产蛋鸡玉米代谢能和氨基酸可利用率的评定

Evaluation of Metabolizable Energy and Amino Acid Availability of Corn in Laying Hens

本试验旨在评定产蛋鸡对30种不同来源玉米的代谢能和氨基酸可利用率，并探索适合于产蛋鸡饲料营养价值评定的方法。选择240只体重（1.60±0.10） kg、产蛋率90％的罗曼蛋鸡，按单因素完全随机设计分为30个组，每组8个重复，每个重复1只鸡。采用“训饲＋全收粪法”测定30种不同来源玉米的代谢能和氨基酸可利用率，并在此基础上利用傅里叶近红外光谱（NIRS）分析技术建立玉米代谢能和氨基酸可利用率的NIRS预测模型。结果表明：1）产蛋鸡采食含89％玉米的30种试验饲粮后，产蛋率由90.00％下降至76.77％，蛋重、蛋白高度、蛋黄颜色、哈夫单位、蛋壳强度、蛋壳厚度、蛋黄比重的平均值分别为51.17 g、7.06 mm、8.08、84.62、3.80 kgf、31.14×10^-2 mm、28.15％。2）30种不同来源玉米的代谢能范围为11.37～16.91 MJ/kg，平均值为15.26 MJ/kg；总氨基酸可利用率范围为82.06％～93.23％，平均值为87.47％；总必需氨基酸可利用率范围为82.37％～92.97％，平均值为86.70％；必需氨基酸中缬氨酸（Val）、异亮氨酸（Ile）、亮氨酸（Leu）、苯丙氨酸（Phe）、赖氨酸（Lys）、组氨酸（His）、精氨酸（Arg）、苏氨酸（ Thr）可利用率的平均值分别为91.59％、74.52％、88.14％、93.13％、77.93％、86.06％、86.78％、95.42％。3）玉米代谢能NIRS预测模型的校正决定系数（ R2cal ）、交叉验证系数（ R2cv ）、外部检验系数（R2val）分别为0.99、0.94、0.95，预测均方根误差（RMSEP）为0.13 MJ/kg；氨基酸可利用率NIRS预测模型的R2cal、R2cv、R2val范围分别为0.91～0.99、0.81～0.98、0.79～0.97，RMSEP范围为0.10％～1.66％。由此得出：1）使用“训饲＋全收粪法”评定产蛋鸡玉米的代谢能和氨基酸可利用率时，未对产蛋鸡的产蛋率和蛋品质产生较大影响；2）不同来源玉米间的代谢能、氨基酸可利用率存在差异；3）通过测定产蛋鸡玉米的代谢能和氨基酸可利用率，可建立起玉米代谢能和氨基酸可利用率的NIRS预测模型。

This study was conducted to evaluate metabolizable energy （ ME） and amino acid availability of 30 corns from different sources, and to explore new methods for determining the nutritive values of feeds suitable for laying hens.A total of 240 Lohmann laying hens with an average body weight of （1.60±0.10） kg and 90%laying rate were randomly divided into 30 groups by single-factor completely random design.Each group had 8 replicates and 1 laying hen in each replicate.ME and amino acid availability of 30 corns from different sources were investigated by the trained feeding combined with a total collection of excreta, and then the near infrared spectroscopy （ NIRS） prediction models of corn ME and amino acid availability were established by Fourier NIRS analysis technique on the basis of determining the ME and amino acid availability of 30 corns from different sources.The result showed as follows：1） the laying rate reduced to 76.77%from 90.00%, and the aver-age values of 30 corns of egg weight, albumen height, yolk color, Haugh unit, egg shell strength, egg shell thickness and yolk relative weight were 76.77%, 51.17 g, 7.06 mm, 8.08, 84.62, 3.80 kgf, 31.14×10^-2mm and 28.15%, respectively, after laying hens fed 30 experimental diets including 89%corn.2） The ME range of 30 corns from different sources was 11.37 to 16.91 MJ/kg, with average value of 15.26 MJ/kg.The ranges of total amino acid availability and total essential amino acid availability of 30 corns from different sources were 82.06%to 93.23%and 82.37%to 92.97%, with average values of 87.47%and 86.70%, respectively.The average values of 30 corns of valine, isoleucine leucine, phenylalanine, lysine, histidine, arginine and threo-nine availability were 91.59%, 74.52%, 88.14%, 93.13%, 77.93%, 86.06%, 86.78%and 95.42%, re-spectively.3） R2cal, R2cv, R2val and root-mean-square error of prediction （RMSEP） of NIRS prediction model for corn ME were 0.99, 0.94, 0.95, 0.13 MJ/kg.The ranges of R2cal, R2cv, R2val and RMSEP of NIRS prediction models for corn amino acid availability were 0.91 to 0.99, 0.81 to 0.94, 0.79 to 0.97, and 0.10%to 1.66%. These results indicate as follows：1） the method of the trained feeding combined with a total collection of ex-creta do not cause bigger effects on laying rate and egg quality when evaluating ME and amino acid availability of corn for laying hens;2） there are variations of ME and amino acid availability between corns from different sources;3） the NIRS prediction models of corn ME and amino acid availability are established according to the values of corn ME and amino acid availability in laying hens.