Use of Soy Protein Concentrate in Pre-Starter and Starter Diets for Broilers

Two experiments were carried out to evaluate the effect of using soy protein concentrate (SPC) in pre-starter and starter diets for broilers. In the first experiment, 600 male Cobb broilers, between one and 40 days of age, were distributed in a completely randomized design, with four treatments and six replications of 25 birds each. Treatments were ofered to broilers in the pre-starter and starter diets and consisted of inclusion of soy protein concentrate (0,3,6 and 9%) in diets. The parameters evaluated were: body weight gain, feed conversion ratio, consumption of ration, enzyme production in the pancreas, villus: crypt ratio, leukocyte count and immunoglobulin A (IgA) dosage. Aimed to determine the coefficient of nutrient metabolization of feeds, 144 male Cobb chicks were distributed, between 14 and 21 days of age, with four treatments and six replications of six birds per experimental unit. Treatments were the same as in the first experiment. The use of 3 and 9% of SPC did not affect weight gain, feed intake, feed conversion or viability of the poultry. The use of 6% of SPC provided an increase in trypsin activity, villi length and crypt depth; the coefficient of dry matter metabolization increased linearly; but no differences were observed in performance or immunological parameters.


INTRODUCTION
Some of the most relevant anti-nutritional factors (ANF) present in soybean meal that affects the development and performance of monogastrics include α-galactosides -raffinose, stachyose and verbascose -and antigenic factors -glycinin and β-conglycinin (Nunes et al., 2001;Jankowski et al., 2009).The presence of non-digestible glycoside substrate in the intestinal lumen involves changes in chyme viscosity, compromising the integrity of gut mucus and increasing the passage rate, reducing the efficiency of nutrient absorption (Góes & Ribeiro, 2002;Batal & Parsons, 2003).Enzymatic production and morphological structure of the digestive system of poultry, change significantly with age and frequency of feed intake (Souza et al., 2005).Allergic factors cause hypersensitivity reactions, promoting a shortening of the intestinal villus and decreasing the ratio between the villi length and crypt depth (Thomaz et al., 2011).The presence of ANF in the diets of broilers from one to 21 days of age implies a reduction in feed intake, impaired digestion and utilization of nutrients (Feng et al., 2007;Kim et al., 2010).
Soy protein concentrate is an ingredient that may be used in poultry diets in replacement of soybean meal, as it contains certain benefits due to its processing: lower levels of oligosaccharides such as raffinose, stachyose and verbascose, higher crude protein, digestible and metabolizable energy compared to soybean meal.Scottá et al. (2013), verified that the average true amino acids digestibility coefficient, of essential and non essential SPC for broilers, were respectively 95,21% e and 94,22% for the soy protein concentrate.Bansemer et al. (2015) verifyed that inclusion of SPC in fish diets, decreased globet cell numbers.This suggests SPC having an effect on reducing mucus production in the intestine.
This study was carried out to evaluate the effect of soy protein concentrate inclusion in pre-starter and starter diets of broilers on: performance, enzyme production in the pancreas, gut integrity, nutrient metabolization and immunological parameters.

Animals and housing
The experimental protocol was submitted to the Ethics Committee for Use of Animals in Research (CEUA -Universidade Federal de Goiás) and approved under number 066/12.
Seven hundred forty-four (744) one-day-old male Cobb broilers were used in the two experiments.
For experiment 1, six hundred (600) one-day-old male Cobb broilers were distributed in a completely randomized design with four treatments (0, 3, 6 and 9% of soy protein concentrate -Table 1 e and 2), and six replications of 25 birds/experimental unit.At 22 days old, the broilers received the same diet, without soy protein concentrate (Table 3).
The birds were housed in a shed with 2.45 m headroom, inside 2.02 x 1.36 x 0.73 m boxes, provided with bed of rice husks (5 cm), tubular feeder and pendulum drinker.Environmental control consisted of heating hoods, negative ventilation system, nebulizer, side curtains, two thermometers, a moisture sealer and 23 hours of light/day.The temperature was recorded in the morning (8 a.m.) and mortality 2x/day (8 a.m. and 4 p.m.).In experiment 1 the variables studied were: body weight gain, feed intake, feed conversion ratio and viability, leukocyte count and immunoglobulin A (IgA) dosage, pancreas weight, amylase and trypsin activity, villi length and crypt depth of small intestine.
One hundred forty-four (144) male Cobb broilers, 14 to 21 days of age, were distributed in a completely randomized design with four treatments, six replications of six birds/experimental unit.Treatments consisted of four levels of inclusion of soy protein concentrate (0, 3, 6 and 9%) in the diet.This group of 144 broilers was housed and managed under the same condition of experiment 1 and received a basal diet until 14 days of age, without soy protein concentrate, according to recommendations of Rostagno et al. (2011).At 14 days of age, the birds were housed in metabolic cages and the experimental diets were administered (Table 2).In experiment 2 the coefficients of metabolizable dry matter, crude protein, ether extract and ash were calculated.The birds were housed in metabolic cages of 0.77 x 0.74 x 0.23 m in size, fitted with trays for excreta collection, food and water trough type; in a shed 2.00 m in height, provided with heating hoods, side curtains, two thermometers and 23 hours of light/day.The average temperatures (minimum and maximum) recorded during the experimental period were 23 and 31°C in Experiment 1; 24 and 28°C in Experiment 2, respectively.

Diets and feeding
Treatments consisted of four inclusion levels of soy protein concentrate (0, 3, 6 and 9%) in pre-starter and starter diets.
The experimental diets were iso-nutrient and isoenergetic and were formulated according to Rostagno et al. (2011) recommendations, and followed a feeding program divided into pre-starter phase (1-7 days of age) and starter phase (8-21 days of age) (Tables 1  and 2).The birds received the same diet from 22 to 40 days of age (Table 3).Feed and water were provided ad libitum.

Performance analysis
Body weight gain, feed intake, feed conversion ratio and viability of birds were calculated according to Sakomura & Rostagno (2007).Weights were taken on the 1 st , 7 th, 21 st and 40 th days of the experiment.

Data collection
For leukocyte count and immunoglobulin A (IgA) dosage, samples were obtained from two birds per replication at 21 days of age.Blood samples were collected in a tube with heparin or without anticoagulante (respectively for WBC and IgA), from the femoral vein; for serum, the samples were centrifuged at 5000 rpm for three minutes and frozen at -20° C. The total leukocyte count followed the protocol set by Natt & Herrick (1952); the specific leukocyte count was obtained as per Garcia-Navarro (2005).The serum IgA These birds were euthanized by cervical dislocation for collection of the pancreas and small intestine.
After weighing, the pancreas was frozen in liquid nitrogen, homogenized on ice and the supernatant was extracted to measure total protein (Bradford, 1976), trypsin (Kunitz, 1947) and amylase content (CNPG amylase kit, Labtest ® ).
The pancreas and intestines were collected and weighing in analytical balance with three decimal places of accuracy.The pancreas was frozen in liquid nitrogen and homogenized on ice; the supernatant was extracted to measure total protein (Bradford, 1976), trypsin (Kunitz,1947) and amylase content (CNPG amylase kit, Labtest.).Enzyme assays were performed in the enzymology laboratory and Physiology of Digestion, the Institute of Biological Sciences (ICB II), Federal University of Goias.
The small intestine was cut between the proventriculus and the cecum-colic junction, and representative sample from its different anatomical regions was collected (duodenum -bounded by the contour of the pancreas, jejunum -near the Meckel diverticulum and ileum -near the cecumcolic junction).Tissue samples were fixed in 10% neutral buffered formalin for histological processing (Macari et al., 2002), for 24 hours and then processed for paraffin embedding according to the routine protocols; 5 micron section were then stained with Harris haematoxylin and eosin, the slides mounted in Entellan.Microphotographs were collected using a magnification of 50x in a Leica DM2500 optic microscope, and retrospectively analyzed in order to obtain data corresponding to 30 readings of villus and crypts per intestinal segment.

Excreta collection
The birds were acquired from the same batch of experiment 1.By 14 days of age were managed according to the breed manual, and at 14 days of age were allocated in metabolic cages for experimental period.Both experiments occur simultaneously.
The experimental period was eight days -four days for adaptation to the cages and experimental diets, and four days to excreta collection.
Excreta were homogenized and a sample of 500g each replication collected and identified.Samples were pre-dried in a forced ventilation oven at 65°C for 72 hours.The air dried percentage of excreta was determined as a relation of the weight after and before drying.After this step, the samples were ground in a Wiley mill, and stored in labeled plastic bags.Analyzes of dry matter were held, nitrogen, lipids and ashes according to the methodology described in Silva & Queiroz (2009).In order to calculate the metabolizable coefficients of dry matter, crude protein, ether extract and ashes, the following equations were used: CMDM= Dry matter intake (g) -Dry matter excreted (g)/ Dry matter intake (g) x 100 CMCP = Nitrogen intake (g) -Nitrogen excreted (g) / Nitrogen intake (g) x 100 CMEE = Ether extract intake (g) -Ether extract excreted (g) / Ether extract intake (g) x100 CM Ash% = ash intake (g) -ash excreted (g) / ash intake (g) x 100

Statistical analysis
The results were analyzed by ANOVA.Variables related to performance, gut integrity and enzymology had their means compared by the Scott-Knott test.The coefficients of metabolizability underwent polynomial regression.Hematological parameters were subjected to the Kruskal-Wallis test, with the difference between treatments assessed by Bonferroni test.We used R software (R Development Core Team, 2011).We adopted α = 0.05.

RESULTS AND DISCUSSION
Inclusion of increased levels of SPC in pre-starter and starter diets did not affect the final body weight, body weight gain, feed intake, feed conversion ratio and viability of broilers (Table 4), suggesting that it did not influence broilers performance in any of the rearing periods surveyed (1 to 7, 1 to 21 and 1 to 40 days of age).
No effect of SPC inclusion on broiler performance in any period studied (1 to 7, 1 to 21 and 1 to 40 days of age) was observed.Reduction in anti-nutritional factors does not appear to be the determining factor for improved performance.The results disagreed with Trindade Neto et al. (2007), who reported low feed intake associated with the presence of trypsin inhibitors, and with Thomas et al. (2011) who associated reduced growth with digestive disorders resulting from transient hypersensitivity reaction caused by glycinin and β-conglycinin allergenic proteins.Siugzdaite et al. (2008) found significant improvements in the performance of piglets with 10% inclusion of SPC in the weaning diet; Lenehan et al. (2007), with 14-21% inclusion of SPC; Bertol et al. (2001) with 50% replacement of soybean meal by SPC in the nursery phase of pigs diet.These reports suggest that levels of inclusion of SPC proposed in the present study may not have been sufficient to demonstrate improvements in performance.
The inclusion of 3 to 9 % of SPC in the pre-starter and starter diets showed no decrease in pancreatic weight (Table 5), although higher activities of trypsin and amylase were recorded with the use of 3, 6 and 9% of SPC, respectively.Li et al. (1991a and1991b) hypothesized that the reduction in anti-nutritional factors, provided by the use of SPC, deletes their antigenic power, minimizing the transient hypersensitivity and improving growth.These reports, combined with the results obtained herein, corroborate the theory that the levels proposed in the study were not sufficient to demonstrate improvements in performance.
Diets without SPC resulted in increased small intestine weight and higher villus, crypt ratio in the jejunum (Table 6).The treatments with 6 and 9% inclusion of SPC provided greater villi length and crypt depth in the duodenum, jejunum and ileum, indicating greater absorption area which resulted in better utilization of nutrients in the diets.
There are very few reports on the effects of glycinin and β-conglycinin on the mucous secretion and moisture in the excreta of birds, which does not make it possible to discard them as predisposing factors to inflammatory reactions observed in broilers between seven and ten days of age (Ortiz, 2009).Cortés (2012) tested the effect of a mono competent protease on anti-nutritional factors of soybean, and a reduction in the presence of these factors was proven.A linear relationship was observed for villi length and a cubic relationship to crypt depth, with increased surface area for absorption and villus: crypt ratio at 14 days of age.This author drew attention to the fact that the reduction of the stimulus on the pancreatic secretion and increasing the absorptive surface of the intestine brought about a shift in nutrient utilization for performance and carcass yield with lower levels of fat.
The use of 3 to 9% SPC did not affect leukocyte count and the dosage of immunoglobulin A in blood (Table 7).Based on these results, it can be inferred that the reduction of allergenic dietary factors was not enough, which explains the lack of difference in performance and in pancreatic weight, despite increased CMDM.
In Brazil there is paucity of data on reference levels for hematological and biochemical values in broiler chickens, showing the importance of studies that include such evaluations in several experimental situations (Minafra, 2010).Furthermore, the method commonly used, immune histochemistry, enables the determination of dimeric IgA and observation of leukocytes in slide, contributing to greater accuracy in the discussion about the relevance of antigenic factors in catabolic processes associated with the immune response.In the present study the coefficient of metabolization of dry matter increased linearly and the coefficient of metabolization of ash decreased linearly according to higher SPC inclusion in diets (Table 8).According to Batal & Parsons (2003), the increase in the use of nutrients in SPC is due to the removal of soluble nonstarch oligosaccharides.
Soluble non-starch oligosaccharides negatively interfere with the absorption of minerals (Arruda, 2003).Although SPC has a higher concentration of minerals due to processing (Miranda, 2012), the metabolization of nutrients decrease in the content of α-galactosides, the excretion of minerals (Fischer et al., 2002) causing CMA increase -contradicting the results.
The coefficients of metabolization of crude protein and ether extract were not affected by using SPC.Pinheiro et al. (2008), comparing the metabolization of different soybean subproducts also found no significant differences for the CMCP and CMEE diets low in fiber.Methods that allow the exclusion of nitrogen from uric acid in the calculation of the CMCP and to enable inclusion of the percentage of complex lost minerals with organic matter in the calculation of CMA, contribute to a greater accuracy in the analysis of results.
In conclusion, the inclusion of 6% of soy protein concentrate increased trypsin activity, the villi length and crypt depth in the small intestine, suggesting an improvement in the process of digestion and nutrient absorption, yet insufficient to show an increase in performance parameters.Similar to what occurs with other species (pigs, cattle and fish), studies using SPC for broilers from one to 21 days of age should be encouraged.The use of soy protein concentrate does not affect broiler performance and can be used in prestarter and starter diets until 9%.

Table 1 -
Composition of experimental diets of pre-starter phase (1-7 days old)

Table 4 -
Performance of broilers from one to 40 days, fed increasing levels of SPC in the pre-starter and starter diets (Mean ±SEM)* LF = lack of fit; CV = coefficient of variation, SEM = standard error of mean *mean of 6 replicates, with 25 broilers/replicate.Vasconcelos LG,

Table 5 -
Weight and enzymology of the pancreas of broilers at 21 days of age, fed increasing levels of soy protein concentrate in the pre-starter and starter diets (Mean ±SEM)

Table 6 -
Weight and intestinal histomorphometry of broilers from 21 days of age, fed increasing levels of soy protein concentrate in the pre-starter and starter diets (Mean ±SEM)

Table 7 -
Leukocyte count (total and specific) and dosage of immunoglobulin A in broilers at 21 days of age, fed increasing levels of Soy protein concentrate in the pre-starter and starter diets (Mean ±SEM)