We then develop MutComputeX: a structure-based self-supervised residual neural network (3DResNet) trained to generalize at protein:non-protein interfaces, which is used to generate activity-enriched Nb4OMT designs from an ML-generated protein-cofactor-substrate structure.
A generalist transcription factor, RamR, is evolved into a highly sensitive biosensor for 4NB that precisely discriminates against the non-methylated precursor norbelladine, and the biosensor is then used to monitor the activity of norbelladine 4’-O-methyltransferase (Nb4OMT) from the daffodil Narcissus pseudonarcissus in Escherichia coli. Here, we synergize the development of custom biosensors with machine learning(ML)-guided protein design to improve microbial fermentation of the branchpoint AA 4′-O-methylnorbelladine (4NB). The industrial application of such pathways could be greatly accelerated by augmenting high-throughput screens with genetic biosensors 17, 18, 19, 20, and using machine learning to guide protein design 21, 22, 23, 24, yielding enzymes and pathways with improved stability and activity.
Furthermore, semi-synthetic methods have been proposed using characterized enzymes to generate advanced intermediates 16. While the complete biosynthetic pathway for any AA with therapeutic value has not yet been elucidated, recent studies have characterized early pathway enzymes responsible for the biosynthesis of 4’-O-Methylnorbelladine, the last common intermediate before AA pathway branches diverge 15. Recently, long plant pathways have been reconstituted into microbial hosts for the production of therapeutic benzylisoquinoline alkaloids 11, 12, tropane alkaloids 13, and monoterpene indole alkaloids 14. In an effort to improve galantamine production, agricultural techniques are currently being tested to boost daffodil-sourced yields 9, 10.Ī promising alternative to amaryllidaceae alkaloid extraction from plants is microbial fermentation. Due to galantamine’s challenging synthesis, global supplies largely rely on isolating the low quantities (0.3% dry weight) that accumulate in harvested daffodils, ultimately resulting in an expensive ($50,000/kg) and environmentally-dependent supply chain 7, 8. One of the most notable AAs is galantamine, a selective and reversible acetylcholinesterase inhibitor that is a licensed treatment for mild to moderate symptoms of Alzheimer’s disease and a promising scaffold for drug design 5, 6.
Among the approximate ~600 reported Amaryllidoideae alkaloids (AAs), those derived from the lycorine, haemanthamine, and narciclasine scaffolds have been used as lead molecules in anticancer research 1, 2, 3, 4. Similar content being viewed by othersĪlkaloids produced by the Amaryllidoideae subfamily of flowering plants have great therapeutic promise, including anticancer, fungicidal, antiviral, and acetylcholinesterase inhibition properties. A solved crystal structure elucidates the mechanism behind key beneficial mutations. Functional enzyme variants are identified that yield a 60% improvement in product titer, 2-fold higher catalytic activity, and 3-fold lower off-product regioisomer formation. A structure-based residual neural network (MutComputeX) is subsequently developed and used to generate activity-enriched variants of a plant methyltransferase, which are rapidly screened with the biosensor. Directed evolution is used to develop a highly sensitive (EC 50 = 20 μM) and specific biosensor for the key Amaryllidaceae alkaloid branchpoint 4’-O-methylnorbelladine. Here, we propose an efficient biosensor-machine learning technology stack for biocatalyst development, which we apply to engineer an Amaryllidaceae enzyme in Escherichia coli. Due to their difficult synthesis they are regularly sourced by extraction and purification from the low-yielding daffodil Narcissus pseudonarcissus. Amaryllidaceae alkaloids, such as the Alzheimer’s medication galantamine, are complex plant secondary metabolites with recognized therapeutic value. A major challenge to achieving industry-scale biomanufacturing of therapeutic alkaloids is the slow process of biocatalyst engineering.