RIKEN BioResource Research Center

A completely asexual morph of Splanchospora is revealed for the first time. This genus is characterised by clavate-ellipsoid, thick-walled ascospores strongly constricted at the submedian septum, the presence of paraphyses in the conidiomata, and globose to ellipsoid, 1-celled conidia. Through phylogenetic analyses based on the small subunit (SSU) and large subunit (LSU) of the nuclear ribosomal DNA, the nuclear ribosomal internal transcribed spacer (ITS) region, the second largest RNA polymerase II subunit (RPB2) gene, and the translation elongation factor 1-alpha (TEF1) gene, the genus is clarified to belong to the family Neohendersoniaceae (Pleosporales, Dothideomycetes). The type species, Splanchospora ampullacea, is neotypified and found to be a species complex. Because of this, we describe four new Splanchospora species, S. fulviconidia sp. nov., S. microspora sp. nov., S. oblongiconidia sp. nov., and S. pseudomicrospora sp. nov. from the twigs of lindens in Japan. All Splanchospora species examined show slight differences in conidial morphology, such as size, shape, and pigmentation. Species-level analyses suggest that this genus, previously considered to be monotypic, in fact harbours a remarkable diversity of species existing on lindens worldwide.

Archaea and Bacteria are two fundamentally distinct domainsoflife that share prokaryotic traits, yet differ markedly in molecularand cellular architecture. While many archaeal species identifiedthus far have been found in extreme environments, recent metagenomicstudies have revealed their widespread presence in moderate habitats,including soils, oceans, and even the human microbiome. However, archaearemain less well characterized than bacteria, largely due to the technicalchallenges associated with culturing and identifying these microorganisms.In this study, we present a culture-independent method for discriminatingarchaea from bacteria at the single-cell level using Raman spectroscopycombined with machine learning. We constructed a Raman spectral dataset comprising 22 prokaryotic species (11 archaea and 11 bacteria)and developed a domain-level Archaea–Bacteria (AB) classifierusing the LightGBM tree-based machine learning algorithm. Our AB classificationmodel achieved an average classification accuracy of 89.1% and a sensitivityof 98.1% on eight representative species (including two independentheld-out test species) with minimal data size and preprocessing. Wealso compared its performance to convolutional neural networks withtransfer learning, a widely used deep learning approach. Our methodprovides a robust analytical framework for archaeal detection andrepresents a valuable addition to the microbiological toolkit, particularlyfor studying unculturable or low-abundance archaeal populations incomplex microbial communities.

The spatial organization and transcriptional activity of transcription factors are increasingly recognized to be regulated by liquid-liquid phase separation (LLPS). Whether GATA1, the master regulator of erythropoiesis, undergoes LLPS and how this process influences erythroid development have remained unknown. Here, we show that GATA1 forms dynamic nuclear condensates in HEK293T, HUDEP2 progenitors, and erythroleukemia cells, as well as concentration-dependent droplets invitro. These condensates exhibit hallmark properties of LLPS, including fusion behavior, sensitivity to 1,6-hexanediol, and rapid fluorescence recovery after photobleaching (FRAP). Domain deletion and mutational analyses revealed intrinsically disordered region 2 (IDR2) as the primary driver of GATA1 LLPS. Importantly, the congenital anemia-associated R307H mutation and dephosphorylation-mimicking S310A substitution within IDR2 disrupted droplet formation and abolished FRAP recovery. Mechanistically, loss of LLPS reduced GATA1 chromatin occupancy, impaired its assembly with cofactors such as FOG1 and LMO2, and abrogated activation of erythroid gene promoters in luciferase assays, leading to defective terminal differentiation. Together, these findings uncover a previously unrecognized mechanism by which GATA1 regulates erythropoiesis through LLPS and highlight modulation of its IDR2 domain and post-translational modifications as potential therapeutic strategies for erythroid-related disorders.

Background: Plasmid-mediated horizontal transfer of antimicrobial resistance genes (ARGs) is a major driver of resistance dissemination across clinical and environmental settings. Urban rivers flowing through densely populated megacities represent critical interfaces where human-associated and environmental microbiomes intersect; however, the genetic structures and functional characteristics of resistance plasmids circulating in such environments remain insufficiently resolved. Methods: In this study, we conducted detailed genomic and phenotypic analyses of 11 ARG-bearing plasmids previously captured from the Tama River, an urban river flowing through the Tokyo megalopolis. These plasmids belonged to IncN, IncU, IncQ2γ, IncC, and IncPγ groups. Whole-plasmid sequencing, comparative genomic analyses, conjugation assays, and antimicrobial susceptibility testing were employed to characterize plasmid backbones, accessory resistance regions, mobile genetic elements, and conjugative transferability. Results: A total of 11 plasmids belonging to five major incompatibility groups (IncN, IncC, IncU+IncQ2γ, and IncP) were analyzed. These plasmids collectively encoded ARGs conferring resistance to five major antimicrobial classes, including aminoglycosides, β-lactams, tetracyclines, chloramphenicol, and mercury, and frequently harbored class 1 integrons, ISCR1 elements, and Tn3-derived inverted-repeat miniature elements (TIME). Notably, two plasmids (IncN and IncC) exhibited high structural similarity to clinically reported plasmids from geographically distant regions, whereas multiple IncP plasmids and one multi-replicon IncU+IncQ2γ plasmid displayed accessory-region architectures characteristic of environmental plasmids and broad host-range transferability. Antibiotic susceptibility testing demonstrated that these plasmids substantially increased resistance levels in hosts. Conclusions: This study reveals that urban river environments can harbor both clinically related and environmentally unique multidrug resistance plasmids, shaped by diverse mobile genetic elements. By providing nucleotide-level structural and functional evidence, this study highlights urban rivers as potential ecological hubs linking clinical and environmental resistance plasmid pools and supports the importance of continued monitoring of resistance plasmids in megacity-associated river systems.

Accurate prediction of compound-protein interactions (CPIs) is crucial for chemical biology and drug discovery. Despite recent advancements, existing deep learning (DL)-based CPI models often struggle to simultaneously achieve high generalization performance, quantify prediction confidence, and ensure explainability. Here, we propose ChemGLaM, a chemical genomics language model designed to address these three crucial challenges, thereby enabling reliable and explainable CPI predictions. ChemGLaM integrates independently pre-trained chemical and protein language models through an interaction block with a cross-attention mechanism, achieving near state-of-the-art performance in predicting novel CPIs at a low computational cost. Incorporating uncertainty estimation and attention visualization enables ChemGLaM to enhance the success rate of virtual screening and to provide molecular insights into CPIs. To demonstrate the practical impact of ChemGLaM, we constructed a publicly available database containing large-scale CPI predictions for every possible pairing between all 20,434 human proteins and all 11,455 drugs and validated its practical applicability in a case study on amyotrophic lateral sclerosis. ChemGLaM marks an important step forward in addressing the challenges of AI-driven CPI exploration and drug discovery.Scientific ContributionThis study established a unified CPI prediction framework that simultaneously achieves high generalization performance, confidence quantification, and explainability. We leveraged this framework to create a community resource by constructing a comprehensive CPI database and demonstrated its practical utility by successfully prioritizing hit compounds and deconvoluting their targets in a phenotypic screening for amyotrophic lateral sclerosis.

Abstract Fae.ca.li.bac.te'ri.um. N.L. masc./fem. adj. faecalis , pertaining to feces; Gr. neut. dim. n. baktêrion , a small rod; N.L. neut. n. Faecalibacterium , rod from feces, as this bacterium is present in high numbers in feces in the colon, its presumed habitat. Bacillota / Clostridia / Eubacteriales / Oscillospiraceae / Faecalibacterium The genus Faecalibacterium consists of nine species at the time of writing. They are Gram‐stain‐negative, nonmotile, catalase‐negative, and strictly anaerobic rods. They are gut inhabitants, and all known species originate from feces of animals. They produce butyrate as the major end product from carbohydrate metabolism and are considered as the most predominant butyrate producing bacteria in the human gut. They also produce anti‐inflammatory molecules. Several strains in the genus possess heterogeneous copies of the 16S rRNA gene, resulting in difficulty in species‐level identification. Housekeeping genes, including recA and rpoA , are suitable gene markers for the identification of Faecalibacterium . DNA G + C content (mol%) : 55.9–59.0 (Genome sequence). Type species : Faecalibacterium prausnitzii Duncan et al. 2002 VP (basonym: “ Bacteroides praussnitzii ” Hauduroy et al. 1937, synonym: Fusobacterium prausnitzii Moore and Holdeman 1970 AL ).

The pseudoautosomal region (PAR) of mammalian sex chromosomes is a small region of sequence identity shared by the X and Y chromosomes that allows pairing, crossover, recombination, and proper segregation of sex chromosomes. Although the mouse PAR sequence was largely a mystery, we have recently obtained the complete PAR sequences of the C57BL/6J and CAST/EiJ strains. Here, we report the complete PAR sequence of a Japanese wild mouse-derived strain, MSM/Ms, and compare the PARs of the three strains. There are considerable differences in the size of PARs between strains (MSM/Ms PAR is only about 20% the length of C57BL/6J PAR) and numerous amino acid substitution variants were found in the PAR genes. High GC-content exons and short introns are common features of the PAR genes and are likely a consequence of maintaining the functions of the encoded proteins during rapid evolution of the mouse PAR, whose recombination frequency in male meiosis is ∼100 times higher than the autosomal average.

Human cellular models used as materials in 183 articles published in *Human Cell* during 2025 comprise a total of 313 samples (Table1, Tables S1, S2). This highlights current trends in research involving human cells and raises concerns about the use of these materials.

In all vertebrates studied to date, a rise(s) in intracellular calcium is indispensable for successful fertilization and further embryonic development. Recent studies demonstrated that zinc is ejected to the extracellular milieu, the 'zinc spark', and follows the first few calcium rises of fertilization. However, the role of the zinc sparks in fertilization and development, as well as the supporting influx mechanism(s), remains unknown. In this study, we focused on zinc transporters Slc39a10/Zip10 which were expressed in mouse oocytes through follicular development and investigated the oocyte-specific deficient mice for Slc39a10/Zip10 (Slc39a10 cKO: Slc39a10flox/flox Gdf9Cre/+). Slc39a10 mRNA or SLC39A10/ZIP10 protein was expressed throughout folliculogenesis in the oocyte or plasma membrane, respectively. The number of ovulated oocytes was examined in Slc39a10 cKO mice, and no change from the number of oocytes was observed. Slc39a10 cKO oocytes decreased zinc level in the oocytes but did not affect maturation and metaphase II spindle formation. Fertilization-induced calcium oscillations were present in Slc39a10 cKO oocytes, but zinc sparks were not observed. Despite other events of egg activation proceeding normally in Slc39a10 cKO oocytes, embryo development into 4 cells and beyond was compromised. We show here for the first time that the zinc transporter ZIP10 contributes to zinc homeostasis in oocytes and embryos, highlighting the role of labile zinc ions in early development.

The MoG+ (Mouse Genome database with high added value), which has been operational since 2019, provides detailed visualization of genomic variations in mouse experimental strains, including wild-derived inbred strains, in particular those maintained at RIKEN BioResource Research Center. Here, we report on the enhancement of MoG+ by inclusion of the latest genome reports and by incorporation of structural variation (SV) information from recent analyses. The latter included long-read sequencing studies of the disease model strains FLS/Shi, NC/Nga, STR/OrtCrlj, JF1/Ms, and MSM/Ms. These studies described SNPs (4,482,628 to 19,644,769), short indels (726,646 to 2,391,782), and SVs such as insertions (32,949 to 131,311), deletions (28,259 to 102,226), and inversions (32 to 164). The new version of the database, which is named MoG+3.0, includes a feature that allows users to visually observe variants in the five strains. Through enhancement of the functionality of the database, SVs have been incorporated and visualized, allowing users to visually examine variants that were difficult to detect using only short-read-based resequencing data. The inclusion of the new variant data, along with enhanced features such as visualization, is expected to serve as a valuable resource for studies of disease and phenotype in experimental mice.