Abstract:
163 Enzymatic degradation of non_starch polysaccharides in vegetable proteins in poultry diets A. Kocher Animal Science, University of New England, Ar midale NSW 2351 akocher2@metz.une.edu.au Summary Vegetable proteins such as soybean meal, canola meal and lupins are widely used as the main protein source in poultry diets. However, there is only limited information on the nutritive or anti_nutritive effects of indigestible non_starch polysaccharides in vegetable proteins. This review discusses a number of experiments which investigated the effects of feed enzymes on the utilisation of non_starch polysaccharide (NSP) in lupins, soybean meal, canola meal and sunflower meal included at a minimum of 350 g/kg in semi_purified broiler diets. It was shown that enzymes with a high level of polygalacturonase activity were able to depolymerise part of the soluble and insoluble pectic polysaccharides in vegetable proteins. However, significant effects on broiler performance and nutrient digestibility were only evident when enzymes were included at very high levels in the diet. Keywords: vegetable proteins, non_starch polysaccharides, feed enzymes, broilers crude protein (CP) content varies from 22% CP in field peas to over 50% in some soybean meals (SBM). The occurrence and distribution of various anti_nutritive factors (ANF) such as protease inhibitors, tannins, phytic acid, alkaloids and indigestible carbohydrates affect nutrient utilisation. Mechanical processing like dehulling can be used to reduce tannins and NSP. Heat treatment is commonly used to inactivate proteolytic ANF. The success of heat treatment depends on the fine balance between the inactivation of ANF and the loss of heat sensitive amino acids (van der Poel 1989). The removal of heat stable ANF located in the cotyledon has proven to be more difficult. The addition of microbial phytases to hydrolyse plant phytate has proven highly effective in improving phosphorus availability and reducing the anti_nutritive effects of phytate (Touchburn et al. 1999). The application of feed enzymes to reduce the amount of indigestible carbohydrates has shown to be more challenging than the application of phytases. Enzymes are target specific. An understanding of the target substrates for feed enzymes coupled with improvements of microbiological technology have led to many investigations of the effects of feed enzymes in broiler diets. Most of the work has concentrated on the use of glycanases in cereal_based diets which has been the subject of several reviews (Chesson 1987; Classen and Campbell 1990; Annison and Choct 1991). As a result, diets based on wheat or barley commonly include enzyme products such as xylanases, _glucanases and cellulases. A number of studies on the effects of multi_activity enzyme products in broiler diets containing high levels of vegetable proteins show inconclusive results (Brenes et al. 1993a; Classen et al. 1993; Roth Maier and Kirchgessner 1994). The two main reasons for the inconsistency are a lack of clear understanding of the anti_nutritive effects of NSP in vegetable proteins and the inability of currently available feed enzymes to depolymerise these carbohydrates. This paper discusses the use of exogenous feed enzymes in broiler diets containing high levels of commonly used vegetable proteins such as lupins, SBM, Introduction The worldwide production of oilseed crops and grain legumes in 2000 was estimated to be in excess of 280 million tonnes (FOASTAT 2000). The introduction of a total ban on the use of animal protein sources in feedstuffs by the European Union in 1999 (Adams 2000), and the increased availability of cost effective meals as by_products from the production of vegetable oil, placed greater emphasis on the inclusion of high levels of vegetable protein from various sources in animal feed. Beans, peas, lupins, peanut and soybeans are all leguminosae whereas oilseeds such as rape or canola, sunflower, palm and cotton belong to a wider variety of botanical families. Despite their differences, proteins from all these plant sources are referred to in this paper as vegetable proteins. The values for apparent metabolisable energy (AME) of vegetable proteins are generally lower compared to those of cereal grains, and the average Recent Advances in Animal Nutrition in Australia, Volume 13 (2001) 164 Kocher, A. canola meal (CM),and sunflower meal (SFM) with some references to field peas, faba beans, and some non_ conventional ingredients including peanut meal and palm kernels. Content and chemical structure of indigestible carbohydrates in vegetable proteins The NSP content of vegetable proteins used in poultry diets varies according to their plant origin, the variety, the degree of processing, and subsequently on the proportion of NSP_rich hull in the final product. The total NSP content in vegetable proteins ranges from 180 g/kg DM in peas and canola meal to over 350 g/kg DM in some lupin species (Table 1). The major NSP components of the cell wall are pectic polysaccharides, which include rhamnogalacturonan, arabinans, galactans and arabinogalactans (Arora 1983). Rhamnogalacturonans are characterised by a linear chain of _(1_4)_D_ galacturonic acid units interrupted with rhamnopyranosyl residues and substituted with numerous sidechains consisting of L_arabinose, D_galactose and D_xylose (Bacic et al. 1988). In addition, neutral polysaccharides lacking the galacturonic acid backbone, such as arabinans, galactans and arabinogalactans are present as separate polysaccharides or as sidechains on the uronic acid backbone. Arabinogalactans found in legumes are mainly type I polymers containing D_galactose units (_1_4 linked) with L_arabinose units as sidechains (Aspinall and Cottrell 1971). In contrast, those found in canola or rapeseed meal (arabinogalactan type II) are characterised by the present of 1_3 and 1_6 linked _galactopyranose units with terminal residues of arabinofuranose and doubly branched galactose residues (Siddiqui and Wood 1972). Other neutral polysaccharides in vegetable proteins include cellulose, xylans, arabinoxylans and glucoxylans. These structures are predominantly found in the hull fraction with only a small proportion present in the cotyledon. The NSP described thus far are widely distributed and are found in most vegetable proteins. Polysaccharides that are more specific to one legume or oilseed meal are the galactomannans found in soybean hulls (Aspinall and Whyte 1964), an a_galacturonan which only occurs in sunflower seeds (Ziko and Bishop 1966), linear mannans in palm cake and coconut cake (Bach Knudsen 1997) and galactomannans found in copra meal (Saittagaroon et al. 1983). Effects of exogenous enzyme addition on the utilisation of NSP in vegetable proteins Enzymes used in diets containing high levels of vegetable proteins are mostly multi_glycanase products containing polygalacturonase (pectinase), their main activity being the hydrolysis of pectins present in vegetable products; a range of other enzymes, such as cellulase and hemicellulases targeting neutral NSP, are also used. Lupins Several authors reported significant improvement in broiler performance and weight gain when commercial enzyme products were added to diets containing 45% dehulled Lupinus albus (Brenes et al. 1993a; Roth Maier and Kirchgessner 1994). However, these studies made no reference to improvements in AME or nutrient digestibility. Recently, we (Kocher et al. 2000b) compared the effects of three commercial enzyme products on the nutritive value of whole seed L. angustifolius and L. albus. This study failed to show any effects on bird performance or AME in any of the experimental diets. However, in diets with L. angustifolius the addition of a multi_activity enzyme product containing cellulase, _glucanase and pectinase activities resulted in a marked increase in digesta viscosity as a result of increased amounts of soluble NSP released by the enzyme. The increase in soluble NSP was due to an increase in rhamnose, arabinose and galactose, which led to the conclusion that the enzyme was capable of solubilising galactans, arabinogalactans and rhamnogalacturonans of L. angustifolius, releasing smaller soluble fragments into the digesta (Table 2). The values in Table 2, and in Table 3, are expressed as g NSP per kg acid_insoluble ash (AIA); these enable direct comparisons of the effects of the enzymes. In diets with L. albus the addition of the same enzyme product had no effect on digesta viscosity or the concentration of soluble NSP but increased the digestibility of insoluble NSP in the ileum. The increased digestibility was directly related to increased digestibility of glucose, xylose and arabinose leading to the conclusion that the addition of this enzyme depolymerised the cellulose and xylan and arabinoxylan mainly present in the hull fraction (Table 2). Although Table 1 Non_starch polysaccharides (NSP) in vegetable proteins (g/kg DM). Vegetable protein Soybean meal Canola meal2 Sunflower meal 2 3 1 Soluble NSP 63 15 10 31 14 52 50 2 Insoluble NSP 154 139 199 336 319 129 140 3 L. angustifolius L. albus 3 Pea (P. sativum)1 Beans (V. faba) 1 1 Bach Knudsen 1997; Kocher et al. 2000a; Kocher et al. 2000b Table 2 Ileal viscosity and neutral sugar composition in ileal digesta of birds fed diets containing Lupinus angustifolius and L. albus with or without enzyme supplementation. Values are g/kg acid_insoluble ash (AIA; see text). Viscosity Lupin Enzyme Control Enzyme A Enzyme B Enzyme C mPa 11.6 11.3 11.6 32.8 b b b a Soluble NSP 673.1bc 670.5bc 704.4 b Insoluble NSP 5812.4ab 5934.3ab 6467.7a 6080.6a Rhamnose 23.3 21.8 24.3 39.8 b b b a Soluble NSP Arabinose 85.2 82.5 87.6 b b b Xylose 43.4 31.9 40.8 65.6 ab b b a Mannose 32.1 32.4 39.6 36.4 Galactose 491.2 505.5 511.3 888.0 b b b a Glucose 53.9 50.9 51.8 57.0 a a a a L. angustifolius L. angustifolius L. angustifolius L. angustifolius 1119.9a 139.1a Insoluble NSP Enzymatic degradation of non_starch polysaccharides in poultry diets Rhamnose Arabinose 1061.9a 992.4ab 971.9ab 896.8b Xylose 1098.6a 936.9 887.2 886.8 b b b Mannose 60.8 54.3 48.6 42.6 Galactose 1802.5 1831.2 1793.0 1664.5 a a a a Glucose 2328.0a 1904.3 1559.5 ab bc L. albus L. albus L. albus L. albus a,b,c Control Enzyme A Enzyme B Enzyme C 5.3 5.2 5.1 c c c 487.9 404.6 533.5 ef f def 5754.5 5190.1 ab bc c c 63.3 62.8 70.9 50.9 a a a b 4782.6 4471.8 6.1bc 565.3bcd 1442.6c Values with unlike superscripts differ significantly (P<0.05) Table 3 Feed conversion ratio, nitrogen corrected apparent metabolizable energy (AMEn), ileal protein digestibility and amounts of free sugars, soluble and insoluble NSP in the ileum of broiler chickens fed on diets containing soybean meal without enzyme supplementation or with enzyme supplementation at different inclusion levels. FCR Diet Control Enzyme A normal Enzyme A (5x) g feed/g gain 1.72 1.70 1.70 AMEn MJ/kg DM 12.84 12.89 13.04 b b a Ileal protein digestibility 0.82 0.80 0.86 b bc a Ileal digesta (g/kg AIA) Free sugars 939 1073 1205 b ab a Soluble NSP 298 279 272 a a ab 1 Insoluble NSP1 3980 3627 3601 a b b 165 166 Kocher, A. the sugar composition of the two lupin species is similar, differences in the amount of soluble NSP and their ability to increase digesta viscosity is the most likely reason for the different responses of the two lupin species to the same enzyme. Soybean Addition of enzymes to a corn_SBM broiler diet resulted in a significant improvement in weight gain and FCR as a result of increased ileal digestibility of crude protein (CP), starch and fat (Zanella et al. 1999). Similarly, Marsman et al. (1997) reported a significant improvement in ileal digestibility of CP and NSP in broiler diets with corn_SBM. It was concluded from both studies that the added enzyme products not only had cell wall degrading activities but also exhibited protease activity, which explained the improved nutrient digestibility. We showed recently that the effects of enzyme addition to SBM_based broiler diets not only depended on the enzyme product but also on the inclusion level (Kocher et al. 2001b). The inclusion of a multi_activity commercial enzyme product with hemicellulase, pectinase and cellulase activities at the supplier 's recommended dosage (400 ppm) had no effects on FCR, AME, CP and NSP digestibility. However, when the same enzyme was added at five times the recommended dosage level a significant improvement in AME, protein and NSP digestibility was observed (Table 3). Analysis of NSP content in ileal digesta revealed a significant increase in free sugars correlated to a significant decrease in insoluble NSP, indicating a partial depolymerisation by the enzyme of the main NSP of SBM. The improvement in nutrient digestibility can be a direct result of the enzymatic breakdown of the cell walls and the release of entrapped nutrients as well as the presence of small amounts of exogenous proteases in this enzyme product (Marsman et al. 1997). On the other hand, Irish and Balnave (1993) found that the addition of two multi_activity enzyme preparations to corn/ wheat_SBM diets resulted in a significantly poorer growth compared to an unsupplemented control diet. It was concluded that NSPs were broken down; the presence of large amounts of indigestible low molecular weight NSP results in fluid retention in the small intestine and can adversely affect the absorption of nutrients (Wiggins 1984). with 400 g/kg CM, resulting in a significant increase in the NSP digestibility in the ileum (Slominski and Campbell 1990). Similar, we (Kocher et al. 2000a) found a reduction in soluble and insoluble NSP in the jejunum and a tendency towards improved FCR on adding the same glycanase. It was suggested that the enzyme depolymerised soluble NSP in the upper intestine and possibly disrupted the cell_wall matrix leading to easy access of endogenous proteolytic enzymes to digest entrapped proteins. A second study (Kocher et al. 2001a) showed that the additional protein resulted in additional muscle growth, and subsequently increased the dressed yield of the broiler. Sunflower meal Sunflower meal can be used at high inclusion levels (up to 350 g/kg) to replace SBM in broiler diets despite its high NSP content (Rad and Keshavarz 1976). These NSP are highly susceptible to enzymatic degradation in vitro (D�sterh�ft et al. 1993b). However, Gerendai et al. (1997), Sherif et al. (1997) and Rebole et al. (1997) failed to show any improvements in broiler performance or AME when commercially available multi_activity enzymes were included in broiler diets. In our study (Kocher et al. 2000a), we demonstrated that the addition of such an enzyme preparation based on pectinase and hemicellulase to broiler diets significantly improved the digestibility of NSP in the jejunum and the digestibility of protein in the ileum. The lack of response in growth performance and AME in this study was explained by the fact that birds fed SFM regardless of enzyme addition were close to their genetic growth potential. Field peas and faba beans In contrast to the vegetable proteins discussed so far field peas and faba are rich in starch (over 50% of DM in the cotyledon). The structure of the starch granules, their location within the cell wall matrix and the complex mixture of NSP make starch less susceptible to endogenous enzymes of poultry (Longstaff and McNab 1987) and indicates scope for improving nutrient digestibility when exogenous enzymes are added to the diet. However, studies on the effects of enzymes on the nutritive value of beans or peas are scarce and give confusing results. Results of these studies showed either no effect on broiler or laying hen performance or failed to demonstrate a clear relationship between the addition of enzymes and any observed improvement in animal performance (Brenes et al. 1993b; Igbasan and Guenter 1997; Igbasan et al. 1997; Keller and Jeroch 1997). Canola meal Canola meal is a suitable protein source for poultry diets despite an NSP content of 195 g/kg (Bell 1993). Slominski and Campbell (1990) and Simbaya et al. (1996) showed the potential of exogenous enzymes high in polygalacturonase to enhance the digestion of these NSP of CM in vitro, by rapidly hydrolysing the soluble NSP fraction followed by a slow degradation of the insoluble NSP fraction. These findings were confirmed when the same enzyme was added to a laying hen diet Other vegetable proteins The NSP found in copra meal, coconut meal or palm kernels consist predominantly of mannans (galactomannans and linear mannans); therefore the concentration of mannanolytic activities in enzyme Enzymatic degradation of non_starch polysaccharides in poultry diets 167 preparations determines the extent to which NSP are solubilised (D�sterh�ft et al. 1993a). Most of the research on the utilisation of NSP in vegetable protein in the presence of feed enzymes has concentrated on SBM, CM and sunflower meal with little or no data available on the use of enzymes in other products. Pluske et al. (1997) reported some evidence that the addition of a commercially available mannanase to diets including 20% copra meal improved the liveweight gain of broilers measured over a period of 42 days Brenes, A., Marquardt, R.R., Guenter, W. and Rotter, B.A. (1993a). Effect of enzyme supplementation on the nutritional value of raw, autoclaved, and dehulled lupins (Lupinus albus) in chicken diets. Poultry Science 72, 2281_2293. Brenes, A., Rotter, B.A., Marquardt, R.R. and Guenter, W. (1993b). The nutritional value of raw, autoclaved and dehulled peas (Pisum sativum L.) in chicken diets as affected by enzyme supplementation. Canadian Journal of Animal Science 73, 605_614. Chesson, A. (1987). Supplementary enzymes to improve the utilization of pig and poultry diets. In: Recent Advances in Animal Nutrition, pp.71_89. (eds. W. Haresign and D.J.A. Cole). Butterworths, London. Classen, H.L., Balnave, D. and Bedford, M.R. (1993). Reduction of legume antinutritional factors using biotechnological techniques. Proceedings of the 70th Recent Advances of Research in Antinutritional Factors in Legume Seeds, Wageningen, pp. 5516. Pudoc Wageningen, Netherlands. Classen, H.L. and Campbell, G.L. (1990). Improvement in feed utilization through the use of enzyme products. Proceedings of the 2nd Australian Poultry Symposium, Sydney, Australia, pp. 1_8. D�sterh�ft, E.M., Bonte, A.W., Venekamp, J.C. and Voragen, A.G.J. (1993a). The role of fungal polysaccharides in the hydrolisis of cell wall materials from sunflower and palm_kernel meals. World Journal of Microbiology and Biotechnology 9, 544_554. D�sterh�ft, M., Bonte, A.W. and Voragen, A.G.J. (1993b). Solubilisation of non_starch polysaccharides from oil_ seed meals by polysaccharide_degrading enzymes. Journal of the Science of Food and Agriculture 63, 211_220. FOASTAT (2000). Food and Agriculture Organisation of the United Nations, 2000. http://apps.fao.org/ default.htm. Last updated 7/2/2001. Accessed 13/2/2001. Gerendai, D., Sherif, K.E.L. and Gippert, T. (1997). The effect of Kemzyme and Phylacell enzyme preparation on the utilisation of broiler feeds containing sunflower meal. Proceedings of the 9th Australian Poultry Science Symposium, Sydney, Australia, pp. 211_214. Igbasan, F.A. and Guenter, W. (1997). The influence of micronization, dehulling, and enzyme supplementation on the nutritional value of peas for laying hens. Poultry Science 76, 331_337. Igbasan, F.A., Guenter, W. and Slominski, B.A. (1997). The effect of pectinase and alpha_galactosidase supplementation on the nutritive value of peas for broiler chickens. Canadian Journal of Animal Science 77, 537_539. Irish, G.G. and Balnave, D. (1993). Non_starch polysaccharides and broiler performance on diets containing soyabean meal as the sole protein concentrate. Australian Journal of Agricultural Research 44, 1483_1499. Keller, T. and Jeroch, H. (1997). Effect of different microbial carbohydrases in its combination of preparation and application rate on contents of Conclusions It is evident that commercial enzyme products have some effects on the utilisation of NSP in poultry diets containing high levels of vegetable proteins. However, NSP of vegetable proteins do not exhibit strong anti_ nutritive effects in contrast to NSP in wheat and barley and therefore the addition of enzymes do not always clearly affect bird growth and performance. The consequence and benefits of enzyme addition can often only be seen after extensive analyses of feed and digesta. In order to develop enzyme products to improve the nutritive value of vegetable proteins, a better understanding of the structural changes in NSP is necessary. References Adams, C.A. (2000). Nutricines as alternatives to antibiotics. In: Meeting the Challenges of Intensive AnimalProduction, pp. 1_10. Kemin Industries (Asia), Singapore. Annison, G. and Choct, M. (1991). Anti_nutritive activities of cereal non_starch polysaccharides in broiler diets and strategies minimizing their effects. World's Poultry Science Journal 47, 232_242, 257, 260. Arora, S.K. (1983). Legume carbohydrates, In: Chemistry and Biochemistry of Legumes, pp. 1_50 (ed. S.K. Arora). Edward Arnold, London. Aspinall, G.O. and Cottrell, I.W. (1971). Polysaccharides of soybeans. VI. Neutral polysaccharides from cotyledon meal and chemistry of cell wall polysaccharides. Canadian Journal of Chemistry 49, 1019_1022. Aspinall, G.O. and Whyte, J.N.C. (1964). Polysaccharides of soybeans. I. Galactomannans from the hull. Journal of the Chemical Society (c) 5, 5058_5063. Bach Knudsen, K.E. (1997). Carbohydrate and lignin contents of plant materials used in animal feeding. Animal Feed Science and Technology 67, 319_338. Bacic, A., Harris, P.J. and Stone, B.A. (1988) Structure and function of plant cell walls, In: The Biochemistry of Plants, Vol. 14, pp. 297_372 (ed. J. Preiss). Academic Press, Inc., London . Bell, J.M. (1993). Factors affecting the nutritional value of canola meal: a review. Canadian Journal of Animal Science 73, 679_697. 168 Kocher, A. n_corrected metabolizable energy of dried peas (Pisum sativum L ) on broiler chicks. Archiv f�r Gefl�gelkunde 61, 131_134. Kocher, A., Choct, M., Morrisroe, L. and Broz, J. (2001a). The effects of enzyme supplementation on the replacement value of canola meal for soybean meal in broiler diets. Australian Journal of Agricultural Research 54, (in press). Kocher, A., Choct, M., Porter, M.D. and Broz, J. (2000a). The effects of enzyme addition to broiler diets containing high levels of canola or sunflower meal. Poultry Science 79, 1767_1774. Kocher, A., Choct, M., Porter, M.D. and Broz, J. (2001b). Effects of feed enzymes on nutritive value of soybean meal by broilers. British Poultry Science (in press). Kocher, A., Hughes, R.J., Choct, M. and Broz, J. (2000b). Effect of food enzymes on utilisation of lupin carbohydrates by broilers. British Poultry Science 41, 75_82. Longstaff, M. and McNab, J.M. (1987). Digestion of starch and fibre carbohydrates in peas by adult cockerels. British Poultry Science 28, 261_285. Marsman, G.J., Gruppen, H., van der Poel, A.F., Kwakkel, R.P., Verstegen, M.W. and Voragen, A.G. (1997). The effect of thermal processing and enzyme treatments of soybean meal on growth performance, ileal nutrient digestibilities, and chyme characteristics in broiler chicks. Poultry Science 76, 864_872. Pluske, J.R., Moughan, P.J., Thomas, D.V., Kumar, A. and Dingle, J. (1997). Releasing energy from rice bran, copra meal and canola in diets using exogenous enzymes. In: Proceedings of the Alltech's 13th Annual Symposium, pp. 81_93 (eds. T.P. Lyons and K.A. Jacques). University Press, Nottingham. Rad, F.H. and Keshavarz, K. (1976). Evaluation of the nutritional value of sunflower meal and the possibility of substitution of sunflower meal for soybean meal in poultry diets. Poultry Science 55, 1757_1765. Rebole, A., Rodriguez, M.L., Alzueta, C., Ortiz, L.T. and Trevino, J. (1999). A short note on effect of enzyme supplement on the nutritive value of broiler chick diets containing maize, soyabean meal and full_fat sunflower seed. Animal Feed Science and Technology 78, 153_158. Roth Maier, D.A. and Kirchgessner, M. (1994). High proportions of white lupins (Lupinus albus L.) and enzyme supplements for fattening chickens. Archiv f�r Gefl�gelkunde 58, 245_248. Saittagaroon, S., Kawakishi, S. and Namiki, M. (1983). Characterisation of polysaccharides of copra meal. Journal of the Science of Food and Agriculture 34, 855_860 Sherif, K.E., Gerendai, D. and Gippert, T. (1997). Complete substitution of sunflower meal for soybean meal with or without enzyme supplementation in broiler rations. Archiv f�r Gefl�gelkunde 61, 8_14. Siddiqui, I.R. and Wood, P.J. (1972). Structural investigation of water_soluble rapeseed (Brassica camperestris) polysaccharides. II Acidic arabinogalactan. Carbohydrate Research 24, 1_9. Simbaya, J., Slominski, B.A., Guenter, W., Morgan, A. and Campbell, L.D. (1996). The effects of protease and carbohydrase supplementation on the nutritive value of canola meal for poultry -- in vitro and in vivo studies. Animal Feed Science and Technology 61, 219_234. Slominski, B.A. and Campbell, L.D. (1990). Non_starch polysaccharides of canola meal: Quantification, digestibility in poultry and potential benefit of dietary enzyme supplementation. Journal of Agricultural and Food Chemistry 53, 175_184. Touchburn, S.P., Sebastian, S. and Chavez, E.R. (1999). Phytase in poultry nutrition. In: Recent Advances in Animal Nutrition, pp. 147_164 (eds. P.C. Garnsworthy and J. Wiseman). University Press, Nottingham. van der Poel, A.F.B. (1989). Effects of processing on antinutritional factors (ANF) and nutritional value of legume seeds for non_ruminant feeding. In: Recent Advances of Research in Antinutritional Factors in Legume Seeds, pp. 213_229. Pudoc Wageningen, Netherlands.. Wiggins, H.S. (1984). Nutritional value of sugars and related compounds undigested in the small gut. Proceedings of the Nutrition Society 43, 69_75. Zanella, I., Sakomura, N.K., Silversides, F.G., Fiqueirdo, A. and Pack, M. (1999). Effect of enzyme supplementation of broiler diets based on corn and soybeans. Poultry Science 78, 561_568. Ziko, V. and Bishop, C.T. (1966). Structure of a galacturonan from sunflower pectic acid. Canadian Journal of Chemistry 44, 1275_1282.