BMC Biology最新文献

Generative artificial intelligence is reshaping how researchers discover protein-binding compounds and develop them into drug candidates. Unlike traditional methods that screen existing molecules, structure-based generative AI designs novel compounds tailored to a protein's three-dimensional binding pocket. This review outlines how these approaches are applied in early drug discovery, focusing on general principles. We categorize methods according to their generative modeling paradigms and their strategies for using structural data to guide molecular design, distinguishing de novo incremental builders from models that generate full structures. We also survey lead-optimization techniques, highlighting a recent shift toward generation-driven medicinal chemistry.

Background: Signal divergence and sensory preferences may cause reproductive isolation and eventually speciation between animal populations. These patterns are particularly well known in systems with acoustic or visual signals. However, relatively few studies have quantified the patterns, causes and consequences of geographical variation of chemical signals, especially in mammalian populations.

Results: We examined Hipposideros armiger (the great Himalayan leaf-nosed bat) and collected secretions from the forehead glands of males. We selected nine colonies across a large geographic area and investigated both the potential causes and patterns of variation in chemical signals from that gland between these colonies. We also examined whether or not males could tell the difference between gland scents from a foreign colony and their own colony using habituation-discrimination tests. To determine whether females preferred the scent of local versus foreign males, we performed two-choice tests. Overall, we found significant colony-level differences in the concentrations of compounds and in general compound categories. We show that symbiotic microbes inhabiting odor glands may explain the observed chemical variation between colonies. Moreover, behavioral studies indicated that males were able to discriminate between gland scents from their own colony and those from a distant colony. Finally, females were more attracted to scents of local males from their own colony compared to scents from foreign colonies.

Conclusions: Our results demonstrate significant geographical variation in colony chemical signals linked to gland symbiotic microbes, with males discriminating colony-specific scents and females preferring the gland scents of local males. This study expands our limited knowledge of geographical variation of mammalian chemical signals and highlights the importance of bacterial communities in shaping chemical divergence.

Background: Vertebrate immune systems exhibit striking evolutionary diversity, yet our understanding remains biased toward mammalian models. Here, we generate a single-cell transcriptome of immune cells from the ecologically and economically important salmonid Salmo trutta fario (river brown trout), a lineage characterized by an ancestral whole-genome duplication (WGD).

Results: Profiling over 83,000 kidney-derived immune cells, we resolved 34 transcriptionally distinct populations, identified core immune lineages, and uncovered novel markers in neutrophils, macrophages, T-cells, and B-cells. We detected pervasive transcriptional divergence between WGD-derived ohnologue pairs, indicating putative sub- and neofunctionalization in immune gene regulation. We further show that the transcriptional identity of immune cells is shaped by rearing history: fish raised in hatcheries-whether for one or multiple generations-showed shifts in immune gene expression across cell types.

Conclusions: Our findings provide baseline data for the healthy brown trout cellular immune system, insight into the evolution of vertebrate immunity, and avenues for understanding the molecular basis of reduced immunological fitness of hatchery-reared fish.

Background: Head and neck cancer, predominantly squamous cell carcinoma, has emerged as a significant global health concern. Growing evidence has established a strong association between dysbiosis of the oral microbiota and both oral and systemic diseases. However, the association between the oral microbiota and head and neck cancer has not yet been fully described. This study aimed to investigate the distinct profiles of the oral microbiota in patients with head and neck cancer and their potential as diagnostic and prognostic biomarkers for head and neck cancer.

Results: Comparative analyses revealed that compared to controls, the oral microbiota of patients with head and neck squamous cell carcinoma (HNSCC) exhibited an increased abundance of anaerobic, biofilm-forming bacteria, and potential pathogens. A machine learning model successfully differentiated HNSCC patients from controls with an area under the curve of 0.902. Key features of this model, such as Peptostreptococcus and Capnocytophaga, were found to be candidate biomarkers for HNSCC, with certain taxa, such as Abiotrophia, serving as prognostic indicators. Although pronounced differences in oral microbiota among HNSCC patients primarily resulted from inter-individual variations, distinct community types were identified, with the type dominated by Proteobacteria being associated with the lowest probability of survival.

Conclusions: Our findings indicate that the oral microbiota may predict HNSCC and may act as a therapeutic target to improve the prognosis of HNSCC. This investigation underscores the crucial role of oral microbial dysbiosis in the etiopathogenesis and clinical prognosis of HNSCC, making a case for further integrative metagenomic and clinical research.

Background: Subchorionic hemorrhage (SCH) is characterized by a fluid-filled hypoechoic area in early pregnancy. This study investigates how laminin subunit β1 (LAMB1) modulates type III collagen (COL3A1) and Rac family small GTPase 1 (RAC1) in SCH coagulation and immunity.

Methods: We recruited ten early-pregnancy SCH patients and ten gestational age-matched women undergoing elective abortion as controls. Proteomic and transcriptomic analyses were performed on decidua and villous tissues to identify differentially expressed proteins (DEPs) and genes (DEGs). Key biomarkers were screened using bioinformatics and machine learning algorithms (LASSO, SVM-RFE). An in vivo SCH-like model was established via LPS induction in pregnant rats, and in vitro experiments were conducted using HTR-8/SVneo trophoblast cells.

Results: Proteomic analysis revealed enrichment in extracellular matrix and coagulation pathways, identifying LAMB1 as a central protein. Transcriptomic data confirmed upregulation of LAMB1 and COL3A1 and downregulation of RAC1 in SCH samples. Clinical blood analysis indicated coagulation abnormalities and Th1/Th2 imbalance in SCH patients. In vivo, LAMB1 knockdown alleviated inflammation, improved pregnancy outcomes, and restored Th2 cytokine expression. In vitro, LAMB1 silencing enhanced trophoblast proliferation, migration, and invasion while reducing pro-inflammatory cytokine levels. Mechanistically, LAMB1 modulated SCH progression via the COL3A1/RAC1 axis.

Conclusions: LAMB1 promotes inflammation and coagulation dysfunction in SCH by regulating COL3A1 and RAC1 expression. Targeting LAMB1 may offer a novel strategy for early diagnosis and therapeutic intervention in SCH to improve pregnancy outcomes.

Background: Crohn's disease (CD) is a chronic, relapsing-remitting gastrointestinal inflammatory condition with a multifactorial etiology. At present, drug therapy is the most important treatment option. However, a substantial number of CD patients experience side effects and/or nonresponse to medical drugs. In part, this might be attributed to the interaction of the intestinal microbiome with xenobiotics, such as medical drugs. The aim of this study was to explore the effect of the common CD drugs budesonide, 6-mercaptopurine (6-MP), as well as tofacitinib on the CD patient's microbiome in vitro.

Results: We performed 16S rRNA gene-based bacterial community profiling and metaproteomic analyses on anaerobic ex vivo incubations of CD patient-derived fecal microbiota (FM) that were exposed to CD drugs or control conditions. Both bacterial community profiling and metaproteomics revealed larger differences in 24-h FM incubations between the five donor-derived FM samples than between the various drug incubations. Incubation of the FM of one of the donors with 6-MP or tofacitinib resulted in a significant alteration in the metaproteome when compared to the control condition, whereas no effect could be observed upon incubation with budesonide. Considering only bacterial proteins detected in at least 80% of either the drug or control FM incubations, 33 proteins were consistently more abundant and 93 less abundant in all five donor-derived samples with 6-MP incubation, distinguishing 6-MP from control conditions. In contrast to metaproteomic analyses, bacterial community profiling only detected a significantly lower relative abundance of Colidextribacter in 15 µg/ml tofacitinib FM incubations. No alterations were detected in overall bacterial richness, diversity, or community structure in response to incubation with any of the drugs.

Conclusions: Tofacitinib and especially 6-MP significantly affect microbial function, but barely microbial composition in vitro. These drug-induced functional changes may subsequently influence host physiology and potentially inflammation in CD. Our findings emphasize the relevance to include functional microbial studies when investigating drug-microbiota interactions. Further research is needed to elucidate the impact of 6-MP-induced microbial alterations on intestinal physiology and inflammation in CD.

Background: The initiation of the classical complement pathway begins with the binding of the globular head of complement component 1q (C1q) to antigen-bound immunoglobulin M (IgM). To investigate the binding mechanism and sites of C1q, a single-chain protein mimetic of the globular head of C1q and variants thereof were designed.

Results: Two approaches were used to generate single-chain globular head C1q variants containing single point mutations potentially altering IgM/C1q binding. The rational protein engineering approach aimed to increase surface charge, considering the negatively charged IgM Cµ3 region and positively charged C1q globular heads. Further, a library of 646 variants with single point mutations in the C1q B-chain loops was designed and expressed using yeast surface display. Three rounds of panning in IgM-coated plates yielded twenty-eight sequenced yeast colonies. The His-tagged wild type variant and six of nine selected variants were stably expressed in Chinese hamster ovary cells and purified using immobilized-metal affinity chromatography. All variants were tested for IgM interaction in competition with serum-derived C1q and in a complement activation assay to evaluate the C1q competition potential of the single-chain globular head proteins.

Conclusions: Expression levels differed among the globular head C1q variants, and SDS-PAGE analysis revealed variations in migration mobility, suggesting conformational differences. Four variants showed enhanced IgM binding compared to the wild type variant indicated by improved C1q displacement in the competitive interaction assay. These results were further supported by an advanced complement activation assay, where these variants significantly inhibited complement activation. These findings underpin the critical role of specific amino acids for IgM/C1q interaction and highlight the potential of engineered C1q as a potent inhibitor or activator of the classical complement cascade.

Background: Insect pollinators, such as bees, provide essential ecosystem services yet face increasing environmental challenges, including pathogens, which can negatively impact host fitness. Central to host defences are immune genes and their products but for many bee species, our understanding of the conservation of the immune gene repertoire, as well as mechanisms that allow for functional diversity, is restricted to a few species.

Results: Here, we perform a pan-clade examination of the canonical immune genes and associated functional gene groups found across 70 bee species, representing six of the seven extant families. We show a high level of conservation of immune genes with all major immune gene groups represented, with elevated copy number variation found in CLIP-domain serine proteases, serpins, and small RNA regulatory proteins across bee species. Using the buff-tailed bumblebee Bombus terrestris, we further show that increases in immune-related gene group size are generally associated with increased nucleotide diversity, and transcriptional divergence among group members suggestive that group size, through mechanisms such as gene duplication, may allow for structural and functional diversity across immune components. However, we find that this pattern is non-linear indicating that gene group expansion is constrained. We also find lineage-specific losses of antimicrobial peptides highlighting that certain immune components may be dispensable or compensated by other elements.

Conclusions: Our analyses show that the genetic components of bee immunity are largely conserved, with duplication and loss highlighted as mechanisms that shape immune diversity, which, collectively, has implications for understanding resilience of bee species to increasing pathogenic threats.

Background: Despite harboring diverse pathogenic viruses, bats rarely exhibit clinical symptoms of the diseases. Previous research has reported evolutionary characteristics in key antiviral gene families in bats, such as natural killer cell receptors, MHC class I genes, and type I interferons, suggesting that bats may possess an immune tolerance that allows them to host viruses asymptomatically. However, this hypothesis is based on limited datasets and requires more comprehensive examinations.

Results: We assembled a chromosome-level reference genome of the Chinese horseshoe bat (Rhinolophus sinicus), a recognized reservoir for SARS-like coronaviruses. By combining this genome with data from 37 other bat species spanning major lineages, we have discovered that the evolutionary signatures of these antiviral gene families exhibit lineage-specific characteristics in gene repertoire, genomic distribution, signaling modes, and expression patterns. Furthermore, we found that the evolutionary diversification of these antiviral gene families is largely influenced by the richness and diversity of viruses, particularly in bats that host more viruses.

Conclusions: Our findings offer insights into the immune adaptations of bats in response to viral infections and reveal a greater interspecies evolutionary heterogeneity in their antiviral immune systems than previously recognized.

Background: Intestinal organoids, three-dimensional cultures derived from intestinal stem cells, serve as a promising model for studying aging. DNA methylation is recognized as a biological clock of aging, and we hypothesized the value of organoid DNA methylation for aging research. To test this, we compared DNA methylation patterns in organoids to those observed in primary aging intestinal epithelium.

Results: Our genome-wide analysis revealed significant DNA methylation changes during organoid establishment and culture. Specifically, 27% of CpG sites exhibited hypomethylation, while 11% gained methylation (hypermethylation). The observed hypermethylation primarily affected CpG islands (14-25% of CpG sites) and corresponded with sites susceptible to age-related methylation in vivo (p < 0.001). Early-passage (P0 and P2) organoids derived from 4- and 24-month-old mice retained aging-specific methylation patterns, with a correlation coefficient of 0.48 (p < 0.001) between in vivo and ex vivo methylation differences. Furthermore, age-related hypermethylation continued to change with passage, with linear modeling using CpG sites that strongly associated with age from a previous dataset indicating that organoids age at an approximate rate of 2 weeks per week in culture. In contrast, hypomethylation primarily (30-36% of CpG sites) occurred in non-aging-associated genomic regions, such as non-promoter, non-CpG island regions (including transposable elements), and showed no correlation with differentiation or aging in vivo.

Conclusions: Our findings suggest that DNA methylation changes in organoids reflect two distinct biological processes: aging, as marked by CpG island hypermethylation, and an unexplained stochastic hypomethylation defect. Additionally, treatment with the DNA methylation inhibitor decitabine reduced the methylation age of organoids from 22 to 10 months, demonstrating the potential of organoids as a model for studying extrinsic influences on age-related DNA hypermethylation.