Antônio Carlos Luperni Horta

Enhanced production of recombinant thermo-stable lipase in Escherichia coli at high induction temperature.

Thermostable microbial lipases are potential candidates for industrial applications such as specialty organic syntheses as well as hydrolysis of fats and oils. In this work, basic biochemical engineering tools were applied to enhance the production of BTL2 lipase cloned in Escherichia coli BL321 under control of the strong temperature-inducible λP(L) promoter. Initially, surface response analysis was used to assess the influence of growth and induction temperatures on enzyme production, in flask experiments. The results showed that temperatures of 30 and 45°C were the most suitable for growth and induction, respectively, and led to an enzyme specific activity of 706,000 U/gDCW. The most promising induction conditions previously identified were validated in fed-batch cultivation, carried out in a 2L bioreactor. Specific enzyme activity reached 770,000 U/gDCW, corresponding to 13,000 U/L of culture medium and a lipase protein concentration of 10.8 g/L. This superior performance on enzyme production was a consequence of the improved response of λP(L) promoter triggered by the high induction temperature applied (45°C). These results point out to the importance of taking into account protein structure and stability to adequately design the recombinant protein production strategy for thermally induced promoters.

Robust artificial intelligence tool for automatic start-up of the supplementary medium feeding in recombinant E. coli cultivations

One of the most important events in fed-batch fermentations is the definition of the moment to start the feeding. This paper presents a methodology for a rational selection of the architecture of an artificial intelligence (AI) system, based on a neural network committee (NNC), which identifies the end of the batch phase. The AI system was successfully used during high cell density cultivations of recombinant Escherichia coli. The AI algorithm was validated for different systems, expressing three antigens to be used in human and animal vaccines: fragments of surface proteins of Streptococcus pneumoniae (PspA), clades 1 and 3, and of Erysipelothrix rhusiopathiae (SpaA). Standard feed-forward neural networks (NNs), with a single hidden layer, were the basis for the NNC. The NN architecture with best performance had the following inputs: stirrer speed, inlet air, and oxygen flow rates, carbon dioxide evolution rate, and CO2 molar fraction in the exhaust gas.

A Heuristic Search for Optimal Parameter Values of Three Biokinetic Growth Models for Describing Batch Cultivations of Streptococcus Pneumoniae in Bioreactors

Simulated annealing (SA) is a stochastic search procedure which can lead to a reliable optimization method. This work describes a dynamic mathematical model for Streptococcus pneumoniaebatch cultivations containing 8 unknown parameters, which were calibrated by a SA algorithm through the minimization of an evaluation function based on the performance of the model on real experimental data. Three kinetic expressions, the Monod, Moser and Tessier equations, commonly employed to describe microbial growth were tested in the model simulations. SA convergence was achieved after 13810 interactions (about 10 minutes of computing time) and the Tessier equation was identified as the kinetic expression which provided the best fit to the cultivation dataset used for parameter estimation. The model comprising the Tessier equation, estimated parameter values supplied by SA and mass balance equations was further validated by comparing the simulated results to 3 experimental datasets from new cultivations carried out in similar conditions.

On-line prediction of the feeding phase in high-cell density cultivation of rE. coli using constructive neural networks

Streptococcus pneumoniae (pneumococcus) is a bacterium responsible for a wide spectrum of illnesses. The surface of the bacterium consists of three distinctive membranes: plasmatic, cellular and the polysaccharide (PS) capsule. PS capsules may mediate several biological processes, particularly invasive infections of human beings. Prevention against pneumococcal related illnesses can be provided by vaccines. There is a sound investment worldwide in the investigation of a proteic antigen as a possible alternative to pneumococcal vaccines based exclusively on PS. A few proteins which are part of the membrane of the pneumococcus seem to have antigen potential to be part of a vaccine, particularly the PspA. A vital aspect in the production of the intended conjugate pneumococcal vaccine is the efficient production (in industrial scale) of both, the chosen PS serotypes as well as the PspA protein. Growing recombinant Escherichia coli (rE. coli) in high-cell density cultures (HCDC) under a fed-batch regime requires a refined continuous control over various process variables where the on-line prediction of the feeding phase is of particular relevance and one of the focuses of this paper. The viability of an on-line monitoring software system, based on constructive neural networks (CoNN), for automatically detecting the time to start the fed-phase of a HCDC of rE. coli that contains a plasmid used for PspA expression is investigated. The paper describes the data and methodology used for training five different types of CoNNs, four of them suitable for classification tasks and one suitable for regression tasks, aiming at comparatively investigate both approaches. Results of software simulations implementing five CoNN algorithms as well as conventional neural networks (FFNN), decision trees (DT) and support vector machines (SVM) are also presented and discussed. A modified CasCor algorithm, implementing a data softening process, has shown to be an efficient candidate to be part of an on-line HCDC monitoring system for detecting the feeding phase of the HCDC process.

Development of a fuzzy system for dissolved oxygen control in a recombinant Escherichia coli cultivation for heterologous protein expression

Abstract One very important bioprocess is the cultivation of recombinant E. coli for expression of heterologous protein. For this, High Cell Density Cultures is one of the most widely used technique. Therefore, researchers from the Chemical Engineering Department of Federal University of Sao Carlos (UFSCar) developed a very useful computer program (SUPERSYS_HCDC) that, among other functions, presents a hybrid system with a PID for agitation and a decision tree for air and oxygen flow rates that controls the percentage of dissolved oxygen in the cultivation (nowadays some commercial controllers also offers this cascade control). However, in particular, delays may occur in the device responsible for air and oxygen injection into the bioreactor, since the decision tree provides no smooth responses. The original system presented operates by introducing steps in the air and oxygen flow rates. Under the light of the above-mentioned facts, fuzzy reasoning was used to develop a fuzzy controller, aiming to improve dissolved oxygen control in recombinant E. coli cultivation for heterologous protein production. At first, fuzzy logic toolbox was used to generate a control algorithm implemented in a MATLAB code. Secondly, the membership function parameters were optimized using ANFIS tool. Finally, in order to perform tests using the fuzzy controller, it was coupled to a neural network model of the process. This was created using artificial neural network toolbox and E. coli cultivation data. Results for oxygen and air flow rates indicated that the trends of aeration required by E. coli cultivation were fulfilled. Using the fuzzy controller, it was possible to maintain the percentage of dissolved oxygen around the set point value of 30%. In general, the fuzzy controller responses were smoother than those provided by the decision tree, in a way that the dissolved oxygen peaks were softened.

An Empirical Investigation of the Use of a Neural Network Committee for Identifying the Streptococcus Pneumoniae Growth Phases in Batch Cultivations

Streptococcus pneumoniaeis a bacterial pathogen that causes many life-threatening diseases and an effective vaccine against this pa-thogen is still subject of research. These bacteria grow with low carbon dioxide production, which hinders the application of exhaust gas composition for on-line process monitoring. This work investigates the proposal of a committee of neural networks for identifying Streptococcus pneumoniaegrowth phases, to be used for on-line state inference. The committee results as well as the accuracy for predicting the culture phases are compared to the results of a unique neural network, for different input variables. The best configuration for the software was: a committee of three NN trained with two input attributes (optical density and mass of alkali solution), 200 epochs of training and log sigmoid as the activation function in the hidden layer as well as in the output layer.