High-level phenol bioproduction by engineered Pichia pastoris in glycerol fed-batch fermentation using an efficient pertraction system

This study aimed to establish a high-level phenol bioproduction system from glycerol through metabolic engineering of the yeast Pichia pastoris (Komagataella phaffii). Introducing tyrosine phenol-lyase to P. pastoris led to a production of 59 mg/L of phenol in flask culture. By employing a strain of P. pastoris that overproduces tyrosine-a precursor to phenol-we achieved a phenol production of 1052 mg/L in glycerol fed-batch fermentation. However, phenol concentrations exceeding 1000 mg/L inhibited P. pastoris growth. A phenol pertraction system utilizing a hollow fiber membrane contactor and tributyrin as the organic solvent was developed to reduce phenol concentration in the culture medium. Integrating this system with glycerol fed-batch fermentation resulted in a 214 % increase in phenol titer (3304 mg/L) compared to glycerol fed-batch fermentation alone. These approaches offer a significant framework for the microbial production of chemicals and materials that are highly toxic to microorganisms.

Neuroevolutive Algorithms Applied for Modeling Some Biochemical Separation Processes

Combining artificial neural networks with evolutive/bioinspired approaches is a technique that can solve a variety of issues regarding the topology determination and training for neural networks or for process optimization. In this chapter, the main mechanisms used in neuroevolution are discussed and some biochemical separation examples are given to underline the efficiency of these tools. For the current case studies (reactive extraction of folic acid and pertraction of vitamin C), the bioinspired metaheuristic included in the neuroevolutive procedures is represented by Differential Evolution, an algorithm that has shown a great potential to solve a variety of problems from multiple domains.