The experiment was a 4×3 factorial design with four MSRC based on the four maize source samples and without or with the addition of complex enzyme A and B, respectively. Four maize samples (M1, M2, M3, and M4) of JINSUI 1# strain were collected from different regions of PR China in 2012. All samples had no mildew, no lumps and no insects. The samples were analyzed for the bulk density, content of dry matter (DM), total starch, amylose, amylopectin, crude protein (CP), gross energy, crude fat, crude ash, Ca, P, neutral detergent fiber, and acid detergent fiber before being used in the experimental diets. The results of chemical analysis of the four maize samples are shown in Table 1. Both enzyme A (Axtra XAP) containing 20,000 U xylanase, 2,000 U amylase and 40,000 U protease per gram and enzyme B (Avizyme 1502) containing 600 U xylanase, 800 U amylase and 8,000 U protease per gram, were sourced from Danisco Animal Nutrition (Marlborough, Wiltshire, UK), and the dose of addition in this study was 100 and 500 g/t, respectively.

The four MSRC with isocaloric and isonitrogenous were formulated to meet the nutrient recommendation according to Feeding Standard of Chicken of the People’s Republic of China (NY/T 33-2004). The ingredient composition and estimated nutrient content of the experimental basal diets are given in Table 2. Dietary protein was provided by the maize supplemented with commercial soybean meal, rapeseed meal, cotton meal and corn gluten meal. Synthetic methionine and lysine were added to the diets as needed to meet the industry standards. Exogenous enzyme A and B were added separately into the four basal diets. No growth promoters or other medications were added to the diets. Titanium dioxide (TiO₂, 0.4%) was added as an indigestible marker. Mash form diets were provided in this trial.

All experimental procedures were reviewed and approved by the Animal Care and Use Committee of Gansu Agricultural University (GAU). The feeding experiment was conducted in the cage pen house of the Animal Research Centre in GAU.

A total of 960 one-day-old Arbor Acres broiler chicks, obtained from a commercial hatchery, were individually weighed and randomly distributed by body weight to the 12 treatments with 8 replicate pens of 10 broilers (Male:female; 50:50). Room temperature was kept at 33ºC to 35ºC during the first week and gradually decreased to 24ºC by the end of the third week. The birds were given free access to feed and water with constant fluorescent illumination. The trial lasted for 21 d.

The DM and CP were determined according to AOAC (2005) procedures. Gross energy was determined using an adiabatic bomb calorimeter (IKA-2000, Staufen, Germany), standardised with benzoic acid. TiO₂ was measured on a UV spectrophotometer following the method of Short et al. (1996).

The apparent fecal and ileal digestibility of nutrients (DM, energy and protein) were calculated by the following formula using the TiO₂ marker ratio in the diet, excreta and digesta:

Where:

The AMEn was determined by correction for zero nitrogen retention by simple multiplication with 36.54 kJ per gram nitrogen retained in the body as described by Hill and Anderson (1958).

Data were analyzed by multivariate analysis of variance (ANOVA) of SPSS 16.0 for Windows (SPSS Inc., Chicago, IL, USA). The model included diet and enzyme as the main effects. Variable means for treatments showing significant differences in the ANOVA were separated by Tukey test. In all analyses, significance was declared at p<0.05.

The BWG and F/G of broilers fed MSRC were significantly different in the current study, which would suggest geographical origin of maize affects the growth performance of broilers. A similar finding was also reported by Brake et al. (2003) that there were significant BWG differences due to maize source at 21 d of age for live birds when birds fed maize soybean based diets. However, differing results were reported by O’neill et al. (2012) and Yegani and Korver (2013) that growth performance profiles of birds did not depend on the geographical source of maize. The lack of consistency in growth improvements among studies may be attributed to variations in the nutritive values, including ME, CP, and amino acid (Cowieson, 2005). In the current study, we observed that huge bulk density differences existed between M1 and M4, and Baidoo et al. (1991) investigated the positive relationships between AMEn and maize kernel density (r = 0.875). If this theory is correct, then M1 had higher AMEn than that of M4, consequently, a similar result was found in the present study. Therefore, we may use the bulk density parameter as the indicator of nutritional value of maize as it is in practical industry. Generally, gross energy parameter of grain is not meaningful in reflecting its nutritional value, but the gross energy value of M1 and M4 coincided with the changes of growth performance in this study. Therefore, gross energy may also as a good indicator for sample collection.

Supplementation of complex enzymes containing xylanase, amylase and protease to MSRC improved broiler’s BWG and FCR in the current study, and the results were similar with the previous findings in broilers fed diets based on either so-called ‘viscous’ grains (wheat, barley, rye, triticale and oats) or low NSP content grain, such as maize (Bedford, 2000; Cowieson, 2005; Choct, 2006; Cowieson and Ravindran, 2008a,b). However, some early research suggested there wasn’t any beneficial effect of enzyme complexes, including amylase and protease, on performance of broilers fed sorghum-soybean meal diet or maize-soybean diet (Mahagna et al., 1995; Douglas et al., 2000; Olukosi et al., 2007; West et al., 2007). Kocher et al. (2003) reported an exogenous complex enzyme product containing xylanase, amylase and protease had little effect on the performance of broiler chickens, and enzyme addition, in some cases, had a negative impact on performance variables (Brufau et al., 2006; Yegani and Korver, 2013). The explanation for the positive effects of enzyme addition in the current study could be that complex enzymes were better matched, so that the xylanase, amylase and protease were able to break down the cell wall matrix, especially the insoluble components, thereby facilitating the release of nutrients encapsulated in cell walls or incorporated into the cell wall itself, resulting in an easier access of digestive enzymes (Cowieson, 2005; Choct, 2006; Francesch and Geraert, 2009). In our study, the reasons why the BWG of birds fed M1 diet containing enzyme A was significantly higher than that of birds fed M3 or M4 diet without enzyme inclusion, and the reverse case for the F/G, were not clear, but it may have been due to an improvement of starch digestion of M1, which contained higher total starch than other maize samples.

The mode of action of enzymes in maize based diets has been linked to improved starch digestibility associated with augmentation of endogenous α-amylase or improved digestion of resistant starches, improved access to cell contents via a reduction in cell wall integrity, modification of the intestinal microbial communities, improved protein solubility and digestibility and a reduction in the inimical effects of maize and/or soy-derived anti-nutritive factors (Cowieson and Ravindran, 2008b). Apparent ileal digestibility of DM, IDE, AME, and AMEn were significantly different for diets containing the four maize samples in the present study. This finding was in accordance with Cowieson (2005) who also reported the AME value of maize can vary by more than 2 MJ/kg from batch to batch making generic energy matrix values for maize inaccurate.

One way to improve the consistency of the nutritive value of maize is to add specific exogenous enzymes to ameliorate the adverse effects of anti-nutritional factors and to improve the digestibility of starch, oil and protein (Cowieson, 2005). In our study, apparent ileal digestibility of DM, CP and gross energy was improved by complex enzyme supplementation that also increased apparent total digestive tract digestibility of DM, CP, gross energy, and AMEn of the diet. Cowieson (2010) reported IDE was increased as a result of addition of xylanase+glucanase in the diet of 21-d-old broilers. Wyatt et al. (1999) reported that addition of an enzyme blend containing xylanase, amylase, and protease increased IDE of maize soybean diets in 28-d-old broiler chickens, which is in agreement with our observations. Rutherfurd et al. (2007) reported that addition of an enzyme mixture of xylanase, amylase, and glucanase in a corn-soy diet significantly increased ileal nitrogen digestibility in 29-d old broiler chickens. Oukosi et al. (2008) found that the combination of phytase with xylanase, amylase, and protease (XAP) in the negative control diet improved total tract retention of all nutrients as well as increased ME compared with the negative control diet, and similar results were also reported by Douglas et al. (2000) who found that ileal digestibility of energy at 21 d of age, AME and AMEn were improved when XAP are used in the maize soybean based diets for broilers. However, it is in contrast to the study of Olukosi et al. (2007) where there was no effect of a mixture of XAP on IDE in 21-d-old broiler chickens. The lack of effects of enzyme treatments on CP and gross energy digestibility in starter phase is in agreement with Mahagna et al. (1995), who did not observe any positive effects of the addition of exogenous amylase and protease on CP digestibility of sorghum-soy diets in 14-d-old broiler chickens. Yegani and Korver (2013) also found the effects of enzyme products on IDE and digestibility of CP and amino acid were not consistent and varied depending on maize sources, enzyme products, and dietary phases. In current study, we also detected the apparent ileal CP digestibility coefficient, N retention, IDE, AME and AMEn could be influenced by the interaction of maize source×enzyme. M1 and M2 diets containing enzymes exhibited better N and energy utilization, which indicated there is correlative between maize quality and effect of enzyme supplementation, high quality maize is seen to be an advantage in improving nutritional value.

Animal manure output has been paid more attention due to its potential impact on the environment, which include effects on surface and ground water quality, soil quality, environment within housing systems, and emissions to the atmosphere. In US, annual manure output from farm animals is estimated to range from 133 (Burkholder et al., 2007) to 300 million metric tons (on a dry weight basis) per year. However, manure was applied on only 15.8 billion acres, which was about 5% of the total acres planted in 2006 (Scott and Kim, 2013). In a broiler’s diet, high level protein (approximately 18% to 23%) is included to meet the requirement of growth, but only 60% to 80% of this is retained as animal protein, with the remaining 20% to 40% of the protein intake excreted in the manure. Substantial quantities of these mineral losses are recycled as landmass organic fertilizers are applied, but excess amounts tend to accumulate in the environment after repetition and time.

The use of appropriate enzyme blends has great potential for improving nutrient availabilities from feedstuffs and for reducing mineral emissions from animal production. Inclusion of phytase in diets increases P digestibility by approximately 20% to 50%, resulting in marked decreases in P excretion (Woyengo and Nyachoti, 2011). In the present study, complex enzymes supplementation improved the DM digestibility and N retention. The possible explaination could be that enzymes hydrolyze storage non-starch polysaccharides, break down various anti-nutritional factors, release more nutrients and increase the availability of nutrients (Ferket et al., 2002).

In conclusion, this study indicated performance profiles could be influenced by maize source, and a combination of amylase, xylanase and protease is effective in improving the available energy and growth performance of broilers fed MSRC diet.