Data-driven rational biosynthesis design: from molecules to cell factories

QM/MM Modeling Aided Enzyme Engineering in Natural Products Biosynthesis

Natural products and their derivatives are widely used across various industries, particularly pharmaceuticals. Modern engineered biosynthesis provides an alternative way of producing and meeting the growing need for diverse natural products. Natural enzymes, on the other hand, often exhibit unsatisfactory catalytic characteristics and necessitate further enzyme engineering modifications. QM/MM, as a powerful and extensively used computational tool in the field of enzyme catalysis, has been increasingly applied in rational enzyme engineering over the past decade. In this review, we summarize recent advances in QM/MM computational investigation on enzyme catalysis and enzyme engineering for natural product biosynthesis. The challenges and perspectives for future QM/MM applications aided enzyme engineering in natural product biosynthesis will also be discussed.

Deep learning-based kcat prediction enables improved enzyme-constrained model reconstruction

Enzyme turnover numbers (kcat) are key to understanding cellular metabolism, proteome allocation and physiological diversity, but experimentally measured kcat data are sparse and noisy. Here we provide a deep learning approach (DLKcat) for high-throughput kcat prediction for metabolic enzymes from any organism merely from substrate structures and protein sequences. DLKcat can capture kcat changes for mutated enzymes and identify amino acid residues with a strong impact on kcat values. We applied this approach to predict genome-scale kcat values for more than 300 yeast species. Additionally, we designed a Bayesian pipeline to parameterize enzyme-constrained genome-scale metabolic models from predicted kcat values. The resulting models outperformed the corresponding original enzyme-constrained genome-scale metabolic models from previous pipelines in predicting phenotypes and proteomes, and enabled us to explain phenotypic differences. DLKcat and the enzyme-constrained genome-scale metabolic model construction pipeline are valuable tools to uncover global trends of enzyme kinetics and physiological diversity, and to further elucidate cellular metabolism on a large scale.

Bottom-up synthetic biology approach for improving the efficiency of menaquinone-7 synthesis in Bacillus subtilis

Background

Menaquinone-7 (MK-7), which is associated with complex and tightly regulated pathways and redox imbalances, is produced at low titres in Bacillus subtilis. Synthetic biology provides a rational engineering principle for the transcriptional optimisation of key enzymes and the artificial creation of cofactor regeneration systems without regulatory interference. This holds great promise for alleviating pathway bottlenecks and improving the efficiency of carbon and energy utilisation.

Results

We used a bottom-up synthetic biology approach for the synthetic redesign of central carbon and to improve the adaptability between material and energy metabolism in MK-7 synthesis pathways. First, the rate-limiting enzymes, 1-deoxyxylulose-5-phosphate synthase (DXS), isopentenyl-diphosphate delta-isomerase (Fni), 1-deoxyxylulose-5-phosphate reductase (DXR), isochorismate synthase (MenF), and 3-deoxy-7-phosphoheptulonate synthase (AroA) in the MK-7 pathway were sequentially overexpressed. Promoter engineering and fusion tags were used to overexpress the key enzyme MenA, and the titre of MK-7 was 39.01 mg/L. Finally, after stoichiometric calculation and optimisation of the cofactor regeneration pathway, we constructed two NADPH regeneration systems, enhanced the endogenous cofactor regeneration pathway, and introduced a heterologous NADH kinase (Pos5P) to increase the availability of NADPH for MK-7 biosynthesis. The strain expressing pos5P was more efficient in converting NADH to NADPH and had excellent MK-7 synthesis ability. Following three Design-Build-Test-Learn cycles, the titre of MK-7 after flask fermentation reached 53.07 mg/L, which was 4.52 times that of B. subtilis 168. Additionally, the artificially constructed cofactor regeneration system reduced the amount of NADH-dependent by-product lactate in the fermentation broth by 9.15%. This resulted in decreased energy loss and improved carbon conversion.

Conclusions

In summary, a "high-efficiency, low-carbon, cofactor-recycling" MK-7 synthetic strain was constructed, and the strategy used in this study can be generally applied for constructing high-efficiency synthesis platforms for other terpenoids, laying the foundation for the large-scale production of high-value MK-7 as well as terpenoids.

Bio-inspired chemical space exploration of terpenoids

Many computational methods are devoted to rapidly generating pseudo-natural products to expand the open-ended border of chemical spaces for natural products. However, the accessibility and chemical interpretation were often ignored or underestimated in conventional library/fragment-based or rule-based strategies, thus hampering experimental synthesis. Herein, a bio-inspired strategy (named TeroGen) is developed to mimic the two key biosynthetic stages (cyclization and decoration) of terpenoid natural products, by utilizing physically based simulations and deep learning models, respectively. The precision and efficiency are validated for different categories of terpenoids, and in practice, more than 30 000 sesterterpenoids (10 times as many as the known sesterterpenoids) are predicted to be linked in a reaction network, and their synthetic accessibility and chemical interpretation are estimated by thermodynamics and kinetics. Since it could not only greatly expand the chemical space of terpenoids but also numerate plausible biosynthetic routes, TeroGen is promising for accelerating heterologous biosynthesis, bio-mimic and chemical synthesis of complicated terpenoids and derivatives.

Deep learning in retrosynthesis planning: datasets, models and tools

In recent years, synthesizing drugs powered by artificial intelligence has brought great convenience to society. Since retrosynthetic analysis occupies an essential position in synthetic chemistry, it has received broad attention from researchers. In this review, we comprehensively summarize the development process of retrosynthesis in the context of deep learning. This review covers all aspects of retrosynthesis, including datasets, models and tools. Specifically, we report representative models from academia, in addition to a detailed description of the available and stable platforms in the industry. We also discuss the disadvantages of the existing models and provide potential future trends, so that more abecedarians will quickly understand and participate in the family of retrosynthesis planning.