【作者】 黄俊；

【导师】 梅乐和；

【作者基本信息】 浙江大学 ， 生物化工， 2006， 博士

【摘要】 γ-氨基丁酸（γ-aminobutyric acid，简称为GABA）是一种天然存在的非蛋白质氨基酸，为哺乳动物中枢神经系统中重要的抑制性神经递质，具有降血压、治疗癫痫、镇静安神、增强记忆、控制哮喘、调节激素分泌、促进生殖、活化肝肾等多种生理功能，其制备和应用广受人们的关注与重视。本文系统地研究了利用短乳杆菌生物法制备GABA的过程及相关工艺。论文在建立γ-氨基丁酸分析方法的基础上，从自然界选育得到了一株GABA高产菌株；通过对摇瓶培养条件和发酵罐培养条件的优化，提高了GABA的发酵产量；采用游离细胞和固定化细胞对催化合成GABA的反应条件进行了优化。此外，对短乳杆菌细胞内催化合成γ-氨基丁酸唯一的关键限速酶—谷氨酸脱羧酶进行了分离纯化，并对其酶学性质进行了初步的研究。首先，建立了GABA的定性和定量方法，纸层析和薄层层析法用于GABA定性分析，高效液相色谱法（HPLC）用于GABA定量分析。采用HPLC定量分析GABA时，以丹磺酰氯作柱前衍生剂，样品的pH高于7.5以及衍生化时间超过20min，衍生化反应稳定；GABA浓度（C）在0.1～2mM之间与峰面积（A）线性关系良好，线性方程为A=313.242·C+8.314，相关系数R为0.9997，GABA的加标回收率为95.30％～106.34％。其次，从未灭菌的新鲜牛奶中筛选到一株产GABA的菌株，记为hjxi-01，发酵72h后GABA的产量6.9g／L。根据菌株hjxj-01的菌落形态以及生理生化特性，鉴定为短乳杆菌Lactobacillus brevis hjxj-01。以Lactobacillus brevis hjxj-01为出发菌株，经UV和⁶⁰Coγ-射线反复诱变处理，以含不同浓度的GABA梯度平板进行筛选，得到一株高产GABA的突变株Lactobacillus brevis hjxj-08119，发酵72hGABA的产量17g／L，比野生菌株产量提高了140％，此高产突变株遗传性状稳定，传代12次，菌株未出现回复突变情况。高产突变株Lactobacillus brevis hjxj-08119已在中国微生物菌种保藏管理委员会（CGMCC）保藏，保藏编号为CGMCCNO.1306。第三，利用Plackett—Burman设计对影响Lactobacillus brevis hjxj-08119发酵生产GABA的15种相关因素进行了效应评价，筛选出具有显著正效应的三个因素：葡萄糖、MnSO₄·4H₂O和L-谷氨酸钠（L-MSG），其他因素对GABA产量无显著影响。然后对具有显著正效应的三个因素进行进一步考察，确定了三个因素的取值范围，以Hybrid设计得到的实验数据作为人工神经网络（ANN）训练样本，建立BP（back propagation）神经网络模型，粒子群（PSO）算法对建立的ANN模型进行全局寻优，得到最佳培养基组成为（g／L）：葡萄糖17.6，酵母膏15，蛋白胨5，乙酸钠3，MgSO₄·7H₂O 0.03，MnSO₄·4H₂O 0.02，NaCl 0.001，FeSO₄·7H₂O 0.001，L-MSG73.3，初始pH6.8，发酵温度30℃，250mL三角瓶中的装液量为50mL／250mL。在此发酵条件下，GABA的发酵产量达到33.42g／L，较突变株Lactobacillus brevishjxi-08119的发酵产量17g／L提高了97％。第四，在3.7升发酵罐中对GABA发酵的操作条件（溶氧和pH控制）以及补料进行研究。结果表明，好氧发酵有利于菌体生长，最大菌体干重可达2.78g／L，而厌氧发酵则有利于产物GABA的生成，发酵72h后GABA的浓度达到23.94g／L。在兼性厌氧条件下，不同pH控制方式下对GABA分批发酵的影响有较大差异，控制pH为5.0时GABA产量最高，达到40.73g／L（72h）。对控制pH5.0的GABA发酵过程进行分析，基于Logistic方程和改进的Luedeking-Piret方程，分两阶段建立控制pH5.0发酵生产GABA的动力学模型，该模型能较好地预测细胞生长、底物消耗以及GABA合成过程。对GABA的补料发酵进行了初步研究，经过四次补料，GABA的最终浓度达到76.36g／L，分别比摇瓶发酵、厌氧发酵以及控制pH5.0的发酵产量提高了128.6％、219％和87.5％。第五，运用ANN模型结合PSO算法对游离短乳杆菌细胞催化合成GABA的反应条件进行了优化，得到的最优催化反应条件为：收集发酵60h的短乳杆菌细胞进行催化，缓冲体系为25mL柠檬酸—磷酸氢二钠缓冲液（100mM，pH4.23），反应液中包括120mM L-MSG，0.83g／L FeSO₄·7H₂O，10μM 5’-磷酸吡哆醛（PLP），2.68g DCW／L，在40℃恒温静置反应5h。在得到的最佳催化条件下，进行了四次实验验证，得到GABA的平均产量为9.34±0.22g／L，预测结果（9.4g／L）与实验情况吻合较好。对响应面（RSM）模型和ANN模型的预测性能进行了比较，结果表明，ANN模型的预测性能稍优于RSM模型。第六，对利用海藻酸钙胶珠包埋短乳杆菌细胞催化合成γ-氨基丁酸的反应特性进行了研究。优化的胶珠中细胞密度为11.2g细胞干重（DCW）／L，反应最适的pH和温度分别为pH4.4和40℃；细胞经固定化后，温度稳定性显著提高。固定化连续进行10批次催化反应后（共80h），GABA的产率仍在50％以上，并且胶珠仍保持较好的完整性。最后，对Lactobacillus brevis hjxj-08119细胞中的谷氨酸脱羧酶（GAD）进行了分离纯化，建立了有效分离纯化GAD的工艺流程，即“溶菌酶处理——→French press细胞破碎——→30％～90％硫酸铵分级盐析——→Q Sepharose FF离子交换层析——→Sephacryl S-200凝胶过滤层析——→Resource Q离子交换层析”。整个工艺活性收率为16.95％，纯化倍数为43.78，比活达到了3.94U／mg（蛋白），由SDS-PAGE得到的GAD分子量约为62kD。对纯化后GAD的酶学性质进行了初步研究，最适反应温度为37℃，最适反应pH为4.4，PLP对酶活力无显著激活作用，酶最大反应速度V_max为6.59U／mg，常数K_m为8.22mM。更多还原

【Abstract】 Gamma-aminobutyric acid （GABA）, a four-carbon nonprotein amino acid, serves as a major inhibitory neurotransmitter in mammalian nervous systems. GABA has several physiological functions such as hypotensive activity, treatment of epilepsy, tranquilizing and allaying excitement, enhancing memory, controlling asthma, regulating hormone secretion, promoting reproduction and activating liver and kidney function. Preparation and application of GABA are concerned.Preparation of GABA by Lactobacillus brevis was studied in detail in this dissertation. Based on determination of GABA analytical method, GABA-producing mutant strain was bred; the fermentation conditions in shake flask and fermenter were optimized; biosynthesis of GABA using free and immobilized Lactobacillus brevis cells was investigated. Furthermore, purification and characterization of glutamate decarboxylase（GAD） which is a unique enzyme synthesizing GABA were studied.First, paper and thin-layer chromatography were used as qualitative analysis of GABA, and high performance liquid chromatography（HPLC） as quantitative determination of GABA. With Dansyl chloride （DNS-Cl） as pre-column derivatization, derivatization reaction was stable when pH of sample was higher than 7.5 and reaction time exceeded 20 min. Peak area （A） had good linearity with GABA concentration（C） when GABA concentration ranged from 0 to 2 mM. The regression equation was A=313.242·C+8.314 and correlation coefficient was 0.9997. Average recoveries were in the range of 95.30%～106.34%.Second, a GABA-yielding strain （hjxj-01） was isolated from fresh milk without pasteurization, and GABA production reached 6.9 g/L after 72h fermentation. The strain hjxj-01 was named as Lactobacillus brevis according to its colony morphology as well as physiological and biochemical properties. Applying ultraviolet and ⁶⁰Co γ-ray to mutagenize Lactobacillus brevis hjxj-01, a mutant strain hjxj-08119 was selectively bred by GABA resistance selection. After 72h fermentation, the mutant strain gave a GABA output of 17 g/L, 140% higher than that of the parent strain （hjxj-01）. After 12 generation, the mutant strain had stable yield of GABA and no reverse mutation. The mutant strain is kept China General Microbiological Culture Collection Center （CGMCC） as Lactobacillus brevis CGMCC NO.1306.Third, artificial neural network （ANN） and particle swarm optimization （PSO） were used to optimize GABA production by Lactobacillus brevis CGMCC NO.1306 in shake flask. Firstly, glucose, sodium glutamate（L-MSG） and MnSO₄·4H₂O, which influencedGABA production positively were screened from 15 related factors by using Plackett-Burman design. The reasonable ranges of these three factors were determined by single factor experiment. Then experimental samples of hybrid design were selected for training ANN, and the ANN was modeled. Finally, based on the ANN model, the optimized condition was predicted by particle swarm optimization（PSO） algorithm. The optimal medium composition in shake flask was determined as follows （g/L）: glucose 17.6, yeast extract 15, peptone 5, CH₃COONa 3, MgSO₄·7H₂O 0.03, MnSO₄·4H₂O 0.02, NaCl 0.001, FeSO₄·7H₂O 0.001, L-MSG 73.3. The fermentation should be performed at 30℃, pH 6.8, and medium volumetric ratio 20%. After 72h fermentation, the yield of GABA reached 33.42 g/L, 97% higher than fermentaion conditions before optimization.Fourth, the effects of operation conditions （aeration and pH） on GABA batch fermentation by Lactobacillus brevis hjxj-08119 in a 3.7 liters stirred fermenter were investigated. The results showed that operation conditions （dissolved oxygen and pH） had the significant effects on GABA production. The aerobic cultivation was advantageous to high level of dry cell weight （2.78 g/L）, but the anaerobic cultivation was advantageous to GABA accumulation that GABA yield reached 23.94 g/L at 72h. To assess the effects of pH on GABA production, three batch processes with pH control at 4.5, 5.0 and 5.5 respectively, were conducted in facultative anaerobic cultivation. The yield of GABA was the highest at pH 5.0 and reached 40.73 g/L at 72h. The kinetic models for two stages were established based on the Logistic and modified Luedeking-Piret equations involving cell growth, product formation and substrate consumption for GABA fermentation process with pH control at 5.0. The parameters of the models were obtained by using Matlab 6.0 software with experimental data and the models. With the evaluated models parameters, the calculated values of the models and experimental data are in a good agreement. The fed-batch fermentation of GABA was preliminarily studied. After four times L-MSG addition, the yield of GABA reached 76.36 g/L at 108h, and the yield which fed-batch fermentation significantly enhanced GABA yield were 128.6%, 219% and 87.5% respectively higher than shake flask fermentation, anaerobic fermentation and fermentation with pH control at 5.Fifth, ANN and PSO were applied to the biotransformation of L-MSG to GABA catalyzed by the free cells of Lactobacillus brevis hjxj-08119. The modeled maximum GABA yield reached 9.4 g/L under the following optimal conditions: 25 mL Na₂HPO₄-citric acid buffer （100 mM, pH 4.23）, 120 mM L-MSG, 0.83 g/LFeSO₄·7H₂O, 10 μM 5’-pyridoxal phosphate（PLP）, the resting cells obtained from a 60-h culture broth, 2.68 g dry cell weight （DCW）/L and without agitation at 40℃ for 5 h. The average value of the four experimentally tested GABA yield was 9.34±0.22 g/L compared with a value of 9.4 g/L by ANN coupling PSO. The prediction capacity between ANN and response surface methodology （RSM） was compared. The results demonstrated a slightly higher prediction accuracy of ANN compared to RSM.Sixth, by entrapping the Lactobacillus brevis cells into Ca-alginate gel beads, the biotransformation conditions of L-MSG to GABA were optimized with the immobilized cells. The optimal cell density in gel beads, reaction pH and temperature were 11.2 g DCW/L, 4.4 and 40℃ respectively. GABA yield still reached more than 50% and Ca-alginate gel beads kept integrity after ten-time recycling （80 h） of the immobilized cells.Finally, effective isolation and purification procedure of GAD from Lactobacillus brevis hjxj-08119, including lysozyme treatment, French press disruption, 30～90% saturation （NH₄）₂SO₄ fractional precipitation, Q Sepharose FF anion-exchange chromatography, Sephacryl S-200 gel filtration chromatography and Resource Q anion-exchange chromatography, was brought forward. Using this protocol, the purified GAD was demonstrated to possess electrophoretic homogeneity via SDS-PAGE. The purification fold, activity recovery and specific activity of GAD were 43.78, 16.95% and 3.94 U/mg（protein）, respectively. The molecular weight of the purified GAD was estimated to be approximately 62 kDa via SDS-PAGE. The optimum pH and temperature of the purified GAD were 4.4 and 37℃, respectively. The V_max and K_m value of the GAD enzyme from Lineweaver-Burk plot was found to be 6.59 U/mg and 8.22 mM. PLP had little effect on the regulation of its activity.更多还原