Mitochondrial ribosome biogenesis depends on RNA helicases such as DDX28, a DEAD-box helicase that plays an essential role during early mitoribosome large-subunit assembly by interacting with 16S rRNA. Here, we demonstrate that the helicase core domain of DDX28 binds sequence and structure specifically to the H88_L stem-loop in 16S rRNA, with the RecA2 domain residue M431 as a key determinant for substrate selectivity. The N-terminal disordered region of DDX28 enhances nonspecific RNA binding but does not contribute to enzymatic activity. Furthermore, DDX28 deficiency disrupts mitochondrial translation, impairs OXPHOS complex assembly, and leads to metabolic dysfunction, including reduced membrane potential, elevated ROS, and suppressed glycolysis. Transcriptomic and metabolomic analyses reveal a compensatory upregulation of ribosome biogenesis genes alongside a dysregulation of the TCA cycle, oxidative phosphorylation, and lipid metabolism. Our integrated structural and functional study establishes DDX28 as an essential factor for mitoribosome assembly with potential links to mitochondrial disorders.

I am a Professor of Biophysics at the University of Science and Technology of China (USTC). I graduated from the Department of Physics, USTC, majoring in Biophysics, in 1965. From 1965 to 1970, I worked as an assistant researcher at the Institute of Chinese Medicine. I joined USTC as an Assistant Professor in 1970 and became a visiting scholar in the Department of Physical Chemistry at the University of Roma from 1979 to1981. My studies on computational biology were mentored by Professor De Santis and supported by the Ministry of Education of China. My studies on bio-NMR were mentored by Professor F. Conti at University of Roma while I simultaneously studied at the CNRS Structural Chemistry Laboratory in Italy. I returned to USTC in 1981 and established a multidisciplinary research team.In 1985 and 1990, I visited the Department of Physical Chemistry, University of Groningen in the Netherlands for six months, under the guidance of Professor H.J.C. Berendsen and Dr. W.F. Van Gunsteren to learn molecular dynamics (MD) simulation of protein and under the guidance of Professor R. Kaptein and Dr. R. Boelens to learn heteronuclear multidimensional NMR experiments for structure determination of proteins in solution. Our laboratory is the first research group to do MD simulation on protein in China and is also one of the pioneer laboratories using NMR to study protein structure in solution in China.

Abstract Histone methylation plays essential roles in modulating chromatin organization and gene expression. H3K23 methylation is a conserved histone modification, yet its biological roles and the enzymes responsible for its deposition remain poorly understood. Here, we show that the loss of set-19 leads to a pronounced reduction in H3K23 methylation in C. elegans , as revealed by quantitative mass spectrometry, western blotting, and immunofluorescence staining. In vitro biochemical assays show that recombinant SET-19 proteins purified from E. coli directly catalyze H3K23 methylation. Genome-wide chromatin immunoprecipitation assays reveal that H3K23me3 is enriched at heterochromatic regions and that loss of set-19 alters H3K23me3 levels, accompanied by derepression of gene expression. Genetic analyses indicate that SET-19 is dispensable for both germline and somatic RNAi as well as transgenerational epigenetic inheritance of RNAi. SET-19 is predominantly expressed in somatic cells and specifically mediates H3K23me3 deposition in somatic tissues. The loss of set-19 causes a developmental delay without affecting fertility. Together, our results identify SET-19 as a somatic H3K23 methyltransferase and link H3K23me3 to gene repression in C. elegans .

Forkhead box protein M1 (FOXM1) is a key transcription factor that regulates cell cycle progression and is frequently overexpressed in human cancers, driving tumor proliferation and therapy resistance. FOXM1 recognizes the canonical forkhead response element (FKH motif, RYAAAYA) through its conserved DNA-binding domain (DBD). Here, we report the high-resolution crystal structure of the FOXM1-DBD in complex with a double-stranded DNA substrate containing two FKH motifs. The structure reveals that FOXM1-DBD adopts the canonical winged-helix fold, with the third α-helix (α3) inserted into the DNA major groove to mediate sequence-specific recognition. Within this helix, Asn283, Arg286, and His287 form an essential triad that engages DNA bases through specific hydrogen bonds and hydrophobic interactions. Using structure-guided mutagenesis of key DNA-interacting residues combined with biophysical validation by isothermal titration calorimetry (ITC) and DNA binding assessment via electrophoretic mobility shift assay (EMSA), we confirm the functional importance of these residues and uncover position-dependent tolerance to base substitutions within the FKH motif. Furthermore, we demonstrate that FOXM1 overexpression promotes cell proliferation and upregulates the transcription of target genes in a DBD-dependent manner. Our findings provide a structural basis for understanding the DNA recognition mechanism of FOXM1 and offer mechanistic insights into how FOXM1 selectively binds to its genomic targets to regulate transcription.

Traditional methods of assessing handwritten characters are often too subjective, inefficient, and lagging in feedback, which makes it difficult for educators to achieve fully objective writing assessments and for writers to receive timely suggestions for improvement. In this paper, we propose a convolutional neural network (CNN) architecture that combines the attention mechanism with multi-scale feature fusion; specifically, the features are weighted by designing a bottleneck layer that combines the Squeeze-and-Excitation (SE) attention mechanism to highlight the important information and by applying a multi-scale feature fusion method to enable the network to capture both the global structure and the local details of Chinese characters. Finally, a high-quality dataset containing 26,800 images of handwritten Chinese characters is constructed based on the application scenario of the writing grade test, covering the common Chinese characters in the writing grade exam; The experimental results show that the proposed method achieves 98.6% accuracy on the writing grade exam dataset and 97.05% on the ICDAR-2013 public dataset, significantly improving recognition accuracy. The constructed dataset and improved model are suitable for application scenarios such as writing grade exams, which helps to improve marking efficiency and accuracy.

TRIM71 NHL Domain is a critical driver of various cellular process and is dysregulated in several medical conditions like non-small cell lung cancer, hepatocellular carcinoma and congenital hydrocephalus. However, its pathways and binding with CDKN1A has not been well studied. To investigate its interaction with CDKN1A, we expressed TRIM71 NHL domain in SF9 (Spodoptera frugiperda) insect cells using the pFastBacTM HT B plasmid, was purified by size exclusion chromatography and its crystal structure was determined successfully (PDB ID: 9JUR). Fluorescence polarization (Kd = 0.42 ± 0.04 μM) and EMSA confirmed strong and specific binding to CDKN1A mRNA, indicating its role in repressing CDKN1A expression to promote cancer cell proliferation. To further delve into its therapeutic implication, we screened a library of 2517 phytochemicals from 48 medicinal plants to identify potential natural inhibitors of the TRIM71 NHL domain. Epigallocatechin Gallate and Cyanidin 3-O-galactoside demonstrated binding affinities of -9.1 kcal/mol and -9.0 kcal/mol, respectively, while SPR confirmed their affinities with Kd values of 3.2 μM and 17.3 μM, accordingly. Molecular dynamics simulations confirmed protein-ligand complexes stability. In summary, human TRIM71 NHL domain crystal structure provides a foundation for understanding its structural features while exploring two potential inhibitors for therapeutic applications.

ZBTB20, a C2H2 zinc finger and broad-complex, tramtrack and bric-à-brac (BTB) domain-containing protein, is crucial for organ development and metabolic homeostasis. Its functionality is dependent on its DNA-binding zinc fingers, and heterozygous mutations within these regions are linked to Primrose syndrome, which is characterized by various physical and developmental abnormalities. However, the molecular basis underlying ZBTB20 zinc finger recognition of DNA remains largely unknown. Here, we present the crystal structure of ZBTB20 zinc fingers 1-4 (ZF1-4) in complex with the mouse alpha-fetoprotein (AFP) promoter in the region spanning positions -104 to -90. In combination with calorimetric analysis, we established that ZF1-3 is essential for the recognition of the AFP promoter and identified key residues involved in DNA binding. Furthermore, our data allow us to correlate Primrose syndrome mutations with alterations in DNA-binding efficacy. Overall, our study provides mechanistic insights into the physiological and pathological roles of ZBTB20 zinc fingers.

Chromatin contains not only heterochromatin(HC) and euchromatins(EC) but also facultative heterochromatin(f HC), which experience the dynamic remodeling between HCs and ECs by different regulators. The regulation of f HCs involves lots of different cell functions, like genomic stability and gene transcription. Heterochromatin protein 1(HP1) recognizes methylated H3K9 and reshapes the chromatin into the f HCs through liquid–liquid phase separation(LLPS). Among the three members of the HP1 family, HP1α can condensate by itself and HP1β forms granules with the help of TRIM28, while the HP1γ cannot phase separation alone either and the coordinator is still unclear. So, in this study, we investigated the molecular mechanism of how HP1γ interacts with TRIM66 through Px Vx L motif. Based on that, we examined the key regions that controlled the TRIM66-HP1γ co-phase separation behaviors both in vitro and in vivo. Furthermore, we proved that the liquid granules of TRIM66-HP1γ and chromatin highly correlated with H3K9me3 sites, which indicated the relationship with DNA damage response. Finally, combined with our previous study, we proposed the system for how TRIM66 remodeled the chromatin into compressed f HC through the TRIM66-HP1γ-H3K9me3 axis with liquid–liquid phase separation.

Biomolecular condensates are formed through phase separation of biomacromolecules such as proteins and RNAs. These condensates exhibit liquid-like properties that can futher transition into more stable material states. They form complex internal structures via multivalent weak interactions, enabling precise spatiotemporal regulations. However, the use of inconsistent and non-standardized terminology has become increasingly problematic, hindering academic exchange and the dissemination of scientific knowledge. Therefore, it is necessary to discuss the terminology related to biomolecular condensates in order to clarify concepts, promote interdisciplinary cooperation, enhance research efficiency, and support the healthy development of this field.

NSUN6 preferentially catalyzes the methylation of cytosine nucleotides in mRNA substrates, which enhances transcription. Dysregulation of NSUN6 catalysis drives the oncogenesis of certain cancers. In this study, we determined the crystal structure of human NSUN6 in complex with its S-adenosyl-L-methionine analog and a bound NECT-2 3'-UTR RNA substrate at 2.9 Å resolution. The complex structure reveals how NSUN6 recognizes the specific CUC[CU]A consensus motif of the substrate and facilitates the methyl transfer from S-adenosyl-L-methionine (SAM) to mRNA. By combining the structural data with nuclear magnetic resonance (NMR)-based fragment screening, a virtual screening, and a further comprehensive biochemical verification, we identified thiamine disulfide as a non-SAM analog lead compound that competes with the CUC[CU]A substrate for binding to NSUN6. Our findings pave the way for the discovery of potent inhibitors for the treatment of NSUN6-driven cancers in the future.

Telomeres are the complexes composed of repetitive DNA sequences and associated proteins located at the end of chromatin. As a result of the DNA replication ending issue, telomeric DNA shortens during each cell cycle. The shelterin protein complex caps telomeric ends and forms a high-order protein-DNA structure to protect telomeric DNA. The stability of telomeres is critical for cellular function and related to the progression of many human diseases. Telomeric repeat-containing RNA (TERRA) is a noncoding RNA transcribed from telomeric DNA regions. TERRA plays an essential role in regulating and maintaining the stability of telomeres. Heterogeneous nuclear ribonucleoproteins (hnRNPs) are RNA-binding proteins associated with complex and diverse biological processes. hnRNPA1 can recognize both TERRA and telomeric DNA. Previous research reported that hnRNPA1, TERRA, and POT1, a component of the shelterin complex, work coordinately and displace replication protein A from telomeric single-stranded DNA after DNA replication, promoting telomere capping to preserve genomic integrity. However, the detailed molecular mechanism has remained unclear for >20 years. Here, our study revealed the molecular structure through which the hnRNPA1 UP1 domain interacts with TERRA and identified critical residues on the interacting surface between UP1 and TERRA. Furthermore, we proved that nucleic acids significantly increase the phase-separating ability of hnRNPA1, while disrupting the UP1-TERRA interaction extraordinarily affects hnRNPA1 droplet formation both in vitro and in vivo. Taken together, these data reveal the molecular mechanism of the phase separation of hnRNPA1 and TERRA and the potential contribution of the droplets to maintaining genomic stability.

Human guanylate kinase (GMPK) as the sole enzyme for GDP biosynthesis plays pivotal roles in antiviral prodrug activation and tumorigenesis. Despite its biological significance, the catalytic mechanism remains poorly understood. Here, we resolve crystal structures of GMPK in free and GMP-bound form, revealing the interdomain motions of GMPBD and LID relative to the CORE domain. Biochemical assays demonstrate potassium's dual functionality in substrate recognition and phosphoryl transfer catalysis. Structural analyses uncover intradomain conformational motion within the LID domain and essential interactions for ADP/ATP binding. Notably, the cooperative ATPγS binding potentiated by prior GMP binding are structurally elucidated. Three key complexes, pre-reaction state (GMP/ATPγS), transition state (AlF4- mimic), and post-reaction state (GDP/ADP), collectively delineate the reversible catalytic pathway. This comprehensive structural characterization of GMPK's dynamic landscape establishes a foundation for developing conformation-specific inhibitors through structure-guided drug design.

Germ granules are biomolecular condensates present in most animal germ cells. One function of germ granules is to help maintain germ cell totipotency by organizing mRNA regulatory machinery, including small RNA-based gene regulatory pathways. The C. elegans germ granule is compartmentalized into multiple subcompartments whose biological functions are largely unknown. Here, we identify an uncharted subcompartment of the C. elegans germ granule, which we term the E granule. The E granule is nonrandomly positioned within the germ granule. We identify five proteins that localize to the E granule, including the RNA-dependent RNA polymerase (RdRP) EGO-1, the Dicer-related helicase DRH-3, the Tudor domain-containing protein EKL-1, and two intrinsically disordered proteins, EGC-1 and ELLI-1. Localization of EGO-1 to the E granule enables synthesis of a specialized class of 22G RNAs, which derive exclusively from 5' regions of a subset of germline-expressed mRNAs. Defects in E granule assembly elicit disordered production of endogenous siRNAs, which disturbs fertility and the RNAi response. Our results define a distinct subcompartment of the C. elegans germ granule and suggest that one function of germ granule compartmentalization is to facilitate the localized production of specialized classes of small regulatory RNAs.

The E3 ligase HOIP is vital for the NF-κB pathway and is implicated in cancer and immunity. However, it remains challenging to achieve high selectivity by directly targeting the conserved catalytic RBR domain of HOIP. Herein, we identified four low-molecular-weight compounds that bind to an uncharacterized pocket of the HOIP PUB domain (HOIPPUB). The complex structure facilitated the discovery of the first single-digit micromolar ligand of HOIPPUB, tolfenamic acid, which exhibited over 30-fold selectivity due to the low sequence identity of the uncharacterized pocket of HOIPPUB. Although tolfenamic acid did not block the substrate recognition and linear ubiquitination activity of HOIP, a ligand of the uncharacterized PUB pocket of HOIP (LUPH), by chemical linking pomalidomide with tolfenamic acid, degraded HOIP, reduced NEMO ubiquitination and p65 phosphorylation, and eventually inhibited NF-κB activation and breast cancer cell proliferation. Our work proposes an alternative strategy to target the nonfunctional pocket of the PUB domain with high sequence diversity to promote HOIP degradation, rather than targeting the conserved RBR domain to block the catalytic function of HOIP.

In eukaryotes, end-binding (EB) proteins serve as a hub for orchestrating microtubule dynamics and are essential for cellular dynamics and organelle movements. EB proteins modulate structural transitions at growing microtubule ends by recognizing and promoting an intermediate state generated during GTP hydrolysis. However, the molecular mechanisms and physiochemical properties of the EB1 interaction network remain elusive. Here we show that EB1 formed molecular condensates through liquid-liquid phase separation (LLPS) to constitute the microtubule plus-end machinery. EB1 LLPS is driven by multivalent interactions among different segments, which are modulated by charged residues in the linker region. Phase-separated EB1 provided a compartment for enriching tubulin dimers and other plus-end tracking proteins. Real-time imaging of chromosome segregation in HeLa cells expressing LLPS-deficient EB1 mutants revealed the importance of EB1 LLPS dynamics in mitotic chromosome movements. These findings demonstrate that EB1 forms a distinct physical and biochemical membraneless-organelle via multivalent interactions that guide microtubule dynamics.

LIM homeodomain transcription factor 1‐alpha (LMX1a) is a neuronal lineage‐specific transcription activator that plays an essential role during the development of midbrain dopaminergic (mDA) neurons. LMX1a induces the expression of multiple key genes, which ultimately determine the morphology, physiology, and functional identity of mDA neurons. This function of LMX1a is dependent on its homeobox domain. Here, we determined the structures of the LMX1a homeobox domain in complex with the promoter sequences of the Wnt family member 1 (WNT1) or paired like homeodomain 3 (Pitx3) gene, respectively. The complex structures revealed that the LMX1a homeobox domain employed its α3 helix and an N‐terminal loop to achieve specific target recognition. The N‐terminal loop (loop1) interacted with the minor groove of the double‐stranded DNA (dsDNA), whereas the third α‐helix (α3) was tightly packed into the major groove of the dsDNA. Structure‐based mutations in the α3 helix of the homeobox domain significantly reduced the binding affinity of LMX1a to dsDNA. Moreover, we identified a nonsyndromic hearing loss (NSHL)‐related mutation, R199, which yielded a more flexible loop and disturbed the recognition in the minor groove of dsDNA, consistent with the molecular dynamics (MD) simulations. Furthermore, overexpression of Lmx1a promoted the differentiation of SH‐SY5Y cells and upregulated the transcription of WNT1 and PITX3 genes. Hence, our work provides a detailed elucidation of the specific recognition between the LMX1a homeobox domain and its specific dsDNA targets, which represents valuable information for future investigations of the functional pathways that are controlled by LMX1a during mDA neuron development.

Stress granules (SGs) are membrane-less organelles (MLOs) or cytosolic compartments formed upon exposure to environmental cell stress-inducing stimuli. SGs are based on ribonucleoprotein complexes from a set of cytoplasmic proteins and mRNAs, blocked in translation due to stress cell-induced polysome disassembly. Post-translational modifications (PTMs) such as methylation, are involved in SG assembly, with the methylation writer PRMT1 and its reader TDRD3 colocalizing to SGs. However, the role of this writer-reader system in SG assembly remains unclear. Here, we found that PRMT1 methylates SG constituent RNA-binding proteins (RBPs) on their RGG motifs. Besides, we report that TDRD3, as a reader of asymmetric dimethylarginines, enhances RNA binding to recruit additional RNAs and RBPs, lowering the percolation threshold and promoting SG assembly. Our study enriches our understanding of the molecular mechanism of SG formation by elucidating the functions of PRMT1 and TDRD3. We anticipate that our study will provide a new perspective for comprehensively understanding the functions of PTMs in liquid-liquid phase separation driven condensate assembly.

Immunotherapy, including immune checkpoint inhibitors and adoptive cell transfer, has obtained great progress, but their efficiencies vary among patients due to the genetic and epigenetic differences. Human MEX3B (hMEX3B) protein is an RNA-binding protein that contains two KH domains at the N-terminus and a RING domain at its C-terminus, which has the activity of E3 ubiquitin ligase and is essential for RNA degradation. Current evidence suggests that hMEX3B is involved in many important biological processes, including tumor immune evasion and HLA-A regulation, but the sequence of substrate RNA recognized by hMEX3B and the functional molecular mechanisms are unclear. Here, we first screened the optimized hMEX3B binding sequence on the HLA-A mRNA and reported that the two tandem KH domains can bind with their substrate one hundred times more than the individual KH domains. We systematically investigated the binding characteristics between the two KH domains and their RNA substrates by nuclear magnetic resonance (NMR). Based on this information and the small-angle X-ray scattering (SAXS) data, we used molecular dynamics simulations to obtain structural models of KH domains in complex with their corresponding RNAs. By analyzing the models, we noticed that on the KH domains’ variable loops, there were two pairs of threonines and arginines that can disrupt the recognition of the RNA completely, and this influence had also been verified both in vitro and in vivo. Finally, we presented a functional model of the hMEX3B protein, which indicated that hMEX3B regulated the degradation of its substrate mRNAs in many biological processes. Taken together, our research illustrated how the hMEX3B protein played a key role in translation inhibition during the immune response to tumor cells and provided an idea and a lead for the study of the molecular mechanism and function of other MEX3 family proteins.