Agentic systems for drug discovery have demonstrated autonomous synthesis planning, literature mining, and molecular design. We ask how well they generalize. Evaluating six frameworks against 15 task classes drawn from peptide therapeutics, in vivo pharmacology, and resource-constrained settings, we find five capability gaps: no support for protein language models or peptide-specific prediction, no bridges between in vivo and in silico data, reliance on LLM inference with no pathway to ML training or reinforcement learning, assumptions tied to large-pharma resources, and single-objective optimization that ignores safety-efficacy-stability trade-offs. A paired knowledge-probing experiment suggests the bottleneck is architectural rather than epistemic: four frontier LLMs reason about peptides at levels comparable to small molecules, yet no framework exposes this capability. We propose design requirements and a capability matrix for next-generation frameworks that function as computational partners under realistic constraints.
Target-specific peptides, such as conotoxins, exhibit exceptional binding affinity and selectivity toward ion channels and receptors. However, their therapeutic potential remains underutilized due to the limited diversity of natural variants and the labor-intensive nature of traditional optimization strategies. Here, we present CreoPep, a deep learning-based conditional generative framework that integrates masked language modeling with a progressive masking scheme to design high-affinity peptide mutants while uncovering novel structural motifs. CreoPep employs an integrative augmentation pipeline, combining FoldX-based energy screening with temperature-controlled multinomial sampling, to generate structurally and functionally diverse peptides that retain key pharmacological properties. We validate this approach by designing conotoxin inhibitors targeting the $α$7 nicotinic acetylcholine receptor, achieving submicromolar potency in electrophysiological assays. Structural analysis reveals that CreoPep-generated variants engage in both conserved and novel binding modes, including disulfide-deficient forms, thus expanding beyond conventional design paradigms. Overall, CreoPep offers a robust and generalizable platform that bridges computational peptide design with experimental validation, accelerating the discovery of next-generation peptide therapeutics.
The systematic discovery of effective drug combinations is a challenging problem in modern pharmacology, driven by the combinatorial growth of potential pairings and dosage configurations. Network medicine, modeling diseases and drugs as interconnected modules of the human protein-protein interactome, has emerged as a new paradigm for understanding disease mechanisms and drug action. In this work, we propose a quantum annealing-based algorithm for identifying effective drug combinations. Underlying our approach is the biologically motivated principle of `Complementary Exposure', which posits that therapeutic drug combinations target distinct yet complementary regions of a disease module. We translate this into a quadratic unconstrained binary optimisation problem. We test our method for Diabetes Mellitus, Rheumatoid Arthritis, Asthma, and Brain Neoplasms, relying on experimentally validated drug combinations for these diseases. Our simulated quantum annealing experiments reveal that low-energy configurations align with biologically plausible combinations, demonstrating the algorithm's ability to generate novel predictions for drug combinations.
Paolo Madeddu, Styliani Goulopoulou, David Wambeke
Endothelial cells release various vasorelaxing molecules, such as nitric oxide and prostacyclin, along with defined factors that induce hyperpolarization of vascular smooth muscle cells through the opening of calcium-sensitive potassium channels. Potassium channel-dependent vasorelaxation is prevalent in microvessels and can partially compensate for deficiencies in other vasodilatory mechanisms. Enhancing this backup vasorelaxant mechanism may aid the treatment of microvascular disorders, such as cerebral small vessel disease and preeclampsia, a pregnancy-specific hypertensive syndrome, which is characterized by systemic endothelial dysfunction. The development of pharmacological potassium channel openers has encountered significant challenges, including issues of specificity, safety concerns, and off-target effects. This study critically evaluates the advantages and drawbacks of integrating hyperpolarization into a holistic vasorelaxant strategy for managing ischemic disease through single or combination drug therapies.
Designing therapeutic peptides with tailored properties is hindered by the vastness of sequence space, limited experimental data, and poor interpretability of current generative models. To address these challenges, we introduce PepThink-R1, a generative framework that integrates large language models (LLMs) with chain-of-thought (CoT) supervised fine-tuning and reinforcement learning (RL). Unlike prior approaches, PepThink-R1 explicitly reasons about monomer-level modifications during sequence generation, enabling interpretable design choices while optimizing for multiple pharmacological properties. Guided by a tailored reward function balancing chemical validity and property improvements, the model autonomously explores diverse sequence variants. We demonstrate that PepThink-R1 generates cyclic peptides with significantly enhanced lipophilicity, stability, and exposure, outperforming existing general LLMs (e.g., GPT-5) and domain-specific baseline in both optimization success and interpretability. To our knowledge, this is the first LLM-based peptide design framework that combines explicit reasoning with RL-driven property control, marking a step toward reliable and transparent peptide optimization for therapeutic discovery.
Clare R. Harwood, Clare R. Harwood, David A. Sykes
et al.
IntroductionThe β2-adrenoceptor (β2AR) is a class A G protein-coupled receptor (GPCR). It is therapeutically relevant in asthma and chronic obstructive pulmonary disease (COPD), where β2AR agonists relieve bronchoconstriction. The β2AR is a prototypical GPCR for structural and biophysical studies. However, the molecular basis of agonist efficacy at the β2AR is not understood. We hypothesised that the kinetics of GPCR–G protein interactions could play a role in determining ligand efficacy. By studying a range of agonists with varying efficacy, we examined the relationship between ligand-induced mini-Gs binding to the β2AR and ligand efficacy, along with the ability of individual ligands to activate the G protein in cells.MethodsWe used NanoBRET technology to measure ligand-induced binding of purified Venus-mini-Gs to β2AR-nLuc in membrane preparations under both equilibrium and kinetic conditions. In addition, we examined the ability of these β2AR agonists to activate the heterotrimeric Gs protein, measured using the Gs-CASE protein biosensor in living cells. This assay detects a reduction in NanoBRET between the nano-luciferase (nLuc) donor on the Gα subunit and Venus acceptor on the Gγ upon Gs protein activation.ResultsThe 12 β2AR agonists under study revealed a broad range of ligand potency and efficacy values in the cellular Gs-CASE assays. Kinetic characterisation of mini-Gs binding to the agonist β2AR complex revealed a strong correlation between ligand efficacy values (Emax) and mini-Gs affinity (Kd) and its association rate (kon). In contrast, there was no correlation between ligand efficacy and reported ligand dissociation rates (or residence times).ConclusionThe association rate (kon) of the G protein to the agonist β2AR complex is directly correlated with ligand efficacy. These data support a model in which higher-efficacy agonists induce the β2AR to adopt a conformation that is more likely to recruit G protein. Conversely, these data did not support the role of agonist binding kinetics in determining the molecular basis of efficacy.
Vasiliki Koumaki, Eleni Voudanta, Aikaterini Michelaki
et al.
<b>Background:</b> Carbapenemase-producing Enterobacterales (CPEs) represent a significant global health threat, particularly in the context of nosocomial infections. The current study constitutes a retrospective epidemiological survey that aimed to provide updated data on the prevalence and characteristics of carbapenemases among carbapenem-resistant Enterobacterales (CREs) in a Greek tertiary hospital in Athens during and after the COVID-19 pandemic. <b>Results</b>: A total of 2021 non-duplicate CPE clinical isolates were detected. A significant increase in the number of carbapenemase-positive Enterobacterales was revealed during the study period (<i>p</i> < 0.05). KPC remained the predominant carbapenemase type through all four years of the survey, representing 40.7%, 39.9%, 53.5%, and 45.7% of the CPE isolates, respectively. However, a rapid transition from VIM to NDM metal-β-lactamase types was revealed, changing the epidemiological image of carbapenemases in the hospital setting. Notably, among the CPEs, antimicrobial resistance rates were significantly raised in the post-COVID-19 period (2022 and 2023) compared to the first study year (2020) for almost all the tested antibiotics, including those characterized as last-resort antibiotics. <b>Methods</b>: CREs were identified and subjected to screening for the five most prevalent carbapenemase genes [<i>Klebsiella pneumoniae</i> carbapenemase (KPC), Verona integron-borne metallo-β-lactamase (VIM), New Delhi metallo-β-lactamase (NDM), imipenemase (IMP), and oxacillin-hydrolyzing (OXA-48)] using a lateral flow immunoassay, and the CREs recovered from blood cultures were analyzed using a FilmArray system. Their clinical and epidemiological characteristics, as well as their antimicrobial susceptibility profiles, were also subjected to analysis <b>Conclusions</b>: Given this alarming situation, which is exacerbated by the limited treatment options, the development of new, effective antimicrobial agents is needed. The continued monitoring of the changing epidemiology of carbapenemases is also imperative in order to undertake rational public health interventions.
Camila Queraltó, Iván L. Calderón, Isidora Flores
et al.
<i>Clostridioides difficile</i> is a Gram-positive bacterium recognized for its ability to produce toxins and form spores. It is mainly accountable for the majority of instances of antibiotic-related diarrhea. <b>Background.</b> Bacterial persister represent a minor fraction of the population that shows temporary tolerance to bactericidal agents, and they pose considerable medical issues because of their link to the rise of antibiotic resistance and challenging chronic or recurrent infections. Our previous research has shown a persister-like phenotype associated with treatments that include pefloxacin. Nonetheless, the mechanism is still mostly unclear, mainly because of the difficulty in isolating this small group of cells. <b>Objectives.</b> To enhance the understanding of <i>C. difficile</i> persister cells, we made an enrichment and characterization of these cells from bacterial cultures during the exponential phase under pefloxacin treatment and lysis treatment. <b>Results.</b> We demonstrate the appearance of cells with lower metabolism and DNA damage. Furthermore, we noted the participation of toxin–antitoxin systems and Clp proteases in the generation of persister cells. <b>Conclusions.</b> This work demonstrates the formation of <i>C. difficile</i> persister cells triggered by a lethal concentration of pefloxacin.
Previously reported nanodosimetric measurements of therapeutic-energy carbon ions penetrating simulated tissue have produced results that are incompatible with the predicted mean energy of the carbon ions in the nanodosimeter and previous experiments with lower energy monoenergetic beams. The purpose of this study is to explore the origin of these discrepancies. Detailed simulations using the Geant4 toolkit were performed to investigate the radiation field in the nanodosimeter and provide input data for track structure simulations, which were performed with a developed version of the PTra code. The Geant4 simulations show that with the narrow-beam geometry employed in the experiment, only a small fraction of the carbon ions traverse the nanodosimeter and their mean energy is between 12 % and 30 % lower than the targeted values. Only about one-third or less of these carbon ions hit the trigger detector. The track structure simulations indicate that the observed enhanced ionization cluster sizes are mainly due to coincidences with events in which carbon ions miss the trigger detector. In addition, the discrepancies observed for high absorber thicknesses of carbon ions traversing the target volume could be explained by assuming an increase in thickness or interaction cross-sections in the order of 1 %. The results show that even with strong collimation of the radiation field, future nanodosimetric measurements of clinical carbon ion beams will require large trigger detectors to register all events with carbon ions traversing the nanodosimeter. Energy loss calculations of the primary beam in the absorbers are insufficient and should be replaced by detailed simulations when planning such experiments. Uncertainties of the interaction cross-sections in simulation codes may shift the Bragg peak position.
Small molecules play a pivotal role in modern medicine, and scrutinizing their interactions with protein targets is essential for the discovery and development of novel, life-saving therapeutics. The term "bioactivity" encompasses various biological effects resulting from these interactions, including both binding and functional responses. The magnitude of bioactivity dictates the therapeutic or toxic pharmacological outcomes of small molecules, rendering accurate bioactivity prediction crucial for the development of safe and effective drugs. However, existing structural datasets of small molecule-protein interactions are often limited in scale and lack systematically organized bioactivity labels, thereby impeding our understanding of these interactions and precise bioactivity prediction. In this study, we introduce a comprehensive dataset of small molecule-protein interactions, consisting of over a million binding structures, each annotated with real biological activity labels. This dataset is designed to facilitate unbiased bioactivity prediction. We evaluated several classical models on this dataset, and the results demonstrate that the task of unbiased bioactivity prediction is challenging yet essential.
De novo molecule generation allows the search for more drug-like hits across a vast chemical space. However, lead optimization is still required, and the process of optimizing molecular structures faces the challenge of balancing structural novelty with pharmacological properties. This study introduces the Deep Genetic Molecular Modification Algorithm (DGMM), which brings structure modification to the level of medicinal chemists. A discrete variational autoencoder (D-VAE) is used in DGMM to encode molecules as quantization code, mol-gene, which incorporates deep learning into genetic algorithms for flexible structural optimization. The mol-gene allows for the discovery of pharmacologically similar but structurally distinct compounds, and reveals the trade-offs of structural optimization in drug discovery. We demonstrate the effectiveness of the DGMM in several applications.
Aiswarya Premchandar, Ruiji Ming, Abed Baiad
et al.
Cystic fibrosis (CF) is a monogenic disease caused by mutations in the CF transmembrane conductance regulator (CFTR) gene. Premature termination codons (PTCs) represent ∼9% of CF mutations that typically cause severe expression defects of the CFTR anion channel. Despite the prevalence of PTCs as the underlying cause of genetic diseases, understanding the therapeutic susceptibilities of their molecular defects, both at the transcript and protein levels remains partially elucidated. Given that the molecular pathologies depend on the PTC positions in CF, multiple pharmacological interventions are required to suppress the accelerated nonsense-mediated mRNA decay (NMD), to correct the CFTR conformational defect caused by misincorporated amino acids, and to enhance the inefficient stop codon readthrough. The G418-induced readthrough outcome was previously investigated only in reporter models that mimic the impact of the local sequence context on PTC mutations in CFTR. To identify the misincorporated amino acids and their ratios for PTCs in the context of full-length CFTR readthrough, we developed an affinity purification (AP)-tandem mass spectrometry (AP-MS/MS) pipeline. We confirmed the incorporation of Cys, Arg, and Trp residues at the UGA stop codons of G542X, R1162X, and S1196X in CFTR. Notably, we observed that the Cys and Arg incorporation was favored over that of Trp into these CFTR PTCs, suggesting that the transcript sequence beyond the proximity of PTCs and/or other factors can impact the amino acid incorporation and full-length CFTR functional expression. Additionally, establishing the misincorporated amino acid ratios in the readthrough CFTR PTCs aided in maximizing the functional rescue efficiency of PTCs by optimizing CFTR modulator combinations. Collectively, our findings contribute to the understanding of molecular defects underlying various CFTR nonsense mutations and provide a foundation to refine mutation-dependent therapeutic strategies for various CF-causing nonsense mutations.
Zhiyuan You,1 Jialin Zhang,1 Yifeng Xu,1 Junhong Lu,1 Renling Zhang,2 Zhujing Zhu,3 Yiqin Wang,1 Yiming Hao1 1Shanghai Key Laboratory of Health Identification and Assessment/Laboratory of TCM Four Diagnostic Information, Shanghai University of Traditional Chinese Medicine, Shanghai, 201203, People’s Republic of China; 2Gastroenterology, Longhua Hospital Shanghai University of Traditional Chinese Medicine, Shanghai, 201203, People’s Republic of China; 3Rheumatology, Longhua Hospital Shanghai University of Traditional Chinese Medicine, Shanghai, 201203, People’s Republic of ChinaCorrespondence: Yiming Hao, Email hymjj888@163.com
V. M. Kosman, M. V. Karlina, V. A. Vavilova
et al.
SCIENTIFIC RELEVANCE. The high prevalence of fungal skin infections motivates expanding the range of sertaconazole products for external use.AIM. The study was a preclinical comparison of the safety, antifungal activity, and pharmacokinetics of Sertaverin® 2% medicated shampoo (VERTEX JSC, Russia) with those of Sertamicol® 2% solution for external use (Glenmark Pharmaceuticals Ltd, India) and Nizoral® 2% shampoo (Janssen Pharmaceuticals N.V., Belgium) approved in the Russian Federation.MATERIALS AND METHODS. In the toxicity study, the medicinal products were applied to the skin of male and female outbred rats at doses of 0.5 or 1.5 mL/animal for 28 days. The authors evaluated the pharmacokinetics of two sertaconazole formulations (shampoo and solution) following a single administration to adult male rats at the same dose. Nizoral® was not used in the pharmacokinetics study because it contains a different active substance, ketoconazole. The minimum inhibitory concentration (MIC) was determined using the serial microdilution method in a wide range of concentrations.RESULTS. The medicinal products did not exhibit any significant toxic effects in laboratory animals after 28 days of repeated dermal application. Plasma sertaconazole concentrations were negligible. Sertaconazole was intensively distributed in the liver, which is a highly vascularised organ, and in the target organ (skin at the site of application). The relative bioavailability of sertaconazole from the shampoo relative to that from the solution for external use was approximately 30% in liver tissues and approximately 363% in skin tissues at the application site. Sertaverin® was comparable to sertaconazole in the active substance form in terms of inhibiting the growth of Malassezia furfur strains. The MICs calculated on the active substance basis were ≤16–64 μg/mL.CONCLUSIONS. With its synergistic dual mechanism of action, broad-spectrum antifungal activity, lipophilic properties, and low systemic absorption, Sertaverin® may provide a more effective and safe alternative to marketed medicinal products for scalp diseases.
Maryam Sanaee, K. Göran Ronquist, Elin Sandberg
et al.
Antibodies, disruptive potent therapeutic agents against pharmacological targets, face a barrier crossing immune-system and cellular-membranes. To overcome these, various strategies have been explored including shuttling via liposomes or bio-camouflaged nanoparticles. Here, we demonstrate the feasibility to load antibodies into exosome-mimetic nanovesicles derived from human red-blood-cell-membranes. The goat-anti-chicken antibodies are loaded into erythrocyte-membrane derived nanovesicles and their loading yields are characterized and compared with smaller dUTP-cargo. Applying dual-color coincident fluorescence burst methodology, the loading yield of nanocarriers is profiled at single-vesicle level overcoming their size-heterogeneity and achieving a maximum of 38-41% antibody-loading yield at peak radius of 52 nm. The average of 14 % yield and more than two antibodies per vesicle is estimated, comparable to those of dUTP-loaded nanovesicles after additional purification through exosome-spin-column. These results suggest a promising route for enhancing biodistribution and intracellular accessibility for therapeutic antibodies using novel, biocompatible, and low-immunogenicity nanocarriers, suitable for large-scale pharmacological applications.
The drug development process necessitates that pharmacologists undertake various tasks, such as reviewing literature, formulating hypotheses, designing experiments, and interpreting results. Each stage requires accessing and querying vast amounts of information. In this abstract, we introduce a Large Language Model (LLM)-based Natural Language Interface designed to interact with structured information stored in databases. Our experiments demonstrate the feasibility and effectiveness of the proposed framework. This framework can generalize to query a wide range of pharmaceutical data and knowledge bases.
Abstract Background Delayed neutrophil apoptosis during sepsis may impact neutrophil organ accumulation and tissue immune homeostasis. Elucidating the mechanisms underlying neutrophil apoptosis may help identify potential therapeutic targets. Glycolysis is critical to neutrophil activities during sepsis. However, the precise mechanisms through which glycolysis regulates neutrophil physiology remain under-explored, especially those involving the non-metabolic functions of glycolytic enzymes. In the present study, the impact of programmed death ligand-1 (PD-L1) on neutrophil apoptosis was explored. The regulatory effect of the glycolytic enzyme, pyruvate kinase M2 (PKM2), whose role in septic neutrophils remains unaddressed, on neutrophil PD-L1 expression was also explored. Methods Peripheral blood neutrophils were isolated from patients with sepsis and healthy controls. PD-L1 and PKM2 levels were determined by flow cytometry and Western blotting, respectively. Dimethyl sulfoxide (DMSO)-differentiated HL-60 cells were stimulated with lipopolysaccharide (LPS) as an in vitro simulation of septic neutrophils. Cell apoptosis was assessed by annexin V/propidium iodide (annexin V/PI) staining, as well as determination of protein levels of cleaved caspase-3 and myeloid cell leukemia-1 (Mcl-1) by Western blotting. An in vivo model of sepsis was constructed by intraperitoneal injection of LPS (5 mg/kg) for 16 h. Pulmonary and hepatic neutrophil infiltration was assessed by flow cytometry or immunohistochemistry. Results PD-L1 level was elevated on neutrophils under septic conditions. Administration of neutralizing antibodies against PD-L1 partially reversed the inhibitory effect of LPS on neutrophil apoptosis. Neutrophil infiltration into the lung and liver was also reduced in PD-L1−/− mice 16 h after sepsis induction. PKM2 was upregulated in septic neutrophils and promoted neutrophil PD-L1 expression both in vitro and in vivo. In addition, PKM2 nuclear translocation was increased after LPS stimulation, which promoted PD-L1 expression by directly interacting with and activating signal transducer and activator of transcription 1 (STAT1). Inhibition of PKM2 activity or STAT1 activation also led to increased neutrophil apoptosis. Conclusion In this study, a PKM2/STAT1-mediated upregulation of PD-L1 on neutrophils and the anti-apoptotic effect of upregulated PD-L1 on neutrophils during sepsis were identified, which may result in increased pulmonary and hepatic neutrophil accumulation. These findings suggest that PKM2 and PD-L1 could serve as potential therapeutic targets.