Biosensor and machine learning-aided engineering of an amaryllidaceae enzyme

Abstract

A major challenge to achieving industry-scale biomanufacturing of therapeutic alkaloids is the slow process of biocatalyst engineering. Amaryllidaceae alkaloids, such as the Alzheimer’s medication galantamine, are complex plant secondary metabolites with recognized therapeutic value. Due to their difficult synthesis they are regularly sourced by extraction and purification from the low-yielding daffodil Narcissus pseudonarcissus. Here, we propose an efficient biosensor-machine learning technology stack for biocatalyst development, which we apply to engineer an Amaryllidaceae enzyme in Escherichia coli. Directed evolution is used to develop a highly sensitive (EC50 = 20 μM) and specific biosensor for the key Amaryllidaceae alkaloid branchpoint 4’-O-methylnorbelladine. 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. 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 solved crystal structure elucidates the mechanism behind key beneficial mutations.

Recent research progress of selenotungstate-based biomolecular sensing materials

“…This phenomenon increases the specific recognition signal, thereby improving the sensitivity of aptamer-based EABs markedly. 13 In order to pursue higher sensitivity and activity of aptamer-based EABs, a conductive medium is required between the aptamer and electrode to facilitate electron transfer of aptamer-based EABs.…”

“…1). 10–14 The precise binding of biorecognition elements leads to interaction fluctuations, which can be subsequently captured by signal transducers and converted into discernible electrical impulses, fluorescence signals, thermal or acoustic signals. 15–17 In order to facilitate the conversion of interaction fluctuations into detectable signals, a variety of nano-functional materials, such as carbon-based materials, 18 metal nanoparticles, 19 MXenes, 20 metal–organic frameworks, 12 covalent organic frameworks, 21 and organic polymers, 22 have been used to manufacture electrochemical and fluorescent biosensors.…”

Engineering a Photoenzyme to Use Red Light

“…To improve upon this starting point, we mutations predicted by MūtCompute, a three-dimensional self-supervised, convolutional neural network, which were previously identied for an enzyme engineered to catalyze an asymmetric alkene hydroalkylation. [27][28][29] We found that mutating threonine 36 to aspartic acid (T36E), increased the yield to 78% with no change in the enantioselectivity of the transformation. Mutation of serine at position 118 to cystine (S118C) led to product formation in 86% yield.…”

Machine learning guided rational design of a non-heme iron-based lysine dioxygenase improves its total turnover number

“…While ML tools can be integrated into engineering workflows, they often require extensive training datasets either from enzyme-specific reaction assays or non-structural genetic data. [ 27 – 32 ] Tailoring of extensive datasets for each specific enzyme still requires burdensome front-work from researchers. Thus, synergizing ML technologies with existing rational design strategies offers a low-barrier solution for identifying potentially beneficial mutations in an engineering campaign.…”