Int J Mol Sci. 2025 Jul 12;26(14):6711. doi: 10.3390/ijms26146711.

ABSTRACT

Parkinson's disease (PD) is the second most common neurodegenerative disease. It primarily affects the motor system but is also associated with a range of cognitive impairments that can manifest early in disease progression, indicating its multifaceted nature. In this paper, we performed a meta-analysis of transcriptomics and proteomics data using MultiOmicsIntegrator to gain insights into the post-transcriptional modifications and deregulated pathways associated with this disease. Our results reveal differential isoform usage between control and PD patient brain samples that result in enriched alternative splicing events, including an extended UTR length, domain loss, and the upregulation of non-coding isoforms. We found that Inositol-Requiring Enzyme 1 (IRE1) is active in PD samples and examined the role of its downstream signaling through X-box binding mRNA 1 (XBP1) and regulated IRE1-dependent decay (RIDD). We identified several RIDD candidates and showed that the enriched alternative splicing events observed are associated with RIDD. Moreover, in vitro mRNA cleavage assays demonstrated that OSBPL3, C16orf74, and SLC6A1 mRNAs are targets of IRE1 RNAse activity. Finally, a pathway enrichment analysis of both XBP1s and RIDD targets in the PD samples uncovered associations with processes such as immune response, oxidative stress, signal transduction, and cell-cell communication that have previously been linked to PD. These findings highlight a potential regulatory role of IRE in PD.

PMID:40724959 | DOI:10.3390/ijms26146711

Int J Mol Sci. 2025 Jul 12;26(14):6698. doi: 10.3390/ijms26146698.

ABSTRACT

Memory impairment, ranging from mild memory impairment to neurodegenerative diseases such as Alzheimer's disease, poses an escalating global health challenge that necessitates multi-targeted interventions to prevent progression. Health functional foods (HFFs), which include bioactive dietary compounds that not only provide basic nutrition but also function beyond that to modulate physiological pathways, offer a promising non-pharmacological strategy to preserve memory function. This review presents an integrative framework for the discovery, evaluation, and clinical translation of biomarkers responsive to HFFs in the context of preventing memory impairment. We examine both established clinical biomarkers, such as amyloid-β and tau in the cerebrospinal fluid, neuroimaging indicators, and memory assessments, as well as emerging nutritionally sensitive markers including cytokines, microRNAs, gut microbiota signatures, epigenetic modifications, and neuroactive metabolites. By leveraging systems biology approaches, we explore how network pharmacology, gut-brain axis modulation, and multi-omics integration can help to elucidate the complex interactions between HFF components and memory-related pathways such as neuroinflammation, oxidative stress, synaptic plasticity, and metabolic regulation. The review also addresses the translational pipeline for HFFs, from formulation and standardization to regulatory frameworks and clinical development, with an emphasis on precision nutrition strategies and cross-disciplinary integration. Ultimately, we propose a paradigm shift in memory health interventions, positioning HFFs as scientifically validated compounds for personalized nutrition within a preventative memory function framework.

PMID:40724948 | DOI:10.3390/ijms26146698

Life (Basel). 2025 Jul 20;15(7):1143. doi: 10.3390/life15071143.

ABSTRACT

Investigations of endogenous plant circular RNAs (circRNAs) in several plant species have revealed changes in their circular RNA profiles in response to biotic and abiotic stresses. Recently, circRNAs have emerged as critical regulators of gene expression. The destructive impacts on agriculture due to plant viral infections necessitate better discernment of the involvement of plant circRNAs during viral infection. However, few such studies have been conducted hitherto. Sobemoviruses cause great economic impacts on important crops such as rice, turnip, alfalfa, and wheat. Our current study investigates the dynamics of plant circRNA profiles in the host Arabidopsis thaliana (A. thaliana) during infections with the sobemoviruses Turnip rosette virus (TRoV) and Rice yellow mottle virus (RYMV), as well as the small circular satellite RNA of the Lucerne transient streak virus (scLTSV), focusing on circRNA dysregulation in the host plants and its potential implications in triggering plant cellular defense responses. Towards this, two rounds of deep sequencing were conducted on the RNA samples obtained from infected and uninfected plants followed by the analysis of circular RNA profiles using RNA-seq and extensive bioinformatic analyses. We identified 760 circRNAs, predominantly encoded in exonic regions and enriched in the chloroplast chromosome, suggesting them as key sites for circRNA generation during viral stress. Gene ontology (GO) analysis indicated that these circRNAs are mostly associated with plant development and protein binding, potentially influencing the expression of their host genes. Furthermore, Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis showed photosynthesis as the most affected pathway. Interestingly, the non-coding exogenous scLTSV specifically induced several circRNAs, some of which contain open reading frames (ORFs) capable of encoding proteins. Our biochemical assays demonstrated that transgenic expression of scLTSV in A. thaliana enhanced resistance to TRoV, suggesting a novel strategy for improving plant viral resistance. Our results highlight the complexity of circRNA dynamics in plant-virus interactions and offer novel insights into potential circRNA-based strategies for enhancing plant disease resistance by modulating the differential expression of circRNAs.

PMID:40724645 | DOI:10.3390/life15071143

Life (Basel). 2025 Jun 25;15(7):1010. doi: 10.3390/life15071010.

ABSTRACT

Neuropeptides (NPs) are a diverse group of signaling molecules involved in regulating key physiological processes such as pain perception, stress response, mood, appetite, and circadian rhythms. Acting as neurotransmitters, neuromodulators, or neurohormones, they play a critical role in modulating and fine-tuning neural signaling networks. Despite their biological significance, identifying NPs through experimental techniques remains time-consuming and resource-intensive. To support this effort, computational prediction tools have emerged as a cost-effective approach for prioritizing candidate sequences for experimental validation. In this study, we propose EnsembleNPPred, an ensemble learning framework that integrates traditional machine learning (ML) models with a deep learning (DL) component. By combining the complementary strengths of these approaches, the model aims to improve generalization and predictive robustness. EnsembleNPPred employs a majority voting mechanism to aggregate the outputs from three classifiers: Support Vector Machine (SVM), Extra Trees (ET), and a CNN-based DL model. When evaluated on independent datasets, EnsembleNPPred demonstrated consistently competitive performance, achieving improvements in both accuracy and sensitivity-specificity balance compared to several existing methods. Furthermore, testing on multiple neuropeptide families from the NeuroPep database yielded an average accuracy of 91.92%, suggesting the model's potential to generalize across diverse peptide classes. These results suggest that EnsembleNPPred may be a useful tool for early-stage neuropeptide candidate identification and for supporting downstream experimental validation.

PMID:40724513 | DOI:10.3390/life15071010

Biomolecules. 2025 Jul 16;15(7):1030. doi: 10.3390/biom15071030.

ABSTRACT

In the original publication [...].

PMID:40723926 | DOI:10.3390/biom15071030

Biomolecules. 2025 Jul 8;15(7):980. doi: 10.3390/biom15070980.

ABSTRACT

The global antibiotic resistance crisis raises concerns about antibiotic use, and alternative strategies are urgently needed. In this context, host defense peptides (HDPs) have rapidly gained interest. However, one of the main obstacles is their production strategy. Chemical synthesis is the most widely used, although it is not scalable and has sequence limitations. A possible alternative is recombinant production, but the strategies used so far have limited efficiency. In this study, we aim to compare the activity and main characteristics of different HDPs produced by both chemical synthesis and by recombinant production, using an approach based on tetramers to ameliorate the production process. The results obtained showed that the production of HDPs as tetrameric peptides by recombinant production in Lactococcus lactis enhanced the peptide activity, with HDPs being much more active in terms of antimicrobial activity, more structurally stable, and nanostructured. Thus, the recombinant strategy described herein, fusing four repetitions of the same peptide, can become a real alternative to produce highly active HDPs through a scalable production process.

PMID:40723852 | DOI:10.3390/biom15070980

Biology (Basel). 2025 Jul 17;14(7):876. doi: 10.3390/biology14070876.

ABSTRACT

Solar lentigines, commonly caused by prolonged ultraviolet exposure, raise the risk of skin disorders and remain challenging to manage due to their complex mechanisms. Understanding the molecular mechanisms driving the progression of solar lentigines is crucial for developing effective protective strategies. In this study, we introduced a novel method, Dynamic Network Driver (DND), which identifies upstream regulators that drive disease progression by integrating the Dynamic Network Biomarker (DNB) approach with network control theory. By applying DND to multi-omics data from solar lentigines subjects, we (1) identified the key drivers associated with solar lentigo progression, with their functions involved in differentiation and dermal-epidermal junction; and (2) highlighted ARNT2 and TBX2 as significant master factors supported by in vitro validation in melanocytes and pigmented 3D living skin equivalent models. These results demonstrate the potency of DND for uncovering the molecular mechanisms behind solar lentigines and informing therapeutic strategies. In summary, the DND approach identified novel drivers of solar lentigo progression, acting as new markers for spot mitigation in 3D spot mimic models.

PMID:40723431 | DOI:10.3390/biology14070876

Biology (Basel). 2025 Jul 2;14(7):801. doi: 10.3390/biology14070801.

ABSTRACT

Climate change poses significant challenges to temperate rice production, particularly affecting grain quality and market acceptance. This review synthesizes current knowledge of climate-induced quality changes, with a focus on the Australian rice industry as a case study with comparisons to other temperate regions. Environmental stressors such as extreme temperatures, variable rainfall, elevated CO2, and salinity disrupt biochemical pathways during grain development, altering physicochemical, textural, and aromatic traits. Different rice classes exhibit distinct vulnerabilities: medium-grain japonica varieties show reduced amylose under heat stress, aromatic varieties experience disrupted aroma synthesis under drought, and long-grain types suffer kernel damage under combined stresses. Temperature is a key driver, with quality deterioration occurring above 35 °C and below 15 °C. Systems biology analyses reveal complex signalling networks underpinning these stress responses, although experimental validation remains limited. The Australian industry has responded by developing cold-tolerant cultivars, precision agriculture, and water-saving practices, yet projected climate variability demands more integrated strategies. Priorities include breeding for stress-resilient quality traits, refining water management, and deploying advanced phenotyping tools. Emerging technologies like hyperspectral imaging and machine learning offer promise for rapid quality assessment and adaptive management. Sustaining high-quality rice in temperate zones requires innovation linking physiology with practical adaptation.

PMID:40723360 | DOI:10.3390/biology14070801

Biology (Basel). 2025 Jun 28;14(7):785. doi: 10.3390/biology14070785.

ABSTRACT

The number of Fritillaria species native to the Alps has long been debated, and observational biases due to the short flowering periods and the scattered distributions of endemic Fritillaria populations along the mountain range have probably made the task of botanists more complicated. Moreover, previous phylogenetic studies in Fritillaria have considered alpine taxa only marginally. To test species boundaries within the F. tubaeformis species complex and to study their phylogenetic relationships, intra- and inter-specific genetic variability of sixteen samples belonging to four Fritillaria species was carried out in different localities of the Maritime and Ligurian Alps, with extensions to the rest of the Alpine arc. The combined use of five plastid DNA markers (matK, ndhF, rpl16, rpoC1, and petA-psbJ) and nrITS showed that F. tubaeformis and F. burnatii are phylogenetically independent taxa, fully confirming morphological and morphometric divergences and, that F. burnatii is not related phylogenetically to the central European F. meleagris. Our phylogenetic study also supports the separation of F. tubaeformis from F. moggridgei, pointing to environment/ecological constraints or reproductive barriers as possible causes of their distinct evolutionary status. Our analysis also showed that the mountain endemic F. involucrata is not closely related to F. tubaeformis, contrasting with previous studies. The phylogenetic analysis of the nrITS region supports a close relationship between F. burnatii and F. moggridgei, but with low statistical support.

PMID:40723345 | DOI:10.3390/biology14070785

Antioxidants (Basel). 2025 Jun 30;14(7):811. doi: 10.3390/antiox14070811.

ABSTRACT

Herpes simplex virus type 1 (HSV-1) is a widespread pathogen responsible for recurrent infections, primarily affecting the skin and mucous membranes. With the aim of targeting both the viral infection and the associated inflammatory response, biotechnologically produced Lavandula angustifolia Miller (L. angustifolia) extract, rich in rosmarinic acid, was incorporated into liposomal formulations intended for topical application. Lavender is known for its strong anti-inflammatory, antioxidant, wound-healing, and antiviral properties. However, its low stability under certain conditions limits its therapeutic potential. Four different formulations were developed: conventional liposomes, glycerosomes, hyalurosomes, and glycerohyalurosomes. The vesicles were characterized for size, stability, and entrapment efficiency. Glycerosomes were the smallest (~58 nm), while the other formulations ranged around 77 nm, all maintaining a highly negative surface charge, ensuring stability and reduced aggregation. Glycerol-containing formulations demonstrated superior stability over 12 months, while liposomes and hyalurosomes increased their size after only two months. Entrapment efficiency reached up to 100% for most vesicles, except for glycerohyalurosomes (~54%). In vitro studies on Normal Human Dermal Fibroblasts (NHDFs) demonstrated that all formulations were biocompatible and enhanced cell viability under oxidative stress. Glycerosomes, hyalurosomes, and glycerohyalurosomes exhibited significant anti-inflammatory activity by reducing MMP-1 and IL-6 levels in LPS-stimulated fibroblasts. Furthermore, these preliminary results highlighted promising antiviral activity against HSV-1 of the obtained formulations, particularly when applied during or post-infection. Overall, these phospholipid vesicles offer a dual therapeutic approach, combining antioxidant, anti-inflammatory, and antiviral effects, positioning them as promising candidates for the treatment of HSV-induced skin lesions and related inflammatory conditions.

PMID:40722915 | DOI:10.3390/antiox14070811

Antioxidants (Basel). 2025 Jun 27;14(7):799. doi: 10.3390/antiox14070799.

ABSTRACT

This study aimed to evaluate the microbiological quality and functional properties of Moringa oleifera Lam. leaves from plants cultivated in Sicily, with the objective of exploring their potential use in functional food production. Precision agriculture techniques, including unmanned aerial vehicle-based multispectral remote sensing, were used to determine the optimal harvesting time for M. oleifera. After harvesting, leaves were dried using a smart solar dryer system based on a wireless sensor network and milled with a laboratory centrifugal mill to produce powdered M. oleifera leaves (PMOLs). Plate counts showed no colonies of undesired microorganisms in PMOLs. The MiSeq Illumina analysis revealed that the class Alphaproteobacteria was dominant (83.20% of Relative Abundance) among bacterial groups found in PMOLs. The hydroalcoholic extract from PMOLs exhibited strong redox-active properties in solution assays and provided antioxidant protection in a cell-based lipid peroxidation model (CAA50: 5.42 μg/mL). Additionally, it showed antiproliferative activity against three human tumour epithelial cell lines (HepG2, Caco-2, and MCF-7), with GI50 values ranging from 121.03 to 237.75 μg/mL. The aromatic profile of PMOLs includes seven phytochemical groups: alcohols, aldehydes, ketones, esters, acids, terpenes, and hydrocarbons. The most representative compounds were terpenes (27.5%), ketones (25.3%), and alcohols (14.5%). Results suggest that PMOLs can serve as a natural additive for functional foods.

PMID:40722903 | DOI:10.3390/antiox14070799

Biomedicines. 2025 Jul 14;13(7):1717. doi: 10.3390/biomedicines13071717.

ABSTRACT

Background: Immune checkpoint inhibitors (ICIs) have transformed cancer treatment, yet severe immune-related adverse events (irAEs) often necessitate immunotherapy discontinuation and cause life-threatening complications. Circulating plasma proteins, dynamically accessible and functionally linked to immunity, may predict and offer novel targets for irAEs. Methods: Leveraging multi-omics integration, we conducted bidirectional two-sample Mendelian randomization (MR) using protein quantitative trait loci (pQTLs) from 4998 plasma proteins and genome-wide association data of irAE phenotypes. A causal inference framework combining colocalization analysis, multivariable MR (MVMR) adjusting for body mass index (BMI) confounding, and mediation MR elucidated BMI-independent pathways. Systems biology approaches including tissue-specific expression profiling, pathway enrichment, and protein interaction network analysis revealed spatial and functional drivers of irAE pathogenesis. Results: Proteome-wide MR mapping identified eight plasma proteins (CCL20, CSF1, CXCL9, CD40, TGFβ1, CLSTN2, TNFSF12, TGFα) causally associated with all-grade irAEs, and five (CCL20, CCL25, CXCL10, ADA, TGFα) with high-grade irAEs. Colocalization prioritized CD40/TNFSF12 (all-grade) and ADA/CCL25 (high-grade) as therapeutic targets (PPH4 > 0.7). CXCL9/TNFSF12 (all-grade) and CCL25 (high-grade) exerted BMI-independent effects, suggesting intrinsic immune dysregulation mechanisms. Tissue-specific gene expression patterns, CSF1, TGFβ1 in lung, TNFSF12 in the ileum may explain organ-specific irAE vulnerabilities. High-grade irAEs correlated with compartmentalized immune dysregulation and IL-17/immunodeficiency pathway activation. Conclusions: This study establishes the causal atlas of plasma proteins in irAE pathogenesis, bridging biomarker discovery with actionable therapeutic targets. These advances align with next-generation immunotherapy goals: maximizing efficacy while taming the immune storm.

PMID:40722787 | DOI:10.3390/biomedicines13071717

Diagnostics (Basel). 2025 Jul 17;15(14):1807. doi: 10.3390/diagnostics15141807.

ABSTRACT

Heart failure (HF) is a global health burden characterized by high morbidity and mortality, necessitating advancements in diagnostic and therapeutic approaches. Molecular diagnostics, encompassing genomics, transcriptomics, proteomics, metabolomics, and epigenetics, offer unprecedented insights into HF pathogenesis, aiding early diagnosis, risk stratification, and personalized management. This state-of-the-art review critically examines recent developments in molecular diagnostics in HF, evaluates their translational potential, and highlights key challenges in clinical implementation. Emerging tools such as liquid biopsy, multi-omics integration, and artificial intelligence (AI)-driven platforms are explored. We propose strategies to enhance clinical translation, equity in access, and utility in guiding treatment, thereby advancing precision cardiovascular medicine.

PMID:40722556 | DOI:10.3390/diagnostics15141807

Bioengineering (Basel). 2025 Jul 15;12(7):764. doi: 10.3390/bioengineering12070764.

ABSTRACT

Generative image models have revolutionized artificial intelligence by enabling the synthesis of high-quality, realistic images. These models utilize deep learning techniques to learn complex data distributions and generate novel images that closely resemble the training dataset. Recent advancements, particularly in diffusion models, have led to remarkable improvements in image fidelity, diversity, and controllability. In this work, we investigate the application of a conditional latent diffusion model in the healthcare domain. Specifically, we trained a latent diffusion model using unlabeled histopathology images. Initially, these images were embedded into a lower-dimensional latent space using a Vector Quantized Generative Adversarial Network (VQ-GAN). Subsequently, a diffusion process was applied within this latent space, and clustering was performed on the resulting latent features. The clustering results were then used as a conditioning mechanism for the diffusion model, enabling conditional image generation. Finally, we determined the optimal number of clusters using cluster validation metrics and assessed the quality of the synthetic images through quantitative methods. To enhance the interpretability of the synthetic image generation process, expert input was incorporated into the cluster assignments.

PMID:40722456 | DOI:10.3390/bioengineering12070764

Brain Sci. 2025 Jul 16;15(7):756. doi: 10.3390/brainsci15070756.

ABSTRACT

Parkinson's disease (PD) is a progressive neurodegenerative disorder marked by dopaminergic neuronal loss, α-synuclein aggregation, and chronic neuroinflammation. Recent evidence suggests that exosomal microRNAs (miRNAs)-small, non-coding RNAs encapsulated in extracellular vesicles-are key regulators of PD pathophysiology and promising candidates for biomarker development and therapeutic intervention. Exosomes facilitate intercellular communication, cross the blood-brain barrier, and protect miRNAs from degradation, rendering them suitable for non-invasive diagnostics and targeted delivery. Specific exosomal miRNAs modulate neuroinflammatory cascades, oxidative stress, and synaptic dysfunction, and their altered expression in cerebrospinal fluid and plasma correlates with disease onset, severity, and progression. Despite their translational promise, challenges persist, including methodological variability in exosome isolation, miRNA profiling, and delivery strategies. This review integrates findings from preclinical models, patient-derived samples, and systems biology to delineate the functional impact of exosomal miRNAs in PD. We propose mechanistic hypotheses linking miRNA dysregulation to molecular pathogenesis and present an interactome model highlighting therapeutic nodes. Advancing exosomal miRNA research may transform the clinical management of PD by enabling earlier diagnosis, molecular stratification, and the development of disease-modifying therapies.

PMID:40722346 | DOI:10.3390/brainsci15070756

BMC Microbiol. 2025 Jul 28;25(1):456. doi: 10.1186/s12866-025-04175-1.

ABSTRACT

BACKGROUND: The genus Enterobacter, in the family Enterobacteriaceae, is of both clinical and environmental importance. This genus has undergone frequent taxonomic changes, making it challenging to identify taxa even at genus level. This study aimed to design Enterobacter genus-specific primers that can be used for simple PCR identification of large sets of putative Enterobacter isolates.

RESULTS: Comparative genomic approaches were employed to identify genes that were universally present on Enterobacter genomes but absent from the genomes of other members of the family Enterobacteriaceae, based on an initial set of 89 genomes. The presence of these genes was further confirmed in 4,276 Enterobacter RefSeq genomes. While no strictly genus-specific genes were identified, the hpaB gene demonstrated a restricted distribution outside of the genus Enterobacter. Semi-nested primers were designed for hpaB and its flanking gene hpaC (hpaBC) and evaluated on 123 strains in single-tube PCR reactions. All taxa showing positive reactions belonged to the genus Enterobacter. For Enterobacter strains the PCR yielded two amplicons at 110 bp and at 370 bp, while strains only displaying the 110 bp amplicon were classified as Leclercia pneumoniae. A blind-test on 120 strains accessioned as Enterobacter sp. from the USDA-ARS culture collection (NRRL), revealed that one third of the strains had an incorrect genus assignment. Comparison of gene trees of the hpaBC fragment sequences with marker genes frequently used for single-gene barcoding or multi-locus sequence analysis (MLSA) further demonstrated its potential for preliminary species identification.

CONCLUSIONS: The nested PCR assay represents a rapid and cost-effective approach for preliminary identification of Enterobacter species. As the primer design was based on large-scale genomic comparison, including currently undescribed species clades, it will remain valid even after taxonomic changes within the genus.

PMID:40722002 | DOI:10.1186/s12866-025-04175-1

Mol Biol Rep. 2025 Jul 28;52(1):765. doi: 10.1007/s11033-025-10880-x.

ABSTRACT

BACKGROUND: Circadian rhythms are intrinsic 24-h biological cycles that regulate key physiological processes, including skin cell proliferation, DNA repair, and barrier homeostasis. Disruption of these rhythms accelerates skin aging, compromises barrier integrity, and increases susceptibility to oxidative stress. Protocatechuic acid (PCA) is a naturally occurring compound with antioxidant and anti-inflammatory properties; however, its role in regulating circadian rhythms has not been previously explored. Therefore, this study aimed to investigate the potential of PCA to regulate the circadian rhythm within keratinocytes and the broader effects of PCA on skin physiology.

METHODS AND RESULTS: This potential of PCA as a circadian rhythm modulator in human epidermal keratinocytes was investigated. PCA enhanced circadian activity in a dose-dependent manner, as evidenced by increased amplitude of basic helix-loop-helix ARNT like 1 (BMAL1)-driven bioluminescence. In silico docking revealed strong binding affinity of PCA to retinoic acid-related orphan receptor alpha (RORα), a core clock regulator, suggesting a molecular mechanism of action. PCA also modulated core clock gene expression. Under oxidative stress conditions, PCA reduced reactive oxygen species (ROS) levels and upregulated antioxidant enzymes, including catalase and superoxide dismutase 1. Additionally, PCA promoted skin barrier integrity by increasing structural protein and ceramide-related gene expression and enhanced cellular longevity markers, such as cyclin-dependent kinase inhibitor 1B (CDKN1B) and telomerase reverse transcriptase (TERT).

CONCLUSIONS: These findings demonstrate that PCA functions as a multifunctional agent that modulates circadian rhythms, reduces oxidative stress, and supports skin barrier homeostasis and cellular longevity. Overall, PCA shows strong potential as a therapeutic candidate for treating skin disorders associated with circadian disruption and oxidative damage.

PMID:40721875 | DOI:10.1007/s11033-025-10880-x

Nat Methods. 2025 Jul 28. doi: 10.1038/s41592-025-02723-1. Online ahead of print.

ABSTRACT

Protein language models embed protein sequences for different tasks. However, these are suboptimal at learning the language of protein interactions. We developed an interaction language model (iLM), Sliding Window Interaction Grammar (SWING) that leverages differences in amino-acid properties to generate an interaction vocabulary. SWING successfully predicted both class I and class II peptide-major histocompatibility complex interactions. Furthermore, the class I SWING model could uniquely cross-predict class II interactions, a complex prediction task not attempted by existing methods. Using human class I and II data, SWING accurately predicted murine class II peptide-major histocompatibility interactions involving risk alleles in systemic lupus erythematosus and type 1 diabetes. SWING accurately predicted how variants can disrupt specific protein-protein interactions, based on sequence information alone. SWING outperformed passive uses of protein language model embeddings, demonstrating the value of the unique iLM architecture. Overall, SWING is a generalizable zero-shot iLM that learns the language of protein-protein interactions.

PMID:40721872 | DOI:10.1038/s41592-025-02723-1

Cell Death Dis. 2025 Jul 28;16(1):567. doi: 10.1038/s41419-025-07823-6.

ABSTRACT

Epigenetic pathways, including DNA methylation, non-coding RNAs, histone modification, and chromatin remodeling, control spatiotemporal gene expression and tightly coordinate the activities during embryogenesis. Emerging evidence indicates that epigenetic regulators are critically required for the maintenance of normal lung development and that the epigenetic marks are altered in lung cells during disease progression. In this review, we focus on the recent studies that have yielded insights into how the levels and patterns of epigenetic regulators are changed, and how these regulators contribute to the regulation of various stages during lung development. A deeper understanding of these epigenetic mechanisms could offer novel therapeutic targets for preventing fetal lung diseases.

PMID:40721598 | DOI:10.1038/s41419-025-07823-6

Nat Commun. 2025 Jul 28;16(1):6923. doi: 10.1038/s41467-025-61897-6.

ABSTRACT

Fumarate hydratase (FH), a key node of mitochondrial metabolism, is also a tumour suppressor. Despite its prominent roles in tumourigenesis and inflammation, its regulation remains poorly understood. Herein, we show that histone deacetylase 6 (HDAC6) regulates FH activity. In triple-negative breast cancer cells, HDAC6 inhibition or knockdown results in alterations to mitochondrial cristae structure, as detected by live-cell super-resolution STED nanoscopy and electron microscopy, along with the release of mitochondrial DNA. Mass-spectrometry immunoprecipitation reveals multiple mitochondrial HDAC6-interactors, with FH emerging as a top hit. Super-resolution 3D-STORM shows HDAC6 interactions with FH in mitochondrial networks, which increases after perturbation of HDAC6 activity with BAS-2. Treatment with BAS-2 leads to fumarate accumulation by 13C glucose labelling, along with downstream succination of proteins and cell death. Together, these results identify HDAC6 inhibition as a regulator of endogenous FH activity in tumour cells, and highlight it as a promising candidate for indirectly targeting tumour metabolism.

PMID:40721560 | DOI:10.1038/s41467-025-61897-6