Effect of diets with different contents of sunflower meal without or with exogenous enzymes supplementation on gastrointestinal tract response of growing turkeys

The present experiment was conducted to investigate the effect of a different dietary content of sunflower meal (SFM) and the efficiency of an non-starch polysaccharide (NSP)-degrading enzyme preparation on growth and gut function of young turkeys. A total of 1512 one-day-old male turkey poults were randomly assigned to 8 dietary treatments, with 7 pens per treatment and 27 birds per pen. Experimental diets with a different content of SFM (0, 7, 14 and 21%; SFMQ, SFM 7, SFM ) 4 and SFM 2 1 groups, respectively) were administered in two variants, with and without NSP-degrading enzymes (E+ and EQ treatments, respectively). Diets fed to the turkeys for 8 weeks were isonitrogenous and isocaloric, but they differed substantially with regard to crude fibre content (in average 2.98, 3.97, 4.64 and 5.64% in the SFM 0, SFM 7, SFM l 4 and SFM 2 1 groups, respectively). The enzyme preparation applied to a diet caused a tendency towards lower ileal viscosity (P=0.099) and a significant decrease in caecal total volatile fatty acids concentration, * Supported by the State Committee for Scientific Research, Grant No. 311 058 31 4 Corresponding author: e-mail: j.juskiewicz@pan.olsztyn.pl JUSKIEWICZ J. ET AL. 469 despite of the observed increase in activities of bacterial a-glucosidase, a-galactosidase and /?-galactosidase in the E+treatment. At the same time, two-way ANOVA revealed that following the dietary inclusion of SFM at the amount of 14 and 21%, a significant decrease was observed in final body weight, small intestine and caecal tissue mass, caecal digesta mass, as well as the rate of bacterial production of volatile fatty acids in the caeca. Such an effect was not recorded when SFM was applied at a dose of 7%. In conclusion, sunflower meal rich in crude fibre added at the level of 14-21% to a diet for growing turkeys may induce undesirable processes manifested in the decrease in relative mass of small intestine and caecal tissues, as well as potent inhibition of the fermentation processes in the caeca. Our study showed additionally that high fibre sunflower meal could be used at a dose of up to 7% without any adverse effects on the gastrointestinal physiology of the growing turkeys.


INTRODUCTION
Soyabean meal (SBM) has been reported to presumably be the only common protein supplement that is typically included in poultry rations with no limitation as to the quantity used (Bach Knudsen, 2001).In contrast, other studies have shown that dietary SBM added at a high level may lead to a decrease in intestinal absorptive surface area and an undesirable increase in litter moisture concentration in young turkeys due to relatively high amounts of raffinose family oligosaccharides (Juskiewicz et al., 2009).Having in mind that effect as well as the fact that protein is often one of the most expensive components of poultry diets, nutritionists have started a search for alternative ingredients which have potential as cost-effective protein sources, sometimes underutilized in poultry production (Mushtaq et al., 2006).It is well known that sunflower oil is considered one of the most healthful vegetable oils for humans, thus the availability of sunflower meal (SFM), as a by-product, is relatively high (Vieira et al., 1992;Stodolak et al., 2009).SFM obtained from processed intact sunflower seeds is rich in protein but also has a high content of fibre.Moreover, diets formulated with SFM can be deficient in lysine.The latter could be overcome by an appropriate supplementation.But high level of fibre, causing low dietary energy values, may excessively reduce the time of feed passage throughout the digestive system and diminish nutrients utilization (Wenk, 2001).Among poultry, turkeys seem to be most sensitive to a high content of crude fibre in a diet.Experiments concerning SFM in turkeys' feeding are sparse, thus the recommended level of dietary SFM for growing turkeys have not been explicitly established.Moreover, studies focused not only on the growth and performance response but with a deeper look into physiology have increasingly been in great demand.In the light of the above, the aim of this study was to investigate the growth and the physiological response of the gastrointestinal tract of young turkeys (to 8 wk of age) to diets with different contents of sunflower meal, applied without or with NSP-degrading enzymes.As the experimental diets differed substantially in non-digestible carbohydrate fractions, great attention was paid to the development and bacterial fermentation processes in the caeca of the birds.

Birds and housing
The experimental procedure was approved by the Local Ethics Committee.The study was conducted at the research laboratory of the Department of Poultry Science, University of Warmia and Mazury in Olsztyn.The experimental materials comprised a total of 1512 heavy-type Big 6 turkeys purchased as one-day-old male poults at the "Grelavi" Hatchery in K^trzyn.The birds were randomly assigned to 8 dietary treatments, with 7 pens per treatment and 27 birds per pen, and they were raised on deep litter.At the end of 8-wk period, birds were weighed, and feed consumption was recorded.Weight gain and feed efficiency were calculated.For performance indices, each pen was considered an experimental unit.Light was provided for 16 h per day.Indoor temperature was 32°C at the beginning of the experiment and 22°C at the end of week 8.At the beginning of the study, poults were administered Aviffa-RTI -live vaccine against infectious rhinotracheitis.During two rearing periods (weeks 1-4 and 5-8), birds were fed isoenergetic diets containing 28 and 26% crude protein, respectively, covering turkeys nutrient requirement of (NRC, 1994).They had free access to feed and water.

Diets
Commercially available soyabean (from a local feed company) and sunflower (from WIOJL-AGRO Co., Winnica, Ukraine) meals were used in this experiment.The composition of experimental diets with a different content of SFM (0, 7, 14 and 21%) is given in Table 1.The levels of other components, including SBM, were determined so as to meet the nutrient requirement of turkeys (NRC, 1994) aged 1-4 and 5-8 weeks.Each diet was prepared in two variants, with and without non-starch polysaccharide (NSP)-degrading enzymes.The enzymatic preparation supplied, U: pectinase 500, cellulose 40U, xylanase 1600, glucanase 800, mannanase 200, galactanase 20, and other minor enzymes per kg diet (Superzyme CS, Canadian Bio-Systems Inc., Calgary, AB, Canada).The lysine, arginine, methionine, threonine, tryptophan, mineral and vitamin contents were similar in all dietary treatments.The average content of crude fibre in the diets was approximately 2. 98, 3.97, 4.64 and 5.64 in the SFM 0 , SFM 7 , SFM 14 and SFM 21 groups, respectively.In terms of total NSP the aforementioned diets contained approximately 11.17, 11.93, 12.68 and 13.43 of total NSP, respectively.77.9 90.8 106.7 1 for 1 to 4 and 5 to 8 week of feeding the vitamin and mineral premix supplied per kg of diet, IU: vit.A 15.000, and 13.000; mg: vit.E 40 and 35, respectively.For 1 to 8 week of feeding the vitamins and mineral premix supplied per kg of diet, mg: Se 0.3, Mn 150, Zn 90, Fe 60, Cu 15,1 1, diclazuril 1, IU: vit.D 3 4,500; mg: vit K 3 2.5, thiamin 3.5, riboflavin 10, vit.B 6 6, vit B 12 0.03, folic acid 2, biotin 0.36, niacin 75, pantothenic acid 21, choline 600 For chemical analysis, the samples were ground to pass through a 0.5 mm sieve.Feed ingredient, diet or digesta samples were analysed in duplicate for dry matter, crude protein (CP), fat, and crude fibre (CF) using AO AC (2005) methods 934.01, 976.05, 920.39 and 978.10, respectively.NSP were determined by gasliquid chromatography using the procedure described by Slominski and Campbell (1990).Briefly, 100 mg samples were boiled with 2 ml dimethylsulphoxide for 1 h and then incubated at 45°C overnight with a sodium acetate buffer solution (pH 5.2) of starch-degrading enzymes: amylase, pullulanase and amyloglucosidase (Sigma Chemical Co., St Louis, MO).Ethanol was then added and the mixture was left for 1 h at room temperature before being centrifuged at 1990 g, for 10 min at 21°C.The supernatant was discarded and the dried residue was dissolved in 1 ml of 12 M sulphuric acid and incubated for 1 h at 35°C.Six ml of water and 5 ml of myoinositol (internal standard) solution were then added and the mixture was boiled for 2 h.One ml of the hydrolysate was taken and neutralized with 12 M ammonium hydroxide, reduced with sodium borohydride, and acetylated with acetate anhydride in the presence of 1-methylimidazole.Component sugars were separated using a SP-2340 column and a Varian CP 3380 Gas Chromatograph (Varian Canada Inc., Mississauga, Ontario, Canada).

Sample collection and chemical analyses
After 8 weeks of feeding, 10 turkeys that represented an average body weight for each group were killed by cervical dislocation.After laparatomy, segments of the digestive tract (crop, ventriculus, small intestine, caeca and colon) were removed and weighed.As soon as possible after euthanasia (about 20 min) pH was measured in each segment using a microelectrode and pH/ION meter (model 301, Hanna Instruments, Vila do Conde, Portugal).Samples of ileal (middle, 1/3 section of ileum) and caecal contents were used for immediate analysis (ammonia, dry matter, viscosity, volatile fatty acids -VFA), while the rest of the caecal digesta was transferred to tubes and stored at -70°C.The ventriculus (gizzard), small intestine, caeca and colon were flushed with water, blotted on filter paper and weighed as the tissue mass.
Dry matter of digesta was determined at 105°C.The total contents of the small intestine were collected, mixed on a vortex, and centrifuged at 7.211 g for 10 min.The supernatant fraction (0.5 ml) was placed in a Brookfield LVDV-II+cone-plate rotational viscometer (CP40; Brookfield Engineering Laboratories Inc., Stoughton, MA, USA) and the viscosity of all samples was measured at a fixed temperature of 39°C and at a shear rate of 60 per min.The viscosity value was recorded as an apparent viscosity.In fresh caecal digesta, ammonia was extracted, trapped in a solution of boric acid in the Conway's dishes, and determined by direct titration with sulphuric acid (Hofirek and Haas, 2001).
Caecal digesta samples were subjected to VFA analysis using gas chromatography (Shimadzu GC-2010, Shimadzu, Kyoto, Japan).The samples (0.2 g) were mixed with 0.2 ml formic acid, diluted with deionized water and centrifuged at 7.211 g for 10 min.The supernatant was loaded onto a capillary column (SGE BP 21.30 m x 0.53 mm) using an on-column injector.The initial oven temperature was 85°C and was raised to 180°C by 8°C/min and held there for 3 min.The temperatures of flame ionization detector and the injection port were 180 and 85°C, respectively.The sample volume for GC analysis was 1 |il.Caecal VFA pool size was calculated as the sum of VFA concentration in digesta and caecal digesta mass.
The bacterial glycolytic activity in the caecal digesta was measured by the rate of p-or o-nitrophenol release from their nitrophenylglucosides according to the modified method of Djouzi and Andrieux described by Juskiewicz et al. (2006).The following substrates were used: p-nitro-phenyl-a-Dglucopyranoside (for a-glucosidase), and />nitrophenyl-/?-D-glucopyranoside (for yff-glucosidase),p-nitrophenyl-a-D-galactopyranoside(a-galactosidase),o-nitrophenyl-/?-D-galactopyranoside (/?-galactosidase), and p-nitrophenyl-/^D-glucuronide (for /^-glucuronidase).The reaction mixture contained 0.3 ml of a substrate solution (5 mM) and 0.2 ml of a 1:10 (v/v) dilution of the caecal sample in 100 mM phosphate buffer (pH 7.0) after centrifiigation at 7.211 g for 15 min.Incubation was carried out at 37°C and p-nitrophenol was quantified at 400 nm and at 420 nm (o-nitrophenol concentration) after the addition of 2.5 ml of 0.25 M cold sodium carbonate.The enzymatic activity (a-and/?-glucosidase, a-and/?-galactosidase, and^-glucuronidase) was expressed as |imol product formed per min (IU) per g of digesta.

Statistical analysis
The STATISTICA software package version 8.0 (StatSoft Corp., Krakow, Poland) was used to determine if variables differed among treatment groups.Twoway ANOVA was performed to assess the effects of the dietary level of sunflower meal (0, 7, 14 and 21%), of dietary addition of enzyme mixture preparation (without and with supplementation) and of the interaction between SFM level and NSP-degrading enzyme addition (D x E) (Snedecor and Cochran, 1989).When the ANOVA indicated significant treatment effects, means were separated using Duncan's multiple range test.In the Tables, results are presented as mean values with pooled standard errors.Data were checked for normality before statistical analysis was performed.Differences were considered to be significant at P<0.05.

RESULTS
After 8 weeks, the body weights of turkeys fed diets containing 14 and 21% SFM were significantly lower, compared with the control group (Table 2).Moreover, the birds fed the diet containing 21% of SFM were also significantly lighter than those from the SFM 7 treatment.A tendency (P=0.068)towards lower diet intake in SFM groups was observed during eight weeks of feeding.Neither SFM levels nor enzymatic supplementation affected the feed conversion ratio during the experimental period covering 8 weeks.At 8 weeks of age, regardless of the addition of NSP-degrading enzymes, the inclusion of SFM to a diet caused a tendency towards lower ventriculus pH (P=0.069) and relative tissue mass (P=0.089), when compared to the control treatment without SFM addition (Table 3).The turkeys fed diets containing 14 and 21% SFM were characterized by a significantly lower small intestinal tissue weight than those fed the 0 and 7% SFM diets (PO.05).Dietary SFM had no effect on ileal digesta pH, dry matter concentration nor viscosity value (P<0.05).When diets supplementation with the enzyme preparation was taken into consideration, the effect of this additive on indices of upper gastrointestinal tract functioning was negligible, except numerically lower ileal viscosity (1.26 vs 1.36 mPa-s; P=0.099).After 8 weeks of experimental feeding, the lowest caecal tissue mass was found in turkeys fed the diet with highest SFM content (SFM 21 < SFM Q and SFM 7 ; P<0.05; Table 4).The mass of the caecal digesta was noted to be the highest in the control treatment and was decreasing as follows: SFM 0 a > SFM y b > SFM 14 C > SFM 21 C .The low dietary level of SFM (7%) elicited a significant increase in caecal dry matter concentration as compared to the other groups (P<0.05).The caecal ammonia concentration as well as colonic pH and tissue mass were not affected by the dietary treatments, either by SFM or enzyme preparation addition.The dietary application of the NSP-degrading enzymes mixture resulted in a significant increase in the pH of caecal digesta (E+ > E Q ; PO.05).The activity of bacterial /?-glucosidase and /^-glucuronidase was not affected by the level of dietary SFM or the application of exogenous enzymes (Table 5).On the other hand, both main enzymes (Table 5).On the other hand, both main experimental factors (SFM and enzyme preparation) significantly influenced the activities of bacterial a-glucosidase, a-galactosidase and /?-galactosidase in the caecal digesta.
When the level of SFM in a diet was taken into account, the highest activity of a-glucosidase was observed in turkeys fed diets with low and moderate SFM content (SFM ?, SFM 14 > SFM Q , SFM 21 ; PO.05).The birds from the SFM ? and SFM 21 groups displayed the highest and the lowest a-galactosidase activity, respectively (in both cases P<0.05 vs other treatments).The SFM 21 group was also characterized by the lowest /?-galactosidase activity (PO.05 vs SFM Q and SFM 7 groups).The supplementation of diets with the NSP-degrading enzyme preparation resulted in enhanced activities of bacterial a-glucosidase, a-galactosidase and /?-galactosidase in the caecal digesta of the turkeys (PO.05).Regardless the addition of the NSP-degrading enzyme preparation, the increased SFM content in a diet up to 14-21% caused a significant decrease in the total concentration as well as content (pool) of VFA in the caeca in comparison to the control and SFM 7 treatments (Table 6).It was mainly due to the enhanced concentration of acetic acid and -to a lower extent -of valeric acids (SFM 0 and SFM ?vs SFM 14 and SFM 21 groups; PO.05).Moreover, the SFM ?treatment was accompanied by the highest caecal propionate concentration as compared with the other groups (PO.05).When the enzyme preparation applied to a diet was taken into consideration, the E+ treatment caused a significant decrease in the total VFA and butyric acid as well as an increase in isovaleric acid concentrations (P<0.05vs E Q treatment).An interaction between SFM content and enzymatic supplementation affected the caecal concentrations of acetic and propionic acids.In contrast to the control treatment, the addition of enzymes to diets with a low (7%), moderate (14%) and high (21%) sunflower meal content did not affect C 2 caecal concentration significantly (PO.05).The concentration of this acid in the SFM 0 E 0 subgroup was significantly higher than in the birds fed the SFM Q E+ diet.Alike differences were noted between subgroups SFM y E 0 and SFM 7 E+ with regard to the concentration of caecal propionic acid.

DISCUSSION
At the termination of the performance part of this study we observed that the body weight (BW) of 8-wk turkeys fed diets with a relatively high sunflower meal (SFM) content may be depressed in comparison to the BW of control birds receiving the soyabean meal-based diet.Therefore, a special attention was paid to the gastrointestinal tract response of growing turkeys fed diets with different contents of SFM added at the expense of soyabean meal and partly of wheat.The efficacy of NSP-degrading enzymes application was assessed as well.
As for the applied enzyme supplementation, the data obtained revealed that the enzymatic mixture added to a diet caused a tendency towards lower ileal viscosity (P=0.099) and a significant decrease in caecal total volatile fatty acids (VFA) concentration, despite an increase in the activities of bacterial a-glucosidase, a-galactosidase and /?-galactosidase observed in the E+ treatment.The decrease in VFA concentration was probably the main reason for an increase in caecal pH value observed in E-supplemented treatment.The minor effect of the enzyme preparation on ileal viscosity was in accordance with a relatively small content of soluble NSP in experimental diets.It should be emphasized, however, that the viscosity values observed in the current study were relatively low (1.22-1.44 mPa-s) and lower than the values of 3.0 to 5.0 mPa-s observed in broiler chickens and growing turkeys fed wheat-based diets (Juskiewicz et al., 2005;Meng et al., 2005).Furthermore, such low intestinal viscosity values have been reported to have a minimal, if any, effect on growth performance of broilers (Liang and Liu, 1988).
Following the increasing dietary level of SFM in a diet, a considerable decrease in relative tissue mass of ventriculus, small intestine, and caeca, expressed as a percentage of BW, was observed after 8 weeks of feeding in turkeys exposed to the higher doses of SFM.It is common knowledge that it is not only the level of dietary fibre that is of paramount importance for gastrointestinal tract (GIT) segments development, but the type or the source of fibre also plays a significant role in digestion and absorption (Wenk, 2001).In comparison to soyabean meal, sunflower meal used in this study contained more total NSP (27.0 vs 17.9%), more lignin (8.6 vs 0.5%), and more crude fibre (15.3 vs 4.8%).SBM exceeded SFM only in respect of the content of water-soluble NSP (1.05 vs 0.78%).Therefore, diets with SFM were characterized by an increasing level of crude fibre, lignin and total NSP, but a lower content of water-soluble NSP (Table 1).Our findings are in contradiction with the suggestion made by Dusterhoft et al. (1992) that the presence in the sunflower seeds of a substantial amount of cell-wall material, mainly cellulose, pectic polysaccharides and 4-0-methyl-glucuroxylans, with small amounts of glucomannans and fucoxyloglucans, could exert a trophic effect on the intestinal mucosa, increasing the intestinal size.Other authors reported a different response of intestinal organs to sunflower products.Arija et al. (2000) showed the shortening and thickening of the jejunal villi caused by the inclusion of full-fat sunflower kernels in chicken diets.In turn, Brenes et al. (2008) observed a decrease in the size of the small intestine (particularly the duodenum) and caeca in chickens fed 15% high oleic acid sunflower seeds.One explanation to the lower weight of GIT segments, observed in our study, could be faster passage of digesta through the gastrointestinal tract in the SFMfed birds, as supported by the significantly decreased digesta mass in the caeca.Another reason for the foregoing effect might be the lower content of soluble polysaccharides in diets containing sunflower meal in comparison to the control SBM-containing diet.It is well established that water-soluble polysaccharides may evoke a considerable increase in the weight of caecal tissue and digesta in monogastric animals, in a big part due to an increased bacterial number and greater fermentation in the lower parts of GIT (Juskiewicz and Zduhczyk, 2004).Some researchers reported that a diet supplementation with non-digestible carbohydrate fractions may result in an enhanced glycolytic activity of microorganisms living in the large intestines of monogastric animals (Monsan and Paul, 1995).In our experiment, the highest activities of bacterial a-glucosidase, a-galactosidase and /?-galactosidase were observed when SFM 7 treatment was applied./?-galactosidase, a-galactosidase and a-glucosidase activities can improve the fermentation of lactose, raffinose family oligosaccharides and resistant starch, leading to VFA and lactic acids which are the source of energy for the tissues.However, the activities of potentially-harmful /^-glucuronidase and /?-glucosidase were similar in all groups, which should be considered as a positive effect of sunflower meal addition to a diet for turkeys.Both /^-glucuronidase and /?-glucosidase have been implicated in the generation of mutagens or carcinogens in the hindgut (Campbell et al., 1997).It could be assumed that well-balanced content of dietary insoluble and soluble non-digestible carbohydrates may positively affect the activity of enzymatic microbiota.
When large amounts of fermentable carbohydrates were fed, the increased caeca weights (especially tissue) were considered as a positive physiological response associated with some beneficial changes in the large intestine, e.g., acidification of digesta, increased VFA contents (Jankowski et al., 2009).In the present study, the application of diets containing increasing levels of SFM was indeed in reverse relation to VFA concentration and pool.In the presence of different fermentable carbohydrates, the VFA pool size rather than the individual fatty acid concentration has been postulated to be the best indicator of the intensity of colonic fermentation (Campbell et al., 1997).In the 8-wk-old turkeys, the VFA pool in the birds fed the SFM 14 and SFM 21 treatments was more than 2+fold lower than that of the birds fed the SFM 7 and SFM Q diets.Our recent study on growing turkeys demonstrated that, in contrast to dietary crude fibre applied at the doses of 3.5 or 5.3%, the different dietary level of soyabean a-gactosides (0.06, 0.93, 1.69 and 2.33%) caused a distinct response of turkeys caeca (Zduhczyk et al., in press).In that study we concluded that soyabean raffinose-family oligosaccharides seemed to greatly affect the development and physiology of the GIT, and that their presence in the amount of approximately 1% in a diet for growing turkeys ought to be considered as nutritionally advisable.The present experiment proved that sunflower meal added at the level of 14 or 21% to a diet delivered excessive amounts of crude fibre, lignin, and water-insoluble non-starch polysaccharides, which in turn inhibited the fermentation processes in the caeca and, to some extent, impaired the development of the upper GIT segments in growing turkeys.

CONCLUSIONS
In conclusion, sunflower meal (SFM) rich in crude fibre and lignin added at the level of 14-21% to a diet for growing turkeys may induce undesirable processes in the gastrointestinal tract development manifested in decreased relative mass of small intestine and caecal tissues, as well as in potent inhibition of the fermentation processes in the caeca.Having in mind a worsening in birds' growth observed during the performance stage of the study, it should be concluded that the dietary SFM level at 14% and more seems to be an overdose for young growing turkeys.On the other hand, our study showed also that high fibre SFM could be an effective alternative substitute of soyabean meal for growing turkeys when applied at the dietary level of up to 7%.

Table 1 .
Composition of experimental diets

Table 2 .
The effect of different content of sunflower meal 1 (SEM) in diets on the growth performance of turkeys aged 8 weeks of age

Table 3 .
Indices of upper GIT functioning in turkeys fed diets with increasing levels of sunflower meal 1 (SFM) without or with inclusion of feed enzymes (E)

Table 4 .
Indices of lower GIT functioning in turkeys fed diets with increasing levels of sunflower meal 1 (SFM) without or with inclusion of feed enzymes (E) 'SFM was applied at following dietary levels: 0, 7, 14 and 21% (SFM 0 , SFM 7 , SFM, 4 , and SFM 2] respectively); 2 SEM, standard error of the mean (SD for all birds divided by the square root of turke; number); values within each column with the same superscript letter are not different at P<0.05

Table 5 .
Activity of bacterial enzymes in caecal digesta (U/g) of turkeys fed diets with increasing levels of sunflower meal 1 (SFM) without or with inclusion of feed enzymes (E)

Table 6 .
Concentration (umol/g fresh digesta), total content (umol/kg body weight) of volatile fatty acids (VFA) in the caeca of turkeys fed experimental diets