Nutrient digestibility in broiler chickens fed diets containing high levels of soybean oil is affected by the source of fiber

Abstract This study evaluated the effects of including soluble and insoluble dietary fiber sources in soybean oil-containing diet for broiler chickens on coefficient of apparent ileal digestibility (AID) of amino acids, coefficient of apparent retention (AR) of nutrients, and nitrogen-corrected apparent metabolizable energy (AMEn). A total of 180 broilers were divided into 30 groups and fed 3 diets from day 14 to 21 of age. The diets were corn–soybean meal containing 6.8% soybean oil without or with soybean hulls (SBH) or sugar beet pulp (SBP) at 8.0% as sources of insoluble and soluble fiber, respectively. The SBH and SBP contained 6.0% and 14.7% of soluble dietary fiber and 60.5% and 37.0% of insoluble dietary fiber, respectively. The AID of indispensable amino acids was decreased (P < 0.05) due to dietary inclusion of SBH, but not of SBP. Dietary SBH and SBP reduced the AR of gross energy (by 14% and 8%, respectively) and AMEn (by 20% and 14%, respectively). Overall, inclusion of SBH or SBP at 8.0% in a corn–soybean meal-based diet for broilers that contains high level of soybean oil might not improve nutrient digestibility. However, SBP has less detrimental effects on nutrient digestibility than SBH.


Introduction
Fibrous feedstuffs are relatively inexpensive and available in large quantities in countries with agro-based economies (Abdel-Hafeez et al. 2018). Furthermore, the inclusion of moderate amounts of fiber in broiler diets results in improved gut health (Jiménez-Moreno et al. 2010) and digestive capacity (Jiménez-Moreno et al. 2010). However, fibrous feedstuffs have lower energy value compared with starchy feedstuffs such as cereal grains (Hervik and Svihus 2019). Thus, highfiber diets are supplemented with fat to meet dietary energy requirement (Bakker 1996). The effect of dietary fiber on the digestive capacity of broilers depends on the physicochemical characteristics of the fiber source. For instance, viscous soluble fiber decreased nutrient digestibility (Owusu-Asiedu et al. 2006), whereas non-viscous fiber did not reduce nutrient digestibility (Jiménez-Moreno et al. 2013). Also, oat hulls containing lignified insoluble fiber were more effective in improving digestive capacity of gizzard in broilers than sugar beet pulp (SBP) containing soluble fiber (Jiménez-Moreno et al. 2013), implying that the effectiveness of the fiber is dependent on the degree of insolubility and lignification.
Soybean is the major oilseed crop that is grown in the United States of America. Thus, soybean oil and soybean hulls (SBH), which are co-products from soybean processing plants, are available in large quantities for inclusion in livestock feeds. More than 70% of the fiber in SBH is insoluble (Lo 1989). Thus, SBH is a rich source of insoluble fiber. The SBP is one of the fibrous feedstuffs that has a high content of soluble fiber , which is most widely used to formulate animal feeds (Fishman et al. 2011). However, there is little information about the effect of combining soybean oil with SBH or SBP on the digestibility of nutrients and dietary energy value for poultry (Tejeda and Kim 2020). The SBH and SBP are less lignified than, for example, oat hulls, implying that relatively larger amounts are required to effectively stimulate the development of the digestive system. Unsaturated fatty acids reduce the flow of digesta in gastrointestinal tract (Omidi et al. 2020). Soybean oil has a high content of unsaturated fatty acids (Jiménez-Moreno et al. 2009). Thus, it was hypothesized that SBH or SBP in relatively larger amounts can complement soybean oil in improving nutrient digestibility due to reduced digesta passage rate, whereas the fibrous feedstuffs can increase digestive capacity. It was also hypothesized that SBH is more effective in shifting nutrient digestibility from hindgut fermentation toward ileal digestibility because of its relatively high content of insoluble fiber that increases the digestive capacity of the gizzard, whereas SBP is more effective in increasing energy retention because of its high fermentability due to its relatively high content of soluble fiber. The objective of this study was to investigate the effects of adding SBH or SBP to broiler diets containing a high level of soybean oil on apparent ileal digestibility (AID) and retention of nutrients, and nitrogen-corrected apparent metabolizable energy (AMEn) value.

Ethical approval
Experimental procedures were reviewed and approved by the Institutional Animal Care and Use Committee at South Dakota State University (18-087E), and the birds were handled in accordance with standards for animal care and use contained within the Guide for the Care and Use of Agricultural Animals in Agricultural Research and Teaching (FASS 2010).

Experimental diets
Three diets were fed to broilers in the present study. The diets included corn-soybean meal-soybean oil-based diet (control) without or with 80 g kg −1 of SBH or SBP (Table 1). The diets were formulated to meet the nutrient recommendations of the Ross 308 strain (Aviagen 2014) and contained titanium dioxide (3 g kg −1 ) as an indigestible marker. Both the SBH and SBP were sourced from South Dakota State University's Feed Mill (Brookings, SD, USA).
Birds, housing, and experimental procedure A total of 180 1-day-old male broiler chicks of Ross 308 strain were acquired from a commercial hatchery. The chicks were distributed to electrically heated Petersime battery brooders (Petersime Incubator Co., Gettysburg, OH, USA) so that each cage housed 14 birds or less. Room temperature was in accordance with the programs recommended for the Ross strain. Light was provided for 24 h daily throughout the experiment. Chicks were fed drug-free commercial starter diet (3047.67 kcal kg −1 metabolizable energy, 220 g kg −1 crude protein (CP), 10 g kg −1 Ca, and 4.5 g kg −1 non-phytate P) from day 1 to day 14. On day 14, the birds were distributed in 30 cages (6 birds/cage) and group weighed. The three experimental diets were randomly assigned to the cages in a completely randomized design (10 cages per diet) and fed from days 14 to 21 of age. Fresh water and the feed were offered ad libitum throughout the experiment. Body weight gain, feed intake, and feed conversion ratio were determined from days 14 to 21 of age. On days 19 and 20, excreta samples were collected from each cage and stored frozen at -20 • C for later laboratory analyses. On day 22, all the six birds from each treatment were euthanized via cervical dislocation, and the entire ileum content (between Meckel's diverticulum and 2 cm above the ileal-caecal junction) was squeezed out and stored frozen at -20 • C for later laboratory analyses.

Sample preparation and analyses
Daily excreta samples were pooled for each cage, and ovendried for 4 days at 60 • C. Ileal digesta samples were freezedried. The dried excreta and ileal digesta samples along with diet samples and feedstuffs were finely ground in a centrifugal mill (model ZM200; Retsch GmbH, Haan, Germany) through a 0.75 mm screen. All samples were analyzed for dry matter (DM) and CP (N × 6.25). Samples were further analyzed as follows: SBH and SBP for soluble dietary fiber (SDF) and insoluble dietary fiber (IDF); diet samples for organic matter (OM), gross energy (GE), amino acids (AAs), and titanium; ileal digesta samples for AAs and titanium; and excreta samples for OM, GE, and titanium.
Samples were analyzed for DM and CP according to AOAC (2007) methods 930.15 and 990.03, respectively; and for GE using an adiabatic bomb calorimeter (model 1261; Parr Instrument Co., Moline, IL, USA), and benzoic acid was used as a standard. The SDF and IDF contents in the sub-samples and the undigested residues were measured using the Megazyme  (Myers et al. 2004). The ash values of samples used for the analysis of titanium were used for the calculation of the OM content of the samples.

Calculation and statistical analysis
The coefficient of AID and coefficient of apparent retention (AR) of energy and nutrients were calculated using the indicator method (Stein et al. 2007) using the following equation: % Apparent nutrient digestibility or retention where N f is the nutrient concentration in the excreta or ileal digesta (% DM); N d is the nutrient concentration in the diet (% DM); M d is the titanium concentration in the diet (% DM); and M f is the titanium concentration in the excreta or ileal digesta (% DM). The AMEn value for diets was calculated as AME value (per gram diet) minus 8.22 multiplied by grams of N retained per gram diet according to Hill et al. (1960).

Statistical analysis
Data were subjected to analysis of variance using the MIXED procedure of SAS (version 9.4; SAS Institute Inc., Cary, NC, USA). The model used was as follows: where y i is the observed value for a particular trait, μ is the common mean, T i is the fixed effect of treatments (diets), and e i represents random error. Lsmeans were separated by least significant difference test. To test the hypotheses, P < 0.05 was considered significant, and tendencies were reported when 0.05 ≤ P ≤ 0.10. For all response variables, initial body weight was added as covariate; however, it was nonsignificant, and removed from the final model. c We use titanium oxide as the indigestible marker in our laboratory for digestibility studies. However, we had issues with the equipment for analyzing titanium when starting this study. Therefore, we included chromic oxide to have the option of analyzing for chromium in case the equipment for titanium analysis was not functional during our experiment.

Analyzed compositions of diets
The SBH had greater content of TDF and IDF than the SBP. However, the SBP had greater content of SDF than the SBH ( Table 2). The CP content of the control diet was slightly greater than that for the SBH diet while it was lower than that of the SBP diet (Table 3). The control diet had a higher content of AAs than the SBP diet or SBH diet. The SBP and SBH diets did not differ in AA composition.

Growth performance
Effects of dietary inclusion of SBH or SBP on growth performance of broilers are shown in Table 4. Initial body weight, final body weight, and feed intake were unaffected by dietary SBH or SBP. Body weight gain of birds fed SBH diet or SBP diet tended to be lower (P = 0.05) than that of the birds fed the control diet. Feed conversion ratio was depressed (P < 0.05) by dietary inclusion of SBH or SBP.

Apparent ileal digestibility of nutrients
Effects of dietary inclusion of SBH or SBP on AID of DM, OM, and AA are shown in Table 5. The AID of DM for SBH diet was lower (P < 0.05) than that for the control diet or SBP diet. Control diet and the SBP diet did not differ in the AID of DM. The AID of indispensable AAs (Arg, Gly, His, Ile, Leu, Lys, Phe, Ser, Thr, Try, and Val) was decreased (P < 0.05) by dietary inclusion of SBH. The AID of dispensable AAs (Ala, Asp, Cys, Glu, Pro, and Tyr) was also decreased (P < 0.05) by dietary inclusion of SBH. Dietary inclusion of SBP did not affect the AID of all AAs measured in this study except for Met. The AID of Met was lowered (P < 0.05) by dietary inclusion of SBH or SBP; however, the reduction in AID of Met was more (P < 0.05) pronounced when SBH was included in the diet than the SBP was included in the diet.

Apparent retention of nutrients
Effects of dietary inclusion of SBH and SBP on AR of GE and nutrients in broilers are shown in Table 6. Dietary inclusion of SBH or SBP reduced AR of DM, OM, and GE, and AMEn value (P < 0.05). The AR of DM, OM, and GE, and AMEn values for SBH-containing diet were lower (P < 0.05) than those for SBPcontaining diet. Dietary inclusion of SBH decreased (P < 0.05) AR of CP; however, dietary inclusion of SBP did not affect AR of CP.

Discussion
In the current study, nutrient digestibility in broilers fed diets containing relatively high amounts of soybean oil (∼7%) and fibrous feedstuffs (8% of SBH or SBP) was determined at 21 days of age. It should be noted that development of digestive enzymes in gastrointestinal tract of broilers increases from hatch up to around 14 days of age (Nitsan et al. 1991;Sell et al. 1991). Noy and Sklan (1995) reported that lipase secretion in small intestine of broilers was low at hatch, but increased 20-fold between days 4 and 21 post-hatch. The digestibility of AAs (Barua et al. 2021) and ether extract (Tancharoenrat et al. 2013) increased with an increase in age of broilers from hatch up to around 21 days of age and did not further increase  with increase in age to days 35 or 42 post-hatch. Also, an increase in dietary level of ether extract from 3.32% to 8.21% did not affect apparent total tract digestibility of ether extract in broilers at 40 days of age (Attia et al. 2021). Thus, it is highly unlikely that the age at which nutrient digestibility measurement was done and level of fat in diets fed in the current study limited the digestibility of fat and other nutrients. Thus, the results of the present study can be gen-eralized to broilers at age higher than 21 days. In addition, fiber can improve digestive capacity of gizzard in broilers by increasing the flow of digestive enzymes and acids in the gizzard and by increasing the retention of feed in the gizzard (Mateos et al. 2012). Fiber is indigestible by endogenous enzymes of poultry, and its fermentation in the upper part of the gastrointestinal tract (crop, proventriculus, and gizzard) is expected to be so minimal, implying that the amount of  dietary fiber that reaches the gizzard, and hence the effects of dietary fiber on enzyme and acid secretions and retention of feed in the gizzard, is unlikely to be affected by the age of the birds. Therefore, it is also highly unlikely the age of the birds at which nutrient digestibility measurement was done in the current study limited the effects of fiber on nutrient digestibility. The SBP and SBH are co-products of sugar beet and soybean processing plants. Nutrient composition of co-products is affected by several factors including species, processing conditions, and type of components extracted or removed from the parent raw materials . Therefore, the differences between SBH and SBP regarding fiber content and fiber solubility are attributed to the differences in the aforementioned factors.
The decrease in body weight gain of broilers due to dietary inclusion of SBH or SBP could partly be attributed to reduction in AMEn and digestible AA contents of diets due to the dietary inclusion of the fibrous feedstuffs as evidenced by greater AA values for control diet than for diets containing SBH or SBP. The SBH and SBP have lower digestible energy and AA contents than corn and soybean meal that they replaced in the control diet. The presence of larger amounts of dietary fiber (soluble or insoluble) in the gastrointestinal tract increases the size of the gizzard and intestines to compensate physical distension caused by the swelling and increase in the bulk of the digesta (Jiménez-Moreno et al. 2010). Several studies have indeed shown increased gizzard size due to inclusion of fibers in the diet. For instance, Sacranie et al. (2012) reported that adding 15% coarse hulls (consisting of equal weights of hulls from oat and barley) to the diet of broiler chickens led to an increase in gizzard weight. Jiménez-Moreno et al. (2009) also observed that the inclusion of 3% oat hulls or SBP increased gizzard weight in broilers. An increase in the size of gizzard of broiler chickens fed diet containing wood shavings has also been reported in the study of Amerah et al. (2009). These changes can result in an increase in energy requirement for tissue synthesis in gastrointestinal tract, leading to increased dietary energy requirement for maintenance at the expense of growth (Nyachoti et al. 2000). Thus, the decrease in body weight gain of broilers due to dietary inclusion of SBH or SBP could also partly be attributed to an increase in utilization of dietary energy for maintenance at expense of body weight gain. In contrast to the results of the present study, Pettersson and Razdan (1993) reported that feeding broilers with diets containing SBP at 2.3% increased body weight and feed intake at days 14 and 21 of age. Also, Ndou (2018) observed that feeding diets containing oat hulls increased body weight gain and feed intake of pigs. In agreement with our results, Tejeda and Kim (2020) reported that inclusion of 6.06% purified cellulose or 19.33% SBH had adverse effects on body weight gain and feed conversion ratio of broilers on 20 days of age. However, it should be noted that in the current study, birds were fed experimental diets for a short period of time (7 days) because the main of objective of the study was to determine the effects of dietary SBH and SBP on energy and nutrient digestibility. The feeding period may not have been long enough for establishing the actual effects of SBH and SBP on growth performance of broilers fed diets that contain relatively high level of soybean oil.
Inclusion of fat with high content of unsaturated fatty acids in diets for poultry reduces passage rate of digesta in the small intestine (Omidi et al. 2020), which results in increased nutrient digestibility (Mateos and Sell 1980). The IDF such as that present in SBH can stimulate the development of the digestive system and thereby increase the digestive capacity of poultry by stimulating the development and function of the gizzard (Hetland et al. 2005), and by increasing the secretion of HCl, bile acids, and digestive enzymes (Mateos et al. 2012), which leads to improved nutrient digestibility (Scapini et al. 2018). The SDF has higher water-holding capacity in digestive tract, and hence SDF can increase digesta bulkiness, the size of the digestive tract, and time of retention of digesta in the gizzard (Jiménez-Moreno et al. 2010). The SDFs such as β-glucans that are present in oat can increase both digesta bulkiness and viscosity; an increase in digesta viscosity leads to reduction in nutrient digestibility (Hooda et al. 2011). However, SDFs present in SBP can increase digesta bulkiness, but with minimal effects on digesta viscosity (Flis et al. 2017). Thus, it had been hypothesized that the inclusion of SBH or SBP in the diet that contains a relatively higher level of soybean oil would result in increased ileal digestibility of AAs and OM due to increased digestive capacity. However, in the current study, dietary SBH or SBP did not increase AA and OM digestibilities. It should be noted that the digestibility of CP and AAs in SBH or SBP is expected to be lower than that of corn or soybean meal, which were partially replaced with SBH or SBP in the control diet. Thus, the failure of dietary SBH or SBP to the increase AID of AAs in the current study could be partly attributed to the lower AA digestibility of SBH or SBP and the relatively higher level of inclusion of these feedstuffs in the diets. Also, the failure of dietary SBH or SBP to increase AID of AAs could be attributed to the fact that large amount of dietary fiber results in increased abrasion of intestinal mucosal cells (Bergner et al. 1975), leading to increased secretion of mucin in intestinal lumen (Hedemann et al. 2006). Mucin is composed of AAs such as serine, threonine, and cysteine (Saadatmand et al. 2019) and constitutes a high proportion of endogenous protein at the terminal ileum (Li et al. 1994). Another acceptable explanation can be that non-starch polysaccharides are substrates for microbial activity that lead to deconjugation of bile acids (Ndou 2018). The deconjugated glycine can escape re-absorption and enter the large intestine (Li et al. 1994). Finally, the failure of dietary SBH or SBP to increase the AID of AAs could be attributed to the adsorption of dietary and endogenous proteins by the fiber, thereby reducing AA digestion and absorption (Bergner et al. 1975). Results from in vitro studies demonstrated that fiber can adsorb proteolytic enzymes, and thereby reduce the activities of these enzymes (Schneeman 1978).
In contrast to our findings, Jiménez-Moreno et al. (2013) reported improved AID of CP and starch due to inclusion of oat hulls in the diet of broilers at 25, 50, or 75 g kg −1 . The results from the current study are also in contrast to those from the study of Tejeda and Kim (2020), who reported that inclusion of 127.2 or 193.3 g kg −1 SBH in the diet of broilers increased digestibility of most AAs. It should be noted that in the study of Tejeda and Kim (2020), the increasing inclusion of SBH in the diets was accommodated by reducing corn, soybean meal, and sand, and increasing soybean oil and synthetic AAs such as lysine. In the current study, the inclusion of SBH in the diet was accommodated only by reducing corn and soybean meal. As previously mentioned, the ileal digestibility of AAs in pigs and poultry was increased by the dietary inclusion of oils (such as soybean oil) that are rich in unsaturated fatty acids. Also, the inclusion of synthetic Lys in diets for broilers resulted in increased ileal digestibility of other AAs (Selle et al. 2007). Sand is inert (indigestible); thus, the higher AA digestibility in broilers due to dietary inclusion of SBH in the study of Tejeda and Kim (2020) could partly have been due to the higher content of soybean oil and synthetic AAs and lower content of sand in the SBH-containing diets than in the control diet.
The decrease in AID of AAs due to dietary inclusion of SBH, but not SBP, could partly be due to higher lignin content in the former than in the latter. Nutrient digestibility of feedstuffs is negatively related to the degree of lignification of the same feedstuffs (Van Soest 1994). It could also be attributed to the differences between insoluble and soluble fibers with regard to inducing ileal endogenous N losses. The IDF is more abrasive than SDF, and hence the former compared with the latter scraps more mucin from intestinal mucosa during its passage in the gastrointestinal tract (Montagne et al. 2003).
Furthermore, IDF increases intestinal passage rate (Vahouny and Cassidy 1985), which limits nutrient digestibility (Mroz et al. 1986). The lack of effect of dietary inclusion of SBP on AID of DM could partly be attributed to the lack of effect of the same feedstuff on AA digestibility, whereas the decrease in AID of DM due to dietary inclusion of SBH could partly be attributed to the decrease in AID of AAs due to dietary inclusion of the same feedstuff.
Results from previous studies demonstrated increased AR of nutrient and energy in broilers due to dietary inclusion of oat hulls or SBP (Jiménez-Moreno et al. 2009). Also, Gonzalez-Alvarado et al. (2007) observed that total tract AR of DM, OM, and ether extract, and AMEn, increased when 30 g kg −1 of oat hulls or soy hulls were included in broilers, which were contrary to the results from the current study in which dietary SBH or SBP decreased AR of CP and GE. The reduction in AR of OM, GE, and CP, and hence dietary AMEn value, by the dietary SBH could partly have been due to reduction in AID of DM and AAs by the inclusion of the same feedstuff in the diet. It could also have been due to the lower fermentability of fiber in SBH than in soybean meal. The cell wall of endosperm (that is present in soybean meal) is thinner and it is fermented faster than the cell wall of SBH (van Laar et al. 1999). It is not clear why dietary SBP reduced AR of DM, OM, and GE since it did not significantly reduce the AID of DM and OM. It is known that fiber is less digestible in the small intestine than starch and AAs, which are the major components of corn and SBM. The SBP has a relatively high content of SDF, which is highly fermentable in the hindgut. Thus, the amount of reduction in dietary energy or nutrient digestibility due to dietary inclusion of SBP is expected to be greater at the ileal level than at the total tract level. It is also known that an increase in the availability of fermentable fiber in the hindgut can result in increased excretion of N via feces at the expense of the excretion via urine, and that N that is excreted via urine is generally more volatile than N that is excreted via feces. Thus, a relatively higher proportion of N that is excreted via urine can be lost during sample processing (e.g., drying) and analysis, leading to overstimulation of AR of N and hence AR of DM and GE for diets with less fermentable fiber content. However, the AR of N for the SBP did not differ from that of the control diet. Thus, it appears that fiber and other components of SBP that escape digestion in the small intestine are less fermentable in the hindgut of broilers than fiber and other components of corn and soybean meal that escape digestion in the small intestine; or dietary SBP increases excretion of N and other energy-yielding components via urine, leading to reduced AR of GE and CP. However, the exact mechanisms by which dietary SBP could reduce AR of OM, GE, and CP without affecting the AID of DM, OM, or CP ought to be established. The greater AR of GE and CP, and AMEn values, for the SBP diet than for the SBH diet could be attributed to the higher SDF content in SBP than in SBH. SDF is more fermentable than insoluble fiber (Agyekum and Nyachoti 2017). Furthermore, the fiber in SBH is more lignified than fiber in SBP--fiber fermentability is negatively correlated with its degree of lignification (Van Soest 1994).

Conclusions
In conclusion, SBH negatively affected AA digestibility and AMEn values, whereas SBP only negatively affected dietary AMEn value. The magnitude by which dietary SBP reduced dietary AMEn value was less than that by which dietary SBH reduced dietary AMEn value. Thus, inclusion of SBH or SBP at a relatively high level (80 g kg −1 ) in the corn-soybean mealbased diet for broilers that contains relatively high levels of soybean oil (70 g kg −1 ) might not improve nutrient digestibility. However, SBP compared with SBH has less detrimental effects on nutrient digestibility when it is included at the high level in the corn-soybean meal-based diet for broilers that contains high level of soybean oil.