Biology subjects

Ryu, A.

Publications and source records attributed to Ryu, A..

3 recordsLinked to original sources

bioRxiv2026

Host membrane cholesterol constrains constitutive signaling of the oncogenic KSHV GPCR ORF74

Cholesterol is a major structural component of the plasma membrane and a key allosteric regulator of G protein-coupled receptors (GPCRs), yet its role in controlling constitutive receptor activity remains poorly understood. Virally encoded GPCRs provide an ideal system to address this question because many exhibit constitutive signaling that promotes viral persistence and pathogenesis, although how excessive receptor activation is restrained remains unknown. Here, we identify a previously unrecognized cholesterol-dependent allosteric mechanism by which the oncogenic Kaposi's sarcoma-associated herpesvirus (KSHV) GPCR ORF74 constrains its constitutive activity. CryoEM structural analysis reveals a cholesterol-binding pocket formed by transmembrane helices 3, 5, and 6 that is present only in the inactive receptor. Cholesterol binding restrains the outward movement of transmembrane helix 6, stabilizes the inactive conformation, and suppresses spontaneous receptor activation, as supported by molecular dynamics simulations. Mechanistically, replacement of the canonical DRY motif with a non-canonical VRY motif exposes the conserved R1433.50 residue, creating a membrane-facing cholesterol-binding interface that couples membrane cholesterol to receptor conformational control. Consistent with this model, cellular cholesterol depletion enhances ORF74 signaling, whereas disruption of cholesterol binding impairs receptor stabilization. Together, our findings uncover a previously unrecognized mechanism by which a viral GPCR exploits host membrane cholesterol to regulate its constitutive activity, suggesting that persistent viruses optimize constitutive signaling by coupling receptor activity to host lipid-dependent allosteric regulation.

microbiology↗

bioRxiv2026

EVEE: Interpretable variant effect prediction from genomic foundation model embeddings

Scientific foundation models learn high-dimensional representations from diverse data modalities, yet what they encode and how to extract that knowledge remain open questions. Here we show that probing the internal representations of Evo 2, a 7-billion-parameter genomic foundation model, enables accurate and interpretable genetic variant effect prediction. We introduce a covariance-based probe that captures second-order structure from Evo 2 sequence embeddings to predict variant pathogenicity across variant types and functional consequences, matching or exceeding specialized predictors within their domains. To ground these predictions in known biological mechanisms, we train a complementary panel of probes on existing annotations to detect which genomic properties are disrupted by a variant. This categorized evidence is then integrated with each variants genomic context through a language model to generate variant-specific mechanistic hypotheses. Our pathogenicity predictions correlate with experimental measures of variant function, clinical penetrance, and biobank disease associations while the mechanistic hypotheses are consistent with expert reviews, known mechanism classes, and downstream molecular readouts. We release pathogenicity scores, disruption profiles, and contextualized interpretations for 4.2 million variants from the ClinVar database as an open resource through the Evo Variant Effect Explorer (EVEE). More broadly, this structured probing approach offers a general framework for interrogating foundation models across scientific disciplines and grounding their outputs in existing domain concepts.

genomics↗

bioRxiv2024

Landscape profiling of PET depolymerases using a natural sequence cluster framework

Since the demonstration that rapid polyethylene terephthalate (PET) decomposition using enzymes is feasible, a number of efficient depolymerases have been reported with the aim of resolving the plastic pollution issues. However, sporadic studies on enzymes with PET hydrolysis activity hinder the understanding of the distribution of potential PETases hidden in natures repertoire, and subsequently, the identification of potent enzymes. Here, we present the clustering of 1,894 PETase candidates, which include the majority of known PETases, and describe their profiling. An archipelago landscape of 170 lineages shows distribution of 289 representative sequences with features associated with PET-degrading capabilities. A birds-eye view of the landscape identifies three highly promising yet unexplored PETase lineages and two potent PETases, Mipa-P and Kubu-P. The engineered Mipa-PM19 and Kubu-PM12 variants exhibit both high PET hydrolysis activity and thermal stability. In particular, Kubu-PM12 outperformed the engineered benchmarks in terms of PET depolymerization in harsh environments, such as with high substrate load and ethylene glycol as the solvent. Our landscape framework and the identified variants assist in the understanding of how biological processes respond to solid-state and non-natural PET plastics.

systems biology↗