Biology subjects

Arnold, F.

Publications and source records attributed to Arnold, F..

7 recordsLinked to original sources

bioRxiv2026

EZSolver: Template-free prediction of polar enzymatic mechanisms via bidirectional flow matching and search

Predicting enzymatic reaction mechanisms is critical for understanding enzyme function and for designing and discovering new enzymes. Current computational predictors rely on deterministic, rule-based dictionaries, which perform well on in-distribution tasks but fail to generalize to out-of-distribution (OOD) chemistry. To address this limitation, we present EZSolver, a template-free, generative framework for polar enzymatic mechanism prediction. Powered by a flow matching predictor (EZFlow) and navigated by an evaluator-guided bidirectional beam search, EZSolver learns the chemistry of electron redistribution instead of memorizing rigid templates. Evaluated across diverse enzyme classes, EZSolver achieves a 60.0% accuracy and an 84.6% chemical plausibility rate for full mechanism prediction of unseen polar enzymatic reactions. While rule-based models collapse without predefined templates, EZSolver successfully extrapolates chemical knowledge to infer uncatalogued pathways, as demonstrated during rigorous OOD benchmarking. By illuminating enzymatic chemical mechanisms, EZSolver helps pave the way for automated prediction of enzyme function and discovery and design of novel biocatalysts for sustainable chemistry. TOC Graphic O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=152 SRC="FIGDIR/small/737313v1_ufig1.gif" ALT="Figure 1"> View larger version (30K): org.highwire.dtl.DTLVardef@1b0ad92org.highwire.dtl.DTLVardef@535a71org.highwire.dtl.DTLVardef@56e846org.highwire.dtl.DTLVardef@1ab76ec_HPS_FORMAT_FIGEXP M_FIG C_FIG

bioinformatics↗

bioRxiv2025

Exploiting TGF-β-mediated Stromal Programming in Homologous Recombination-Deficient Pancreatic Cancer

The tumor microenvironment (TME) actively contributes to pancreatic ductal adenocarcinoma (PDAC) pathogenesis through dynamic bidirectional tumor-stroma interactions. Here, we demonstrate that homologous recombination-defective (HRD) tumor epithelium reprograms the TME in a genotype-specific manner to enhance cancer aggressiveness. Using genetically engineered mouse models, pancreatic stellate cell (PSC) and cancer-associated fibroblast (CAF) co-culture systems, single-nucleus multiomics, and human PDAC models, we show that tumoral loss of ATM serine/threonine kinase drives CAFs toward SMA+ myofibroblastic differentiation, independently of P53 status. These myCAFs, in turn, promote cancer aggressiveness and chemoresistance. Mechanistically, ATM deficiency increases reactive oxygen species and contractility signaling, enhancing TGF-{beta}1 secretion. Pharmacological TGF-{beta} inhibition reverses myCAF differentiation, sensitizes tumors to chemotherapy, and impairs tumor progression in both murine and human ATM-null models. Our findings reveal that ATM-deficient tumors shape a cancer-promoting niche via TGF-{beta} signaling and identify dual targeting of intrinsic and extrinsic vulnerabilities as a promising precision oncology strategy. SIGNIFICANCEHRD pancreatic cancers reprogram the tumor microenvironment in a genotype-specific manner through TGF-{beta}-driven myCAF-enrichment. Targeting this stromal axis alongside platinum-based chemotherapy improves therapeutic efficacy in ATM-deficient models. These findings highlight the need to integrate epithelial genotype and stromal context for truly personalized treatment strategies in PDAC.

cancer biology↗

bioRxiv2025

Novel peptide targeting CXCR4 disrupt tumor-stroma crosstalk to eliminate migrating cancer stem cells

Pancreatic ductal adenocarcinoma (PDAC) is one of the most aggressive and metastatic malignancies worldwide. Migrating cancer stem cells (miCSCs) marked by CD133+CXCR4+ expression drives metastasis but lacks effective drug targets. Here, we show that activated pancreatic stellate cells secrete the CXCR4 ligand CXCL12 to foster stemness, epithelial-to-mesenchymal transition (EMT), and chemoresistance. Protein interaction network analyses links CXCL12/CXCR4 signaling axis and the downstream transcription factor BMI1. Knockdown experiments confirmed the BMI1s role in (mi)CSCs maintenance and survival. Novel CXCR4 inhibitors, i.e., the endogenous human peptide EPI-X4 and its derivatives (e.g., JM#21) strongly inhibited the in vitro migration of miCSCs. In particular, the most potent EPI-X4 derivate JM#21 sufficiently suppressed EMT, stemness, and self-renewal of human PDAC cell lines. In addition, JM#21 sensitized cell lines towards gemcitabine and paclitaxel. Overall, our study reveals that (mi)CSCs are enhanced and maintained via a tumor-stroma crosstalk through BMI1, ultimately promoting metastases and therapeutic resistance in PDAC. Peptide targeting of the CXCL12/CXCR4/BMI1 signaling axis via JM#21 could enhance PDAC combination therapies, offering a promising strategy against this deadly cancer. SynopsisThe study identifies a tumor-stroma interaction mediated by pancreatic stellate cells (PSCs) secreting CXCL12, which binds to CXCR4 on (mi)CSCs, fostering stemness, epithelial-to-mesenchymal transition (EMT), and chemoresistance. The CXCL12/CXCR4 axis activates the downstream BMI1 transcription factor, crucial for migration and stemness maintenance. O_FIG O_LINKSMALLFIG WIDTH=173 HEIGHT=200 SRC="FIGDIR/small/641126v1_ufig1.gif" ALT="Figure 1"> View larger version (50K): org.highwire.dtl.DTLVardef@14e117org.highwire.dtl.DTLVardef@c5db35org.highwire.dtl.DTLVardef@1ba7d2dorg.highwire.dtl.DTLVardef@702ccf_HPS_FORMAT_FIGEXP M_FIG C_FIG O_LICXCL12 enhances (mi)CSC populations and metastatic potential through CXCR4 signaling. C_LIO_LIBMI1 is identified as a pivotal downstream effector linking CXCR4 to EMT and stemness. C_LIO_LIJM#21 effectively blocks CXCL12-induced migration, EMT, and stemness in vitro, demonstrating superior efficacy compared to other CXCR4 inhibitors. C_LIO_LIEncapsulation of JM#21 in silica nanoparticles enhances its stability and delivery, reducing chemoresistance and miCSC populations in co-culture systems. C_LIO_LICombining JM#21 with chemotherapy significantly impairs colony formation and CSC-mediated drug resistance. C_LI

cancer biology↗

bioRxiv2024

Evaluation of Machine Learning-Assisted Directed Evolution Across Diverse Combinatorial Landscapes

Various machine learning-assisted directed evolution (MLDE) strategies have been shown to identify high-fitness protein variants more efficiently than typical wet-lab directed evolution approaches. However, limited understanding of the factors influencing MLDE performance across diverse proteins has hindered optimal strategy selection for wet-lab campaigns. To address this, we systematically analyzed multiple MLDE strategies, including active learning and focused training using six distinct zero-shot predictors, across 16 diverse protein fitness landscapes. By quantifying landscape navigability with six attributes, we found that MLDE offers a greater advantage on landscapes which are more challenging for directed evolution, especially when focused training is combined with active learning. Despite varying levels of advantage across landscapes, focused training with zero-shot predictors leveraging distinct evolutionary, structural, and stability knowledge sources consistently outperforms random sampling for both binding interactions and enzyme activities. Our findings provide practical guidelines for selecting MLDE strategies for protein engineering.

bioengineering↗

bioRxiv2023

Glycosylation differentially affects immune cell-specific tetraspanins CD37 and CD53

Tetraspanin proteins play an important role in many cellular processes as they are key organizers of different receptors on the plasma membrane. Most tetraspanins are highly glycosylated at their large extracellular loop, but the function of this post-translational modification remains largely unstudied. In this study we investigated the effects of glycosylation of CD37 and CD53, two tetraspanins important for cellular and humoral immunity. Broad and cell-specific repertoires of N-glycosylated CD37 and CD53 were observed in human B cells. We generated different glycosylation mutants of CD37 and CD53 and analyzed their localization, nanoscale organization and partner protein interaction capacity. Abrogation of glycosylation in CD37 revealed the importance of this modification for CD37 surface expression, whereas neither surface expression nor nanoscale organization of CD53 was affected by its glycosylation. CD37 interaction with its known partner proteins, CD20 and IL-6R, was not affected by glycosylation, other than via its changed subcellular localization. Surprisingly, glycosylation was found to inhibit the interaction between CD53 and its partner proteins CD45 and CD20. Together, our data show that tetraspanin glycosylation affects their function in immune cells, which adds another layer of regulation to tetraspanin-mediated membrane organization.

cell biology↗

bioRxiv2022

MicroED structure of a protoglobin reactive carbene intermediate

Microcrystal electron diffraction (MicroED) is an emerging technique which has shown great potential for describing new chemical and biological molecular structures. [1] Several important structures of small molecules, natural products and peptides have been determined using ab initio methods. [2] However, only a couple of novel protein structures have thus far been derived by MicroED. [3, 4] Taking advantage of recent technological advances including higher acceleration voltage and using a low-noise detector in counting mode, we have determined the first structure of an Aeropyrum pernix protoglobin (ApePgb) variant by MicroED using an AlphaFold2 model for phasing. The structure revealed that mutations introduced during directed evolution enhance carbene transfer activity by reorienting an alphahelix of ApePgb into a dynamic loop making the catalytic active site more readily accessible. After exposing the tiny crystals to substrate, we also trapped the reactive iron-carbenoid intermediate involved in this engineered ApePgbs new-to-nature activity, a challenging carbene transfer from a diazirine via a putative metallo-carbene. The bound structure discloses how an enlarged active site pocket stabilizes the carbene bound to the heme iron and, presumably, the transition state for formation of this key intermediate. This work demonstrates that improved MicroED technology and the advancement in protein structure prediction now enables investigation of structures that were previously beyond reach.

biochemistry↗

bioRxiv2020

Pancreatic cancer stem cells (CSCs) are regulated via a feedback-loop of telomerase activity and stemness factors

To date, it is still unclear how cancer stem cells (CSCs) regulate their stemness properties, and to what extent they share common features with normal stem cells. Telomerase regulation is a key factor in stem cell maintenance. In this study, we investigate how telomerase regulation affects cancer stem cell biology in pancreatic ductal adenocarcinoma (PDAC), and delineate the mechanisms by which telomerase activity and CSC properties are linked. Using primary patient-derived pancreatic cancer cells, we show that CSCs have higher telomerase activity and longer telomeres than bulk tumor cells. Inhibition of telomerase activity, using genetic TERT-knockdown or pharmacological inhibitor (BIBR1532) resulted in CSC marker depletion in vitro, and reduced tumorigenicity in vivo. Furthermore, we identify a positive feedback loop between stemness factors (KLF4, SOX2, OCT3/4, NANOG) and telomerase, which is essential for the self-renewal of pancreatic CSCs. Disruption the balance between telomerase activity and stemness factors, eliminates CSCs via induction of DNA damage and apoptosis, opening future perspectives to avoid CSC driven therapy resistance and tumor relapse in PDAC patients.

cancer biology↗