I. INTRODUCTION
Metabolizable energy (ME), digestible amino acids (dAA) and available phosphorus (avP) are the largest and most expensive components in broiler diets, considerable fractions of which still pass through the digestive tract undigested and are lost via excretion (Ravindran et al., 2013), hence, emphasizing the need for higher nutrient usage efficiency than current practice. Ability of carbohydrases to degrade non-starch polysaccharides and liberate caged starch and protein (Cozannet et al., 2017) and of phytase to degrade phytate to increase availability of P and Ca (Amerah et al., 2014) have been established. The drive to further increase nutrient usage efficiency necessitates higher efficiency of exogenous enzymes usage. Combined usage of both carbohydrases and phytase in diets is hypothesised to enable significant reductions in ME, dAA, avP, and Ca with no adverse effects on growth performance and feed efficiency of the birds. Efficacy of a global enzyme solution, which consists of a multi-carbohydrase and phytase complex (MCPC), to release ME, dAA, avP, and Ca was evaluated throughout performance and feed efficiency effects on broilers from 10 to 42 d. The objective of the current study was to evaluate the effect of MCPC on growth performance and feed efficiency of broilers fed diets with different levels of ME, dAA and avP.
II. MATERIALS AND METHODS
The trial was conducted on 3,840 Ross 308 broiler chicks, housed using a partial block design (Box-Behnken) with 20 birds per pen and 192 pens and fed common starter diet from 0 to 10 d and one of 24 grower and finisher experimental diets from 10 to 28 d and 28 to 42 d, respectively. Diets were corn-wheat-wheat bran-soybean meal-soy hulls-based and included a positive control (PC) diet (12.98 MJ ME, 11.1 g/kg dLys, 3.5 g/kg avP, and 7.5 g/kg Ca and 13.19 MJ ME, 8.8 g/kg dLys, 3.0 g/kg avP, and 6.0 g/kg Ca during grower and finisher phase, respectively) and 11 negative control (NC) diets based on optimized factorial combinations of 3 reduced ME levels (-4.0, -6.5, -9.0%) × 3 reduced dAA levels (-4.0, -6.5, -9.0%) × 2 reduced avP levels (balanced with Ca; PC-36, -70%) with MCPC supplementation at 0 (NC-) or 100 g/ton of feed (NC+). The associated levels were dLys (10.0, 9.6, 9.3, 9.1 g/kg), ME (13.06, 12.54, 12.21, 11.89 MJ/kg), Ca (6.8, 5.8, 4.7 g/kg) and avP (3.3, 2.1, 1.2 g/kg). The MCPC (Rovabio® Advance Phy, Adisseo France SAS, Antony France) consisted of a multicarbohydrase from Talaromyces versatilis (xylanases / beta-glucanases / arabinofuranosidases 1,250 / 850 / 9,250 U/kg of feed) and a phytase (1,000 FTU/kg of feed) sourced from Buttiauxiella spp.
The growth performance and feed efficiency, BW and feed intake were measured at 10, 28, and 42 d for calculation of mortality-corrected 10 to 42 BWG, FI, and FCR. Generated data were analyzed using Proc NLIN. Experimental diets allowed precise evaluation of ME, dLys and avP on performance (i.e. BWG, FI and FCR). MCPC effect was estimated using the equation established and converted into ME, dAA and avP improvement. Response surface was fitted by first-, second-, or third-degree polynomial regressions. Pen mean was used as the experimental unit and a 5% level of probability was considered to be significant.
III. RESULTS
Performance results are summarized in Table 1, FI, BWG and FCR varied largely by 15, 17 and 4% coefficient of variation, respectively. A model was developed based on individual results. The surface response for BW and FCR is presented for 2 levels of avP. Equations suggested for the control diet a large impact of avP on animal performance with correlation coefficient equal to 0.84, 0.91 and -0.75 for FI, BWG and FCR (P < 0.001). Decrease of avP from 0.33 to 0.12% for control diets resulted in decrease by 25, 33 and 11% of FI, BWG, and FCR, respectively. As a consequence, avP effect on FI was quadratic with an inflection point at 0.26% avP. Secondly, ME and dAA had an important role on intake and feed efficiency. Decrease of ME by 9% was associated with decrease of 12% of BWG. Decrease of dAA by 9% was associated with decrease of 3% of FCR. Interaction between ME and dAA was also important with largest impact of dAA deficiency in high ME diets compared with low ME diet.
The effect of MCPC was highly significant for all parameters with 16, 20 and -4% for intake, gain and FCR, respectively. Significant interactions were found between diet and MCPC. Minimum-maximum variation values were 0 - 44, 2 - 53 and -2 - -6% for FI, BWG and FCR, respectively. Primary discrimination between enzyme responses was related with avP reformulation. Using MCPC, the correlation between avP and FI BWG and FCR became not significant (r = -0.34, 0.43 and -0.54 respectively). In contrast to unsupplemented MCPC diets, ME and dAA became primary drivers of FI and BWG (r = - 0.89 and 0.74, P < 0.001, respectively). The MCPC restored FI and BWG at levels not significantly different from those observed for PC. Therefore, FCR was only partly restored for diets reduced in ME, dAA and avP. Also, MCPC supplementation only completely alleviated the increased FCR by NC1+, NC2+ and NC4+ diets; partially alleviated the increased FCR by NC6+, NC8+ and NC11+; with no effect of the enzyme usage for FCR of other NC+ diets, relative to the PC and respective NC- diets.
Based on the previous relationship established between animal performance and control diet nutrient content, ME dAA and avP released by MCPC might be estimated at 0.56 MJ/kg and 0.06% units and 0.15% units, respectively. The established relationship between diet nutrient content (ME, dAA, avP) and the MCPC effect precisely described the broiler response with high R square between predicted and observed values for BWG, FI and FCR (R² = 0.98, 0.99 and 0.81, respectively).
IV. DISCUSSION
This study is a primary work to further define the interactions among nutrients for broiler performance, as demonstrated by Sharma et al. (2017). The fitted models clearly illustrate a main deleterious effect of P deficiency on each indicator of growth performance. In broilers, P deficiency results in a loss of appetite (Underwood and Suttle, 1999) and reduced growth. Restoration of FI by MCPC supplementation indicates the effectiveness of phytase to release phosphorus and reach FI similar to those of animal fed at the requirement or event be above the requirement. This finding is similar to those of Letourneau et al. (2010) who found a positive effect of phytase enzyme on FI and the effect was negatively correlated with the amount of avP in the diet. Interaction was also found with dietary phytate but this effect was not tested in present experiment.
The AA requirements of growing broilers have been frequently estimated using an empirical method (Mack et al., 1999). These studies evaluating requirements did not allow evaluation of interaction between digestible amino acid requirement and other nutrient such as avP and ME. The results suggest an interaction between avP and dAA for FCR and ME and dAA for FI. Inclusion of MCPC in the diet with effects on BWG and FCR of the birds might be partly associated with dLys and other amino acid release. These finding are in accordance with previous results obtained with phytase (Walk et al, 2013) or with carbohydrase (Cozannet et al., 2017).
Finally, the broiler chicken seems immune to the effects of variable diet energy level on general growth and development. Contrary to the observations of Newcombe and Summers (1984), results from the current study suggest that the broiler has a remarkable ability to control energy intake when offered diets of varying energy content. As expected, negative relationships have been found between ME content of the diet and FI. This finding is in line with previous results of Leeson et al. (1996). Hence, exogenous enzymes remove the nutrient encapsulating effect of NSP in broiler diets, thereby improving nutrient access for endogenous enzymes and enhancing overall feed digestibility (i.e., starch, fat, and CP; Meng et al., 2005). As a result, the ME content of the diets also improved via gross energy from digestible nutrients including amino acids previously described.
V. CONCLUSION
This study affirmed that significant reductions in ME, dAA and avP are essential to optimize the benefits of substrate degrading efficacy of enzymes on performance and efficiency of broilers. Therefore, interrelationships among dietary nutrients might also be considered. Imbalance among nutrients might be associated with higher decrease of performance than expected and use of additives might even result in detrimental effects accentuating imbalance, hence the inconsistent effect of the MCPC. This study presents an interesting design for evaluation of animal requirement and MCPC matrix value. Equations of present experiment were the result of one single trial and required further data set to be validated. Many miscellaneous points have not been considered in connection with missing freedom degree. In the following development of such disposal, it might be interesting to test such as amounts of Ca and their interactions with avP or to amounts of substrates such as arabinoxylans or phytate.
Abstract presented at the 30th Annual Australian Poultry Science Symposium 2019. For information on the next edition, check out http://www.apss2022.com.au/. 
