end . The values for IAA end were obtained using an Nfree diet in combination with an indigestible marker according to the equation: where IAA end = IAA end in g/kg DM intake. The SID of CP and AA were calculated using the equation: where SID D = SID of CP and AA in yeast products and soy bean meal.

Statistical analysis
Homogeneity of variances and normal distribution of data were confirmed using the UNIVARIATE procedure of the SAS software package (SAS Inst., Inc., Cary, NC, USA). Out liers were identified using the MIXED procedure of SAS, where the data with studentized residuals greater than 3.0 were considered outliers and excluded from further statisti cal analyses. The data were subjected to mixed modeling using diet as a fixed effect and pig and period as random ef fects. The individual pig was the experimental unit. The significant differences between treatments based on a ttest were set at α = 0.05 using the algorithm for letterbased rep resentation of all pairwise comparison according to Piepho [19].

RESULTS
The pigs remained healthy and readily consumed their assay diets throughout the experiment. Two outliers were detected for the pigs fed with soybean meal in period 2 and whole yeast in period 3. The AID and SID of CP and AA were con sidered outliers as influential and hence excluded from further statistical analysis. Therefore, the number of observations included in the model was 5 for whole yeast and soybean meal, and 6 for autolyzed yeast.

Chemical composition
The analyzed chemical compositions of soybean meal, whole yeast and autolyzed yeast are shown in Table 2. The CP con tents were lower in whole yeast and autolyzed yeast (341.6 and 334.3 g/kg DM, respectively) than in soybean meal (467.8 g/kg DM). The ether extract contents were lower in soybean meal (13.2 g/kg DM) than in whole yeast and autolyzed yeast (81.6 and 89.8 g/kg DM, respectively), whereas the crude fiber contents were greater in soybean meal (50.8 g/kg DM) than in both yeast products (8.8 g/kg DM). The NDF contents were greatest in whole yeast (165.0 g/kg DM) and lowest in autolyzed yeast (68.6 g/kg DM) with intermediate contents for soybean meal (137.4 g/kg DM), whereas the ADF con Table 1. Composition of assay diets for determination of apparent and standardized ileal digestibility of crude protein and amino acids in whole yeast, autolyzed yeast and soybean meal (%, as-fed basis)

Item Assay diet N-free
Test product 47.86 -Corn starch 1) 36.42 66.28 Dextrose 2) 9.00 20.00 Cellulose 3) 2.00 5.00 Soybean oil 4) 2.00 4.00 Calcium carbonate 0.99 -Monodicalcium phosphate 5) 0.60 3.00 Vitamin and mineral premix 6) 0  tents were greater in soybean meal (76.8 g/kg DM) than in whole yeast and autolyzed yeast (21.7 and 2.2 g/kg DM, re spectively). The ash contents were greatest in whole yeast and autolyzed yeast (244.5 and 256.0 g/kg DM, respectively) compared to soybean meal (68.5 g/kg DM). The contents of all AA samples were greatest in soybean meal, ranging for indispensable AA from 5.3 g/kg DM for methionine to 34.4 g/kg for leucine. For yeast products, the contents of indis pensable AA in whole yeast ranged from 3.0 g/kg DM for histidine to 24.9 g/kg DM for arginine, whereas those values in autolyzed yeast ranged from 3.0 g/kg DM for histidine to 25.8 g/kg DM for arginine. There were relatively small differ ences in chemical composition and AA between whole yeast and autolyzed yeast.

Apparent and standardized ileal CP and AA digestibility of whole yeast, autolyzed yeast, and soybean meal
The AID and SID of CP and almost all AA were lowest (p< 0.05) in whole yeast compared to those digestibilities in auto lyzed yeast and soybean meal (Tables 3, 4). The AID and SID of CP in autolyzed yeast were not different from those digestibilities in soybean meal. In comparison, the AID of histidine, threonine, cystine and serine was greatest (p<0.05) in soybean meal, but the AID of alanine was greatest (p<0.05) in autolyzed yeast. The SID of threonine and serine was greater (p<0.05) in soybean meal than in autolyzed yeast, whereas the SID of alanine and glycine was greater (p<0.05) in autolyzed yeast than in soybean meal. For other AA, no differences in the AID and SID were observed.

DISCUSSION
In the present study, the yeast products derived from sugar cane ethanol production had greater contents of fat and ash, but lower contents in CP and almost AA compared to those digestibilities in brewer' s yeast [10,17], torula' s yeast [11], single cell protein [17] and yeasts from grain ethanol coproducts [20]. The variation in nutrient contents of the yeast products may have been due to the different substrates [4] and the specific yeast strains [7,11]. For example, Saccharomyces cerevisiae (brewer's yeast) uses cereal grain as a substrate [21], whereas Candida utilis (torula's yeast) uses wood pulp from paper manufacturing as a substrate [4,22], resulting in different chemical compositions. As shown in the tabulated values [17], the CP contents were 46.52% and 51.17% DM for brewer's yeast and torula's yeast, respectively. The indis pensable AA ranged from 0.74 for methionine to 3.22% DM for lysine in brewer's yeast, and from 0.64 for methionine to 3.41% DM for leucine in torula's yeast. In the present study, sugarcane molasses was used as a substrate and fermented with Saccharomyces cerevisiae. The lower nutrient contents in sugarcane molasses than in cereal grain or wood pulp may have resulted in the relative low contents of CP and AA in the whole yeast and autolyzed yeast samples compared to other yeast products. The AID and SID of CP and AA were lower in whole yeast, but for autolyzed yeast were within the range of published digestibility for brewer's yeast [10], torula's yeast [11], and yeast extract [12,13]. As the yeast cell has a rigid cell wall con taining mainly glucans and mannans and a minor structure of chitin and protein [23], the lack of endogenous enzymes in the small intestine against these component [8] might have lowered the digestibility of CP and AA in whole yeast. Alter natively, the adjusted conditions of pH and temperature in industrial processing can cause cell autolysis that releases nutrients (such as CP and AA) from the interior of the yeast cell [7], resulting in greater AID and SID of CP and AA in autolyzed yeast than in whole yeast. The similar range of nu trient digestibility in autolyzed yeast compared to other yeast products suggested that Saccharomyces cerevisiae extracted from sugarcane ethanol production had good quality and was the same as brewer' s yeast, torula' s yeast and yeast extract. Notably, the AID and SID of CP and AA in autolyzed yeast were greater than those digestibility in single cell protein [17]. One possible explanation is that the single cell protein may be produced from bacterial protein containing high nucleo tide contents, which are poorly digested by pigs [24].
The autolysis process can produce a small size of yeast cell debris with thinner intracellular layers [25]. This phenomenon may be more beneficial to improve nutrient digestion and absorption of autolyzed yeast for host animal. In the present study, the degradation of yeast cell wall can be observed by lower NDF and ADF contents in autolyzed yeast than in whole yeast. According to previous studies, the low NDF con tents can reduce endogenous AA losses [26], and increase retention time of digesta in the gastrointestinal tract of pigs [27]. In contrast, greater NDF and ADF contents may induce the specific endogenous CP and AA losses [28], resulting in impaired ileal digestibility of CP and AA [29]. Therefore, the small protein molecule and the relatively low fiber fraction in autolyzed yeast might be the reasons for enhancing AID and SID of CP and AA in growing pigs.
Soybean meal is widely used as a major protein source for pigs [30]. The chemical composition of brewer's yeast is comparable to the nutrient composition of dehulled soybean meal [4]. Likewise, the AID and SID of CP and AA in brewer' s yeast [10] were within the ranges of those digestibility re ported in soybean meal [17]. The AID and SID of CP and almost all the AA in soybean meal in this study were close to the tabulated digestibility [17]. Compared with soybean meal, the digestibility of CP and AA were lower in whole yeast, but were close to those digestibility for autolyzed yeast. This result suggested that autolyzed yeast can be used as an alternative protein feedstuff for inclusion in pig diets. Therefore, the re duction in soybean imports could be partly compensated by 2) Each least squares mean represents 5 observations except for autolyzed yeast (6 observations). a-c Within a row, least squares mean with a common superscript are not different at p < 0.05.
using the autolyzed yeast derived from sugarcane ethanol production.

CONCLUSION
There were no significant differences in the chemical com position (except for NDF and ADF) between whole yeast and autolyzed yeast; however, whole yeast had low digest ibility of CP and AA due to the presence of a rigid cell wall and no digestive enzymes in the pigs. Autolysis could dam age the cell wall and release intracellular nutrient contents; thus, the AID and SID of CP and AA were greater in auto lyzed yeast than in whole yeast. The information obtained on the SID of CP and AA in both yeast products could be used to provide accurate estimates of the bioavailability of CP and AA in feed formulations. Therefore, yeast products derived from sugarcane ethanol production are suitable as an alternative protein source for pig diets. However, further research is needed to assess the effect of dietary supplemen tation with autolyzed yeast on growth performance, carcass quality and the gut microbiota composition of pigs.