Data from: Growth profiling, kinetics and substrate utilization of low-cost dairy waste for production of β-cryptoxanthin by Kocuria marina DAGII

Dairy industry produces enormous amount of cheese whey compromising of major milk nutrients but remains unutilized all over the globe. The present study investigates the production of β-Cryptoxanthin (β-CRX) by Kocuria marina DAGII using cheese whey as substrate. Response surface methodology (RSM) and artificial neural network (ANN) was implemented to obtain the maximum β-CRX yield. Significant factors viz. yeast extract, peptone, cheese whey and initial pH were the input variables in both the optimizing studies and β-CRX yield and biomass were taken as output variables. The ANN topology of 4-9-2 was found to be optimum when trained with feed-forward back propagation algorithm. Experimental values of β-CRX yield (17.14 mg/L) and biomass (5.35 g/L) were compared and ANN predicted (16.99 mg/L and 5.33 g/L respectively) values were found to be more accurate compared to RSM predicted values (16.95 mg/L and 5.23 g/L respectively). Detailed kinetic analysis of cellular growth, substrate consumption and product formation revealed that growth inhibition took place at substrate concentrations higher than 12%(v/v) of cheese whey. Han and Levenspiel model was the best fitted substrate inhibition model that described the cell growth in cheese whey with a R2 and MSE of 0.9982 and 0.00477%, respectively. The potential importance of this study lies in the development, optimization, modelling and characterization of a suitable cheese whey supplemented medium for increased β-CRX production.

乳制品行业会产生大量干酪乳清（cheese whey），其富含牛乳的主要营养成分，但在全球范围内均未得到充分利用。本研究以干酪乳清为底物，利用海洋库特氏菌（Kocuria marina）DAGII合成β-隐黄质（β-Cryptoxanthin，β-CRX）。为获得最高的β-隐黄质产率，本研究采用响应面法（Response Surface Methodology，RSM）与人工神经网络（Artificial Neural Network，ANN）开展优化研究。显著影响因素包括酵母提取物、蛋白胨、干酪乳清与初始pH值，两项优化方法均以这些参数作为输入变量，而以β-隐黄质产率与生物量作为输出变量。当采用前馈反向传播算法（Feed-forward Back Propagation Algorithm）进行训练时，4-9-2拓扑结构的人工神经网络为最优拓扑。将β-隐黄质产率（17.14 mg/L）与生物量（5.35 g/L）的实验值，分别与人工神经网络预测值（依次为16.99 mg/L与5.33 g/L）及响应面法预测值（依次为16.95 mg/L与5.23 g/L）进行对比，结果显示人工神经网络的预测结果更为准确。对细胞生长、底物消耗与产物生成的详细动力学分析表明，当干酪乳清浓度高于12%(v/v)时，会发生细胞生长抑制现象。Han-Levenspiel模型是拟合干酪乳清中细胞生长的最优底物抑制模型，其决定系数（Coefficient of Determination，R²）与均方误差（Mean Squared Error，MSE）分别为0.9982与0.00477%。本研究的潜在价值在于，开发、优化、建模并表征了一种适配的干酪乳清补加培养基，以提升β-隐黄质的合成产量。