Sistema automático de supervisão e controle de cultivos de alta densidade celular de E. coli recombinante

Abstract

High cell density cultivations of recombinant E. coli are a fast and economical way to produce recombinant proteins. Through this bioprocess, products with high added value and pharmaceuticals of great importance such as insulin, human and bovine growth hormone, protein antigens for formulation of vaccines, enzymes, among others, are obtained. However, keeping these cultivations within the desired conditions becomes a major challenge, since some variables such as dissolved oxygen concentration (DOC) and substrate concentration are difficult to control. Therefore, the development and implementation of an automatic monitoring and control tool are key requirements for the performance of high density cultivation. The present work has as main objectives to study feeding strategies for high cell density cultivation of recombinant Escherichia coli and develop a computational tool capable of ensuring the implementation of the chosen strategies, performing the monitoring, control and supervision of the cultivations. Fed batch cultivations were carried out under the supervision of the tool in a 5 L in-house bioreactor, equipped with sensors for temperature, dissolved oxygen, pH, pressure and biomass (sensor that measures the concentration of viable cells based on permittivity measurements), peristaltic pumps and connected to the gas analyzer. The tool was developed with LabView 8.0 and MatLab 6.5, being the acquisition and communication with the different bioreactor accessories via compact Field Point. Twenty two fed-batch cultivations with 5 different clones of E. coli, BL21(D3) expressing the enzyme penicillin G acylase (PGA) as well as antigenic proteins of S. pneumoniae (PspA3, PspA245 and PspA4Pro) and E. rhusiopathiae (SpaA) were performed during the development of the tool and the studies of feeding strategy. Both defined medium (HDF modified) as complex medium (ZYM-5052 modified), usually having glycerol as main carbon source and IPTG or lactose as inducers were used. In all cultivations, samples were collected to quantify the concentration of cells (dry weight method in filter of 0.22 m and optical density at 600 nm), organic acids, glucose, glycerol and lactose (HPLC) as well as protein expression (densitometry and NIPAB method for PGA) and plasmid stability (plating). The tool SUPERSYS_HCDCR (registered as a free software) developed, implemented and validated in the performed cultivations, carries out the basic functions of bioreactor supervision software, such as monitoring and data acquisition of pressure, temperature, pH, DOC, fraction of CO2 and O2 in the outlet gas as well as real-time estimate of the respiratory quotient, the rate of oxygen consumption and CO2 production. However, it also has the following special features, including: i) automatic control of air and oxygen flow according to cellular demand, ii) automatic activation of the feed pump at the end of the batch; iii) automatic control of feeding flow rate as function of the specific growth rate inferred in real time; iv) automatic control of feeding flow rate constrained by the concentration of dissolved oxygen, v) audible alarms indicating failures in the process; vi) failure messages sent via email; vii) automatic control of dissolved oxygen concentration; viii) control of the bioreactor pressure; and ix) control of bath temperature. Regarding the studies of feeding strategies aimed at biomass productivity increase in high cell density cultivations of recombinant E. coli, using the supervision tool developed together with changes in the composition of the synthetic culture medium available in the literature, a cellular concentrations greater than 150 g/L was achieved in less than 24 hours of cultivation, corresponding to a productivity of 9.2 g/Lh. This value, which is higher than the reported in the literature, was obtained without acetate accumulation and allowing high production of recombinant protein.