Biology subjects

Woldring, D. R.

Publications and source records attributed to Woldring, D. R..

5 recordsLinked to original sources

bioRxiv2025

Comparing the evolvability of an ancestrally reconstructed and modern adenylate kinase

Directed evolution transformed protein engineering by providing a customizable framework for generating enzymes with improved catalytic performance across diverse functions. Yet modern enzymes often stall during directed evolution because populations become trapped on local fitness peaks. Researchers have suggested that ancestral enzymes offer better starting points because they are typically more thermostable. Here we propose and experimentally test an alternative explanation that does not depend on ancestral thermostability. We posit that ancestrally reconstructed sequences are unusually evolvable because they are inferred from the evolutionary lineages that survived to produce extant proteins. Less evolvable ancestors, and the trajectories emanating from them, disappeared by extinction and therefore do not contribute to reconstructed ancestors. Using thermophilic ancestral and modern adenylate kinases, we performed independent single-round selection experiments for activity in vivo and in vitro. In both settings, the ancestral enzyme tolerates a larger number of mutations, yielding more viable variants with greater genetic diversity than its modern descendants. Because mutational robustness promotes evolvability, these results support an intrinsic evolvability of reconstructed ancestral sequences that makes them superior launch points for directed evolution.

biochemistry↗

bioRxiv2025

Tuning Yeast Glycosylation Proximal to the FLS2-flg22 Binding Interface enables Functional Yeast Surface Display under Induced ER Stress

Pattern recognition receptors such as FLAGELLIN SENSING 2 (FLS2) are central to plant immunity and attractive targets for engineering broader detection of bacterial phytopathogens for application in pest management (and diagnostics) in food crops, and sustainable agriculture practices. However, evaluating numerous FLS2 variants for altered pathogen sensing specificity directly in plants is slow and low throughput, and have been seldom optimized for heterologous display systems. Here, we established conditions that enabled Arabidopsis thaliana FLS2 ectodomain expression on the surface of Saccharomyces cerevisiae and evaluated binding to its cognate ligand, flg22. We show how yeast high-mannose glycosylation of the FLS2 ectodomain contributes to inefficient folding and loss of detectable flg22 binding in standard yeast surface display conditions. Substitutions at all N-glycosylation motifs compromised surface expression, indicating that some glycosylation is required for trafficking. We tuned the extent of glycosylation using tunicamycin, an N-linked glycosylation inhibitor, in combination with thermal stress to modulate ER quality control. Under these conditions, we observed a reproducible subpopulation of cells with improved flg22 binding despite reduced overall expression, and we confirmed flg22 selectivity against non-FLS2 proteins using both flow cytometry and magnetic bead-based enrichment. Guided by structural modeling of high-mannose glycans on the FLS2 ectodomain, we then substituted asparagines at selected N-glycan sites to serine. We identified a key glycan site variant, N388S, which lies proximal to the flg22 binding interface and increased the binding population size under stress conditions. Binding assays against FLS2 variants and a reported non-binding variant affirmed that FLS2 selectivity was specific to FLS2 display, showing that all variants maintained low affinity interaction with flg22. Together, these results point to FLS2 display conditions, not only glycosylation state, as an underlying limitation to detect true flg22 interactions which will require more sensitive approaches to confidently resolve true binding populations. For Table of Contents Use Only O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=144 SRC="FIGDIR/small/690463v2_ufig1.gif" ALT="Figure 1"> View larger version (38K): org.highwire.dtl.DTLVardef@1556b0eorg.highwire.dtl.DTLVardef@e764d1org.highwire.dtl.DTLVardef@18bdcfborg.highwire.dtl.DTLVardef@1592114_HPS_FORMAT_FIGEXP M_FIG C_FIG

synthetic biology↗

bioRxiv2025

VesicleVoyager: In vivo selection of surface displayed proteins that direct extracellular vesicles to tissue-specific targets

The development of technologies for screening proteins that bind to specific tissues in vivo and facilitate delivery of large cargos remains challenging, with most approaches limited to cell culture systems that often yield clinically irrelevant hits. To overcome this limitation, we developed a novel molecular screening platform using an extracellular vesicle (EV) display library. EVs are natural molecular carriers capable of delivering diverse cargos, which can be engineered to enhance specificity and targeting through surface modifications. We constructed an EV-display library presenting monobody repertoires on EV surfaces, with genetic cargo inside the EVs corresponding to the displayed proteins. These libraries were screened for tissue specific delivery through serial passage in mice via sequential intravenous administration in and recovery of tissue-selected EVs and amplification of their encapsulated monobody genes at each passage. Our results demonstrated successful selection of tissue-specific targeting proteins, as revealed by fluorescence and bioluminescence imaging followed by DNA sequencing. To understand the stochastic relationship between displayed proteins and packaged genes, we developed a Markov chain model that quantified selection dynamics and predicted enrichment patterns despite the imperfect correlation between phenotype and genotype. This EV-based monobody screening approach, combined with mathematical modeling, is a significant advancement in targeted drug delivery by leveraging the natural capabilities of EVs with the selection of targeting proteins in a physiologically relevant environment.

bioengineering↗

bioRxiv2024

EvoSeq-ML: Advancing Data-Centric Machine Learning with Evolutionary-Informed Protein Sequence Representation and Generation

From protein structure prediction to novel protein generation, challenging protein engineering tasks have been made possible by advancements in machine learning (ML). While largely driven by ML architecture refinements, these advancements in ML-based protein engineering campaigns have left the impact of data curation underexplored. In light of the growing wealth of labeled sequence data, data-centric advances (e.g. prioritizing improvements in ML protein engineering tools through the curation of high-quality, domain-specific training data) are increasingly preferred over model-centric advancements. Implementing datasets that accurately reflect biological complexity and diversity has been shown to improve the efficiency of training protein engineering ML tools. Here, we evaluate an ancestral sequence reconstruction (ASR)-informed data augmentation strategy for training generative and representation-learning models in protein engineering. Using ethylene-forming enzyme (EFE) as a model system, we show that variational autoencoder models trained on ancestral and near-ancestral sequence datasets generate variants with improved predicted and experimentally measured thermostability relative to variants generated from modern-sequence training data. All experimentally tested ancestral and ML-generated EFEs produced detectable ethylene, although ML-generated variants showed reduced activity relative to wild-type EFE, indicating that the approach more strongly captured stability-associated features than catalytic optimization. We further evaluated ASR-enriched sequence sets for fine-tuning ESM2 representations in endolysin and lysozyme C stability-classification tasks, where ancestral representations were competitive with modern-sequence fine-tuning in selected settings. Overall, this work supports ASR-informed data augmentation as a promising strategy for stability-oriented protein sequence generation and motivates future work to couple ancestral sequence diversity with explicit functional selection.

bioinformatics↗

bioRxiv2024

Predicting Inhibitors of OATP1B1 via Heterogeneous OATP-Ligand Interaction Graph Neural Network (HOLI-GNN)

Organic anion transporting polypeptides (OATPs) are membrane transporters crucial for drug uptake and distribution in the human body. OATPs can mediate drug-drug interactions (DDIs) in which the interaction of one drug with an OATP impairs the uptake of another drug, resulting in potentially fatal pharmacological effects. Predicting OATP-mediated DDIs is challenging, due to limited information on OATP inhibition mechanisms and inconsistent experimental OATP inhibition data across different studies. This study introduces Heterogeneous OATP-Ligand Interaction Graph Neural Network (HOLIgraph), a novel computational model that integrates molecular modeling with a graph neural network to enhance the prediction of drug-induced OATP inhibition. By combining ligand (i.e., drug) molecular features with protein-ligand interaction data from rigorous docking simulations, HOLIgraph outperforms traditional DDI prediction models which rely solely on ligand molecular features. HOLIgraph achieved a median balanced accuracy of over 90 percent when predicting inhibitors for OATP1B1, significantly outperforming purely ligand-based models. Beyond improving inhibition prediction, the data used to train HOLIgraph can enable the characterization of protein residues involved in inhibitory drug-OATP interactions. We identified certain OATP1B1 residues that preferentially interact with inhibitors, including I46 and K49. We anticipate such interaction information will be valuable to future structural and mechanistic investigations of OATP1B1. Scientific ContributionHOLIgraph introduces a new paradigm for DDI prediction by incorporating protein-ligand interactions derived from docking simulations into a graph neural net framework. This approach, enabled by recent structural breakthroughs for OATP1B1, represents a significant departure from traditional models that rely only on ligand features. By achieving high predictive accuracy and uncovering mechanistic insights, HOLIgraph sets a new trajectory for computational tools in drug design and DDI prediction.

bioengineering↗