Search Results (228)

Artificial intelligence has advanced from merely predicting static protein structures to modeling equilibrium conformational ensembles. It now concurrently forecasts structure and binding affinity and actively participates in candidate selection during the initial stages of drug discovery. Foundation models introduced between 2024 and 2026, including BioEmu, AlphaFlow, DiG, Boltz-2, Chai-1, NeuralPLexer, and explicit-solvent prediction systems such as SuperWater, have begun to address issues previously identified as fundamental concerns in earlier critiques of AI in drug discovery. Nevertheless, many of these models are presently accessible only as preprints and require validation through independent peer review. Evidence indicates a shift in the primary bottleneck from representation challenges to validation difficulties. However, this transition remains incomplete and heavily dependent on context. The risks associated with AI-enabled drug discovery are increasingly not solely about the models’ capacity to accurately represent ensembles, but also about whether the evidentiary standards used to validate AI-derived predictions keep pace with the rapidity with which these predictions are generated and employed. This article introduces a four-tier validation framework designed to align the extent of computational and experimental evidence with the translational and regulatory risks associated with various artificial intelligence (AI) applications within the molecular sciences. These applications include machine learning (ML) models that analyze sequences, structures, conformational ensembles, protein–ligand complexes, and molecular dynamics trajectories. Tier 1 addresses the internal reproducibility of ML inference when applied to molecular inputs; Tier 2 pertains to the robustness of molecular-science benchmarks such as CASP, CASF-2016, PoseBusters, and OpenFE; Tier 3 involves prospective experimental validation against biophysical and biochemical measurements; and Tier 4 encompasses clinical and translational calibration within physiologically based pharmacokinetic (PBPK) and quantitative systems pharmacology (QSP) frameworks. This validation hierarchy functions as an explicit conceptual guide, serving as a framework rather than a regulatory requirement. It is firmly grounded in established principles derived from ICH Q8/Q9/Q10, the FDA model-informed drug development (MIDD) approach, the EMA reflection paper on AI in the medicinal product lifecycle, and the EU AI Act. The manuscript further incorporates recent evidence from ensemble-aware AI, prospective docking, free-energy campaigns, and clinical-stage AI-derived candidates. It concludes with specific recommendations pertaining to lifecycle governance, uncertainty reporting, and the adoption of harmonized evidentiary templates for AI/ML applications in the molecular sciences. Full article

Neurodegenerative diseases such as Alzheimer’s disease (AD) and Parkinson’s disease (PD) represent some of the most complex and therapeutically challenging disorders in modern medicine. Despite decades of research, the traditional one drug–one target paradigm has largely failed to deliver disease-modifying therapies. Increasing evidence suggests that these complex diseases arise from interconnected pathological networks involving protein aggregation, oxidative stress, mitochondrial dysfunction, neuroinflammation, and synaptic loss. In this context, natural products (NPs) have re-emerged as a promising pipeline for next-generation therapeutics. Unlike conventional small molecules, NPs inherently exhibit polypharmacology, targeting multiple pathways simultaneously. Recent advances (2019–2026) demonstrate a paradigm shift, from crude NPs and single-mechanism compounds toward engineered derivatives, network pharmacology, and multi-target drug design. Using AD and PD as case studies, this review critically evaluates how NPs are redefining drug discovery by highlighting key emerging NPs, translational strategies, and future directions. Full article

Bioequivalence (BE) studies are essential to file an abbreviated new drug application (ANDA) against an innovator drug product. Conventional BE studies can be complex, time-consuming, and operationally challenging, particularly for products with long half-life drugs, high variability, or formulation complexity. Advances in quantitative modeling and simulation have expanded the role of model-generated information in generic drug development from a supportive role toward providing critical regulatory evidence. Model-Integrated Bioequivalence (MIBE) represents a focused application of this paradigm in which mechanistic or empirical models are used to directly support BE determination. While physiologically based pharmacokinetic (PBPK) and physiologically based biopharmaceutics modeling (PBBM) approaches have been widely discussed in the literature, increasing attention is being directed toward population pharmacokinetic (POP-PK) modeling for MIBE implementation, particularly when mechanistic assumptions are uncertain or extensive in vitro characterization is impractical. This review provides a contemporary overview of MIBE in generic drug development, with a specific emphasis on POP-PK-based approaches. Key quantitative modeling frameworks are discussed along with evolving regulatory perspectives that support the integration of model-based evidence for BE assessment. We illustrate six diverse hypothetical case examples covering different formulations, a variety of BE scenarios and using MIBE to answer specific regulatory questions on BE. Collectively, this manuscript addresses an important topic of MIBE for complex and non-complex generic formulations and may provoke thinking among the generic companies to use such approaches in the regulatory context to enable faster and timely approval to bring the necessary medicines to the market at a rapid pace. Full article

Advanced oxidation processes based on photo-Fenton chemistry have gained increasing attention as effective treatment alternatives for the removal of pharmaceutical contaminants from water and wastewater systems. However, large-scale implementation remains constrained by operational requirements, limited mineralization efficiency, and challenges associated with process stability and selectivity. This review provides a critical assessment of recent advances (2022–2025) in conventional photo-Fenton and hybrid systems, including photocatalysis/photo-Fenton and sono-photo-Fenton processes, with emphasis on their performance in water and wastewater treatment applications. The removal of non-steroidal anti-inflammatory drugs, antibiotics, pharmaceutical mixtures, and real wastewater matrices is analyzed considering catalyst configuration, irradiation sources, oxidant utilization, and operating conditions relevant to practical treatment scenarios. Conventional homogeneous Fe²⁺/H₂O₂ systems enable rapid contaminant degradation but typically require acidic conditions and show limited mineralization efficiency. In contrast, iron-complexed and heterogeneous catalysts allow operation under near-neutral pH and visible-light irradiation, improving applicability in realistic water treatment systems. Hybrid photocatalysis/photo-Fenton processes enhance treatment efficiency through synergistic generation of reactive oxygen species, while ultrasound-assisted systems further intensify oxidation rates and contaminant removal. Special attention is given to oxidation mechanisms, catalyst stability, transformation products, and toxicity evolution to identify the key factors controlling treatment performance. Finally, current technological limitations, operational challenges, and design considerations for process integration, scale-up, and sustainable implementation in water and wastewater treatment are discussed. Full article

Natural product research has experienced substantial growth over the past two decades, driven by a renewed appreciation for the structural complexity and biological relevance of compounds derived from nature. Technological advances in separation science, spectroscopic characterization, and high-sensitivity bioassays have collectively restored natural products to a position of prominence in modern drug discovery efforts. Nature remains the most prolific source of bioactive molecular diversity, drawing from microorganisms, plants, and marine life to offer a vast reservoir of structurally novel scaffolds whose pharmacological potential remains largely unexplored. Effective extraction and isolation remain foundational to natural product research, as the quality and purity of isolated compounds directly govern the reliability of downstream biological evaluation. Recent years have witnessed remarkable innovation in this space, spanning green and designer solvent systems, pressurized and ultrasound-assisted extraction platforms, supercritical fluid techniques, and integrated purification workflows that dramatically reduce processing time while improving compound recovery and analytical throughput. Particularly noteworthy is the growing application of artificial intelligence and machine learning tools for solvent selection, extraction optimization, and metabolite dereplication, which in combination with advanced phase-separation strategies and informatic platforms have substantially expanded the scope of detectable and characterizable metabolites within complex biological matrices. This review summarizes recent progress in extraction and isolation methodologies supporting natural product research, with particular emphasis on combinatorial extraction strategies, next-generation solvent systems, and AI-driven applications that have collectively improved operational efficiency, selectivity, and analytical output over the past five years. Full article

Background/Objectives: Biorelevant in vitro dissolution testing is used increasingly to predict complex mechanisms in the gastrointestinal (GI) tract that determine oral bioavailability. However, the limited use of non-compendial systems is driven by the lack of widely accepted, standardized validation frameworks. This ongoing gap continues to restrict their adoption relative to United States Pharmacopeia (USP) apparatus. While the physiological relevance and biopredictive capabilities of the tiny-TIM and TIM-1 in vitro GI models have been demonstrated in previous studies, their inter-laboratory reproducibility has not been systematically established. Therefore, this international ring study evaluates the reproducibility of in vitro simulations of GI transit and absorption of paracetamol in fasted- and fed-state conditions in tiny-TIM and TIM-1. Methods: Three laboratories used TIM-1 and five used tiny-TIM to simulate oral administration of a 500 mg paracetamol solution to a healthy adult. Paracetamol solution was selected as a well-characterized and widely available BCS I compound to minimize formulation and solubility effects and focus on system performance, enabling the generation of a generic validation dataset for the reproducibility of TIM experiments. Results: Paracetamol bioaccessibility profiles were repeatable and reproducible (all pairwise f₂ > 50). Maximum differences in total bioaccessible paracetamol were 0.9% (TIM-1) and 2.8% (tiny-TIM) within laboratories and 3.4 and 5.9% between laboratories. Inter-lab variability at individual time points remained <4.0% (fasted) and 5.2% (fed). Both TIM models produced biopredictive metrics, correctly predicting no food effect on total paracetamol bioaccessibility and capturing delayed t_max. Gastric and intestinal environments showed repeatable pH, temperature, and GI transit characteristics, with fluctuations across transit stages that mirrored reported in vivo patterns. Conclusions: These results demonstrate that TIM systems can reproducibly simulate gastrointestinal conditions across laboratories and generate consistent measurements of drug product performance, despite the complexity of the dynamic processes involved. While this evaluation involving a single BCS I drug solution should not be directly extrapolated to experiments with poorly soluble compounds or different formulations, it supports the use of TIM systems as robust in vitro models in drug product development. This study provides a standardized, inter-laboratory, baseline performance dataset to support regulatory submissions incorporating TIM data and enable more confident interpretation of TIM experiments. Full article

Wound healing is a dynamic and multifaceted biological process involving hemostasis, inflammation, proliferation, and tissue remodeling. Topical therapy is widely preferred for wound management due to its localized action and reduced systemic adverse effects. However, the effective delivery of therapeutic agents is often limited by the skin’s barrier properties, the complex wound microenvironment, and the physicochemical characteristics of drugs. This review highlights the key physicochemical parameters governing topical drug delivery in wound therapy, including drug solubility, molecular size, lipophilicity, vesicle size distribution, surface charge, encapsulation efficiency, lipid composition, ethanol concentration, and vesicle deformability, which collectively influence drug permeation and retention at the wound site. Nanovesicular delivery systems have emerged as promising strategies to overcome these limitations. In particular, ultradeformable vesicles such as ethosomes, transferosomes, and transethosomes have demonstrated enhanced skin permeation and improved drug deposition in periwound tissue due to their flexible membrane structure and optimized physicochemical properties. This review systematically discusses the composition, preparation techniques, and critical formulation parameters of these vesicular systems that determine their stability, elasticity, and permeation performance. Furthermore, their applications in delivering anti-inflammatory drugs, antimicrobial agents, bioactive phytochemicals, and regenerative therapeutics for different wound types are examined. Widely used in vitro, ex vivo, and in vivo evaluation methods, including permeation studies and wound healing models such as excision, burn, infected, and diabetic wounds, are also summarized. Finally, the review outlines current challenges related to formulation standardization, physicochemical characterization, safety assessment, and large-scale production, while highlighting the future potential of ultradeformable vesicles as next-generation nanocarriers for advanced wound healing therapies. Full article

Background/Objectives: The mycobacterial complex cell envelope serves as a formidable barrier against host immunity and antibiotics. Lipomannan (LM) and lipoarabinomannan (LAM) are key structural components of the mycobacterial envelope and potent immunomodulators. The mycobacterial lipoarabinomannan biosynthesis mannosyltransferase MptC modifies the multiple α-(1→2)-linked branched mannan residues of LAM in the mycobacteria. However, the role of MptC in mycobacterial infectivity, antibiotic susceptibility and host immune regulation remains poorly understood. Methods: An mptC (also named MSMEG_4247) knockout Mycobacterium smegmatis mc²-155 (M. smeg) strain (designated as M. smegΔmptC) was generated using CRISPR–Cas12a technology. The effects of MptC on bacterial physiology, cell wall permeability, drug sensitivity, immune cell function, and survival during infection are analyzed through glycogen staining, drug sensitivity tests, and cellular and mouse infection models. Results: MptC deficiency results in a loss of LM and increase in LAM synthesis. The M. smegΔmptC mutant strain exhibits enhanced cell wall permeability and reduces hydrophobicity. Functionally, the mptC knockout strain increases the intracellular cytokines (IFN-γ, TNF-a and IL-17) production of T cells in mice. Consequently, results based on both macrophage and mouse infection models demonstrate that the M. smegΔmptC strain has less bacterial loads and higher susceptibility to antibiotic rifampicin. Conclusions: Mannosyltransferase MptC plays an important role in maintaining cell wall integrity (via LM/LAM synthesis), regulating T cell responses, and influencing antibiotic susceptibility in mycobacteria. Full article

Marine algae, microalgae, and Cyanophyceae emerge as sustainable and versatile sources of biomacromolecules for the fabrication of hydrogels with broad biomedical potential. Their phycocolloids, such as alginate, agar, carrageenan, ulvan, and extracellular polysaccharides (EPS), exhibit intrinsic biocompatibility, tunable gelation behavior, and bioactive sulfated structures that support cell viability, tissue regeneration, and therapeutic delivery. This review provides a comprehensive overview of hydrogel fabrication strategies, including physical, chemical, and hybrid crosslinking approaches, and highlights recent advances in composite systems incorporating proteins, glycosaminoglycans, and functional nanomaterials. Applications in skin repair, cartilage and bone regeneration, neural and cardiovascular engineering, and controlled drug delivery are examined, alongside the expanding role of marine-derived hydrogels as bioinks for 3D and 4D bioprinting. Despite their promise, challenges remain related to extract variability, purification complexity, mechanical limitations, and the need for standardized characterization. Future perspectives emphasize genetic engineering of algae and cyanobacteria, development of multifunctional hybrid hydrogels, sustainable large-scale production, and pathways toward clinical translation. Together, these insights position marine-derived hydrogels as next-generation biomaterials with significant potential for regenerative medicine and therapeutic innovation. Full article

Lipid-based delivery systems (LDS), including lipid nanoparticles (LNPs) and liposomes, have become indispensable tools in modern biomedicine owing to their biocompatibility, capacity to encapsulate diverse therapeutic agents, and potential for targeted delivery. Despite their clinical success, conventional batch-based manufacturing methods are hindered by variability, limited scalability, and complex processing steps, slowing their broader translation. Microfluidic technologies offer a transformative solution by enabling precise fluid handling, rapid mixing, and reproducible production of LDS with tunable physicochemical attributes such as particle size, lamellarity, and drug-loading efficiency. This review highlights advances in microfluidic design strategies, including hydrodynamic flow focusing, staggered herringbone mixers, and toroidal micromixers, and evaluates how critical parameters such as flow rate, solvent composition, and lipid concentration influence LDS performance. Furthermore, we discuss the application of microfluidics in drug delivery, nucleic acid therapeutics, and vaccine platforms, underscoring its role in improving scalability, quality control, and clinical translation. Finally, we examine current challenges, including throughput limitations and solvent handling, while outlining future directions for integrating emerging materials and additive manufacturing to optimize LDS fabrication. Collectively, microfluidic platforms provide a promising pathway for next-generation lipid nanomedicines with enhanced precision, reproducibility, and therapeutic efficacy. Full article

Accessing and understanding medication information can be challenging for many people, especially when patient information leaflets (PILs) are long, complex, and printed in small font. This study presents MediChat, an open-source, locally executable chatbot designed to provide reliable, easy-to-read answers to medication-related questions based exclusively on official PILs. MediChat follows a retrieval-augmented generation (RAG) architecture: PILs from the Austrian Medicinal Product Index are received via API, converted to text, split into overlapping chunks, embedded, and stored in a Chroma vector database. From there the top-k relevant chunks are retrieved, and Llama 3.1 generates German responses based on this evidence. The system was evaluated using a hybrid framework. Quantitatively, 200 yes/no questions across ten drugs were answered with 80% accuracy, overall precision 0.977, recall 0.686, F1-score 0.806, and a mean response time of 727 ms. Qualitatively, two personas were used in eight simulated dialogues. Response times were around 1.1–1.3 s, and task completion exceeded 85% with high ratings for relevance and quantity. These results indicate that an open-source RAG chatbot can deliver leaflet-grounded, user-friendly medication information and provide a reproducible template for future healthcare chatbot evaluations. Full article

Background: Type 2 diabetes mellitus (T2DM) is commonly managed with complex pharmacotherapy combined with dietary modification, which increases the risk of clinically relevant drug–food interactions (DFIs). Despite their potential impact on treatment efficacy and safety, patient knowledge of DFIs—particularly in the context of modern therapies such as glucagon-like peptide-1 receptor agonists (GLP-1 RAs)—remains insufficiently explored. Methods: This cross-sectional study assessed knowledge of DFIs among 103 adults with T2DM using a self-administered, expert-validated questionnaire. Data on sociodemographic characteristics, clinical variables, anti-diabetic therapy (including GLP-1 RAs), sources of education, and attendance at dietary consultations were collected. Knowledge scores were calculated based on correct responses and categorized into tertiles (low, moderate, high). Associations were analyzed using non-parametric tests. Multivariable logistic regression was performed to identify independent predictors of moderate-to-high DFI knowledge. Results: Substantial gaps in DFI knowledge were identified, particularly regarding interactions involving dietary fiber, dairy products, grapefruit juice, and nutrient deficiencies associated with long-term pharmacotherapy. Knowledge level was not significantly associated with age, educational attainment, diabetes duration, or GLP-1 RA use. Female sex was associated with higher knowledge in univariate analysis (p = 0.026); however, this association did not remain significant in the multivariable regression model. Attendance at at least one dietary consultation in the previous year was significantly associated with higher knowledge levels (p = 0.041) and remained an independent predictor in multivariable analysis (OR = 2.31; 95% CI: 1.04–5.15; p = 0.039). Most participants reported not receiving prior education on DFIs, while expressing a strong need for more frequent counseling. Conclusions: Patients with T2DM demonstrate insufficient knowledge of clinically relevant DFIs, including selected issues related to GLP-1 RA therapy. Attendance at structured dietary consultations was independently associated with higher levels of DFI knowledge; however, the directionality and causality of this relationship cannot be established. Given the cross-sectional design and the assessment of knowledge rather than behavioral or clinical outcomes, these findings should be interpreted as hypothesis-generating. Further longitudinal and interventional studies are required to determine whether improved DFI knowledge translates into meaningful changes in dietary behavior, treatment adherence, or metabolic outcomes. Full article

The SARS-CoV-2 papain-like protease (PL^pro) is an essential viral enzyme that promotes viral polyprotein processing while simultaneously suppressing the host innate immune response, which makes it a primary target for developing antiviral drugs. The present study employs a comprehensive approach integrating untargeted metabolomic profiling, in silico molecular docking and dynamics simulations, Molecular Mechanics Generalized Born Surface Area (MM-GBSA) energetic assessments, and biochemical enzyme assays. This integrated method aims to discover natural PL^pro inhibitors from two ethnomedicinal plants, Lippia javanica and Acorus calamus, which have long been utilized in African traditional medicine to treat respiratory diseases. Comprehensive metabolite profiling using untargeted Ultra-Performance Liquid Chromatography–Tandem Mass Spectrometry (UPLC-MS/MS) and Global Natural Products Social (GNPS) molecular networking revealed flavonoid glucuronides and phenylpropanoid derivatives as the major constituents in both plant species. In situ histochemical staining further offered spatial validation of phenolic- and lignin-associated tissues, supporting the phenolic-dominated molecular families detected by GNPS molecular networking. In silico evaluation of six selected compounds demonstrated spontaneous and thermodynamically favorable binding to PL^pro, with ΔG_bind values ranging from −5.63 to −6.43 kcal/mol. Catechin-7-glucoside emerged as the lead compound, establishing multiple hydrogen bond networks with Asp164, Gln269, Tyr264, and Asn267, supplemented by hydrophobic engagement with Pro247 and Pro248, and π-π stacking with the blocking loop 2 (BL2 loop). Molecular dynamics simulations confirmed the stability of the protein–ligand complexes. Biochemical enzyme assays confirmed concentration-dependent inhibition of PL^pro proteolytic and deubiquitinating activity by both crude plant extracts and isolated bioactive compounds. However, S-adenosyl-methionine showed comparatively high PL^pro proteolytic activity (IC₅₀ 5.872 µM) compared to catechin-7-glucoside, with an IC₅₀ of 7.493 µM, exhibiting efficacy similar to the reference inhibitor GRL0617. Both the extracts of L. javanica and A. calamus have shown significant inhibitory activity while maintaining cell viability in Human embryonic kidney 293T cell (HEK293T) culture models, indicating a favorable safety profile of the tested concentrations. Based on these results, catechin-based polyphenols and phenylpropanoid derivatives appear as promising lead compounds for the development of PL^pro inhibitors. To progress toward therapeutic use, further work is necessary in pharmacokinetics, structural optimization, and antiviral validation in cell models. Full article

Background: Stimuli-responsive hydrogels (SRHs) are smart polymeric materials that undergo reversible physicochemical changes in response to abiotic cues and externally applied fields, enabling applications in drug delivery, wound healing, and tissue engineering. However, they exhibit limited biological specificity and do not adequately reflect the dynamic, disease-relevant complexity of native tissue microenvironments. Microbe-colonized tissues display distinctive biochemical features driven, shaped by microbial metabolism, including localized pH gradients, short-chain fatty acid production, secretion of quorum-sensing molecules, biofilm formation, and expression of specialized enzymes. These endogenous, spatiotemporally regulated signals are closely linked to host physiology and pathology but remain underutilized in hydrogel design. This review aims to highlight microbiome-responsive hydrogels (MRHs) as a strategy to address this gap. Methods: This study summarizes current engineering approaches, key microbial stimuli, and emerging biomedical applications of MRHs, with emphasis on translational and regulatory challenges. Results: Microbiome-responsive hydrogels (MRHs) address this gap by leveraging microbial metabolic and biochemical cues to induce swelling, degradation, drug release, antibacterial activity, or structural transformation. By directly coupling to microbe-derived stimuli, MRHs offer improved physiological relevance, enhanced local specificity, and new opportunities for precision therapy targeting disease-associated microbial niches. Conclusions: Despite their promise, MRHs remain an early and fragmented field, lacking standardized biological triggers, material design frameworks, and performance evaluation strategies. This review summarizes current engineering approaches, key microbial stimuli, and emerging biomedical applications, with emphasis on translational and regulatory challenges, positioning MRHs as an underexplored platform for next-generation smart biomaterials. Full article

The induction of chirality to obtain enantiopure products of high synthetic value is of great importance across various scientific fields, particularly in the medical area, as it has been demonstrated that the different enantiomers of drugs interact differently with biological receptors. In this context, asymmetric catalysis focuses on the design of catalysts that are easy to synthesize, capable of efficiently and enantioselectively forming C–C bonds, and suitable for reuse in multiple catalytic processes. This work describes the application of a Pd(II) complex coordinated with the R and S forms of proline in direct Aldol, Morita–Baylis–Hillman, and Heck coupling reactions. The catalytic system efficiently promoted the aldol reaction, achieving yields of 80–95%, excellent diastereoselectivities (1:69 syn/anti), and enantiomeric excesses greater than 99%. From a mechanistic perspective, the formation of a transition state is proposed in which a proline molecule generates an enamine that, upon coordination with the metal center, is stabilized through interaction with the intermediate’s double bond. Moreover, the study of the Morita–Baylis–Hillman and Heck coupling reactions highlights the versatility of this type of catalyst. Full article