The Relation Between Starch Digestion Rate and Amino Acid Level for Broiler Chickens 1

Transcription

1 The Relation Between Starch Digestion Rate and Amino Acid Level for Broiler Chickens 1 R. E. Weurding,*,,,2 H. Enting,* and M. W. A. Verstegen *Institute for Animal Nutrition De Schothorst, Lelystad, The Netherlands; Brameco ZON, Eindhoven, The Netherlands; and Animal Nutrition Group, Wageningen University & Research Center, Wageningen, The Netherlands ABSTRACT Digestion coefficients of nutrients give information source did not affect feed intake (2,213 g), but weight about the amount of nutrients available to the animal but not about the rate or site of absorption. Gradual digestion of starch may have an amino acid sparing effect and therefore enhance growth efficiency of broiler chickens. A growth trial was performed with 6,800 broiler chickens from 9 to 30 d of age to investigate interactions between starch digestion rate and amino acid level. Birds were fed either a pea-corn-based diet (slowly digestible starch) or a tapioca-corn-based diet (rapidly digestible starch). Both diets were formulated with five levels of digestible lysine, varying from 8.5 to 11.0 g/kg. The minimal levels of other amino acids varied accordingly. Starch gain was consistently higher for birds on pea-corn diets than for those on tapioca-corn diets (1,426 vs. 1,400 g; P < 0.01). Feed conversion was better (P < 0.01) for birds on pea-corn diets (1.55) than for birds on tapioca-corn diets (1.58). The difference in feed conversion between birds on pea-corn and tapioca-corn diets was greater with lower amino acid levels (0.043) than with higher amino acid levels (0.019) in the diet (P = 0.11). This interaction was more pronounced during the first 9 d of the experiment (P < 0.05). It was concluded that feeding slowly digestible starch improved protein and energy utilization in broiler chickens. (Key words: starch digestion rate, amino acid level, broiler chicken, pea, tapioca) 2003 Poultry Science 82: INTRODUCTION Feed evaluation in monogastric animals is based on digested nutrients. Digestion coefficients of nutrients at the terminal ileum give information about the amount of nutrients available to the animal but not about the site or the synchronization of availability of different nutrients. The major part of starch is digested in the upper part of the small intestine. Diets with rapidly digestible starch may result in elevated plasma glucose levels when other nutrients are not yet absorbed, which may have consequences for protein utilization. Diets with similar amounts of digestible nutrients, but differences in digestion kinetics, may result in different performances. In ruminant nutrition, the importance of the kinetics of carbohydrate and protein digestion in the rumen has long been recognized. In human nutrition, the glycemic index of foodstuffs is used to manipulate glucose absorption rate in order to prevent metabolic disorders or to enhance athlete performance (Brand-Miller, 1999). We have ob Poultry Science Association, Inc. Received for publication February 14, Accepted for publication October 8, Funded by the Dutch feed cooperatives and the Dutch Technology Foundation STW (project no. WPR ). 2 To whom correspondence should be addressed: e.weurding@ agrifirm.nl. served differences in performance of broiler chickens receiving diets that were iso-energetic but different in terms of starch digestion rate (unpublished data). Feeding diets containing slowly digestible starch resulted in a better feed conversion for broiler chickens than feeding diets with rapidly digestible starch. Moreover, an interaction between starch digestion rate and amino acid content was observed. Adding casein and glutamine to a diet containing slowly digestible starch (SDS) did not improve feed conversion ratio (FCR), but when casein and glutamine were added to a diet containing rapidly digestible starch, FCR improved. These results may be related to the efficiency of protein deposition, which is affected by insulin levels. The energy supply to the intestinal wall may also be involved. Vaugelade et al. (1994) stated that intestinal transport of absorbed nutrients coincides with their partial catabolism in the gut. The gastrointestinal tract consumes approximately 20% of all dietary energy to support digestive and absorptive processes. Therefore, metabolic activity of the small intestine also affects the supply of nutrients to other tissues in the body (Cant et al., 1996). Glutamine and glucose are preferentially used to provide energy for the small intestine (Fleming et al., 1991, 1997), but other amino acids have Abbreviation Key: DC = digestion coefficient; FCR = feed conversion ratio; PC = pea-corn; RDS = rapidly digestible starch; SDS = slowly digestible starch; TC = tapioca-corn. 279

2 280 WEURDING ET AL. also been mentioned as energy sources for gut tissues. Diets in which all starch is digested in the upper small intestine will not provide the lower part with glucose for its energy demands. In that case, more amino acids may be oxidized for that purpose. Diets containing starch that is partly digested in the lower small intestine (slowly digestible starch) supply that part with glucose, thereby sparing amino acids from being oxidized. Our hypothesis is that the effect of starch digestion rate on performance is more pronounced with lower amino acid levels in the diet. The objective of this experiment was to investigate whether the effect of amino acid supply on performance of broiler chickens depended on the kinetics of starch digestion. MATERIALS AND METHODS Birds and Housing An experiment was performed with 6,800 sexed 1-d-old male and female broiler chickens of the Cobb 500 strain. 3 Chicks were in two compartments. For this experiment, 20 pens per compartment were used, and in each pen 85 male and 85 female chicks were housed. Both compartments were divided in two blocks of 10 pens each. From Days 0 to 9, chicks were fed a starter diet containing peas, tapioca, and corn. From Days 9 to 30, chicks were fed one of 10 diets, which were randomly assigned to a pen in each block. A photoperiod of 23 h light and 1 h dark was used, and chicks had unrestricted access to feed and water. The experimental protocol was in agreement with the standards for animal experiments and was approved by the Ethical Committee of De Schothorst. Average BW at the start of the experimental period (Day 9) was 230 g. This experiment was part of a larger experiment in which the response of birds to varying levels of slowly digestible starch was investigated (R. E. Weurding, H. Enting, and M. W. A. Verstegen, unpublished data). Diets Three starch sources were selected for the experiment: peas, corn, and tapioca pellets. The selection was based on average in vitro starch digestion rates measured in our laboratory. Pea starch is digested slowly (0.6/h), corn starch gradually (1.0/h), and starch from tapioca pellets is digested rapidly (1.7/h). On the basis of the in vitro starch digestion rate of the feedstuffs, diets with two different in vitro starch digestion rates were formulated: a peacorn diet (PC) and a tapioca-corn diet (TC). Within each in vitro starch digestion rate, feeds with five different digestible lysine levels were made so that two digestion rates five lysine levels were obtained. Feeds with intermediate lysine levels were produced by mixing feeds with the highest and lowest lysine levels. The minimum ratio between digestible essential amino acids and digestible lysine in 3 Cobroed, Lievelde, The Netherlands. all diets met or exceeded amino acid standards used in Dutch practice. Diets were conditioned with steam at 55 to 58 C for ±10 s and subsequently pelleted through a mm die (exit temperature of pellets varied from 58 to 62 C). Compositions of the diets are given in Table 1. Analysis Each experimental diet was analyzed for dry matter, nitrogen (Dumas), and starch (Brunt et al., 1998). Two diets with 9.75 g digestible lysine/kg (PC3 and TC3) were additionally analyzed for particle size distribution, pellet quality, starch (Englyst et al., 1992), and in vitro starch digestion (Weurding et al., 2001). Pellet Quality. Pellet quality (Table 2) was measured by means of durability, hardness, and percentage of fines (Payne et al., 1994). Percentage of fines is determined by screening the pellets. The mesh width was slightly less than the pellet diameter. Durability is the most important aspect of pellet quality and is the ability of pellets to withstand the stresses of handling and delivery without breaking up. Durability was measured by the Pfost tumbling can method. In this method 500 g of screened pellets was tumbled for 10 min at 50 rpm. The sample was then screened again, and the whole pellets were weighed. The percentage of whole pellets remaining was expressed as the durability. Hardness is important to avoid breakdown due to pressure in bulk bins and was measured with the Schleuniger test apparatus (Beumer and Vooijs, 1993). Individual pellets were placed between a moving ram and a flat anvil. The moving ram was pushed against the pellet by an electrically driven spindle with increasing force. The force needed to fracture the pellet was recorded by a force transducer and registered. Particle Size Distribution. Particle size distribution (Table 2) was determined by wet sieve analysis in which 50 g material was put in a beaker, and 250 ml demineralized water was added. After 60 min of soaking, the mix of feed and water was stirred for another 60 min at 500 rpm and subsequently put on top of a set of six sieves (2.5, 2.0, 1.4, 1.0, 0.6, and 0.1 mm). Thirty liters of tap water was sprayed through the system, during which the sieves were vibrated with an amplitude of 1 mm (without interruptions). Vibration was stopped 1 min after the tap was closed. Residues on top of each sieve were determined. In Vitro Starch Digestion. In vitro starch digestion was determined as described by Weurding et al. (2001). In this procedure, which simulates the consecutive digestive processes in the various parts of the broiler alimentary tract, test tubes containing the feed sample, glass balls, a mixture of digestive enzymes and, a buffer solution were incubated in a shaking water bath (37 C, two tubes per sample). After each of nine incubation times (0.25, 0.50, 0.75, 1, 2, 3, 4, 5, and 6 h), aliquots were taken from the tubes, and the amount of released glucose was measured colorimetrically. Starch digestion coefficients (DC) were calculated for each incubation time. Rapidly digestible starch (RDS p ) and slowly digestible starch (SDS p ) fractions for poultry were

3 STARCH DIGESTION RATE AND AMINO ACID LEVEL 281 TABLE 1. Composition of experimental diets Ingredients (g/kg) PC-1 1 PC-5 TC-1 TC-5 Peas Tapioca Corn Sorghum Wheat Soybeans, extracted Rape seed, extracted Sunflower seed, extracted Potato protein Feathermeal, hydrolyzed Fish meal Soybean oil Animal fat Formic acid L-Lysine (25%) L-Lysine + DL-methionine ( %) DL-Methionine (10%) L-Lysine + L-tryptophane (18 + 5%) Ground limestone Monocalcium phosphate Sodium chloride Vitamin-mineral premix Anticoccidial premix Phytase premix Calculated nutrient composition (g/kg) AME broilers (MJ/kg) Dry matter Ash Crude protein Crude fat Crude fiber Starch Digestible lysine Digestible methionine and cysteine Digestioble threonine Digestible tryptophane Digestible arginine Digestible valine Digestible isoleucine Sodium PC1 = pea-corn diet with 8.5 g digestible lysine/kg; PC5 = pea-corn diet with 11.0 g digestible lysine/kg; TC1 = tapioca-corn diet with 8.5 g digestible lysine/kg; TC5 = tapioca-corn diet with 11.0 g digestible lysine/ kg. 2 This premix contained the following microelements (mg/kg): Mn, 7,000; Zn, 3,700; Fe, 4,500; Cu, 1,200; I, 100; Se, 15; vitamin A, 1,000,000 IU; vitamin D 3, 200,000 IU; vitamin E, 2,500 IU; menadione, 125; thiamin, 50; riboflavin, 500; pyridoxine, 300; cyanocobalamin, 1.5; nicotinic acid, 4,000; folic acid, 100; d-pantothenic acid, 800; choline, 20,000; biotin, This premix contained (mg/kg): salinomycin-na, 12, This premix contained 100,000 phytase units/kg. 5 Analyzed values. calculated from the DC as measured after 2 and 4 h (DC 2 and DC 4 ). In vitro starch digestion after 2 and 4 h correlated best with in vivo starch digestion until the posterior jejunum and the posterior ileum, respectively (Weurding et al., 2001). RDS p (%) was defined as 1.16 DC ; SDS p = 1.29 DC RDS; digestible starch (DS p ) = RDS p + SDS p. The DC was plotted against time (t) to give an exponential rate curve (DC t = D (1 e k(d) t ) from which the potentially digestible starch fraction (D, %) and the starch digestion rate (k d, /h) were estimated. The D fraction TABLE 2. Particle size distribution and pellet quality of experimental diets, differing in starch sources Particle size distribution (%) Pellet quality MPS 1 (mm) >2.0 mm >1.4 mm >0.6 mm Hardness (N) % Fines Durability (%) PC TC MPS = mean particle size; PC3 = pea-corn diet with 9.75 g digestible lysine/kg; TC3 = tapioca-corn diet with 9.75 g digestible lysine/kg.

4 282 WEURDING ET AL. TABLE 3. Starch characteristics of pea-corn diets and tapioca-corn diets with 9.75 g digestible lysine/kg (PC3 and TC3, respectively) 1 Starch RDS p SDS p DS p D k d Diet (g/kg) (g/kg starch) (g/kg starch) (g/kg starch) (%) (/h) PC TC , Potential starch digestibility (D) and fractional starch digestion rate (k d ) were estimated from in vitro starch digestion coefficients at different incubation times (DC t ) according to DC t = D (1 e k(d) t ). Rapidly digestible starch (RDS p ), slowly digestible starch (SDS p ), and digestible starch (DS p ) were calculated as explained in the text. represents the asymptote of the digestion curve, and the k d determines the steepness of the curve (a higher k d means that the curve is steeper). Statistical Analysis Data for feed intake, weight gain, and FCR were analyzed by ANOVA. The effect of starch source (SS), digestible lysine content (LYS, LYS 2, LYS 3, and LYS 4 ), and the interaction between these factors on performance were tested according to the following model: Y ijkl = µ + W9 + BLOCK i + SS j + LYS k + LYS 2 k + LYS 3 k + LYS 4 k + (SS * LYS*) jk + e ijkl where Y = weight gain, feed intake, or feed conversion ratio; µ = overall mean; W9 = covariate, BW at Day 9 (start of experiment); BLOCK = block (i = 1,, 4); SS = starch source (j = 1, 2); LYS = digestible lysine content (k = 1,, 5); SS LYS = interaction between SS and LYS, LYS 2, LYS 3, and LYS 4 ; and e = residual error term. RESULTS In Vitro Starch Digestion Diets PC3 and TC3 had estimated potentially digestible starch fractions of 100% (Table 3). The contrast in starch digestion rate (k d ) between PC3 (1.05/h) and TC3 (1.99/ h) was as planned. The goodness-of-fit of starch digestion curves is shown in Figure 1. Predicted total starch digestion FIGURE 1. In vitro starch digestion of a pea-corn diet (PC3) and a tapioca-corn diet (TC3) at various incubation times and starch digestion curves fitted to the model (DC t = D (1 e k(d)*t ). n = 2. was high for both diets (98% for PC3 and 104% for TC3). The predicted amount of slowly digestible starch was 183 g/kg starch (61 g/kg feed) for PC3 and 132 g/kg starch (44 g/kg feed) for TC3. Starch Digestion Rate, Amino Acid Content, and Performance Days Nine to Thirty. Feed intake was not affected by starch source in the diet (P > 0.05). For each lysine level, weight gain and FCR were consequently higher and lower, respectively, for birds on PC diets (slowly digestible starch) than for birds on TC diets (rapidly digestible starch) (P < 0.01). Average weight gain (across lysine treatments) of birds on PC and TC diets was 1,426 and 1,400 g, respectively. FCR was on average 1.55 and 1.58 for birds on PC and TC diets, respectively. Adding digestible lysine to the diet reduced feed intake according to a second degree line (linear component: P < 0.01 and quadratic component: P < 0.10, Table 4). Pooled feed intakes over PC and TC diets decreased from 2,245 g at 8.50 g digestble lysine to 2,193 g at g digestible lysine (Table 5). Pooled weight gains over PC and TC diets increased linearly from 1,397 to 1,431 g with increasing digestible lysine contents (P < 0.01). FCR in PC and TC diets decreased according to a second degree line (linear component: P < 0.01 and quadratic component: P < 0.10) from at 8.50 g digestible lysine to at g digestible lysine (Table 5). Days Nine to Eighteen. Feed intake was lower (646 vs. 659 g) and weight gain higher (474 vs. 467 g) for birds on PC diets than those on TC diets (P < 0.01). Adding digestible lysine to the diet reduced feed intake according to a second degree line (linear component: P < 0.01 and quadratic component: P < 0.10). Pooled feed intakes decreased from 662 to 649 g with increasing digestible lysine levels (Table 5). Weight gain increased linearly (P < 0.01) with increasing digestible lysine levels across PC and TC diets from 458 to 480 g. An interaction between starch source and digestible lysine content on FCR was observed (P < 0.05). FCR was lower for birds on PC diets than those on TC diets, and the difference was most pronounced with lower digestible lysine contents in the diets. Days Eighteen to Thirty. The main effect of starch source was significant for feed intake, weight gain, and FCR (Table 4). Pooled feed intake (P < 0.10) and weight gain (P < 0.01) across digestible lysine levels were higher and pooled FCR was lower (P < 0.01) for birds on PC diets than those on TC diets. Adding digestible lysine to the

5 STARCH DIGESTION RATE AND AMINO ACID LEVEL 283 TABLE 4. Effect of starch source (SS) and digestible lysine content on performance of broiler chickens Digestible lysine content (g/kg) Effect Period SS SEM SS Lys Lys 2 SS Lys Days 9 30 Weight gain, g PC 1 1,417 1,406 1,425 1,437 1,444 9 ** ** NS NS TC 1,376 1,395 1,400 1,413 1,418 Feed intake, g PC 2,248 2,206 2,198 2,204 2, NS ** NS TC 2,241 2,231 2,201 2,212 2,186 FCR PC ** ** 0.11 TC Days 9 18 Weight gain, g PC ** ** 0.12 NS TC Feed intake, g PC ** ** NS TC FCR PC ** ** ** * TC Days Weight gain, g PC ** * NS NS TC Feed intake, g PC 1,595 1,562 1,554 1,569 1,558 8 ** NS NS TC 1,574 1,564 1,548 1,567 1,533 FCR PC ** ** NS NS TC PC = pea-corn diet; TC = tapioca-corn diet; FCR = feed conversion ratio. 2 n = 4 (four replicates per treatment, each replicate is a pen containing 85 male and 85 female birds). ** = P < 0.01; * = P < 0.05; = P < 0.10; NS = not significant (P > 0.10). diet reduced feed intake (P < 0.01), increased weight gain (P < 0.05), and decreased FCR (P < 0.01) linearly when pooled across PC and TC diets. DISCUSSION The positive effect of slowly digestible starch on performance was clearly confirmed in this experiment. Birds on PC diets grew faster and more efficiently than those on TC diets (P < 0.01) (Table 4). However, numerically the improvement was only 2%. It should be realized that we assumed that diets only differed in starch digestion rate and digestible lysine levels. Confounding differences in diets such as digestibility of components other than starch, N-retention, and protein efficiency ratio were assumed to be negligible. Riesenfeld et al. (1982) observed that more than one-third of the glucose absorbed during a meal is converted to lactate in the intestinal wall. This way, the peak glucose influx is buffered. Lactate from the intestinal wall and from the muscles is delivered to the liver where it can be converted to glucose again or to fatty acids in situations of excessive glucose. It is likely that more glucose is converted to lactate after rapid starch digestion than after slow starch digestion. This may partly explain the positive effect of slowly digestible starch on feed efficiency. Within starch treatments, feed intake was higher for birds on the diets with 8.50 g/kg digestible lysine compared to diets with higher levels of digestible lysine. This effect appeared in the second period (Days 18 to 30) and was also observed for the whole experimental period. A curvilinear relation between digestible lysine content and feed intake was observed from Days 9 to 30 (Table 5). An interaction between starch source (starch digestion rate) and digestible lysine content was observed for FCR from Days 9 to 18 (P < 0.05). No interaction between these effects was observed for weight gain. We (unpublished data) observed a similar interaction between starch digestion rate and digestible amino acid content on FCR of broiler chickens for untreated and expander-treated peas and corn during the grower phase (Days 14 to 30; P < 0.05). Birds on diets with slowly digestible starch in this previous experiment had lower FCR than birds on diets with rapidly digestible starch. Because the difference in the current experiment was more pronounced for birds on diets with low amino acid levels, this finding indicates that amino acid supply and glucose supply are unbalanced in diets with rapidly digestible starch. This result may mean that a diet with synchronized starch and protein digestion (e.g., a more gradual starch digestion) results in better performance. The explanation may be found in hormonal responses to glucose absorption, which affect protein deposition. It may be hypothesized that gradual starch digestion results in a lower, but longer lasting insulin peak than rapid starch digestion. This hypothesis is supported by data from Björck et al. (2000) who reported a correlation between the glycemic index and the insulinemic index. Elevated insulin levels are required for amino acid transport and uptake by body cells (Fox, 1996). On the other hand, asynchrony of starch and protein digestion may increase the oxidation of amino acids to meet the energy demand of gut tissues (Vaugelade et al., 1994; Fleming et al., 1997). When glucose is metabolized in the gut tissues of the posterior part of the small intestine, as may be the case when diets with slowly digestible starch are fed, amino acids may be spared and can thus be used for muscle growth. Both weight gain and FCR improved with higher digestible lysine contents in the diet. Although 11.0 g/kg

6 284 WEURDING ET AL. TABLE 5. Effect of starch source and digestible lysine content on performance of broiler chickens (pooled values) Digestible lysine content (g/kg) Starch source SEM 1 PC TC SEM 1 Days 9 30 Weight gain, g 1,397 1,401 1,412 1,425 1, ,426 1,400 3 Feed intake, g 2,245 2,218 2,199 2,208 2, ,211 2,214 3 FCR 2 1, Days 9 18 Weight gain, g Feed intake, g FCR Days Weight gain, g Feed intake, g 1,585 1,563 1,551 1,568 1, ,568 1,557 3 FCR , See Table 4 for statistics. 2 Feed conversion ratio. digestible lysine is well above the normal requirement for growing broiler chickens, the lysine requirement may have been higher for the birds in this experiment. Some leg problems occurred in the preliminary period. These were associated with a growth depression of the birds. After a quick recovery following the supply of an additional vitamin-mineral mix via water bowls, the birds compensated for the growth depression and therefore required more lysine than in a normal situation. This may explain that weight gain and FCR were still improving with the last lysine step. However, improvement with this last step was only marginal compared to the other steps. Our data suggest that the plateau value for FCR is not the same for birds on PC and TC diets. This finding implies that the improved protein efficiency due to slowly digestible starch may not be the only factor explaining the improved feed efficiency. Energy efficiency may also be improved by feeding slowly digestible starch. A more or less continuous glucose supply enables more direct utilization of glucose for processes in the body. Less energy consuming conversions of glucose to glycogen or fat and back to glucose are needed in that situation. Based on the results of this experiment we conclude that feeding slowly digestible starch improves feed efficiency when compared to feeding rapidly digestible starch. The major part of this improvement can be attributed to an improved protein utilization. The protein-sparing effect, however, cannot explain the total improvement, and it can be hypothesized that energy utilization is also improved because of prolonged elevated plasma glucose levels. REFERENCES Beumer, H., and A. J. Vooijs TNO-report number B Optimaliseren van het pelleteren van veevoeders. Fase II. Deel II.: experimenteel onderzoek. TNO-IGMB, Wageningen, The Netherlands. Björck, I., H. Liljeberg, and E. Östman Low glycaemicindex foods. Br. J. Nutr. 83(Suppl. 1):S149 S155. Brand-Miller, J. C The glycemic index of foods: Implications for the food industry. Food Austr. 51: Brunt, K., P. Sanders, and T. Rozema The enzymatic determination of starch in food, feed and raw materials of the starch industry. Starch 50: Cant, J. P., B. W. McBride, and W. J. Croom, Jr The regulation of intestinal metabolism and its impact on whole animal energetics. J. Anim. Sci. 74: Englyst, H. N., S. M. Kingman, and J. H. Cummings Classification and measurement of nutritionally important starch fractions. Eur. J. Clin. Nutr. 46(Suppl. 2):S33 S50. Fleming, S. E., M. D. Fitch, S. DeVries, M. L. Liu, and C. Kight Nutrient utilization by cells isolated from rat jejunum, cecum and colon. J. Nutr. 121: Fleming, S. E., K. L. Zambell, and M. D. Fitch Glucose and glutamine provide similar proportions of energy to mucosal cells of rat small intestine. Am. J. Physiol. 273:G Fox, S. I Page 588 in Human Physiology. Wm. C. Brown Publishers, Dubuque, IA. Payne, J., W. Rattink, T. Smith, and T. Winowiski Pages in The Pelleting Handbook. Borregaard Lignotech, Sarpsborg, Norway. Riesenfeld, G., A. Geva, and S. Hurwitz Glucose homeostasis in the chicken. J. Nutr. 112: Vaugelade, P., L. Posho, B. Darcy-Vrillon, F. Bernard, M. T. Morel, and P. H. Duee Intestinal oxygen uptake and glucose metabolism during nutrient absorption in the pig. Proc. Soc. Exp. Biol. Med. 207: Weurding, R. E., A. Veldman, W. A. G. Veen, P. J. van der Aar, and M. W. A. Verstegen In vitro starch digestion correlates well with rate and extent of starch digestion in broiler chickens. J. Nutr. 131: