Biology subjects

Yang, B.

Publications and source records attributed to Yang, B..

At least 91 records · Page 5Linked to original sources

bioRxiv2025

Mapping the ISR Landscape in Cognitive Disorders via single-cell multi-omics

Persistent activation of the integrated stress response (ISR) is a major driver of cognitive decline in both neurodevelopmental and neurodegenerative disorders. Using a new mouse model (Ppp1r15bR658C mice) that mimics the persistent ISR activation and cognitive decline observed in humans, we generated the first single-cell ISR atlas of the brain. By integrating single-cell RNA-seq and single-cell ATAC-seq with proteomics, we discovered that distinct brain cell types respond differently to persistent ISR activation and elicit cell-type-specific ISR programs. Interestingly, chromatin accessibility analyses revealed that the ISR downstream factor ATF4 is a key ISR effector in GABAergic neurons, while AP-1 (JUNB) is implicated in glutamatergic neurons. More importantly, selective deletion of ATF4 in GABAergic neurons--but not in glutamatergic neurons--impacts ISR-mediated cognitive decline in Ppp1r15bR658C mice, demonstrating that different neuronal subtypes rely on unique ISR downstream effectors to regulate mnemonic processes. Furthermore, we defined a comprehensive molecular signature of persistent ISR activation, which we showed could serve as a biomarker for cognitive dysfunction across neurodevelopmental, neurodegenerative disorders and normal aging. This multi-omic framework provides a key platform for exploring and validating new scientific hypotheses, significantly advancing our understanding of ISR-related brain disorders.

neuroscience↗

bioRxiv2025

Construction and characterization of coronavirus nonstructural protein 3-host protein interaction networks unravel an important role of cleavage and polyadenylation specificity factor 6 in regulation of viral RNA replication

Coronavirus nonstructural protein 3 (nsp3) plays a crucial role in viral replication and immune evasion. However, functional and proteomic characterization of this protein, especially the interaction networks between nsp3 from different coronaviruses and host cell factors, is hindered by its huge size, complex structural feature and the presence of multiple transmembrane domains. In this study, we report the application of a high-performance cytoplasmic expression system to efficiently and accurately express the full-length nsp3 from betacoronavirus severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and gammacoronavirus infectious bronchitis virus (IBV), to investigate their interactions with host proteins using proteomics approaches. Our study identified 1,150 host proteins that interact with IBV nsp3 and 920 with SARS-CoV-2 nsp3. Among them, 658 are shared by the two nsp3 proteins. Further validation and preliminary characterization of seven selected candidates, DDX5, DDX39, DHX9, elF4A3, SRRT and CPSF6, demonstrated the reproducibility and reliability of the proteomics data. More interestingly, an important regulatory role of the nsp3-CPSF6 interaction in the replication and transcription of IBV gRNA and sgRNA was unraveled. The construction of nsp3-host protein interaction networks from two distantly related coronaviruses would have provided a foundation for future studies of host cell factors in the regulation of coronavirus replication and pathogenesis.

microbiology↗

bioRxiv2025

Crafting Mathematical Models for Type 2 Diabetes Progression: Leveraging Longitudinal Data

Mathematical modeling is a powerful quantitative tool to investigate the pathogenesis of type 2 diabetes (T2D). Most modeling work on the progression of T2D has been formulated by modifying the pioneering model of Topp et al, which established the paradigm of glucotoxicity as the main driver of pathogenesis. However, certain parameter values in the Topp model deviate from physiological data, leading to predictions that deviate from clinical scenarios. Moreover, the simple structure of the model limits its explanatory capacity for clinical data. Leveraging a four-dimensional longitudinal dataset from Southwest Native Americans who progressed from normal glucose tolerance to T2D, we developed a series of models, starting with a minimally modified version of the Topp model and iteratively incorporating additional model elements to account for new biological mechanisms until optimal data fit was achieved. The notable variability of the individual trajectories was overcome by the non-linear mixed-effect modeling approach. Despite the absence of a discernible common trend among the individual trajectories of each variable, the model effectively captured the diverse glucose-insulin dynamics of individuals progressing to T2D. The reliability of the model was reinforced by its successful cross-validation against a subset of individuals progressing only to prediabetes. The systematic model selection process aided in navigating the trade-off between model complexity and practicability, culminating in a robust framework to address controversial questions in the diabetes field in future research.

systems biology↗

bioRxiv2025

Genome modeling and design across all domains of life with Evo 2

All of life encodes information with DNA. While tools for sequencing, synthesis, and editing of genomic code have transformed biological research, intelligently composing new biological systems would also require a deep understanding of the immense complexity encoded by genomes. We introduce Evo 2, a biological foundation model trained on 9.3 trillion DNA base pairs from a highly curated genomic atlas spanning all domains of life. We train Evo 2 with 7B and 40B parameters to have an unprecedented 1 million token context window with single-nucleotide resolution. Evo 2 learns from DNA sequence alone to accurately predict the functional impacts of genetic variation--from noncoding pathogenic mutations to clinically significant BRCA1 variants--without task-specific finetuning. Applying mechanistic interpretability analyses, we reveal that Evo 2 autonomously learns a breadth of biological features, including exon-intron boundaries, transcription factor binding sites, protein structural elements, and prophage genomic regions. Beyond its predictive capabilities, Evo 2 generates mitochondrial, prokaryotic, and eukaryotic sequences at genome scale with greater naturalness and coherence than previous methods. Guiding Evo 2 via inference-time search enables controllable generation of epigenomic structure, for which we demonstrate the first inference-time scaling results in biology. We make Evo 2 fully open, including model parameters, training code, inference code, and the OpenGenome2 dataset, to accelerate the exploration and design of biological complexity.

genomics↗

bioRxiv2025

CDKN2ALow cancer cells outcompete macrophages for microenvironmental zinc to drive immunotherapy resistance

Anti-PD1 therapies are primarily thought to rely on functional T cell responses; yet tumors with limited T cell infiltration can still benefit, suggesting alternative mechanisms contribute to therapeutic efficacy. Indeed, we found that myeloid-rich, T cell-poor tumor models respond to anti-Pd1, and this is dependent on a cancer cell-macrophage crosstalk mediated by cancer cell Cdkn2a expression. Mechanistically, we found that cancer cells with decreased Cdkn2a expression (Cdkn2aLow), which occurs in [~]50% of all human cancers, reorganize zinc compartmentalization by upregulating the zinc importer Slc39a9 at the plasma membrane. Increased cancer cell plasma membrane Slc39a9 leads to intracellular zinc accumulation in cancer cells and depletion of zinc in the tumor microenvironment (TME), resulting in zinc-starved tumor-associated macrophages (TAMs) with reduced phagocytic activity. Restoring zinc availability in TAMs--via dietary supplementation or Slc39a9 knockdown in cancer cells--reprograms TAMs to a pro-phagocytic state and sensitizes Cdkn2aLow tumors to anti-Pd1 therapy. Remarkably, Slc39a9 knockdown tumors respond to anti-Pd1 in Rag1-/- mice, and co-injection of zinc-replete macrophages is sufficient to drive an anti-Pd1 response in immunodeficient mice, demonstrating the T cell-independent nature of this response. Clinically, TAMs from CDKN2ALow cancer patients show reduced zinc and phagocytosis gene signatures. Moreover, patients with lower circulating zinc levels have significantly worse time-to-event outcomes than those with higher levels. Together, these findings uncover a previously unrecognized mechanism by which Cdkn2aLow cancer cells outcompete TAMs for zinc, impairing their function and limiting anti-Pd1 efficacy. They also provide evidence that macrophages alone, without T cells, can enhance anti-PD1 response through zinc-mediated reprogramming of phagocytosis.

cancer biology↗

bioRxiv2025

SRS microscopy identifies inhibition of vitellogenesis as a mediator of lifespan extension by caloric restriction in C. elegans

The molecular mechanisms of aging are not fully understood. Here, we used label-free Stimulated Raman scattering (SRS) microscopy to investigate changes in proteins and lipids throughout the lifespan of C. elegans. We observed a dramatic buildup of proteins within the body cavity or pseudocoelom of aged adults that was blunted by interventions that extend lifespan: caloric restriction (CR) and the reduced insulin/insulin-like growth factor signaling (IIS) pathway. Using a combination of microscopy, proteomic analysis, and validation with mutant strains, we identified vitellogenins as the key molecular components of the protein buildup in the pseudocoelom. Vitellogenins shuttle nutrients from intestine to embryos and are homologous to human apolipoprotein B, the causal driver of cardiovascular disease. We then showed that CR and knockdown of vitellogenins both extend lifespan by >60%, but their combination has no additional effect on lifespan, suggesting that CR extends the lifespan of C. elegans in part by inhibiting vitellogenesis. The extensive dataset of more than 12,000 images stitched into over 350 whole-animal SRS images of C. elegans at different ages and subjected to different longevity intervention will be a valuable resource for researchers interested in aging.

cell biology↗

bioRxiv2025

Molecular Basis of PAC1R Allosteric Modulation with Lipids in Membranes

The pituitary adenylate cyclase-activating polypeptide receptor I (PAC1R) represents a highly sought-after therapeutic target for chronic pain, migraine, and post-traumatic stress. As a class B G protein-coupled receptor (GPCR), the PAC1R is highly expressed throughout the neuronal and central nervous system membranes, with the receptor subject to hormone activation and subsequent signal transduction. Despite its desirable indication, small molecule agonists for PAC1R have been notoriously difficult to develop due to competition with PACAP. For this reason, allosteric activation of the receptor has emerged as a promising pathway. To probe potential allosteric sites, herein, we present a study of the receptor in biomimetic concentrations of lipid membranes. Our results reveal that cholesterol recognizes two canonical and two non-canonical binding sites at PAC1R, which may influence critical residues in the transmembrane domain and PAC1R activation. Also, our simulations suggest the glycolipid GM3 interacts with PAC1R in both the extracellular and transmembrane domains. These lipid binding hotspots may hold high potential for advancing our understanding of class B GPCR signaling and the discovery of new molecules targeting PAC1R. Statement of SignificanceThe PAC1R is a highly sought after therapeutic target in the GPCR family. Understanding its natural process signaling is highly interesting in identifying new modes of control and is currently not well established. This work uses coarse-grained molecular dynamics simulations to examine diverse PAC1R models interactions with their endogenous lipid bilayer in compositions that match the regions in the brain where the receptor is expressed. Two lipids, cholesterol and GM3, have been previously identified in similar receptors as allosteric modulators and were specifically examined in this study. This work also showcases multiple new lipid binding sites at transmembrane and extracellular sites highly implicated in PAC1R signaling.

biophysics↗

bioRxiv2025

SpaIM: Single-cell Spatial Transcriptomics Imputation via Style Transfer

Spatial transcriptomics (ST) technologies have revolutionized our understanding of cellular ecosystems. However, these technologies face challenges such as sparse gene signals and limited gene detection capacities, which hinder their ability to fully capture comprehensive spatial gene expression profiles. To address these limitations, we propose leveraging single-cell RNA sequencing (scRNA-seq), which provides comprehensive gene expression data but lacks spatial context, to enrich ST profiles. Herein, we introduce SpaIM, an innovative style transfer learning model that utilizes scRNA-seq information to predict unmeasured gene expressions in ST data, thereby improving gene coverage and expressions. SpaIM segregates scRNA-seq and ST data into data-agnostic contents and data-specific styles, with the contents capture the commonalities between the two data types, while the styles highlight their unique differences. By integrating the strengths of scRNA-seq and ST, SpaIM overcomes data sparsity and limited gene coverage issues, making significant advancements over 12 existing methods. This improvement is demonstrated across 53 diverse ST datasets, spanning sequencing- and imaging-based spatial technologies in various tissue types. Additionally, SpaIM enhances downstream analyses, including the detection of ligand-receptor interactions, spatial domain characterization, and identification of differentially expressed genes. Released as open-source software, SpaIM increases accessibility for spatial transcriptomics analysis. In summary, SpaIM represents a pioneering approach to enrich spatial transcriptomics using scRNA-seq data, enabling precise gene expression imputation and advancing the field of spatial transcriptomics research.

bioinformatics↗

bioRxiv2025

Murine gut microbiota dysbiosis via enteric infection modulates the foreign body response to a distal biomaterial implant

The gut microbiota influences systemic immunity and the function of distal tissues, including the brain, liver, skin, lung, and muscle. However, the role of the gut microbiota in the foreign body response (FBR) and fibrosis around medical implants is largely unexplored. To investigate this connection, we perturbed the homeostasis of the murine gut microbiota via enterotoxigenic Bacteroides fragilis (ETBF) infection and implanted the synthetic polymer polycaprolactone (PCL) into a distal muscle injury. ETBF infection in mice led to increased neutrophil and {gamma}{delta} T cell infiltration into the PCL implant site. ETBF infection alone promoted systemic inflammation and increased levels of neutrophils in the blood, spleen, and bone marrow. At the PCL implant site, we found significant changes in the transcriptome of sorted fibroblasts but did not observe gross ETBF- induced differences in the fibrosis levels after 6 weeks. These results demonstrate the ability of the gut microbiota to mediate long-distance effects such as immune and stromal responses to a distal biomaterial implant. Significance StatementThe foreign body response to implants leads to chronic inflammation and fibrosis that can be highly variable in the general patient population. Here, we demonstrate that gut dysbiosis via enteric infection promoted systemic inflammation and increased immune cell recruitment to an anatomically distant implant site. These results implicate the gut microbiota as a potential source of variability in the clinical biomaterial response and illustrate that the local tissue environment can be influenced by host factors that modulate systemic interactions.

bioengineering↗

bioRxiv2025

Dose-dependent sensitivity of human 3D chromatin to a heart disease-linked transcription factor

Dosage-sensitive transcription factors (TFs) underlie altered gene regulation in human developmental disorders, and cell-type specific gene regulation is linked to the reorganization of 3D chromatin during cellular differentiation. Here, we show dose-dependent regulation of chromatin organization by the congenital heart disease (CHD)- linked, lineage-restricted TF TBX5 in human cardiomyocyte differentiation. Genome organization, including compartments, topologically associated domains, and chromatin loops, are sensitive to reduced TBX5 dosage in a human model of CHD, with variations in response across individual cells. Regions normally bound by TBX5 are especially sensitive, while co-occupancy with CTCF partially protects TBX5-bound TAD boundaries and loop anchors. These results highlight the importance of lineage-restricted TF dosage in cell-type specific 3D chromatin dynamics, suggesting a new mechanism for TF-dependent disease.

genomics↗

bioRxiv2024

3D Structure-based Generative Small Molecule Drug Design: Are We There Yet?

Structure-based drug design (SBDD) plays a crucial role in preclinical discovery. Recently, structure-based generative algorithms have been developed to streamline the SBDD process, by generating novel, drug-like molecule designs based on the binding pocket structure of target protein. However, there is no effective metric to evaluate the chemical plausibility of molecules designed by these algorithms, which can limit further applications. In this study, we introduce two new metrics for assessing the chemical plausibility of generated molecules. Along with additional analyses, we demonstrate that these algorithms can generate chemically implausible structures with certain property distributions that differ from those of known drug-like molecules. We also compare results with high-throughput virtual screening hits for three protein targets (the c-SRC kinase, the Smoothened receptor, and the Dopamine D1 receptor). These metrics and analysis methods described here offer valuable tools for model developers and users to assess the chemical plausibility and drug-likeness of generated molecules, ultimately enhancing the use of structure-based generation in drug discovery. TOC O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=169 SRC="FIGDIR/small/629537v3_ufig1.gif" ALT="Figure 1"> View larger version (48K): org.highwire.dtl.DTLVardef@1b5de09org.highwire.dtl.DTLVardef@1045f4corg.highwire.dtl.DTLVardef@c8055org.highwire.dtl.DTLVardef@bd4e86_HPS_FORMAT_FIGEXP M_FIG C_FIG

bioinformatics↗

bioRxiv2024

Geometry-based BERT: an experimentally validated deep learning model for molecular property prediction in drug discovery

Various deep learning based methods have significantly impacted the realm of drug discovery. The development of deep learning methods for identifying novel structural types of active compounds has become an urgent challenge. In this paper, we introduce a self-supervised representation learning framework, i.e., GEO-BERT. GEO-BERT considers the information of atoms and chemical bonds in chemical structures as the input, and integrates the positional information of the three-dimensional conformation of the molecule for training. Specifically, GEO-BERT enhances its ability to characterize molecular structures by introducing three different positional relationships: atom-atom, bond-bond, and atom-bond. By benchmarking study, GEO-BERT has demonstrated optimal performance on multiple benchmarks. We also performed prospective study to validate the GEO-BERT model, with screening for DYRK1A inhibitors as a case. Two potent and novel DYRK1A inhibitors (IC50: <1 M) were ultimately discovered at a hit rate of 10%. Taken together, we have developed the Geometry-based BERT model for molecular property prediction and proved its practical utility in early-stage drug discovery. Graphical Abstract GEO-BERT is a model pretrained on large-scale drug molecule data, while improving the accuracy of property prediction by using three-dimensional structural information within the molecule. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=113 SRC="FIGDIR/small/630211v1_ufig1.gif" ALT="Figure 1"> View larger version (24K): org.highwire.dtl.DTLVardef@1e9bc54org.highwire.dtl.DTLVardef@4b6832org.highwire.dtl.DTLVardef@a5d8eeorg.highwire.dtl.DTLVardef@44f7be_HPS_FORMAT_FIGEXP M_FIG C_FIG

pharmacology and toxicology↗

bioRxiv2024

GlioME: A Novel Organoid Model Retaining the Glioma Mi-croenvironment for Personalized Drug Screening and Thera-peutic Evaluation

Glioma is an aggressive brain tumor with a poor prognosis. Establishing an in vitro culture model that closely replicates the cellular composition and microenvironment of the original tumor has been challenging, limiting its clinical applications. Here, we present a novel approach to generate glioma organoids with a microenvironment (GlioME) from patient-derived glioma tissue. These organoids maintain the genetic and epigenetic characteristics of the primary tumor and preserve cell-to-cell interactions within the tumor microenvironment, including resident immune cells. Bulk RNA sequencing, whole exome sequencing, and DNA methylation analysis were used to confirm the molecular similarities between the organoids and primary glioma tissues. Immunofluorescence and flow cytometry were used to assess immune cell viability, comparing GlioME with floating glioma organoids. GlioME exhibited high responsiveness to chemotherapy and targeted therapy, demonstrating its potential for therapeutic screening applications. Notably, GlioME accurately predicted patient response to the recently approved MET inhibitor, vebreltinib. Thus, this organoid model provides a reliable in vitro platform for glioma microenvironment-related research and clinical drug screening.

cancer biology↗

bioRxiv2024

Multimodal normative modeling in Alzheimer Disease with introspective variational autoencoders

Normative models in neuroimaging learn patterns of healthy brain distributions to identify deviations in disease subjects, such as those with Alzheimers Disease (AD). This study addresses two key limitations of variational autoencoder (VAE)-based normative models: (1) VAEs often struggle to accurately model healthy control distributions, resulting in high reconstruction errors and false positives, and (2) traditional multimodal aggregation methods, like Product-of-Experts (PoE) and Mixture-of-Experts (MoE), can produce uninformative latent representations. To overcome these challenges, we developed a multimodal introspective VAE that enhances normative modeling by achieving more precise representations of healthy anatomy in both the latent space and reconstructions. Additionally, we implemented a Mixture-of-Product-of-Experts (MOPOE) approach, leveraging the strengths of PoE and MoE to efficiently aggregate multimodal information and improve abnormality detection in the latent space. Using multimodal neuroimaging biomarkers from the Alzheimers Disease Neuroimaging Initiative (ADNI) dataset, our proposed multimodal introspective VAE demonstrated superior reconstruction of healthy controls and outperformed baseline methods in detecting outliers. Deviations calculated in the aggregated latent space effectively integrated complementary information from multiple modalities, leading to higher likelihood ratios. The model exhibited strong performance in Out-of-Distribution (OOD) detection, achieving clear separation between control and disease cohorts. Additionally, Z-score deviations in specific latent dimensions were mapped to feature-space abnormalities, enabling interpretable identification of brain regions associated with AD pathology.

neuroscience↗

bioRxiv2024

Peptide Information Compression Technology (PICT): A New Frontier in Therapeutic Peptide Discovery

Therapeutic peptides hold tremendous market potential, yet the discovery methods for active peptides lag considerably behind those for small molecules and biological macromolecule drugs. Due to their respective limitations, traditional peptide compound synthesis and display libraries fail to meet the growing demands of the peptide drug market. Here, we introduce a novel Peptide Information Compression Technology (PICT) and, based on this technology, a large-scale physical peptide library and its accompanying high-throughput screening (HTS) platform. By screening for inhibitors of TSLP/TSLPR/IL7R and antagonists of MC3R at both the molecular and cellular levels, we have identified active molecules in the single-digital micromolar and nanomolar ranges, respectively. This demonstrates the immense application value of Peptide Information Compression Technology in the efficient discovery of peptide drugs.

molecular biology↗

bioRxiv2024

N1-methylpseudouridine mRNA modification enhances efficiency and specificity of gene overexpression by preventing Prkra-mediated global translation repression

In vitro transcribed messenger RNA (IVT mRNA) has emerged as a pivotal tool in mRNA-based therapies and has been extensively employed in gene function studies and genetic tool applications. However, the IVT process generates double-stranded RNA (dsRNA) by-products that are recognized by dsRNA sensors, triggering innate immune responses. In this study, we comprehensively analyzed the detrimental effects of dsRNA by-products on early zebrafish embryos, revealing that these by-products induce cell necrosis and delay maternal-zygotic transition by reducing global translation efficiency via Prkra, a recently identified dsRNA sensor in pluripotent cells. Importantly, we demonstrate that N1-methylpseudouridine (m1{varphi}) modification of IVT mRNAs effectively mitigates these adverse effects, as m1{varphi}-modified dsRNAs exhibit significantly lower binding affinity to the Prkra dimer. Our findings underscore a previously overlooked challenge in the use of IVT mRNA in early embryos and offer a robust solution to enhance the fidelity of mRNA applications. Furthermore, we elucidate that m1{varphi} modification minimizes the dsRNA-induced stress response in pluripotent cells through a distinct mechanism.

developmental biology↗

bioRxiv2024

Quantitative mapping of pseudouridines in bacteria RNA

RNA pseudouridylation is one of the most prevalent post-transcriptional modifications, occurring universally across all organisms. Although pseudouridines have been extensively studied in bacterial tRNAs and rRNAs, their presence and role in bacterial mRNA remain poorly characterized. Here, we used a bisulfite-based sequencing approach to provide a comprehensive and quantitative measurement of bacteria pseudouridines. As a proof of concept in E. coli, we identified 1,954 high-confidence sites in 1,331 transcripts, covering almost 30% of the transcriptome. Furthermore, pseudouridine mapping enabled the detection of differentially expressed genes associated with stress response that were unidentified using conventional RNA-seq approach. We also demonstrate that in addition to pseudouridine profiling, our approach can facilitate the discovery of previously unidentified transcripts. As an example, we identified a small RNA transcribed from the antisense strand of tRNA-Tyr which represses expression of distal genes. Finally, we mapped pseudouridines in oral microbiome samples of human subjects, demonstrating the broad applicability of our approach in complex microbiomes. Altogether, our work highlights the advantages of mapping bacterial pseudouridines and provides a tool to study posttranscription regulation in microbial communities.

microbiology↗

bioRxiv2024

Arabidopsis SUFB and CLP protease regulate SUFBC2D to manipulate iron-sulfur cluster biosynthesis in chloroplast

Iron-sulfur (Fe-S) clusters are essential cofactors for Fe-S proteins. SUFBC2D complex is the scaffold responsible for Fe-S cluster assembly in chloroplasts. However, the regulatory mechanism on SUFBC2D remains elusive. In this study, we report that the transcription of SUFB responds rapidly to leaf senescence, whereas the transcription of SUFC and SUFD does not. Intriguingly, their protein contents remain stable during leaf senescence. We further found that leaf death was occurred only when SUFB RNAi was induced, and SUFB and SUFC contents decreased much faster in the SUFB-RNAi lines than in the SUFC-RNAi lines, indicating that SUFB had a faster turnover rate than SUFC. Moreover, overexpressing SUFB increased the contents of SUFC and SUFD, and SUFBC2D, whereas overexpressing SUFC did not increase SUFB and SUFD. Our findings reveal that SUFB stabilizes SUFC and SUFD via forming SUFBC2D, whereas SUFC lacks this function. Furthermore, SUFB expression was sharply downregulated when the plants were subjected to iron deficiency, whereas SUFC and SUFD expression was not. Interestingly, the contents of all three SUF members decreased, indicating that plants degrade SUFBC2D in response to iron deficiency by downregulating SUFB transcription. We subsequently studied the degradation mechanism of SUFBC2D. Our results indicated that all SUFs are substrates of the caseinolytic protease (CLP) because they all accumulated in the CLP impaired mutant, and they physically interact with CLPS1, the substrate recognition adaptor of CLP. Collectively, our findings provide novel insights into how plants regulate SUFBC2D complex via SUFB to adapt to leaf senescence and iron deficiency.

plant biology↗