Explore or exploit? A model-based screening strategy for PETase secretion by Corynebacterium glutamicum.

Area of Science:

• Biotechnology
• Synthetic Biology
• Protein Engineering

Background:

• Extracellular protein production simplifies downstream processing but requires optimizing protein, signal peptide, and host combinations.
• Predicting optimal secretion strategies is difficult, necessitating extensive screening of microbial strain libraries.

Purpose of the Study:

• To develop and apply an automated screening and Bayesian modeling framework for identifying optimal signal peptides for extracellular secretion of polyethylene terephthalate degrading enzymes (PETases) in Corynebacterium glutamicum.
• To accelerate the early-stage screening process for bioprocess development by focusing on the most promising strain candidates.

Main Methods:

• Screening of Bacillus subtilis signal peptides for Sec secretion of leaf-branch compost cutinase (LCC) and polyester hydrolase mutants in Corynebacterium glutamicum.
• Development of a fully automated high-throughput screening process coupled with a Bayesian statistical modeling framework to quantify uncertainty and rank candidates.
• Assessing enzyme activity via 4-nitrophenyl palmitate degradation assays.

Main Results:

• Identification of the most suitable signal peptide for each PETase within two rounds of screening.
• Demonstration of scalability for LCC secretion from microliter-scale cultivation to laboratory-scale bioreactors.
• Quantification of batch effects and biological errors, improving decision-making in strain selection.

Conclusions:

• An integrated experiment-modeling loop significantly accelerates early-stage screening for extracellular protein production.
• The developed approach enables efficient use of large strain libraries within a Design-Build-Test-Learn framework.
• This methodology streamlines bioprocess development by rapidly identifying high-performing microbial strains for target protein secretion.