- Email: [email protected]
In Vitro Evaluation of Nonstarch Polysaccharide Digestibility of Feed Ingredients by Enzymes
In Vitro Evaluation of Nonstarch Polysaccharide Digestibility of Feed Ingredients by Enzymes V. Malathi1 and G. Devegowda Department of Poultry Science, University of Agricultural Sciences, Hebbal, Bangalore-560 024, India
(Key words: in vitro assay, digestibility, nonstarch polysaccharide, enzymes, viscosity) 2001 Poultry Science 80:302–305
from those in vitro. Hence, the efficacy of feed enzymes when evaluated under in vitro conditions simulating those existing in the gut would be most accurate. A two-stage in vitro digestion assay, in which the feed sample with enzyme is exposed to different pH and proteolytic enzymes present in the gut and simulating the peptic and pancreatic phases, is preferred to predict intestinal viscosity and bird performance over a wide range of enzyme levels (Bedford and Classen, 1993). With the above points in view, the present study was conducted to screen different feed ingredients for important NSP viz., pentosans, pectins, cellulose, and total NSP, because the NSP profile for most the feed ingredients screened in this study are not available. We used a twostage in vitro digestion assay to evaluate the NSP digestibility of sunflower meal, soybean meal, deoiled rice bran, and a broiler starter diet with three different enzyme preparations; the enzyme combinations most suitable for each of these ingredients were judged.
INTRODUCTION Despite the availability of the agro-industrial by products in abundance at relatively lower prices, their inclusion in poultry feeds is limited because of presence of nonstarch polysaccharides (NSP) that, by increasing the gut viscosity, affect the growth and performance of birds. It is now well documented that enzyme supplementation breaks these polymeric chains into smaller pieces, reduces the gut viscosity, and hence improves the nutritive value of such feedstuffs (Smits and Annison, 1996). The efficacy of feed enzymes depends on their substrate specificity, activity, and stability. Therefore, there is often great difficulty in selecting potentially useful enzymes available in the market. Among the different enzymatic assay methods, viscometry has been proven to best predict the response of chicks fed enzyme-supplemented, barleybased diets (Rotter et al., 1990). Although in vitro extract viscosities have correlated well with in vivo intestinal viscosity (Rotter et al., 1989; Bedford and Classen, 1993), in vivo results may often vary because the conditions prevailing in the gut are entirely different
MATERIALS AND METHODS Screening for Nonstarch Polysaccharides We screened the most commonly used feed ingredients for poultry (corn, sorghum, finger millet, deoiled ricebran,
Received for publication February 7, 2000. Accepted for publication November 21, 2000. 1 To whom correspondence should be addressed: [email protected].
Abbreviation Key: FPU = filter paper unit; NSP = nonstarch polysaccharide.
e-mail:
302
Downloaded from http://ps.oxfordjournals.org/ at Fresno Pacific University on January 15, 2015
+ β-glucanase from Aspergillus aculeatus) by incubating 0.1 g of the sample with 3 mL of a pepsin-HCl mixture (2,000 U pepsin/mL of 0.1N HCl) for 45 min to simulate the peptic phase of bird digestion. A pancreatin-NaHCO3 mixture (2 mg pancreatin/mL of 1 M NaHCO3) was used for 2 h at 40 C to simulate the pancreatic phase. Digestibility was assessed by measuring the relative viscosity of the digesta supernatent and the total sugars released. Enzyme-I produced the least relative viscosity and highest total sugars in sunflower meal, deoiled rice bran, and broiler starter diet, whereas Enzyme-III was very effective in soybean meal subjected to in vitro digestion. The assay was a convenient and rapid method of screening for effective and stable enzymes.
ABSTRACT Some of the commonly used feed ingredients for poultry (corn, sorghum, finger millet, deoiled ricebran, soybean meal, peanut meal, sunflower meal, and rapeseed meal) were screened for pentosans, cellulose, pectin, and total nonstarch polysaccharides. The ingredient in vitro digestibilities by enzymes were evaluated. Cereal samples screened contained mainly pentosans. Pectin content was rich in oilseed meals. Sunflower meal, soybean meal, deoiled rice bran, and a broiler starter diet were subjected to a two-stage in vitro digestion assay with three different enzyme mixtures viz., Enzyme-I (xylanase + cellulase from Trichoderma viridae), Enzyme-II (xylanase + cellulase + β-glucanase from Huminicola insolens), and Enzyme-III (xylanase + cellulase + pectinase
303
IN VITRO NONSTARCH POLYSACCHARIDE DIGESTIBILITY BY ENZYMES TABLE 1. Total pentosan, cellulose, pectin, and total nonstarch polysaccharide (NSP) content of different feed ingredients1
Ingredient
Total pentosan (%)
Cellulose (%)
Pectin (%)
Total NSP (%)
Corn Sorghum Finger millet Deoiled rice bran Soybean meal Peanut meal Sunflower meal Rapeseed meal
5.35 2.77 3.31 10.65 4.21 6.11 11.01 8.85
3.12 4.21 3.03 15.20 5.75 6.55 22.67 14.21
1.00 1.66 1.76 7.25 6.16 11.60 4.92 8.86
9.32 9.75 9.40 59.97 29.02 29.50 41.34 39.79
Each value represents the mean of triplicate analysis; NSP = Nonstarch polysaccharide.
1
Enzymes Three commercial multi-enzyme mixtures used in this study were Enzyme-I2 derived from Trichoderma viridae, Enzyme-II3 derived from Huminicola insolens, and EnzymeIII4 derived from Aspergillus aculeatus.
Enzyme Activity Measurements The xylanase activity in the enzyme samples was determined to be 900 U/g for Enzyme-I, 680 U/g for EnzymeII, and 450 U/g for Enzyme-III, when assayed on oat spelt xylan5 (John and Schmidt, 1988). One unit was equal to the amount of the enzyme required to liberate 1 µM xylose/min at 40 C and pH 4.8. The cellulase activity was determined by using filter paper as the substrate (Wood and Bhat, 1988) and was found to be 12 filter paper units (FPU)/g for Enzyme-I, 18 FPU/g for Enzyme-II, and 4.5 FPU/g for Enzyme-III. One FPU was equal to the amount of enzyme required to release 1 µM glucose/min from 50 mg filter paper at 50 C and pH 4.8. The pectinase activity was assayed using pectin as the substrate (Collmer et al., 1988) and was found to be 4,500 U/g for Enzyme-III and was absent in the other two enzyme preparations. One unit was equal to the amount of
enzyme required to release 1 µM D-galacturonic acid at 30 C and pH 5.3. In addition, Enzyme-II and Enzyme-III had β-glucanase activities of 75 and 50 U/g, respectively, according to the manufacturer (not assayed in the present study).
Two-Stage In Vitro Digestion Samples of sunflower meal (28% protein), soybean meal (48% protein), deoiled ricebran (14% protein), and broiler starter diet (corn 55% + 31.5% soybean meal + 10% sunflower meal + 3.5% mineral mixture) with Enzyme-I (0.1 and 0.2%), Enzyme-II (0.06 and 0.12%), and Enzyme-III (0.12 and 0.24%) were subjected to a two-stage in vitro digestion assay as described by Bedford and Classen (1993) with slight modifications. One-tenth gram of the ground (1 mm) feed samples with different levels of the three enzyme preparations, in duplicate, were incubated with 3 mL of 0.1N HCl containing 2,000 U pepsin/mL for 45 min at 40 C with occasional vortexing to simulate the peptic or gastric phase. After 45 min, 1 mL of 1 M NaHCO3 solution containing 2 mg pancreatin/mL was added and incubated at 40 C for 2 h with occassional vortexing to simulate the pancreatic or intestinal phase. After each incubation period, the contents were centrifuged at 3,000 rpm for 20 min, and the supernatant was stored on ice before analysis.
Sample Measurements At the end of each phase, relative viscosity of the supernatants was measured using an Ostwald viscometer6 (Choct and Annison, 1992). At the end of the pancreatic phase, total sugars released were measured by the phenolsulfuric acid method (Dubois et al., 1956).
Statistical Analysis 2
Nutrizyme-B, Tetragon Chemie Ltd., Bangalore 560064, India. Biofeed plus-CT, Novo Nordisk enzymes Pvt. Ltd., Bangalore 560066, India. 4 Energex-CT, Novo Nordisk Enzymes Pvt. Ltd., Bangalore 560066, India. 5 Sigma Chemical Co., Bangalore 560048, India. 6 Borosil Glass Works Ltd., Mumbai 400018, India. 3
The values for relative viscosity and total sugars released were subjected to two-way and one-way analysis of variance, respectively, according to Snedecor and Cochran (1968), and the means were compared by a multiple-range test, according to Duncan (1955) at P < 0.05.
Downloaded from http://ps.oxfordjournals.org/ at Fresno Pacific University on January 15, 2015
soybean meal, peanut meal, sunflower meal, and rapeseed meal) for total pentosans, cellulose, pectin, and total NSP content. Total pentosan content in the sample was estimated by the orcinol-iron method as described by Frazer et al. (1956). Cellulose content was estimated colorimetrically as per the procedure outlined by Updegroff (1969). Pectin was extracted and estimated as described by Sadasivam and Manickam (1996). Total NSP was quantified colorimetrically as per the procedure outlined by Englyst and Cummings (1988). Each sample was analyzed in triplicate.
304
MALATHI AND DEVEGOWDA
RESULTS AND DISCUSSION
Two-Stage In Vitro Digestion
Screening for Nonstarch Polysaccharides
TABLE 2. Effect of enzymes on mean relative viscosity of sunflower meal, soybean meal, deoiled rice bran, and broiler starter diet subjected to a two-stage in vitro digestion Sunflower meal Enzyme Control Enzyme-I 0.1% 0.2% Enzyme-II 0.06% 0.12% Enzyme-III 0.12% 0.24% Mean Pooled SEM
Soybean meal
Deoiled rice bran
Peptic phase
Pancreatic phase
Mean (n = 2)
Peptic phase
Pancreatic phase
Mean (n = 2)
Peptic phase
Pancreatic phase
Mean (n = 2)
Peptic phase
Pancreatic phase
Mean (n = 2)
1.56a
1.63a
1.59a
1.44a
1.63a
1.53a
1.87a
2.06a
1.97a
1.44a
1.56a
1.50a
1.22c 1.11c
1.44bc 1.38c
1.33cd 1.25d
1.33b 1.22c
1.50b 1.38c
1.42b 1.29c
1.44de 1.38e
1.68cd 1.63d
1.56d 1.50d
1.22cd 1.16d
1.44bc 1.38c
1.33d 1.27d
1.44ab 1.22c
1.56ab 1.44bc
1.50ab 1.33cd
1.22c 1.16cd
1.44bc 1.37c
1.33c 1.26c
1.63bc 1.56cd
1.77bc 1.68cd
1.70bc 1.62cd
1.31bc 1.25cd
1.50ab 1.44bc
1.41abc 1.35bcd
1.44ab 1.38b 1.34z 0.03
1.56ab 1.44bc 1.49y 0.05
1.50ab 1.41bc
1.16cd 1.11d 1.23z 0.02
1.38c 1.22d 1.42y 0.01
1.27c 1.17d
1.75ab 1.68bc 1.62z 0.02
1.88b 1.75bcd 1.78y 0.03
1.81b 1.72bc
1.44a 1.38ab 1.31z 0.03
1.56a 1.50ab 1.48y 0.04
1.50a 1.44abc
Means within a column with no common superscript differ significantly (P < 0.05). Means within a row and feedstuff with no common superscript differ significantly (P < 0.05).
a–e y,z
Broiler starter diet
Downloaded from http://ps.oxfordjournals.org/ at Fresno Pacific University on January 15, 2015
The pentosan, cellulose, pectin, and total NSP contents recorded are presented in Table 1. Pentosan (Arabinoxylan) content in the ingredients screened ranged from 2.77% (sorghum) to 11.01% (sunflower meal). Pentosan content was very high in deoiled ricebran, peanut meal, sunflower meal, and rapeseed meal as compared to other ingredients. Pectin content in different ingredients ranged from 1.0% (corn) to 11.6% (peanut meal). Pectin content also was found to be higher in oilseed meals, which is in accordance with the findings of Carre and Brillouet (1986). Also, the raffinose series of oligosaccharides have been reported to be present at higher levels in legumes than in cereals (Kuo et al., 1988). Because oilseed meals contain more pectin, oligosaccharides, and also pentosan, which are either partially or completely soluble in the gastro-intestinal tract of the bird, one has to be cautious in including them in the diet at higher levels. The cellulose content ranged from 3.03% (finger millet) to 22.67% (sunflower meal). Also, deoiled ricebran and rapeseed meal contained very high levels of cellulose. Because cellulose is an insoluble fraction, emphasis can be given only when the cellulose-rich ingredients are incorporated in the diets at higher levels. Total NSP content was higher in deoiled ricebran, soyabean meal, peanut meal, sunflower meal, and rapeseed meal than other ingredients. Although the values for different NSP recorded are comparable to those reported by other researchers, the methods used in the present study could only be used for routine screening. Any method currently being used for NSP analysis lacks nutritional validity. The choice of the procedure should be the one that simulates the gut conditions. Future research in this area should focus on developing a nutritionally valid procedure for measuring soluble and insoluble NSP in simulated in vivo conditions, and the same should be standardized and adopted worldwide.
The degree of NSP digestibility of feed ingredients subjected to a two-stage in vitro digestion was indirectly assessed by measuring the relative viscosity and the total sugars released. The beneficial effects of feed enzymes are primarily the reduction in the viscosity and, secondarily, the release of sugars. Viscosity reduction is due to breakdown of NSP into smaller polymers thus preventing them from forming viscous networks. Release of sugars caused by exogenous enzymes is due to two reasons. First, the breakdown of NSP led to release of their respective monosaccharides, and second, the breakdown of NSP released the starch within the endosperm, which was exposed to the endogenous amylase, releasing more glucose. Any given enzyme can be viscosity-reducing, sugar-releasing, or both, depending on type and source. Both levels of Enzyme-I and 0.12% Enzyme-II significantly reduced the relative viscosity of the digesta of sunflower meal, deoiled rice bran, and broiler starter diet compared to other enzymes in the peptic and pancreatic phases (Table 2). This result is attributed to the high cellulase and xylanase activities in these enzymes. In case of soybean meal, least relative viscosity was produced by Enzyme-III, which might have been due to the additional pectinase it contained. Although soybean meal was one of the ingredients used in broiler starter diets, the cellulase + xylanase combination gave better results. The possible reason can be attributed to the resultant diet containing more cellulose (5.79%) and pentosan (5.4%) and comparatively less pectin (2.98%). This result agreed with the in vivo results recorded by Suresh and Devegowda (1996). Enzyme-III did not significantly reduce the relative viscosity compared to the controls in sunflower meal and deoiled rice bran. In general, irrespective of the enzymes, the relative viscosity in the pancreatic phase was more than that in the peptic phase, which is due to that fact that NSP becomes more soluble in higher pH medium, increasing the viscosity (Moore and Hoseney, 1990). This observation is in
IN VITRO NONSTARCH POLYSACCHARIDE DIGESTIBILITY BY ENZYMES
305
TABLE 3. Effect of enzymes on total sugars released from sunflower meal, soybean meal, deoiled rice bran, and broiler starter diet subjected to a two-stage in vitro digestion Total sugars released (mg/mL) Enzyme Control Enzyme-I 0.1% 0.2% Enzyme-II 0.06% 0.12% Enzyme-III 0.12% 0.24% Pooled SEM
Sunflower meal
Soybean meal
Deoiled rice bran
Broiler starter diet
4.24d
3.39e
4.69d
5.16e
6.46b 7.26a
4.01d 4.65ab
6.27ab 6.67a
6.34bc 7.20a
6.25b 6.79ab
4.11cd 4.56abc
5.95b 6.30ab
5.98cd 6.75ab
4.69cd 5.28c 0.18
4.32bcd 4.82a 0.13
5.24c 5.89b 0.12
5.32e 5.66de 0.15
Means within a column with no common superscript differ significantly (P < 0.05). SEM = Standard error of mean; N for each mean = 2.
a–e
REFERENCES Bedford, M. R., and H. L. Classen, 1993. An in vitro assay for prediction of broiler intestinal viscosity and growth when fed rye-based diets in the presence of exogenous enzymes. Poultry Sci. 72:137–143. Carre, B., and J. M. Brillouet, 1986. Yield and cell wall residues isolated from various feedstuffs used for non-ruminant farm animals. J. Sci. Food Agric. 37:341–351. Choct, M., and G. Annison, 1992. Anti-nutritive effect of wheat pentosans in broiler chickens: Role of viscosity and gut microflora. Br. Poult. Sci. 33:821–834. Collmer, A., J. L. Ried, and M. S. Mount, 1988. Methods Enzymol. A 160:329–334.
Dubois, M., K. A. Gilles, J. K. Hamilton, P. A. Rebers, and F. Smith, 1956. Colorimetric method for determination of sugars and related substances. Anal. Biochem. 28:350. Duncan, D. D., 1955. Multiple range and multiple ‘F’ test. Biometrics 11:1–42. Dusel, G., H. Kluge, and H. Jeroch, 1998. Xylanase supplementation of wheat-based rations for broilers: Influence of wheat characteristics. J. Appl. Poult. Res. 7:119–131. Englyst, H. N., and J. H. Cummings, 1988. Improved method for measurement of dietary fibre as nonstarch polysaccharides in plant foods. J. Assoc. Off. Anal. Chem. 71:808–814. Frazer, J. R., M. B. Bravo, and D. C., Holmes, 1956. The proximate analysis of wheat flour carbohydrates. I—methods and scheme of analysis. J. Sci. Food. Agric. 7:577–589. John, M., and J. Schmidt, 1988. Xylanases and β-xylosidases of Trichoderma longinorum. Methods Enzymol. A 160:662–670. Kuo, J. M., J. F. Vanmiddlesworth, and W. J. Wolt, 1988. Content of raffinose oligosaccharides and sucrose in various plant seeds. J. Agric. Food Chem. 36:32–36. Moore, A. M., and R. C. Hoseney, 1990. Factors affecting the viscosity of flour-water extracts. Cereal Chem. 67:78–80. Rotter, B. A., R. R. Marquardt, W. Guenter, and G. H. Crow, 1990. Evaluation of three enzymatic methods as predictors of in vivo response to enzyme supplementation of barley-based diets when fed to young chicks. J. Sci. Food Agric. 50:19–27. Rotter, B. A., R. R. Marquardt, W. Guenter, C. Biliaderis, and C. W. Newman, 1989. In vitro viscosity measurements of barley extracts as predictors of growth responses in chicks fed barleybased diets supplemented with a fungal enzyme preparation. Can. J. Anim. Sci. 69:431–439. Sadasivam, S., and A. Manickam, 1996. Biochemical Methods. 2nd ed. New Age International Publishers, New Delhi, India. Smits, C.H.M., and G. Annison, 1996. Nonstarch plant polysaccharides in broiler nutrition-towards a physiologically valid approach to their determination. World’s Poult. Sci. J. 52:203–221. Snedecor, G. W., and S. W. Cochran, 1968. Statistical Methods. 7th ed. Iowa State University Press, Ames, IA. Suresh, S. C., and G. Devegowda, 1996. Influence of dietary enzymes on the performance of broilers. Page 228 in: Proceedings of the 20th World’s Poultry Congress, New Delhi, India. Updegroff, D. M., 1969. Semi-micro determination of cellulose in biological material. Anal. Biochem. 32:420–424. Wood, T. M., and K. M. Bhat, 1988. Methods for measuring cellulase activities. Methods Enzymol. A 160:87–112.
Downloaded from http://ps.oxfordjournals.org/ at Fresno Pacific University on January 15, 2015
agreement with in vitro and in vivo results reported by Bedford and Classen (1993) and the in vivo results of Dusel et al. (1998). Similar to the relative viscosity, the total sugars released was highest with 0.2% Enzyme-I in all feed samples, except soybean meal wherein 0.24% Enzyme-III released the most total sugars (Table 3). The 0.1% Enzyme-I treatment, which was as effective as 0.2% in reducing relative viscosity, was found to release significantly less total sugars than that of the 0.2% level, which might be due to the fact that EnzymeI was primarily a viscosity-reducing enzyme and secondarily a sugar-releasing enzyme. In conclusion, a xylanase + cellulase combination has been found to be superior in case of sunflower meal, deoiled ricebran, and broiler starter diets; for soybean meal, a pectinase-rich enzyme preparation has been found to give the best results. However, when soybean meal is included at higher levels in the feeds, a xylanase + cellulase + pectinase combination might prove to be better than xylanase + cellulase alone; these results need to be validated by in vivo results. This digestion assay simulating the gut conditions of poultry can be used as a simple and rapid screening technique of feed enzymes for their efficacy and stability, and also the best effective dosage level of an enzyme can be fixed.
(Key words: in vitro assay, digestibility, nonstarch polysaccharide, enzymes, viscosity) 2001 Poultry Science 80:302–305
from those in vitro. Hence, the efficacy of feed enzymes when evaluated under in vitro conditions simulating those existing in the gut would be most accurate. A two-stage in vitro digestion assay, in which the feed sample with enzyme is exposed to different pH and proteolytic enzymes present in the gut and simulating the peptic and pancreatic phases, is preferred to predict intestinal viscosity and bird performance over a wide range of enzyme levels (Bedford and Classen, 1993). With the above points in view, the present study was conducted to screen different feed ingredients for important NSP viz., pentosans, pectins, cellulose, and total NSP, because the NSP profile for most the feed ingredients screened in this study are not available. We used a twostage in vitro digestion assay to evaluate the NSP digestibility of sunflower meal, soybean meal, deoiled rice bran, and a broiler starter diet with three different enzyme preparations; the enzyme combinations most suitable for each of these ingredients were judged.
INTRODUCTION Despite the availability of the agro-industrial by products in abundance at relatively lower prices, their inclusion in poultry feeds is limited because of presence of nonstarch polysaccharides (NSP) that, by increasing the gut viscosity, affect the growth and performance of birds. It is now well documented that enzyme supplementation breaks these polymeric chains into smaller pieces, reduces the gut viscosity, and hence improves the nutritive value of such feedstuffs (Smits and Annison, 1996). The efficacy of feed enzymes depends on their substrate specificity, activity, and stability. Therefore, there is often great difficulty in selecting potentially useful enzymes available in the market. Among the different enzymatic assay methods, viscometry has been proven to best predict the response of chicks fed enzyme-supplemented, barleybased diets (Rotter et al., 1990). Although in vitro extract viscosities have correlated well with in vivo intestinal viscosity (Rotter et al., 1989; Bedford and Classen, 1993), in vivo results may often vary because the conditions prevailing in the gut are entirely different
MATERIALS AND METHODS Screening for Nonstarch Polysaccharides We screened the most commonly used feed ingredients for poultry (corn, sorghum, finger millet, deoiled ricebran,
Received for publication February 7, 2000. Accepted for publication November 21, 2000. 1 To whom correspondence should be addressed: [email protected].
Abbreviation Key: FPU = filter paper unit; NSP = nonstarch polysaccharide.
e-mail:
302
Downloaded from http://ps.oxfordjournals.org/ at Fresno Pacific University on January 15, 2015
+ β-glucanase from Aspergillus aculeatus) by incubating 0.1 g of the sample with 3 mL of a pepsin-HCl mixture (2,000 U pepsin/mL of 0.1N HCl) for 45 min to simulate the peptic phase of bird digestion. A pancreatin-NaHCO3 mixture (2 mg pancreatin/mL of 1 M NaHCO3) was used for 2 h at 40 C to simulate the pancreatic phase. Digestibility was assessed by measuring the relative viscosity of the digesta supernatent and the total sugars released. Enzyme-I produced the least relative viscosity and highest total sugars in sunflower meal, deoiled rice bran, and broiler starter diet, whereas Enzyme-III was very effective in soybean meal subjected to in vitro digestion. The assay was a convenient and rapid method of screening for effective and stable enzymes.
ABSTRACT Some of the commonly used feed ingredients for poultry (corn, sorghum, finger millet, deoiled ricebran, soybean meal, peanut meal, sunflower meal, and rapeseed meal) were screened for pentosans, cellulose, pectin, and total nonstarch polysaccharides. The ingredient in vitro digestibilities by enzymes were evaluated. Cereal samples screened contained mainly pentosans. Pectin content was rich in oilseed meals. Sunflower meal, soybean meal, deoiled rice bran, and a broiler starter diet were subjected to a two-stage in vitro digestion assay with three different enzyme mixtures viz., Enzyme-I (xylanase + cellulase from Trichoderma viridae), Enzyme-II (xylanase + cellulase + β-glucanase from Huminicola insolens), and Enzyme-III (xylanase + cellulase + pectinase
303
IN VITRO NONSTARCH POLYSACCHARIDE DIGESTIBILITY BY ENZYMES TABLE 1. Total pentosan, cellulose, pectin, and total nonstarch polysaccharide (NSP) content of different feed ingredients1
Ingredient
Total pentosan (%)
Cellulose (%)
Pectin (%)
Total NSP (%)
Corn Sorghum Finger millet Deoiled rice bran Soybean meal Peanut meal Sunflower meal Rapeseed meal
5.35 2.77 3.31 10.65 4.21 6.11 11.01 8.85
3.12 4.21 3.03 15.20 5.75 6.55 22.67 14.21
1.00 1.66 1.76 7.25 6.16 11.60 4.92 8.86
9.32 9.75 9.40 59.97 29.02 29.50 41.34 39.79
Each value represents the mean of triplicate analysis; NSP = Nonstarch polysaccharide.
1
Enzymes Three commercial multi-enzyme mixtures used in this study were Enzyme-I2 derived from Trichoderma viridae, Enzyme-II3 derived from Huminicola insolens, and EnzymeIII4 derived from Aspergillus aculeatus.
Enzyme Activity Measurements The xylanase activity in the enzyme samples was determined to be 900 U/g for Enzyme-I, 680 U/g for EnzymeII, and 450 U/g for Enzyme-III, when assayed on oat spelt xylan5 (John and Schmidt, 1988). One unit was equal to the amount of the enzyme required to liberate 1 µM xylose/min at 40 C and pH 4.8. The cellulase activity was determined by using filter paper as the substrate (Wood and Bhat, 1988) and was found to be 12 filter paper units (FPU)/g for Enzyme-I, 18 FPU/g for Enzyme-II, and 4.5 FPU/g for Enzyme-III. One FPU was equal to the amount of enzyme required to release 1 µM glucose/min from 50 mg filter paper at 50 C and pH 4.8. The pectinase activity was assayed using pectin as the substrate (Collmer et al., 1988) and was found to be 4,500 U/g for Enzyme-III and was absent in the other two enzyme preparations. One unit was equal to the amount of
enzyme required to release 1 µM D-galacturonic acid at 30 C and pH 5.3. In addition, Enzyme-II and Enzyme-III had β-glucanase activities of 75 and 50 U/g, respectively, according to the manufacturer (not assayed in the present study).
Two-Stage In Vitro Digestion Samples of sunflower meal (28% protein), soybean meal (48% protein), deoiled ricebran (14% protein), and broiler starter diet (corn 55% + 31.5% soybean meal + 10% sunflower meal + 3.5% mineral mixture) with Enzyme-I (0.1 and 0.2%), Enzyme-II (0.06 and 0.12%), and Enzyme-III (0.12 and 0.24%) were subjected to a two-stage in vitro digestion assay as described by Bedford and Classen (1993) with slight modifications. One-tenth gram of the ground (1 mm) feed samples with different levels of the three enzyme preparations, in duplicate, were incubated with 3 mL of 0.1N HCl containing 2,000 U pepsin/mL for 45 min at 40 C with occasional vortexing to simulate the peptic or gastric phase. After 45 min, 1 mL of 1 M NaHCO3 solution containing 2 mg pancreatin/mL was added and incubated at 40 C for 2 h with occassional vortexing to simulate the pancreatic or intestinal phase. After each incubation period, the contents were centrifuged at 3,000 rpm for 20 min, and the supernatant was stored on ice before analysis.
Sample Measurements At the end of each phase, relative viscosity of the supernatants was measured using an Ostwald viscometer6 (Choct and Annison, 1992). At the end of the pancreatic phase, total sugars released were measured by the phenolsulfuric acid method (Dubois et al., 1956).
Statistical Analysis 2
Nutrizyme-B, Tetragon Chemie Ltd., Bangalore 560064, India. Biofeed plus-CT, Novo Nordisk enzymes Pvt. Ltd., Bangalore 560066, India. 4 Energex-CT, Novo Nordisk Enzymes Pvt. Ltd., Bangalore 560066, India. 5 Sigma Chemical Co., Bangalore 560048, India. 6 Borosil Glass Works Ltd., Mumbai 400018, India. 3
The values for relative viscosity and total sugars released were subjected to two-way and one-way analysis of variance, respectively, according to Snedecor and Cochran (1968), and the means were compared by a multiple-range test, according to Duncan (1955) at P < 0.05.
Downloaded from http://ps.oxfordjournals.org/ at Fresno Pacific University on January 15, 2015
soybean meal, peanut meal, sunflower meal, and rapeseed meal) for total pentosans, cellulose, pectin, and total NSP content. Total pentosan content in the sample was estimated by the orcinol-iron method as described by Frazer et al. (1956). Cellulose content was estimated colorimetrically as per the procedure outlined by Updegroff (1969). Pectin was extracted and estimated as described by Sadasivam and Manickam (1996). Total NSP was quantified colorimetrically as per the procedure outlined by Englyst and Cummings (1988). Each sample was analyzed in triplicate.
304
MALATHI AND DEVEGOWDA
RESULTS AND DISCUSSION
Two-Stage In Vitro Digestion
Screening for Nonstarch Polysaccharides
TABLE 2. Effect of enzymes on mean relative viscosity of sunflower meal, soybean meal, deoiled rice bran, and broiler starter diet subjected to a two-stage in vitro digestion Sunflower meal Enzyme Control Enzyme-I 0.1% 0.2% Enzyme-II 0.06% 0.12% Enzyme-III 0.12% 0.24% Mean Pooled SEM
Soybean meal
Deoiled rice bran
Peptic phase
Pancreatic phase
Mean (n = 2)
Peptic phase
Pancreatic phase
Mean (n = 2)
Peptic phase
Pancreatic phase
Mean (n = 2)
Peptic phase
Pancreatic phase
Mean (n = 2)
1.56a
1.63a
1.59a
1.44a
1.63a
1.53a
1.87a
2.06a
1.97a
1.44a
1.56a
1.50a
1.22c 1.11c
1.44bc 1.38c
1.33cd 1.25d
1.33b 1.22c
1.50b 1.38c
1.42b 1.29c
1.44de 1.38e
1.68cd 1.63d
1.56d 1.50d
1.22cd 1.16d
1.44bc 1.38c
1.33d 1.27d
1.44ab 1.22c
1.56ab 1.44bc
1.50ab 1.33cd
1.22c 1.16cd
1.44bc 1.37c
1.33c 1.26c
1.63bc 1.56cd
1.77bc 1.68cd
1.70bc 1.62cd
1.31bc 1.25cd
1.50ab 1.44bc
1.41abc 1.35bcd
1.44ab 1.38b 1.34z 0.03
1.56ab 1.44bc 1.49y 0.05
1.50ab 1.41bc
1.16cd 1.11d 1.23z 0.02
1.38c 1.22d 1.42y 0.01
1.27c 1.17d
1.75ab 1.68bc 1.62z 0.02
1.88b 1.75bcd 1.78y 0.03
1.81b 1.72bc
1.44a 1.38ab 1.31z 0.03
1.56a 1.50ab 1.48y 0.04
1.50a 1.44abc
Means within a column with no common superscript differ significantly (P < 0.05). Means within a row and feedstuff with no common superscript differ significantly (P < 0.05).
a–e y,z
Broiler starter diet
Downloaded from http://ps.oxfordjournals.org/ at Fresno Pacific University on January 15, 2015
The pentosan, cellulose, pectin, and total NSP contents recorded are presented in Table 1. Pentosan (Arabinoxylan) content in the ingredients screened ranged from 2.77% (sorghum) to 11.01% (sunflower meal). Pentosan content was very high in deoiled ricebran, peanut meal, sunflower meal, and rapeseed meal as compared to other ingredients. Pectin content in different ingredients ranged from 1.0% (corn) to 11.6% (peanut meal). Pectin content also was found to be higher in oilseed meals, which is in accordance with the findings of Carre and Brillouet (1986). Also, the raffinose series of oligosaccharides have been reported to be present at higher levels in legumes than in cereals (Kuo et al., 1988). Because oilseed meals contain more pectin, oligosaccharides, and also pentosan, which are either partially or completely soluble in the gastro-intestinal tract of the bird, one has to be cautious in including them in the diet at higher levels. The cellulose content ranged from 3.03% (finger millet) to 22.67% (sunflower meal). Also, deoiled ricebran and rapeseed meal contained very high levels of cellulose. Because cellulose is an insoluble fraction, emphasis can be given only when the cellulose-rich ingredients are incorporated in the diets at higher levels. Total NSP content was higher in deoiled ricebran, soyabean meal, peanut meal, sunflower meal, and rapeseed meal than other ingredients. Although the values for different NSP recorded are comparable to those reported by other researchers, the methods used in the present study could only be used for routine screening. Any method currently being used for NSP analysis lacks nutritional validity. The choice of the procedure should be the one that simulates the gut conditions. Future research in this area should focus on developing a nutritionally valid procedure for measuring soluble and insoluble NSP in simulated in vivo conditions, and the same should be standardized and adopted worldwide.
The degree of NSP digestibility of feed ingredients subjected to a two-stage in vitro digestion was indirectly assessed by measuring the relative viscosity and the total sugars released. The beneficial effects of feed enzymes are primarily the reduction in the viscosity and, secondarily, the release of sugars. Viscosity reduction is due to breakdown of NSP into smaller polymers thus preventing them from forming viscous networks. Release of sugars caused by exogenous enzymes is due to two reasons. First, the breakdown of NSP led to release of their respective monosaccharides, and second, the breakdown of NSP released the starch within the endosperm, which was exposed to the endogenous amylase, releasing more glucose. Any given enzyme can be viscosity-reducing, sugar-releasing, or both, depending on type and source. Both levels of Enzyme-I and 0.12% Enzyme-II significantly reduced the relative viscosity of the digesta of sunflower meal, deoiled rice bran, and broiler starter diet compared to other enzymes in the peptic and pancreatic phases (Table 2). This result is attributed to the high cellulase and xylanase activities in these enzymes. In case of soybean meal, least relative viscosity was produced by Enzyme-III, which might have been due to the additional pectinase it contained. Although soybean meal was one of the ingredients used in broiler starter diets, the cellulase + xylanase combination gave better results. The possible reason can be attributed to the resultant diet containing more cellulose (5.79%) and pentosan (5.4%) and comparatively less pectin (2.98%). This result agreed with the in vivo results recorded by Suresh and Devegowda (1996). Enzyme-III did not significantly reduce the relative viscosity compared to the controls in sunflower meal and deoiled rice bran. In general, irrespective of the enzymes, the relative viscosity in the pancreatic phase was more than that in the peptic phase, which is due to that fact that NSP becomes more soluble in higher pH medium, increasing the viscosity (Moore and Hoseney, 1990). This observation is in
IN VITRO NONSTARCH POLYSACCHARIDE DIGESTIBILITY BY ENZYMES
305
TABLE 3. Effect of enzymes on total sugars released from sunflower meal, soybean meal, deoiled rice bran, and broiler starter diet subjected to a two-stage in vitro digestion Total sugars released (mg/mL) Enzyme Control Enzyme-I 0.1% 0.2% Enzyme-II 0.06% 0.12% Enzyme-III 0.12% 0.24% Pooled SEM
Sunflower meal
Soybean meal
Deoiled rice bran
Broiler starter diet
4.24d
3.39e
4.69d
5.16e
6.46b 7.26a
4.01d 4.65ab
6.27ab 6.67a
6.34bc 7.20a
6.25b 6.79ab
4.11cd 4.56abc
5.95b 6.30ab
5.98cd 6.75ab
4.69cd 5.28c 0.18
4.32bcd 4.82a 0.13
5.24c 5.89b 0.12
5.32e 5.66de 0.15
Means within a column with no common superscript differ significantly (P < 0.05). SEM = Standard error of mean; N for each mean = 2.
a–e
REFERENCES Bedford, M. R., and H. L. Classen, 1993. An in vitro assay for prediction of broiler intestinal viscosity and growth when fed rye-based diets in the presence of exogenous enzymes. Poultry Sci. 72:137–143. Carre, B., and J. M. Brillouet, 1986. Yield and cell wall residues isolated from various feedstuffs used for non-ruminant farm animals. J. Sci. Food Agric. 37:341–351. Choct, M., and G. Annison, 1992. Anti-nutritive effect of wheat pentosans in broiler chickens: Role of viscosity and gut microflora. Br. Poult. Sci. 33:821–834. Collmer, A., J. L. Ried, and M. S. Mount, 1988. Methods Enzymol. A 160:329–334.
Dubois, M., K. A. Gilles, J. K. Hamilton, P. A. Rebers, and F. Smith, 1956. Colorimetric method for determination of sugars and related substances. Anal. Biochem. 28:350. Duncan, D. D., 1955. Multiple range and multiple ‘F’ test. Biometrics 11:1–42. Dusel, G., H. Kluge, and H. Jeroch, 1998. Xylanase supplementation of wheat-based rations for broilers: Influence of wheat characteristics. J. Appl. Poult. Res. 7:119–131. Englyst, H. N., and J. H. Cummings, 1988. Improved method for measurement of dietary fibre as nonstarch polysaccharides in plant foods. J. Assoc. Off. Anal. Chem. 71:808–814. Frazer, J. R., M. B. Bravo, and D. C., Holmes, 1956. The proximate analysis of wheat flour carbohydrates. I—methods and scheme of analysis. J. Sci. Food. Agric. 7:577–589. John, M., and J. Schmidt, 1988. Xylanases and β-xylosidases of Trichoderma longinorum. Methods Enzymol. A 160:662–670. Kuo, J. M., J. F. Vanmiddlesworth, and W. J. Wolt, 1988. Content of raffinose oligosaccharides and sucrose in various plant seeds. J. Agric. Food Chem. 36:32–36. Moore, A. M., and R. C. Hoseney, 1990. Factors affecting the viscosity of flour-water extracts. Cereal Chem. 67:78–80. Rotter, B. A., R. R. Marquardt, W. Guenter, and G. H. Crow, 1990. Evaluation of three enzymatic methods as predictors of in vivo response to enzyme supplementation of barley-based diets when fed to young chicks. J. Sci. Food Agric. 50:19–27. Rotter, B. A., R. R. Marquardt, W. Guenter, C. Biliaderis, and C. W. Newman, 1989. In vitro viscosity measurements of barley extracts as predictors of growth responses in chicks fed barleybased diets supplemented with a fungal enzyme preparation. Can. J. Anim. Sci. 69:431–439. Sadasivam, S., and A. Manickam, 1996. Biochemical Methods. 2nd ed. New Age International Publishers, New Delhi, India. Smits, C.H.M., and G. Annison, 1996. Nonstarch plant polysaccharides in broiler nutrition-towards a physiologically valid approach to their determination. World’s Poult. Sci. J. 52:203–221. Snedecor, G. W., and S. W. Cochran, 1968. Statistical Methods. 7th ed. Iowa State University Press, Ames, IA. Suresh, S. C., and G. Devegowda, 1996. Influence of dietary enzymes on the performance of broilers. Page 228 in: Proceedings of the 20th World’s Poultry Congress, New Delhi, India. Updegroff, D. M., 1969. Semi-micro determination of cellulose in biological material. Anal. Biochem. 32:420–424. Wood, T. M., and K. M. Bhat, 1988. Methods for measuring cellulase activities. Methods Enzymol. A 160:87–112.
Downloaded from http://ps.oxfordjournals.org/ at Fresno Pacific University on January 15, 2015
agreement with in vitro and in vivo results reported by Bedford and Classen (1993) and the in vivo results of Dusel et al. (1998). Similar to the relative viscosity, the total sugars released was highest with 0.2% Enzyme-I in all feed samples, except soybean meal wherein 0.24% Enzyme-III released the most total sugars (Table 3). The 0.1% Enzyme-I treatment, which was as effective as 0.2% in reducing relative viscosity, was found to release significantly less total sugars than that of the 0.2% level, which might be due to the fact that EnzymeI was primarily a viscosity-reducing enzyme and secondarily a sugar-releasing enzyme. In conclusion, a xylanase + cellulase combination has been found to be superior in case of sunflower meal, deoiled ricebran, and broiler starter diets; for soybean meal, a pectinase-rich enzyme preparation has been found to give the best results. However, when soybean meal is included at higher levels in the feeds, a xylanase + cellulase + pectinase combination might prove to be better than xylanase + cellulase alone; these results need to be validated by in vivo results. This digestion assay simulating the gut conditions of poultry can be used as a simple and rapid screening technique of feed enzymes for their efficacy and stability, and also the best effective dosage level of an enzyme can be fixed.