Effects of Different Dietary Levels of Two Types of Olive Pulp and Exogenous Enzyme Supplementation on the Gastrointestinal tract size, Immunology and Hematology of Broilers

I Department of Animal Science, Rasht Branch, Islamic Azad University, Rasht, 49147-39975, Iran. E-mail: p.sayeh@ yahoo.com, alirezaseidavi@iaurasht.ac.ir. Iran corresponding authors. II Department of Veterinary Science, Rasht Branch, Islamic Azad University, Rasht, 49147-39975, Iran. III Agricultural and Natural Resources Research Center of Guilan, Rasht, 4914739975, Iran. IV Federal Institute of Education, Science and Technology of Northern Minas Gerais, 39480-000, Brazil. E-mail: aziszoo@yahoo. com.br. Brazil corresponding author. V Department of Animal Science, Federal University of Viçosa, 36570-000, Brazil.


INTRODUCTION
Olive pulp (OP) is the dried residue of olive cake (the raw material resulting from the extraction of olive oil).OP is also a good source of several biologically active compounds, and has antioxidant, antifungal, and antibacterial properties (Al-Harthi, 2014).OP is considered a good source of protein, fat, calcium, copper and cobalt; however, its has nutritional value is poor due its low energy, digestible protein, and mineral content, and high lignin content.It is also poor in phosphorus, magnesium, and sodium, but has fair levels of manganese and zinc (Afsari et al., 2013).
In the past, the utilization of crop residues and by-products as alternatives to cereal-soybean meal based diets was not successful, mainly due to their high fiber content and poor digestibility.Exogenous enzymes may be added to broiler diets containing these by-products to aid fiber digestion (carbohydrases) or to solubilize phytic phosphorus (phytase), thereby reducing their negative effects on broiler performance (Choct, 2006).
The ripening stage of the olives at harvest affects the levels of pectic polysaccharides found in the olive cell walls due to the presence of calcium chelating dimers, thereby changing the nutritional value of this by-product (Cardoso et al., 2007).Most of those compounds are effective antioxidants.According to Kidd (2004), antioxidant substances are capable of reducing cellular free radical damage and enhance the immune response of broilers.

Effects of Different Dietary Levels of Two Types of Olive Pulp and Exogenous Enzyme Supplementation on the Gastrointestinal tract size, Immunology and Hematology of Broilers
The occurrence of hydrophilic phenols in virgin olive oil is strictly related to the activities of various endogenous enzymes of olive fruits, as their concentrations in the oil are strongly affected by extraction conditions (Yorulmaz et al., 2011).Crushing and malaxation are the most important critical points of the oil mechanical extraction process because the desirable compounds in OP residue can be inactivated by the crushing process (Clodoveo, 2012).The destoning processes are important to maintain the properties of the OP due to the removal of the enzymes that are contained in the seeds (stones).Lavelli and Bondesan (2005) observed an increase in the total secoiridoid polyphenol (anti-oxidant, anti-microbial, and anti-inflammatory compound) content and in the antioxidant activity of extra virgin olive oils when the fruits were pre-destoned.
OP is rich in fibers, and these contents change as a function of processing.It is known that digestion is affected by the physical and chemical characteristics of the feed (Le Goff and Noblet, 2001).In non-ruminants, the presence of fiber and its different characteristics, dependent of the source, is highly important for gastrointestinal (GIT) development (Macari, 2008).
Studies evaluating diets with olive pulp (destoned, or not) for commercial broilers are scarce.The objective of this experiment was to determine the effect of different dietary levels of processed and unprocessed olive pulp, with enzyme supplementation, on the gastrointestinal tract size, immune and hematological parameters of broiler chickens.

MATERIALS AND METHODS
Six hundred one-day-old Ross 308 male broilers were housed in cages (1.25 × 1.25m).The floor area provided per bird was 0.15 m 2 .Cages were located in a poultry house with thermostatically-controlled side curtains.The cage floor was covered with paper litter, and the birds remained in the cages for the duration of the experiment, which ended at 42 days of age.Each cage of 10 chickens was assigned to a specific dietary treatment group.The diets (Tables 2 and 3) met or exceeded the recommendations of Ross 308 manual (Aviagen, 2009).
The OP was obtained by washing fresh olive fruit with water.Olives were then milled, placed in hot water (80°C) and centrifuged.At this stage, after "water + oil" emulsion was extracted removed from the olives, and the remaining residue was designated as "olive cake" (OC).In the next step, alpha-tocopherol (anti-oxidant) and an anti-fungal toxin-binder (adsorbent) were added to the OC.The OC was then dried at 70° Cusing hot air, resulting in OP.Olive processing consisted of passing the fruits through a sieve (1.5 mm mesh diameter).During this process, part of the stones (seeds) was removed to produce "partly destoned OP".Olive pulp (processed and unprocessed) chemical composition was determined according to the AOAC (1990).

Gastrointestinal traits
On day 42, birds were fasted for 4 h for the complete evacuation of the gut, and three birds per replicate were selected and euthanized by cervical dislocation.The average obtained by the three birds for each carcass and gastrointestinal tract parameter was calculated and considered as one experimental unit.Birds were plucked using the dry plucking method.Carcasses were weighed before and after evisceration.The following gastrointestinal parts were dissected and collected: crop, proventriculus, pancreas, duodenum, ileum, jejunum, left and right ceca, colon, thymus, liver, spleen, bursa of fabricius, brain, kidneys, and testes.The jejunum and the ileum were identified as the intestinal segments cranial and caudal to Meckel's diverticulum, respectively.All dissected organs were also weighed.The length (cm), width (mm), and wall thickness (mm) of the duodenum, ileum, jejunum, left and right ceca, and colon were recorded.All dimensions were rounded to integers.Organ width was measured when organs were completely empty and flattened.The relative weight of all dissected segments of the digestive tract was calculated as a percentage of the completely eviscerated carcass, according to the following formula: [(weight of the component(s)/ eviscerated carcass weight)×100].
Data were submitted to analysis of variance (ANOVA) using a two-way ANOVA procedure (SAS Institute, Inc., 2000), according to the following model: Y ijk = μ+A i +B j +C k +AB ij +AC ik +BC jk +ABC ijk +e ijkl ; where μ = general mean, A i = effect of olive pulp levels, B j = effect of oliver pulp processing, C k = effect of the enzyme inclusion, AB ij = effect of the interaction between olive pulp level and processing, AC ik = effect of the interaction between olive pulp level and enzyme complex inclusion, BC jk = effect of the interaction between olive pulp processing and enzyme inclusion, ABC ijk = effect of the interaction among olive pulp level, processing, and enzyme inclusion, and e ijkl = random residual effect of observation.
After statistical differences were confirmed, the General Linear Model (PROC GLM) was used, and the differences among means (p≤0.05) were evaluated via Duncan's multiple range test (SAS, 2000).

Hematological parameters
Before blood collection, birds were feed-fasted for 4 h in an attempt to allow for the stabilization of the various plasma constituents.Blood was collected in the morning to further reduce the variability of the measured plasma constituents.At 42 days of age, 5 mL venous blood was collected from the ulnaris vein in the wing of one randomly selected bird per replicate.The whole blood sample was transferred from the syringe into a tube coated with 10 mg of the anticoagulant ethylene diamine tetra acetic acid (EDTA).Blood samples were centrifuged at 3000 rpm for 20 minutes to assure separation of the blood cells from the plasma.Plasma was collected and stored at -20°C until plasma constituent analyses were performed according to standard protocols (Weiss and Wardrop, 2010).The levels of the following hematological parameters were determined: uric acid, total cholesterol, triglycerides, very low density lipoprotein (VLDL), high density lipoprotein (HDL), low density lipoprotein (LDL), LDL/ HDL ratio, aspartate amino transferase (AST), alanine amino transferase (ALT), total protein, albumin, and globulin.

Immunological parameters
A vaccination program against avian influenza virus (1 day of age), infectious bronchitis virus (1 and 18 days of age), Gumboro virus (16 day of age), and Newcastle disease virus (1, 6, and 20 days of age) was applied.
Humoral immune response to the Newcastle vaccine was measured in blood samples collected on days 7 and 21 by antibody titering by hemagglutinationinhibition test (HI).Humoral immune response to the avian influenza vaccine was measured in blood samples collected on days 21 and 28 by antibody titering by HI.Humoral immune response to the infectious bronchitis vaccine was measured in blood samples collected on day 25 by antibody titering by ELISA.
Primary immune response was determined in two birds per replicate injected with sheep red blood cells (SRBC) on day 14.Ten mL SRBC were thoroughly mixed with 1 mL phosphate buffered saline (PBS) solution, and the 0.5 mL of the obtained solution was drawn into a syringe and injected under the breast skin.Blood samples were collected in days 24 and 33 to measure the immune response against SRBC.

Gastrointestinal traits
No significant interactions were observed between olive pulp (OP) level and enzyme supplementation, OP level and pulp processing, or enzyme supplementation and pulp processing for the evaluated gut parameters (p>0.05,Table 4, 5 and 6).Crop weight, relative crop weight (CP%), proventriculus weight, and relative proventriculus weight (PV%) were not influenced Ghorbani A, Araújo WAG, Albino LFT

Effects of Different Dietary Levels of Two Types of Olive Pulp and Exogenous Enzyme Supplementation on the Gastrointestinal tract size, Immunology and Hematology of Broilers
by OP dietary levels (p>0.05,Table 4); however, the inclusion of 100g OP/kg increased jejunum relative weight (J%) and length (p≤ 0.05, Table 4).This indicates that up 100 g OP/kg can be included in broiler diets, with no negative effects on gut parameters.Both processes and unprocessed OP have high fiber levels (unprocessed FB= 350g/kg, and processed FB= 256 g/kg, Table 1).High dietary inclusion levels of fiberrich feedstuffs interfere with organ size and weight, as seen in the present study.The inclusion of fibrous by-products in non-ruminant feeds affects the size and weight of digestive organs (Le Goff and Noblet, 2001).Engberg et al. (2004) concluded that whole wheat feeding, which is richer in fiber than ground dehulled wheat, stimulates the function and size of broiler digestive organs.Other researchers observed significant increase in the relative weights of the digestive organs of broilers fed a whole-wheat diet, which is richer in fiber than ground dehulled wheat (Ravindran et al., 2006).

Effects of Different Dietary Levels of Two Types of Olive Pulp and Exogenous Enzyme Supplementation on the Gastrointestinal tract size, Immunology and Hematology of Broilers
Dietary enzyme inclusion and OP processing did not improve any of the studied carcass, cranial gut segment, or liver parameters (p>0.05,Table 4).Malayoğlu et al. (2010) did not verify any influence of enzyme supplementation on broiler internal organ weights either.Engberg et al. (2004) found that whole wheat feeding did not interfere with digestive size in broilers, but reduced the activities of amylase in the pancreatic tissue; on the other hand, the addition of xylanase increased chymotrypsin and lipase activities.Other researchers reported contradictory results.For instance, Lázaro et al. (2004) found that dietary enzyme inclusion reduced crop size in broilers, whereas enzyme supplementation reduced the relative weight of certain segments of the gastrointestinal tract in the study of Józefiak et al. (2007).According to Wang et al. (2005), the inclusion of enzymes in broiler diets decreased the size of the digestive organs and of the gastrointestinal tract.Another explanation for these contradictory results is the type of experimental design that was used.The experimental design used to evaluate enzymes generally take into account their nutritional value, which is discounted from the diets (Choct, 2006).In the present experiment, however, feeds were formulated to supply the broilers' requirements according to the genetic company manual, regardless of the inclusion or not of the enzyme blend, which was added "on top" of the broilers' requirements.Therefore, the nutrients supplied by enzymes exceeded the requirements of the animals, which may explain the lack of digestive organ differences.
The inclusion levels of OP did not affect pancreas absolute or relative weights (P%); duodenum absolute and relative weight (D%), length, or width; jejunum weight, width, or diameter; ileum ansorlute  et al. (2008) observed an increase in PV% and G% in broiler fed raw corn, but not in all of the other visceral organs that were studied, as also found in the present experiment.In their study, González-Alvarado et al. (2008) reported a higher fiber content in raw corn than raw rice, thereby explaining their results.Broilers that were fed with 100 g/kg of OP presented with a higher relative jejunum weight (J%) and longer jejunum length, but a shorter ileum length and small erileum diameter (EC%) (p≤0.05,Table 5).Contrary to what is commonly thought about the digestive effect of fiber in non-ruminants (Macari, 2008), González-Alvarado et al. (2007) reported that high fibrous feedstuff improved the apparent retention of most nutrients, and increased total digestive tract weight; however, they detected a reduction in small intestine (duodenum, jejunum, and ileum) length.On the other hand, González-Alvarado et al. (2008) observed that high fiber diets did not affect the relative length of the intestines, but low fiber diets reduced the presence of digesta in the intestines and increased the presence of feed in the gizzard.Digesta retention affects intestinal tract size and weight due mechanical stimulation and contact of nutrients (Wilfarta et al., 2007).The jejunum is the main site of nutrient absorption in the gastrointestinal tract, and increasing its size is highly

Effects of Different Dietary Levels of Two Types of Olive Pulp and Exogenous Enzyme Supplementation on the Gastrointestinal tract size, Immunology and Hematology of Broilers
desirable and correlated with greater absorption of nutrients (Macari, 2008).Another substance contained in OP that can interfere with gastrointestinal size is tannin.Dietary tannins are thought to reduce diet digestibility and metabolizable energy levels through a direct interaction with proteins and carbohydrates from both exogenous and endogenous sources (Mansoori and Acamovic, 2007).Therefore, OP inclusion in the present experiment may have influenced the size and weight of different parts of the small intestine due to its tannin content.
Processed OP (partial destoning) resulted in lower jejunum absolute and relative weights (J%)and length, and left cecum length compared with unprocessed OP (p≤0.05,Tables 5 and 6).These results may be attributed to the lower fiber content of the processed material.Olive seeds are highly lignified and have a low fat content (García-Ayuso and Luque de Castro, 1999).Dietary fiber affects the development of the different segments of GIT, and the effects differ according to the physical-chemical characteristics of the fiber source.Plant by-products, like OP, are good sources of fiber in broiler diets.The cecum of birds is a known site of allozyme digestion, which is performed by bacteria (Macari, 2008).These bacteria use as substrate fibrous material that is not digested in the most cranial part of the intestine, and are thus stimulated by the fiber content of the diet.Fiber fermentation produces volatile fatty acids, including butyrate, which is consumed by the intestinal wall as an energy source, thereby stimulating the growth of cecum enterocytes (Macari, 2008).González-Alvarado et al. (2007) verified an increase in the cecum weight of broilers fed high fiber by-product diets (oat hulls and soy hulls).Based on these facts, we may infer that the presence of fiber in the unprocessed material may have stimulated the development of the jejunum and cecum of birds.Other factors may also have influenced these variables.Crushing and malaxation are the most important critical points in the olive oil mechanical extraction process (Yorulmaz et al., 2011).Lavelli and Bondesan (2005) observed an increase in the total secoiridoid polyphenol (anti-oxidant, anti-microbial, and anti-inflammatory) content and in the antioxidant activity in extra virgin olive oils when the fruits were pre-destoned.The GIT is the largest organ of the body's immune defense in animals, which is no different for birds (Macari, 2008).The presence of anti-oxidants and anti-microbial substances may influence immune organsize and weight, thus resulting in smaller, less inflammatory, and chemotactic organs.One portion of the cecum is more sensitive to these substances: the cecal tonsils.These segments of the cecum have lymphoepithelial cells, which characterizes them as lymphoid organs (Macari, 2008).Thus, the processing of the OP may contribute to less inflammation of the organ as a whole, consequently explaining their smaller size in birds.
No significant interaction was observed among the evaluated treatments on the studied hematological variables (p>0.05,Table 7).OP levels and ENZ inclusion did not interfere with any of the studied blood parameters, including (p>0.05,Table 7).Uric acid, total cholesterol, high density lipoprotein (HDL), low density lipoprotein (LDL), LDL/HDL ratio, aspartate amino transferase (AST), alanine amino transferase (ALT), total protein, albumin, and globulin levels were not affected by OP processing (p>0.05,Table 7).Processed OP increased the triglyceride and VLDL concentrations in broiler blood (p≤ 0.05, Table 7).No studies on the interference of dietary OP on broiler hematological parameters were found in literature.However, a series of papers studying antioxidant and functional herb compounds, with OP-related properties, have been published.Ginger extract has antioxidative properties, since it can scavenge superoxide anions and hydroxyl radicals, as shown by Saeid et al. (2010).Those authors did not find any influence of the addition of ginger extract to broiler diets on total protein, albumin, globulin, serum HDL, LDL, and VLDL blood levels, except lower cholesterol levels.According to Mehala and Moorthy (2008), blood total cholesterol, HDL, LDL, and triglyceride levels were not affected by the inclusion of Aloe vera (rich in saponins and antimicrobial properties) or Curcuma longa powder in broiler feeds.
The results of the present study are consistent with the findings of Toghyani et al. (2010a), who reported that the inclusion of Thymus valgaris (rich in phenols, essential oils, and saponins) powder at a 10 g/kg diet did not influence blood protein, albumin, total and LDL cholesterol concentrations, and albumin to globulin ratio in broilers.According to Toghyani et al. (2011), serum protein, albumin, and triglycerides were not affected by the dietary addition of garlic (antibacterialrich compound) and cinnamon (rich in antibacterial compounds and anti-inflammatory compounds) powder in broilers.Toghyani et al. (2010b), when evaluating the supplementation of black seed and peppermint (both rich in essential oils) in broiler diets, did not find any effects on serum protein, albumin, triglyceride, LDL, HDL and total cholesterol levels.OP compounds have similar functions and characteristics as those natural organic compounds, including high

Effects of Different Dietary Levels of Two Types of Olive Pulp and Exogenous Enzyme Supplementation on the Gastrointestinal tract size, Immunology and Hematology of Broilers
instability.Harvesting, processing, and crushing affect the stability of OP compounds (Clodoveo, 2012).Some OP compunds present potent antioxidant activity by direct scavenging of reactive oxygen and nitrogen species and by acting as chain-breaking peroxyl radical scavengers, which have a low stability in the presence of air (Russo et al., 2010).Thus, those compounds may not have had the chance to express their properties due to the OP and the broiler diets processing.
The high VLDL and triglyceride serum concentrations observed in the broilers fed the processed OP maybe explained by improvements in energy digestibility.The processing of OP consists in partial destoning, thereby resulting in a lower fiber material.Dietary fiber is responsible for accelerating digesta passage rate (Macari et al., 2008).Due the low fiber content of the diet (Table 1), broilers that were fed the processed OP presented may have presented lower digesta passage rate than those fed with unprocessed OP, remaining longer in the intestinal tract, possibly resulting in better feed digestion.As a consequence, the triglycerides in soybean oil and in the OP processed oil residue were better absorbed.The VLDL are the main lipids that are transported via avian blood, and triglyceride is the main component the principal content of the VLDL (Macari et al., 2008), which may explain its highlevels in the blood of broilers fed the OP diet.
No significant interactions were observed between olive pulp (OP) levels and enzyme supplementation, OP levels and pulp processing, or enzyme supplementation and pulp processing on the studied immunity traits (p>0.05,Table 8).Antibody titers against avian influenza on d21 (lg2), and d 28 (lg2), Newcastle disease on d27 (lg2) or infectious bronchitis (IgT-IBV) were not affected by OP levels, processing, or ENZ inclusion (p>0.05,Table 8).Likewise, total antibodies (IgT) from the humoral immune response to injections with sheep red blood cells (SRBC) on d 7 and 14 after injection were not influenced by OP levels or processing (p>0.05,Table 8).Some authors reported the inefficacy of some herb components to modulate the immune system of broiler modulations.Toghyani et al. (2011) reported that garlic and cinnamon powder had no influence on any immune-related parameters.
In another study, Toghyani et al. (2010a) observed that thyme powder at a 10 g/kg diet did not affect antibody titers against Newcastle, influenza, or SRBC.However, other authors observed that medicinal herbs may affect the immune response of broilers.The addition of a medicinal plant blend (alfalfa, liquorice root, great burdock, and cinnamon) to a broiler diet resulted in improvements in antibody titers against Newcastle disease virus (Khaligh et al., 2011).
All of these previous results support the efficacy of medicinal herbs in the immunological modulation in broilers.So why were these results not observed in this experiment?Duarte et al. (1993), in a classical experiment to elucidate the phytotherapy properties of oleuropeosides, proved that these substances are capable of inhibiting peak contractile forces in electrically driven left atria of guinea pig hearts.However, the oleopeouside doses used in that experiment were much higher compared to their content in the OP evaluated in the present study.Therefore, the low dose-response hypothesis cannot be ruled out, as it may explain the low effects of OP compounds on most of the evaluated hematological parameters.The OP compounds were possibly too diluted in the diets to cause any effect on the broilers' immune system.In an extensive review about the beneficial effects * Means (± standard error) within each column of dietary treatments with no common superscript differ significantly at p<0.05.Means (± standard error) within each column of dietary treatments without superscript did not differ significantly at p>0.05.ENZ= enzyme blend inclusion; OP= olive pulp; p=processed; u=unprocessed; SRBC= Sheep red blood cell at 7 th or 14 th days after injection; Influenza = titer against influenza 21 and 28 days after vaccination and Newcastle= titer against Newcastle disease 7 and 27 days after vaccination.

Table 1 -
Chemical composition of two types of olive pulp used in the experiment.

Table 2 -
Feed ingredients and chemical composition of the starter diets (1 to 21 days of age).

Table 3 -
Feed ingredients and chemical composition of the finisher diets (22-42 days of age).

Table 6 -
Large intestine segments characteristics of experimental broilers (mean±standard deviation) *

Table 8 -
Immune response to vaccination and SRBC injection, and organ weights (mean±standard deviation) *