重组E. coli生产类人胶原蛋白发酵调控策略与500L中试规模放大方法优化

【作者】 薛文娇；

【导师】 范代娣；

【作者基本信息】 西北大学 ， 生物化工， 2009， 博士

【摘要】 本研究旨在建立一套完整的重组Escherichia coli生产类人胶原蛋白(human-like collagen,HLC)的中试规模生产工艺,并通过对各单元操作从实验室规模到中试规模的放大研究,为进一步将该工艺放大到工业生产奠定基础。重组E.coli的高密度培养是HLC生产过程的核心单元,其放大的成功与否直接关系到整个工艺过程。为了明确影响发酵过程的关键因素,从而选择合适的放大准则和方法,本文对发酵过程中二氧化碳和乙酸的影响进行了系统研究。尽管有大量文献表明,二氧化碳对重组E.coli发酵有负面影响,然而关于二氧化碳对重组E.coli,特别是重组蛋白表达方面的定量研究却未见报道。本研究采用二氧化碳脉冲法首次考察了重组E.coli生长不同阶段二氧化碳浓度对重组蛋白生产的影响。研究结果表明,当在分批培养阶段施加浓度为20%(v/v)二氧化碳脉冲时,细胞生长明显受到抑制;而在补料阶段引入该脉冲时,却引起了比生长速率的小幅上升;二氧化碳对HLC表达的抑制只发生在HLC的表达阶段,即在蛋白表达诱导阶段引入浓度为20%(v/v)二氧化碳脉冲3小时,则最终HLC浓度下降33.9%。尽管二氧化碳脉冲的浓度越高,作用时间越长,其相应的抑制或促进作用也越明显,但当二氧化碳脉冲浓度小于等于10%时,对发酵过程却几乎没有影响。然后,针对重组E.coli发酵过程的关键问题-乙酸抑制作了系统研究。研究结果表明,在分批培养阶段乙酸对E.coli细胞有明显的抑制作用,但在补料培养阶段,其影响却并不明显;而乙酸对HLC表达的抑制只发生在HLC的表达阶段。本文首次建立了这一评价重组E.coli细胞生长不同阶段乙酸影响的方法。而建立该法的理论依据就是在葡萄糖饥饿阶段,重组E.coli细胞可以利用乙酸。为此,本研究也首次讨论了在细胞生长不同阶段,利用葡萄糖饥饿以诱导乙酸吸收对发酵过程的影响。在发酵放大过程中,合适的种子扩大培养过程的建立对于成功的放大也至关重要。为了建立最优的种子扩大培养过程,本研究考察了三级种子不同移种阶段和不同种子培养基浓度对发酵过程的影响。结果表明:在对数期后期移种,HLC产率最高;另外,当种子培养基葡萄糖浓度为20g/L时,其后发酵过程所需的培养时间较短,HLC表达量较高,HLC平均产率最高,达到0.518g/L/h。由乙酸的相关研究可知,乙酸对本表达体系具有较强的抑制作用,因而本研究建立了以控制乙酸产生为根本目的的溶氧探测补料技术,该技术具有很强的适应性,可在不同的环境中迅速得到发酵过程的最佳补料方案。采用溶氧探测补料技术,在实验室规模获得的最终细胞干重和HLC浓度分别为69.1g/L和13.1g/L,该值与以前根据比生长速率优化的结果基本一致。而放大到中试规模时,尽管乙酸浓度和实验室规模基本相同,优化诱导时机后获得的HLC表达水平也基本相同,但最终细胞干重和HLC浓度有一定程度下降,分别为51.7g/L和9.6g/L。直接放大后HLC产量下降的原因之一可能是不同规模发酵罐氧传递能力不同所致。针对该问题,通过测定不同规模发酵罐的体积氧传递系数,并建立了体积氧传递系数对操作参数(通气量、搅拌速度和装液量)的关联方程,使得不同规模的发酵罐在氧的传递方面具有可比性。实验室小规模发酵罐k_La值对操作参数的经验关联式为:中试规模发酵罐k_La值对操作参数的经验关联式为:最后,根据体积氧传递系数的关联方程,分别以k_La和p~*k_La为放大基准进行放大。由于本发酵体系对二氧化碳较不敏感,中试规模(500L)中加压操作得到了成功应用。即以p~*k_La为放大基准进行放大规模的不诱导的分批一补料培养,最终获得的细胞干重为77.9g/L,略低于实验室规模培养获得的细胞干重(80.3g/L),也低于以k_La为放大基准获得的中试规模培养的最终细胞干重(90.0g/L)。但以p~*k_La为基准放大,初始装液量(285L)远高于以k_La为基准放大时的初始装液量(105L);诱导培养后,最终细胞干重达到68.4g/L,最终HLC浓度为13.0g/L,基本同实验室规模。此外,本文还建立了HLC中试纯化工艺,并对其中影响蛋白收率和蛋白纯度的主要工艺单元:高压匀浆、沉淀、超滤及离子交换层析进行了放大研究。结果表明,高压匀浆不用改变任何操作参数可直接放大;超滤过程仅改变和膜面积相关参数也可得到满意的放大结果;对于沉淀单元,由于搅拌釜装置变化,在中试规模重新优化了沉淀时间:即沉淀70min后纯化效果最好;保持线速度恒定进行层析单元的放大也获得了较好结果。中试规模纯化工艺得到HLC的纯度为95.4%,收率为65.4%,与实验室规模纯化结果相比(HLC纯度与收率分别为96.4%,71.7%)仅略有下降。更多还原

【Abstract】 The aim of this investigation was to develop a process for production of human-like collagen （HLC） by recombinant Escherichia coli on pilot-scale. In addition, the study would lay a sound basis for scaling-up to production scale further, through scale-up investigations for unit operations of the process.High cell density culture of recombinant E. coli is the key unit for HLC production and its scale-up performance would determine the whole process efficiency. To find the key factors influencing the fermentation process and thus chose the scale-up criteria and method properly, the effects of carbon dioxide and acetate on HLC production was studied firstly.Carbon dioxide plays a major role in both aerobic and anaerobic fermentations. However, few literatures focused on the effects of carbon dioxide on recombinant E. coli culture and target protein production. In this paper, a systematic study was initially carried out with the carbon dioxide pulse injection method to investigate the influence of carbon dioxide on the cultivation system of recombinant E. coli. The CO₂ pulse injection experiments showed that: 1) a 20% CO₂ pulse introduced in the batch cultivation phases inhibited cell growth obviously while that introduced in the fed-batch cultivation phases stimulated cell growth slightly; and 2) inhibitory effect of CO₂ on HLC expression occurred only in the expression phase, where the final HLC concentration decreased by 33.9% under a 3-h 20% CO₂ pulse. Meanwhile, the higher the CO₂ concentration and/or the longer the duration of the CO₂ pulses, the stronger the stimulatory or inhibitory effects. However, the CO₂ concentration below or equal to 10% had little influence on the HLC production.Then, a systematic study was carried out to determine the effects of acetate on the high cell density culture of recombinant E. coli containing HLC mRNA. In this study, the effects of acetate assimilation through a glucose starvation period at different cell growth phases were initially investigated in a fed-batch culture of recombinant E, coli. The results showed that acetate assimilation was observed without negative effects on the process during a glucose starvation period introduced at all defined cell growth phases except for the post-induction phase. In addition, a new method for evaluating the effects of acetate was development based on the above findings: the acetate concentration was controlled by adding acetate into culture media and/or employed a glucose starvation and thus the effects of acetate at different cell growth phase were determined. Experimental results show that obvious acetate inhibition on cell growth occurred in the phases of batch culture while its inhibitory effect on HLC expression occurred only in the expression phase.The development of a suitable inoculum is important to the fermentation process. To optimize the inoculum preparation process, the third inoculum transfer stage and medium were determined. The results showed that HLC productivity was highest when the third inoculum was transferred at lately exponential stage; in addition, when the glucose concentration was 20 g/L, the following fermentation time was shortest and the final HLC concentration was highest, and thus the HLC average productivity was highest （achieved 0.518 g/L/h）.Acetate had a strong negative influence on the HLC production according to this work, so a feed-back controlled method that the probing feeding strategy was developed to control the acetate production. The method was employed to maintain dissolved oxygen concentration at the same level on both laboratory scale and pilot scale fermentations. The final DCW of 69.1 g/L and HLC concentration of 13.1 g/L were obtained on the laboratory scale and the results are similar to that optimized in the previous study. The final DCW of 51.7 g/L and HLC concentration of 9.6 g/L were obtained on the pilot-scale upon optimization of induction timing.The reason for the decrease in the HLC production on the pilot scale may be the lower oxygen transfer rate for the pilot-scale fermentor. Therefore, the volumetric mass transfer coefficient k_La on different scale fermentors was determined and the correlations between the k_La and operation parameters （aeration, stirrer rate and initial medium volume） were developed. And thus the oxygen transfer capacities for different fermentors could be compared. The correlations for k_La on the lab-scale fermentor is,And the correlations for k_La on the pilot-scale fermentor is, Finally, the scale-up fermentations based on k_La and p*k_La were carried out respectively, according to the correlations for k_La calculation. Because the recombinant system was less sensitivity to the CO₂, pressurized culture was successful for the HLC production. That is, for the non-induced cultivation on the pilot-scale based on the constant p*k_La scale-up criteria, the final DCW was 77.9 g/L, which was similar to that （80.3 g/L） obtained on the lab-scale cultivation and lower than the value （90.0 g/L） obtained in the pilot-scale based on the constant k_La scale-up criteria. However, the initial medium volume for the former （285 L） was much higher than the latter （105 L）. The final DCW and HLC concentration were 68.4 g/L and 13.0 g/L repectively, which were similar to those obtained on the lab-scale, for the induced cultivation on the pilot-scale.The process for HLC purification on the pilot-scale was also developed in this study. The major operation units influencing HLC recovery and purity were homogenization, precipitation, ultrafiltration and CM52 chromatography. The scale-up investigation showed that, 1) the operation parameters could be directly transferred to the larger scale for the homogenization; 2) the operation parameters related to the membrane area were increased proportionally for scaling-up the ultrafiltration process; 3) the precipitation time was optimized on the pilot-scale and the optimized precipitation was 70 min, which is little higher than that on the lab-scale; 4) CM52 chromatography scaling-up based on constant linear velocity was successful. The purity and recovery of HLC reached 95.4%, 65.4%, repectively, which were decreased slightly compared to those （the putity and recovery were 96.4%,71.7%, respectively） obtained on the lab-scale.更多还原