Traditional protein engineering methods, such as directed evolution, while effective, are often slow and labor-intensive. Advances in machine learning and automated biofoundry present new opportunities for optimizing these processes. This study devises a protein language model-enabled automatic evolution platform, a closed-loop system for automated protein engineering within the Design-Build-Test-Learn cycle. The protein language model ESM-2 makes zero-shot prediction of 96 variants to initiate the cycle. The biofoundry constructs and evaluates these variants, and feeds the results back to a multi-layer perceptron to train a fitness predictor, which then makes prediction of second round of 96 variants with improved fitness. With the tRNA synthetase as a model enzyme, four-rounds of evolution carried out within 10 days lead to mutants with enzyme activity improved by up to 2.4-fold. Our system significantly enhances the speed and accuracy of protein evolution, driving faster advancements in protein engineering for industrial applications.

The soil bacterium Pseudomonas putida KT2440 is a “generally recognized as safe”-certified strain with robust property and versatile metabolism. Thus, it is an ideal candidate for synthetic biology, biodegradation, and other biotechnology applications. The known genome editing approaches of Pseudomonas are suboptimal; thus, it is necessary to develop a high efficiency genome editing tool. In this study, we established a fast and convenient CRISPR–Cas9 method in P. putida KT2440. Gene deletion, gene insertion and gene replacement could be achieved within 5 days, and the mutation efficiency reached > 70%. Single nucleotide replacement could be realized, overcoming the limitations of protospacer adjacent motif sequences. We also applied nuclease-deficient Cas9 binding at three locations upstream of enhanced green fluorescent protein (eGFP) for transcriptional inhibition, and the expression intensity of eGFP reduced to 28.5, 29.4, and 72.1% of the control level, respectively. Furthermore, based on this CRISPR–Cas9 system, we also constructed a CRISPR–Cpf1 system, which we validated for genome editing in P. putida KT2440. In this research, we established CRISPR based genome editing and regulation control systems in P. putida KT2440. These fast and efficient approaches will greatly facilitate the application of P. putida KT2440.