实验设计法优化可溶性人肿瘤坏死因子αⅠ型受体包涵体的复性工艺  

Optimization of refolding process for soluble tumor necrosis factorα receptor Ⅰ inclusion body by Design of Experiment

作　　者：秦娇荣 蔡峰[1] 赵兆[1] 代云见[1] 李春阳[1] QIN Jiao-rong;CAI Feng;ZHAO Zhao;DAI Yun-jian;LI Chun-yang(Chengdu Institute of Biological Products Co.,Ltd,Chengdu 610023,Sichuan Province,China)

机构地区：[1]成都生物制品研究所有限责任公司重组蛋白质药物及新疫苗研究室,四川成都610023

出　　处：《中国生物制品学杂志》2018年第10期1142-1149,共8页Chinese Journal of Biologicals

摘　　要：目的通过实验设计(Design of Experiment,DoE)方法对重组的可溶性人肿瘤坏死因子αⅠ型受体(soluble tumor necrosis factorαreceptorⅠ,sTNFαRⅠ)包涵体复性工艺中的关键参数进行优化。方法采用DoE软件中的中心复合设计(Central Composite Design)建立二次模型(Quadratic Model),对复性工艺中游离巯基浓度(C_(-SH))和复性液蛋白浓度(C-Pro)2个关键参数进行优化。设置2个响应值:每克湿菌体中所获得的sTNFαRⅠ蛋白收率和每制备10 g sTNFαRⅠ蛋白所需复性体积(V-10 g),通过DoE软件对模型数据进行评估分析。在sTNFαRⅠ收率最高及V_(-10g)最小的前提下,采用二次模型响应面分析预测最优的复性参数值,并通过两次放大试验对优化结果进行验证。结果由2因素5水平2个Block共14组试验结果所建立的DoE响应面二次模型中心点重复性好,试验点均符合正态分布。响应面二次回归分析中,分别以sTNFαRⅠ收率和V_(-10g)作为响应值进行失拟检验的P均>0.05,模型失拟不显著。预测值和真实值接近,模型拟合度好。信噪比均>4,表明该模型有效,可用于复性工艺优化指导。通过以上模型得出:当C_(-Pro)为0.48 mg/m L,复性起始C_(-SH)为8.69 mmol/L时模型愿望值(0.643)最高,sTNFαRⅠ收率为16.38 mg/g,V_(-10g)为36.73 L。通过两次放大试验对优化结果进行验证,sTNFαRⅠ收率分别为11.34和9.72 mg/g,与优化前(收率均值5.1 mg/g)相比,有所提高。结论 DoE软件中心复合设计的二次模型可用于sTNFαRⅠ包涵体复性条件的筛选,且筛选结果为后续工艺放大提供了可靠的实验依据。Objective To optimize the key parameters in refolding process of recombinant soluble human tumor necrosis factor a receptor I （sTNFα R I ） by Design of Experiment （DOE）. Methods Two key technological parameters in refolding process, free sulthydryl concentration （C-SH） and protein concentration （C-pro）, were optimized by using the Central Composite Quadratic Model established by Central Composite Design in DoE software. Two response values, the yield of sTNFetR I of each gram of wet bacteria and refolding volume （V.10g） for preparation of 10 g of sTNFctR I , were set. The experimental data of model were evaluated by using DoE software. Under the premise of the highest yield of sTNFa R I and the minimum V-10g, the optimal refolding parameter values were predicted by response surface of quadratic model. The optimization results were validated by two scaled-up experiments. Results The center points of response surface of quadratic model established by 14 groups of tests of two Blocks at 2 factors and 5 levels showed high reproducibility, and the test points were showed a normal distribution. In regression analysis of response values for quadratic response surface model fitting, the lack of fit P values using sTNFα R I and V-10s as response values were more than 0. 05, indicating that the lack of fit was insignificant. The predicted values are closed to the actual values, indicating high goodness of fit. Adeq Precisior values were more than 4, indicating that the model was effective and could be used to navigate the design space of optimization of refolding process parameters. The results showed that the yield of sTNFα R I was 16. 38 mg / g and the V-10s was 36. 73 L when the C-por was O. 48 mg/mL and the C.sH was 8. 69 mmol/L in refolding process and thehighest desirability of model was 0. 643. However, in validation by two scale-up experiments, the yields of sTNFα R I were 11. 34 and 9. 72 mg/g respectively, which were higher than that before optimization （5. 1 mg/g in average）. Conclusi

关 键 词：实验设计 包涵体 复性 可溶性人肿瘤坏死因子αⅠ型受体 中心复合设计 响应面 

分 类 号：O212.6[理学—概率论与数理统计] R392-33[理学—数学]