SEARCH · bioRxiv Science

Results for “Synthetic Biology”

Search indexed bioRxiv preprints in genomics, neuroscience, cell biology and bioinformatics. Read source abstracts and check manuscript versions; preprints are not peer reviewed.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,207 records · Page 67Linked to original sources

bioRxiv2023

Nonribosomal peptide synthetases require dynamic interaction between modular domains

Nonribosomal peptide synthetases (NRPSs) are large multidomain enzymes for the synthesis of a variety of bioactive peptides in a modular and pipelined fashion. Here, we investigated how the condensation (C) domain and the adenylation (A) domain cooperate with each other for the efficient catalytic activity in microcystin NRPS modules. We solved two crystal structures of the microcystin NRPS modules, representing two newly identified conformations in the NRPS catalytic cycle. Our data reveals that the dynamic interaction between the C and the A domains in these modules are mediated by the conserved "RXGR" motif, and this interaction is important for the adenylation activity. Furthermore, the "RXGR" motif-mediated dynamic interaction and its functional regulation is prevalent in different NRPSs modules possessing both the A and the C domains. This study provides new insight into the catalytic mechanism of NRPSs and should inspire novel ideas in NRPS enzyme engineering in synthetic biology.

biochemistry↗

bioRxiv2023

An NMR-guided approach for identifying co-factors in boosting the Pfu DNA polymerase

With rapid developments of emerging technologies like synthetic biology, the demand for DNA polymerases with superior activities including higher thermostability and processivity has increased significantly. Thus, rational optimization of the performance of DNA polymerase is of great interest. Nuclear magnetic resonance spectroscopy (NMR) is a powerful technique used for studying protein structure and dynamics. It provides the atomic resolution information of enzymes under its functional solution environment to reveal the active sites (hot spots) of the enzyme, which could be further used for optimizing the performance of enzymes. In our previous work, we identified hot spot residues of Pfu DNA polymerase (Pfu pol). We aim to employ these binding hot spots to screen for co-factors of Pfu pol, particularly targeting those molecules exhibiting weak intermolecular interactions. To validate this concept, we first demonstrated the feasibility of utilizing hot spot residues as screening probes for auxiliary factors by employing the well-characterized Tween-20 as a model system. Employing these hot spots as probes, two new co-factors, the heat shock protein TkHSP20 from Thermococcus Kodakaraensis and the chemical chaperone L-arginine, are identified to interact with Pfu pol to boost its performance in amplifying long DNA fragments by enhancing the thermal stability and the processivity of the Pfu pol. This NMR-based approach requires no prior assignment information of target enzymes, guiding the rational exploration of novel co-factors for Pfu pol. Moreover, our approach is not dependent on structural data or bioinformatics. Therefore, it has significant potential for application in various enzymes to expedite the progress in enzyme engineering.

molecular biology↗

bioRxiv2023

Guanylate cyclase activity of TIR1/AFB auxin receptorsin rapid auxin responses

The plant hormone auxin is perceived by TIR1/AFB receptors, which mediate transcriptional reprogramming via a ubiquitination- and cAMP-dependent mechanism. Same receptors mediate also rapid, non-transcriptional growth responses that are mechanistically less well understood. Here, we demonstrate that TIR1/AFB receptors possess not only adenylate cyclase but also guanylate cyclase activity. Auxin perception independently activates cAMP and cGMP production by TIR1/AFBs in vitro and elevates both cAMP and cGMP levels in planta. Genetic and synthetic biology approaches revealed that AFB1 guanylate cyclase activity is required for the rapid cytosolic Ca2+ transients and root growth inhibition. This unprecedented combination of adenylate and guanylate cyclase activities within a single hormone receptor establishes a new paradigm for how one perception mechanism can mediate distinct downstream responses. One-Sentence SummaryTIR1/AFB auxin receptors possess both, adenylate and guanylate cyclase activities, mediating transcriptional reprogramming and rapid Ca2+-dependent growth responses, respectively.

plant biology↗

bioRxiv2023

Proliferation history and transcription factor levels drive direct conversion

The sparse and stochastic nature of reprogramming has obscured our understanding of how transcription factors drive cells to new identities. To overcome this limit, we developed a compact, portable reprogramming system that increases direct conversion of fibroblasts to motor neurons by two orders of magnitude. We show that subpopulations with different reprogramming potentials are distinguishable by proliferation history. By controlling for proliferation history and titrating each transcription factor, we find that conversion correlates with levels of the pioneer transcription factor Ngn2, whereas conversion shows a biphasic response to Lhx3. Increasing the proliferation rate of adult human fibroblasts generates morphologically mature, induced motor neurons at high rates. Using compact, optimized, polycistronic cassettes, we generate motor neurons that graft with the murine central nervous system, demonstrating the potential for in vivo therapies. One Sentence SummaryUsing a systems and synthetic biology approach to study the molecular determinants of reprogramming, we find that proliferation history and transcription factor levels drive cell fate in direct conversion to motor neurons. HighlightsO_LIminimal high-efficiency cocktail allows systematic interrogation of the reprogramming process C_LIO_LIhistory provides a principal axis to distinguish transcription factors influence C_LIO_LIof the transcription factor cocktail impacts reprogramming efficiency and dynamics C_LIO_LIof individual transcription factors differentially influence the rate of reprogramming C_LIO_LIearly hyperproliferation increases direct conversion of adult human fibroblasts C_LIO_LIptimal cocktail allows neurotrophic factor-free reprogramming and in vivo grafting C_LI

cell biology↗

bioRxiv2023

Optimizing ex-vitro one-step RUBY-equipped hairy root transformation in drug- and hemp-type Cannabis

Using synthetic biology techniques to engineer secondary metabolic pathways through hairy root transformation is one of the most advanced approaches used in research. In this study, we optimized an ex-vitro one-step hairy root transformation of the RUBY system in both drug- and hemp-type cannabis, shedding light on its potential applications in secondary metabolite production. Three different strains of A. rhizogenes including (A4, ARqual, and K599) were used. Significant variation in HR induction and transformation efficiency (TE) was observed based on A. rhizogenes strains and seed types. Drug-type seedlings exhibited the highest hairy root induction, increasing by 58.8% compared to hemp-type seedlings. Also, the A4 strain consistently demonstrated the highest transformation efficiency (75%) irrespective of genotype, while the ARqual strain yielded the lowest one (8.33%). In conclusion, our study is the first to present an ex-vitro one-step transformation of both hemp- and drug-type cannabis. In comparison to the in vitro method, our ex-vitro method is simpler, faster, and has a lower risk of contamination, making it an excellent choice for the efficient production of secondary metabolites in cannabis using the CRISPR/Cas system.

plant biology↗

bioRxiv2023

Deciphering the Biosynthetic Potential of Microbial Genomes Using a BGC Language Processing Neural Network Model

Microbial secondary metabolites are usually synthesized by colocalized genes termed biosynthetic gene clusters (BGCs). A large portion of BGCs remain undiscovered in microbial genomes and metagenomes, representing a pressing challenge in unlocking the full potential of natural product diversity. In this work, we propose BGC-Prophet, a language model based on the transformer encoder that captures the distant location-dependent relationships among biosynthetic genes, allows accurately and efficiently identifies known BGCs and extrapolates novel BGCs among the microbial universe. BGC-Prophet is the first ultrahigh-throughput (UHT) method that is several orders of magnitude faster than existing tools such as DeepBGC, enabling pan-phylogenetic screening and whole-metagenome screening of BGCs. By analyzing 85,203 genomes and 9,428 metagenomes, new insights have been obtained about the diversity of BGCs on genomes from the majority of bacterial and archaeal lineages. The profound enrichment of BGCs in microbes after important geological events have been revealed: Both the Great Oxidation and Cambrian Explosion events led to a surge in BGC diversity and abundance, particularly in polyketides. These findings suggest that it is a general but constantly evolving approach for microbes to produce secondary metabolites for their adaptation in the changing environment. Taken together, BGC-Prophet enables accurate and fast detection of BGCs on a large scale, holds great promise for expanding BGC knowledge, and sheds light on the evolutionary patterns of BGCs for possible applications in synthetic biology. HighlightsO_LIBGC-Prophet shows superior performance to existing tools in terms of accuracy and speed. C_LIO_LIBGC-Prophet is the first ultrahigh-throughput (UHT) method that enables pan-phylogenetic screening and whole-metagenome screening of BGCs. C_LIO_LIBGC-Prophet builds the comprehensive profile of BGCs on 85,203 genomes and 9,428 metagenomes from the majority of bacterial and archaeal lineages. C_LIO_LIBGC-Prophet reveals the profound enrichment pattern of BGCs after important geological events. C_LI

bioinformatics↗

bioRxiv2023

Semantical and Topological Protein Encoding Toward Enhanced Bioactivity and Thermostability

Protein engineering is a pivotal aspect of synthetic biology, involving the modification of amino acids within existing protein sequences to achieve novel or enhanced functionalities and physical properties. Accurate prediction of protein variant effects requires a thorough understanding of protein sequence, structure, and function. Deep learning methods have demonstrated remarkable performance in guiding protein modification for improved functionality. However, existing approaches predominantly rely on protein sequences, which face challenges in efficiently encoding the geometric aspects of amino acids local environment and often fall short in capturing crucial details related to protein folding stability, internal molecular interactions, and bio-functions. Furthermore, there lacks a fundamental evaluation for developed methods in predicting protein thermostability, although it is a key physical property that is frequently investigated in practice. To address these challenges, this paper introduces a novel pre-training framework that integrates sequential and geometric encoders for protein primary and tertiary structures. This framework guides mutation directions toward desired traits by simulating natural selection on wild-type proteins and evaluates variant effects based on their fitness to perform specific functions. We assess the proposed approach using three benchmarks comprising over 300 deep mutational scanning assays. The prediction results showcase exceptional performance across extensive experiments when compared to other zero-shot learning methods, all while maintaining a minimal cost in terms of trainable parameters. This study not only proposes an effective framework for more accurate and comprehensive predictions to facilitate efficient protein engineering, but also enhances the in silico assessment system for future deep learning models to better align with empirical requirements. The PyTorch implementation are available at https://github.com/tyang816/ProtSSN.

bioengineering↗

bioRxiv2023

DeepEnzyme: a robust deep learning model for improved enzyme turnover number prediction by utilizing features of protein 3D Structures

Turnover numbers (kcat), which indicate an enzymes catalytic efficiency, have a wide range of applications in fields including protein engineering and synthetic biology. Experimentally measuring the enzymes kcat is always time-consuming. Recently, the prediction of kcat using deep learning models has mitigated this problem. However, the accuracy and robustness in kcat prediction still needs to be improved significantly, particularly when dealing with enzymes with low sequence similarity compared to those within the training dataset. Herein, we present DeepEnzyme, a cutting-edge deep learning model that combines the most recent Transformer and Graph Convolutional Network (GCN) architectures. To improve the prediction accuracy, DeepEnzyme was trained by leveraging the integrated features from both sequences and 3D structures. Consequently, our model exhibits remarkable robustness when processing enzymes with low sequence similarity compared to those in the training dataset by utilizing additional features from high-quality protein 3D structures. DeepEnzyme also makes it possible to evaluate how point mutations affect the catalytic activity of the enzyme, which helps identify residue sites that are crucial for the catalytic function. In summary, DeepEnzyme represents a pioneering effort in predicting enzymes kcat values with superior accuracy and robustness compared to previous algorithms. This advancement will significantly contribute to our comprehension of enzyme function and its evolutionary patterns across species.

bioinformatics↗

bioRxiv2023

Direct Pacbio sequencing methods and applications for different types of DNA sequences

The development of Sanger sequencing and next-generation sequencing methods within the past few years have assisted investigators profile the diversity and relative abundances of heterogenous species in vector preparations. Especially Recombinant adeno-associated viruses (rAAVs), genome editing, and mRNA related research are currently the most prominently investigated platform in different area and essentially use for synthetic biology, gene and cell therapy, food industrial and medicinal pharmer etc. area. However, these types of research related constructs always contain high GC sequences, poly structure, long-length DNA sequences and ITR repeats sequences. Unfortunately, Sanger sequencing and NGS platforms may be inaccessible to investigators with limited resources, require large amounts of input material, or may require long wait times for sequencing and analyses. Recent advances with PacBio sequencing have helped to bridge the gap for quick and relatively inexpensive long-read sequencing needs. Specifically, long-read sequencing methods, like single molecule real-time (SMRT) sequencing, have been used to uncover truncations, chimeric genomes, and inverted terminal repeat (ITR) mutations in vectors. Recombinant adeno-associated virus (raav) is the most prominent platform in the field of current research, and its sequence is characterized by high GC, multi-structure, long sequence, genome, and repeat sequence. Sanger sequencing has certain defects in the detection of recombinant adeno-associated viruses. Meanwhile, Sanger needs to design sequencing primers based on known sequences to determine whether the sequences are correct. When sequence information is incomplete, it can only randomly design primers, obtain a sequence by luck, and then conduct the next round of sequencing. However, PacBios limitations and sample biases are not well-defined for sequencing. And sometimes the accuracy for base calling was low, resulting in a high degree of miscalled bases and false indels. These false indels led to read-length compression; thus, assessing heterogeneity based on read length is not advisable with current PacBio technologies. In this study, we explored the capacity for PacBio sequencing to directly interrogate content to obtain full-length resolution of encapsulated genomes. We found that the PacBio platform can cover the entirety of different type sequences like poly structure, long-length DNA fragment, high GC sequences and repeat sequences, especially the rAAV sequences from ITR to ITR without the need for pre-fragmentation. At the same time, the sequencing process was optimized to complete the sequencing of long difficult plasmids with the fewest plasmids and the fastest time. In summary, the optimization PacBio sequencing and novel bioinformation (BI) analysis method are able to correctly identify truncation hotspots in single-strand and self-complementary vectors using by SMRT sequencing and can serve as a rapid and low-cost alternative for proofing different type of sequences.

molecular biology↗

bioRxiv2023

Optimization of chondroitin production in E. coli using genome scale models

Chondroitin is a natural occurring glycosaminoglycan with applications as a nutraceutical and pharmaceutical ingredient and can be extracted from animal tissues. Microbial chondroitin-like polysaccharides emerged as a safer and more sustainable alternative source. However, chondroitin titers using either natural or recombinant microorganisms are still far from meeting the increasing demand. The use of genome-scale models and computational predictions can assist the design of microbial cell factories with possible improved titers of these value-added compounds. Genome-scale models have been used to predict genetic modifications in Escherichia coli engineered strains that would potentially lead to improved chondroitin production. Additionally, using synthetic biology approaches, a pathway for producing chondroitin has been designed and engineered in E. coli. Afterwards, the most promising mutants identified based on bioinformatics predictions were constructed and evaluated for chondroitin production in flask fermentation. This resulted in the production of 118 mg/L of extracellular chondroitin by overexpressing both superoxide dismutase (sodA) and a lytic murein transglycosylase (mltB). Then, batch and fed-batch fermentations at bioreactor scale were also evaluated, in which the mutant overexpressing mltB led to an extracellular chondroitin production of 427 mg/L and 535 mg/L, respectively. The computational approach herein described identified several potential novel targets for improved chondroitin biosynthesis, which may ultimately lead to a more efficient production of this glycosaminoglycan.

bioengineering↗

bioRxiv2023

Unexplored regions of the protein sequence-structure map revealed at scale by a library of foldtuned language models

Amino-acid sequence space is combinatorially vast, with well-folded proteins distributed sparsely and connected by vanishingly few permissible mutational paths. Novel-in-sequence versions of structures observed in nature promise to sample features such as new binding motifs and active site geometries but are rendered inaccessible to evolution or direct search by the extent of sequence perturbations required. Here we introduce a novel algorithm - termed "foldtuning" - that leverages principles of adversarial learning to drive protein language models (PLMs) to erase detectable homology to natural sequences while preserving a target structure, systematically traversing protein-space without being limited by evolutionary barriers. We build foldtuned PLMs for >700 targets including membrane-bound receptors, redox enzymes, and signaling domains. Foldtuned proteins are diverse and far-from-natural in sequence, filling out structurally-equivalent families defined by fundamental biophysical constraints invisible to traditional sequence-based bioinformatics methods. Experimental characterization demonstrates that foldtuned proteins express stably in vitro and function in vivo. By revealing sequence-structure information at scale beyond evolution, foldtuning promises to accelerate the reconstitution and realization of novel-to-nature systems for synthetic biology problems from therapeutics to catalysis.

bioengineering↗

bioRxiv2023

The mutational landscape of Bacillus subtilis conditional hypermutators suggests how proofreading inherently skews polymerase error rates

Polymerase errors during DNA replication are a major source of point mutations in genomes. The resulting rate of spontaneous mutation also depends on the counteracting activity of DNA repair mechanisms, with mutator phenotypes appearing constantly and allowing for periods of rapid evolution in nature and in the laboratory. Here, we use the Gram-positive model bacterium Bacillus subtilis to disentangle the contributions of DNA polymerase initial nucleotide selectivity, DNA polymerase proofreading, and mismatch repair (MMR) to the mutation rate. To achieve this, we constructed several conditional hypermutators with a proofreading-deficient allele of polC and/or a deficient allele of mutL and performed mutation accumulation experiments. With their wide range of mutation rates and contrasting mutation profiles, these conditional hypermutators enrich the B. subtilis synthetic biology toolbox for directed evolution. Using mathematical models, we investigated how to interpret the apparent probabilities with which errors escape MMR and proofreading, highlighting the difficulties of working with counts that aggregate potentially heterogeneous mutations and with unknowns about the pathways leading to mutations in the wild-type. Aware of these difficulties, the analysis shows that proofreading prevents partial saturation of the MMR in B. subtilis and that an inherent drawback of proofreading is to skew the net polymerase error rates by amplifying intrinsic biases in nucleotide selectivity.

genomics↗

bioRxiv2023

Partial RNA Design

RNA design is a key technique to achieve new functionality in fields like synthetic biology or biotechnology. Computational tools could help to find such RNA sequences but they are often limited in their formulation of the search space. In this work, we propose partial RNA design, a novel RNA design paradigm that addresses the limitations of current RNA design formulations. Partial RNA design describes the problem of designing RNAs from arbitrary RNA sequences and structure motifs with multiple design goals. By separating the design space from the objectives, our formulation enables the design of RNAs with variable lengths and desired properties, while still allowing precise control over sequence and structure constraints at individual positions. Based on this formulation, we introduce a new algorithm, libLEARNA, capable of efficiently solving different constraint RNA design tasks. A comprehensive analysis of various problems, including a realistic riboswitch design task, reveals the outstanding performance of libLEARNA and its robustness.

bioinformatics↗

bioRxiv2024

Insights into Yeast Response to Chemotherapeutic Agent through Time Series Genome-Scale Metabolic Models

BackgroundOrganism-specific genome-scale metabolic models (GSMMs) can unveil molecular mechanisms within cells and are commonly used in diverse applications, from synthetic biology, biotechnology, and systems biology to metabolic engineering. There are limited studies incorporating time-series transcriptomics in GSMM simulations. Yeast is an easy-to-manipulate model organism for tumor research; MethodsHere, a novel approach (TS-GSMM) was proposed to integrate time-series transcriptomics with GSMMs to narrow down the feasible solution space of all possible flux distributions and attain time-series flux samples. The flux samples were clustered using machine learning techniques, and the clusters functional analysis was performed using reaction set enrichment analysis; ResultsA time series transcriptomics response of Yeast cells to a chemotherapeutic reagent - doxorubicin - was mapped onto a Yeast GSMM. Eleven flux clusters were obtained with our approach, and pathway dynamics were displayed. Induction of fluxes related to bicarbonate formation and transport, ergosterol and spermidine transport, and ATP production were captured; ConclusionsIntegrating time-series transcriptomics data with GSMMs is a promising approach to reveal pathway dynamics without any kinetic modeling and detects pathways that cannot be identified through transcriptomics-only analysis. The codes are available at https://github.com/karabekmez/TS-GSMM.

bioengineering↗

bioRxiv2024

Posttranscriptional tuning of gene expression over a large dynamic range in synthetic tobacco chloroplast operons

Achieving balanced gene expression within synthetic operons requires a spectrum of expression levels. Here we investigate the expression of gfp reporter gene in tobacco chloroplasts, guided by variants of the plastid atpH 5 UTR, which harbors a binding site for PPR10, a protein that activates atpH at the post-transcriptional level. Our findings reveal that endogenous tobacco PPR10 confers distinct levels of reporter activation when coupled with the tobacco and maize atpH 5 UTRs in different design contexts. Notably, high GFP expression was not coupled to stabilization of monocistronic gfp transcripts in dicistronic reporter lines, adding to the evidence that PPR10 activates translation via a mechanism that is independent of its stabilization of monocistronic transcripts. Furthermore, the incorporation of a tRNA upstream of the UTR nearly abolishes gfp mRNA (and GFP protein), resulting in a substantial reduction in GFP accumulation. When combined with a mutant atpH 5 UTR, the tRNA leads to an exceptionally low level of transgene expression. Collectively, this approach allows for tuning reporter gene expression across a wide range, spanning from 0.02% to 25% of the total soluble cellular protein (TSP). These findings highlight the toolbox available for plastid synthetic biology applications requiring multigene expression at varying levels.

bioengineering↗

bioRxiv2024

Divisome minimization shows that FtsZ and SepF can form an active Z-ring, and reveals BraB as a new cell division influencing protein in Bacillus subtilis

One of the aims of synthetic biology is the construction of a minimal cell. Since bacteria are the simplest life forms, they are the preferred blueprint for such a cell. In this study, we tried to find the minimal set of proteins required for division of a cell wall containing cell, using the bacterial model system Bacillus subtilis. Bacterial cytokinesis begins with polymerization of the tubulin homologue FtsZ at midcell into the Z-ring, which recruits the late cell division protein that synthesize the division septum. Assembly of FtsZ is carefully regulated, involving a dozen conserved cell division proteins. These proteins are not essential, but removing more than one is in many cases lethal. We made use of known suppressor mutations to find a gene deletion route that eventually enabled us to remove eight conserved cell division proteins: ZapA, MinC, MinJ, UgtP, ClpX, Noc, EzrA and FtsA. Only FtsZ and its membrane anchor SepF appeared to be required for Z-ring formation. Interestingly, SepF is also the FtsZ anchor in archaea, and both proteins may date back to the Last Universal Common Ancestor (LUCA). Curiously, viability was not greatly affected by the multiple deletions, although the frequency of cell division was considerably reduced. However, genome sequencing exposed the accumulation of several suppressor mutations, and revealed an unexpected cell division regulation function for the branched chain amino acid transporter BraB. The implications of these findings for the role of SepF in cell division, and the construction of a minimal cell division machinery are discussed.

microbiology↗

bioRxiv2024

Single-Molecule Force Spectroscopy of Toehold-Mediated Strand Displacement

Toehold-mediated strand displacement (TMSD) is extensively utilized in dynamic DNA nanotechnology and is the foundation for a wide range of DNA or RNA-based reaction circuits. Previous studies have typically relied on bulk fluorescence measurements to investigate the kinetics of TMSD, which only provide effective, bulk-averaged reaction rates and do not resolve the process on the level of individual molecules or even base pairs. In this work, we addressed this limitation by exploring the dynamics of strand displacement processes at the single-molecule level using single-molecule force spectroscopy (SMFS) with an optical trap supported by state-of-the-art coarse-grained simulations. By probing the ends of the hairpin of a toehold structure, we can trigger and observe TMSD in real time with microsecond and nanometer resolution. Using a microfluidic assay where we expose the hairpin to a solution of trigger strands, we find that under load, TMSD proceeds very rapidly, with single step times of 1 s. Introducing mismatches into the invader sequence allows us to tune stability such that invasion and re-invasion occur in equilibrium, even under load. This allows us to study thousands of invasion/re-invasion events on a single molecule and analyze the kinetics of the invasion process. Extrapolating our findings to zero load, we find single step times for DNA invading DNA four times faster than for RNA invading RNA. Moreover, we used force to study the kinetics of DNA invading RNA, a process that in the absence of force would rarely occur. Our results reveal the importance of sequence effects for the TMSD process and have importance for a wide range of applications in nucleic acid nanotechnology and synthetic biology.

biophysics↗

bioRxiv2024

Single-Cell RNA-seq based elucidation of the antidepressant hyperforin biosynthesis de novo in St. John's wort.

Hyperforin is the compound responsible for the effectiveness of St. Johns wort (Hypericum perforatum) as an antidepressant, but its biosynthesis remains unknown. Gene discovery based on co-expression analysis of bulk RNA-sequencing data or genome mining failed to discover the missing steps in hyperforin biosynthesis. Here we sequenced the tetraploid H. perforatum genome. By single-cell RNA-seq, we identified a distinct type of cells, Hyper cells, wherein hyperforin biosynthesis de novo takes place. Through pathway reconstitution in yeast and tobacco, we identify and characterize four transmembrane prenyltransferases to resolve hyperforin biosynthesis. The hyperforin polycyclic scaffold is created by a reaction cascade involving an irregular isoprenoid coupling and a tandem cyclization. Our findings reveal how and where hyperforin is biosynthesized that enables synthetic-biology reconstitution of the complete pathway. These results deepen our comprehension of specialized metabolism at the cellular level, and we anticipate acceleration of pathway elucidation in plant metabolism.

plant biology↗