This review shows how structure-based modeling, data-driven learning, and rational engineering are overcoming long-standing barriers in reconstituting complex plant biosynthetic pathways in microbes.
Plant natural products underpin many modern medicines, flavors, pigments, and nutraceuticals, yet their supply still relies heavily on extraction from crops that grow slowly, yield inconsistently, and compete for land and resources. Chemical synthesis is often impractical because of structural complexity, while plant cell cultures remain difficult to scale. Synthetic biology offers a compelling alternative by reconstructing plant pathways in microorganisms such as yeast or bacteria. However, progress has been constrained by incomplete pathway knowledge, low activity or instability of plant enzymes outside their native cellular context, and metabolic imbalances in host cells. Over the past decade, rapid gains in genomics, structural biology, and computational modeling have created new opportunities to address these bottlenecks. This review situates protein design—spanning enzyme mining, mechanistic analysis, and optimization—as a central driver of the next phase of plant natural product biosynthesis.
A study (DOI: 10.1016/j.bidere.2025.100048) published in BioDesign Research on 18 September 2025 by Yongshuo Ma’s & Yi Shang’s team, Yunnan Normal University, provides an integrated toolkit to discover missing enzymes, decode catalytic mechanisms, and engineer robust, high-performance pathways for plant natural product production in microbial systems.
At the core of the review is a logical framework that follows the enzyme life cycle from discovery to deployment. First, protein design reshapes enzyme mining. Traditional sequence homology searches often fail to identify low-similarity or “orphan” enzymes in plants. Structure prediction, molecular docking, and AI-based clustering now allow researchers to search enzyme space using three-dimensional features and catalytic pockets rather than sequence alone, dramatically expanding the pool of functional candidates. Second, protein design enables mechanistic elucidation. By integrating structure prediction with molecular dynamics and quantum mechanics/molecular mechanics simulations, researchers can visualize how enzymes bind substrates, traverse transition states, and generate product diversity. These insights reveal key residues controlling specificity, regioselectivity, and efficiency—knowledge that was previously inaccessible for many membrane-bound or structurally elusive plant enzymes. Third, the review surveys optimization strategies. Directed evolution remains powerful but resource-intensive; rational and semi-rational design leverage structural and evolutionary information to target mutations more efficiently. Increasingly, AI-assisted approaches learn from multidimensional datasets to predict kinetic parameters, stability, solubility, and host compatibility, accelerating the design–build–test cycle. Finally, the review highlights chimeric protein design and subcellular engineering as ways to solve localization mismatches and folding problems, enabling multi-enzyme assemblies that channel intermediates and boost pathway flux. Beyond methodological advances, the review emphasizes applications and implications. Protein design–enabled synthetic biology has already delivered industrially relevant titers of high-value compounds, from terpenoids and alkaloids to glycosylated phenolics. As AI models become more plant-specific and datasets grow richer, enzyme engineering is shifting from single-protein tuning toward coordinated optimization of entire pathways and host metabolisms. This transition supports more predictable scale-up, reduced environmental footprints, and diversification of bio-based products.
In summary, this review portrays protein design as the engine driving a new era of plant natural product biosynthesis. By uniting AI, structural biology, and synthetic biology, researchers are moving from trial-and-error engineering toward intelligent programming of enzymes and pathways. The long-term vision is clear: microbial factories designed in silico and realized in vivo, capable of producing complex plant molecules sustainably and at scale—fulfilling the promise of “AI design, cellular synthesis” for the bioeconomy of the future.
###
References
DOI
10.1016/j.bidere.2025.100048
Original Source URL
https://doi.org/10.1016/j.bidere.2025.100048
Funding information
This research was supported by the National Natural Science Foundation of China (32488302), Yunnan Basic Research Special Project (202501BC070003).
About BioDesign Research
BioDesign Research is dedicated to information exchange in the interdisciplinary field of biosystems design. Its unique mission is to pave the way towards the predictable de novo design and assessment of engineered or reengineered living organisms using rational or automated methods to address global challenges in health, agriculture, and the environment.