Structural basis of membrane skeleton organization in red blood cells.

The spectrin-based membrane skeleton is a ubiquitous membrane-associated two-dimensional cytoskeleton underneath the lipid membrane of metazoan cells. Mutations of skeleton proteins impair the mechanical strength and functions of the membrane, leading to several different types of human diseases. Here, we report the cryo-EM structures of the native spectrin-actin junctional complex (from porcine erythrocytes), which is a specialized short F-actin acting as the central organizational unit of the membrane skeleton. While an α-/ß-adducin hetero-tetramer binds to the barbed end of F-actin as a flexible cap, tropomodulin and SH3BGRL2 together create an absolute cap at the pointed end. The junctional complex is strengthened by ring-like structures of dematin in the middle actin layers and by patterned periodic interactions with tropomyosin over its entire length. This work serves as a structural framework for understanding the assembly and dynamics of membrane skeleton and offers insights into mechanisms of various ubiquitous F-actin-binding factors in other F-actin systems.

Structure of a TOC-TIC supercomplex spanning two chloroplast envelope membranes.

The TOC and TIC complexes are essential translocons that facilitate the import of the nuclear genome-encoded preproteins across the two envelope membranes of chloroplast, but their exact molecular identities and assembly remain unclear. Here, we report a cryoelectron microscopy structure of TOC-TIC supercomplex from Chlamydomonas, containing a total of 14 identified components. The preprotein-conducting pore of TOC is a hybrid ß-barrel co-assembled by Toc120 and Toc75, while the potential translocation path of TIC is formed by transmembrane helices from Tic20 and YlmG, rather than a classic model of Tic110. A rigid intermembrane space (IMS) scaffold bridges two chloroplast membranes, and a large hydrophilic cleft on the IMS scaffold connects TOC and TIC, forming a pathway for preprotein translocation. Our study provides structural insights into the TOC-TIC supercomplex composition, assembly, and preprotein translocation mechanism, and lays a foundation to interpret the evolutionary conservation and diversity of this fundamental translocon machinery.

Pro-CRISPR PcrIIC1-associated Cas9 system for enhanced bacterial immunity.

The CRISPR system is an adaptive immune system found in prokaryotes that defends host cells against the invasion of foreign DNA1. As part of the ongoing struggle between phages and the bacterial immune system, the CRISPR system has evolved into various types, each with distinct functionalities2. Type II Cas9 is the most extensively studied of these systems and has diverse subtypes. It remains uncertain whether members of this family can evolve additional mechanisms to counter viral invasions3,4. Here we identify 2,062 complete Cas9 loci, predict the structures of their associated proteins and reveal three structural growth trajectories for type II-C Cas9. We found that novel associated genes (NAGs) tended to be present within the loci of larger II-C Cas9s. Further investigation revealed that CbCas9 from Chryseobacterium species contains a novel ß-REC2 domain, and forms a heterotetrameric complex with an NAG-encoded CRISPR-Cas-system-promoting (pro-CRISPR) protein of II-C Cas9 (PcrIIC1). The CbCas9-PcrIIC1 complex exhibits enhanced DNA binding and cleavage activity, broader compatibility for protospacer adjacent motif sequences, increased tolerance for mismatches and improved anti-phage immunity, compared with stand-alone CbCas9. Overall, our work sheds light on the diversity and 'growth evolutionary' trajectories of II-C Cas9 proteins at the structural level, and identifies many NAGs-such as PcrIIC1, which serves as a pro-CRISPR factor to enhance CRISPR-mediated immunity.

A UPR-Induced Soluble ER-Phagy Receptor Acts with VAPs to Confer ER Stress Resistance.

Autophagic degradation of the endoplasmic reticulum (ER-phagy) is triggered by ER stress in diverse organisms. However, molecular mechanisms governing ER stress-induced ER-phagy remain insufficiently understood. Here we report that ER stress-induced ER-phagy in the fission yeast Schizosaccharomyces pombe requires Epr1, a soluble Atg8-interacting ER-phagy receptor. Epr1 localizes to the ER through interacting with integral ER membrane proteins VAPs. Bridging an Atg8-VAP association is the main ER-phagy role of Epr1, as it can be bypassed by an artificial Atg8-VAP tether. VAPs contribute to ER-phagy not only by tethering Atg8 to the ER membrane, but also by maintaining the ER-plasma membrane contact. Epr1 is upregulated during ER stress by the unfolded protein response (UPR) regulator Ire1. Loss of Epr1 reduces survival against ER stress. Conversely, increasing Epr1 expression suppresses the ER-phagy defect and ER stress sensitivity of cells lacking Ire1. Our findings expand and deepen the molecular understanding of ER-phagy.

Structural duality enables a single protein to act as a toxin-antidote pair for meiotic drive.

In sexual reproduction, selfish genetic elements known as killer meiotic drivers (KMDs) bias inheritance by eliminating gametes that do not carry them. The selective killing behavior of most KMDs can be explained by a toxin-antidote model, where a toxin harms all gametes while an antidote provides resistance to the toxin in carriers. This study investigates whether and how the KMD element tdk1 in the fission yeast Schizosaccharomyces pombe deploys this strategy. Intriguingly, tdk1 relies on a single protein product, Tdk1, for both killing and resistance. We show that Tdk1 exists in a nontoxic tetrameric form during vegetative growth and meiosis but transforms into a distinct toxic form in spores. This toxic form acquires the ability to interact with the histone reader Bdf1 and assembles into supramolecular foci that disrupt mitosis in noncarriers after spore germination. In contrast, Tdk1 synthesized during germination of carrier spores is nontoxic and acts as an antidote, dismantling the preformed toxic Tdk1 assemblies. Replacement of the N-terminal region of Tdk1 with a tetramer-forming peptide reveals its dual roles in imposing an autoinhibited tetrameric conformation and facilitating the assembly of supramolecular foci when autoinhibition is released. Moreover, we successfully reconstituted a functional KMD element by combining a construct that exclusively expresses Tdk1 during meiosis ("toxin-only") with another construct that expresses Tdk1 specifically during germination ("antidote-only"). This work uncovers a remarkable example of a single protein employing structural duality to form a toxin-antidote pair, expanding our understanding of the mechanisms underlying toxin-antidote systems.

A meiotic driver hijacks an epigenetic reader to disrupt mitosis in noncarrier offspring.

Killer meiotic drivers (KMDs) are selfish genetic elements that distort Mendelian inheritance by selectively killing meiotic products lacking the KMD element, thereby promoting their own propagation. Although KMDs have been found in diverse eukaryotes, only a limited number of them have been characterized at the molecular level, and their killing mechanisms remain largely unknown. In this study, we identify that a gene previously deemed essential for cell survival in the fission yeast Schizosaccharomyces pombe is a single-gene KMD. This gene, tdk1, kills nearly all tdk1Δ progeny in a tdk1+ × tdk1Δ cross. By analyzing polymorphisms of tdk1 among natural strains, we identify a resistant haplotype, HT3. This haplotype lacks killing ability yet confers resistance to killing by the wild-type tdk1. Proximity labeling experiments reveal an interaction between Tdk1, the protein product of tdk1, and the epigenetic reader Bdf1. Interestingly, the nonkilling Tdk1-HT3 variant does not interact with Bdf1. Cryoelectron microscopy further elucidated the binding interface between Tdk1 and Bdf1, pinpointing mutations within Tdk1-HT3 that disrupt this interface. During sexual reproduction, Tdk1 forms stable Bdf1-binding nuclear foci in all spores after meiosis. These foci persist in germinated tdk1Δ progeny and impede chromosome segregation during mitosis by generating aberrant chromosomal adhesions. This study identifies a KMD that masquerades as an essential gene and reveals the molecular mechanism by which this KMD hijacks cellular machinery to execute killing. Additionally, we unveil that losing the hijacking ability is an evolutionary path for this single-gene KMD to evolve into a nonkilling resistant haplotype.

The USTC co-opts an ancient machinery to drive piRNA transcription in C. elegans.

Piwi-interacting RNAs (piRNAs) engage Piwi proteins to suppress transposons and nonself nucleic acids and maintain genome integrity and are essential for fertility in a variety of organisms. In Caenorhabditis elegans, most piRNA precursors are transcribed from two genomic clusters that contain thousands of individual piRNA transcription units. While a few genes have been shown to be required for piRNA biogenesis, the mechanism of piRNA transcription remains elusive. Here we used functional proteomics approaches to identify an upstream sequence transcription complex (USTC) that is essential for piRNA biogenesis. The USTC contains piRNA silencing-defective 1 (PRDE-1), SNPC-4, twenty-one-U fouled-up 4 (TOFU-4), and TOFU-5. The USTC forms unique piRNA foci in germline nuclei and coats the piRNA cluster genomic loci. USTC factors associate with the Ruby motif just upstream of type I piRNA genes. USTC factors are also mutually dependent for binding to the piRNA clusters and forming the piRNA foci. Interestingly, USTC components bind differentially to piRNAs in the clusters and other noncoding RNA genes. These results reveal the USTC as a striking example of the repurposing of a general transcription factor complex to aid in genome defense against transposons.

An EMC-Hpo-Yki axis maintains intestinal homeostasis under physiological and pathological conditions.

Balanced control of stem cell proliferation and differentiation underlines tissue homeostasis. Disruption of tissue homeostasis often results in many diseases. However, how endogenous factors influence the proliferation and differentiation of intestinal stem cells (ISCs) under physiological and pathological conditions remains poorly understood. Here, we find that the evolutionarily conserved endoplasmic reticulum membrane protein complex (EMC) negatively regulates ISC proliferation and intestinal homeostasis. Compromising EMC function in progenitors leads to excessive ISC proliferation and intestinal homeostasis disruption. Mechanistically, the EMC associates with and stabilizes Hippo (Hpo) protein, the key component of the Hpo signaling pathway. In the absence of EMC, Yorkie (Yki) is activated to promote ISC proliferation due to Hpo destruction. The EMC-Hpo-Yki axis also functions in enterocytes to maintain intestinal homeostasis. Importantly, the levels of the EMC are dramatically diminished in tunicamycin-treated animals, leading to Hpo destruction, thereby resulting in intestinal homeostasis disruption due to Yki activation. Thus, our study uncovers the molecular mechanism underlying the action of the EMC in intestinal homeostasis maintenance under physiological and pathological conditions and provides new insight into the pathogenesis of tunicamycin-induced tumorigenesis.

SAD-1 kinase controls presynaptic phase separation by relieving SYD-2/Liprin-α autoinhibition.

Neuronal development orchestrates the formation of an enormous number of synapses that connect the nervous system. In developing presynapses, the core active zone structure has been found to assemble through liquid-liquid phase separation. Here, we find that the phase separation of Caenorhabditis elegans SYD-2/Liprin-α, a key active zone scaffold, is controlled by phosphorylation. We identify the SAD-1 kinase as a regulator of SYD-2 phase separation and determine presynaptic assembly is impaired in sad-1 mutants and increased by overactivation of SAD-1. Using phosphoproteomics, we find SAD-1 phosphorylates SYD-2 on 3 sites that are critical to activate phase separation. Mechanistically, SAD-1 phosphorylation relieves a binding interaction between 2 folded domains in SYD-2 that inhibits phase separation by an intrinsically disordered region (IDR). We find synaptic cell adhesion molecules localize SAD-1 to nascent synapses upstream of active zone formation. We conclude that SAD-1 phosphorylates SYD-2 at developing synapses, activating its phase separation and active zone assembly.

The ortholog of human REEP1-4 is required for autophagosomal enclosure of ER-phagy/nucleophagy cargos in fission yeast.

Selective macroautophagy of the endoplasmic reticulum (ER) and the nucleus, known as ER-phagy and nucleophagy, respectively, are processes whose mechanisms remain inadequately understood. Through an imaging-based screen, we find that in the fission yeast Schizosaccharomyces pombe, Yep1 (also known as Hva22 or Rop1), the ortholog of human REEP1-4, is essential for ER-phagy and nucleophagy but not for bulk autophagy. In the absence of Yep1, the initial phase of ER-phagy and nucleophagy proceeds normally, with the ER-phagy/nucleophagy receptor Epr1 coassembling with Atg8. However, ER-phagy/nucleophagy cargos fail to reach the vacuole. Instead, nucleus- and cortical-ER-derived membrane structures not enclosed within autophagosomes accumulate in the cytoplasm. Intriguingly, the outer membranes of nucleus-derived structures remain continuous with the nuclear envelope-ER network, suggesting a possible outer membrane fission defect during cargo separation from source compartments. We find that the ER-phagy role of Yep1 relies on its abilities to self-interact and shape membranes and requires its C-terminal amphipathic helices. Moreover, we show that human REEP1-4 and budding yeast Atg40 can functionally substitute for Yep1 in ER-phagy, and Atg40 is a divergent ortholog of Yep1 and REEP1-4. Our findings uncover an unexpected mechanism governing the autophagosomal enclosure of ER-phagy/nucleophagy cargos and shed new light on the functions and evolution of REEP family proteins.

Diamond controls epithelial polarity through the dynactin-dynein complex.

Epithelial polarity is critical for proper functions of epithelial tissues, tumorigenesis, and metastasis. The evolutionarily conserved transmembrane protein Crumbs (Crb) is a key regulator of epithelial polarity. Both Crb protein and its transcripts are apically localized in epithelial cells. However, it remains not fully understood how they are targeted to the apical domain. Here, using Drosophila ovarian follicular epithelia as a model, we show that epithelial polarity is lost and Crb protein is absent in the apical domain in follicular cells (FCs) in the absence of Diamond (Dind). Interestingly, Dind is found to associate with different components of the dynactin-dynein complex through co-IP-MS analysis. Dind stabilizes dynactin and depletion of dynactin results in almost identical defects as those observed in dind-defective FCs. Finally, both Dind and dynactin are also required for the apical localization of crb transcripts in FCs. Thus our data illustrate that Dind functions through dynactin/dynein-mediated transport of both Crb protein and its transcripts to the apical domain to control epithelial apico-basal (A/B) polarity.

O-GlcNAcylation of RPA2 at S4/S8 antagonizes phosphorylation and regulates checkpoint activation during replication stress.

O-linked N-acetylglucosamine (O-GlcNAc) is the most abundant mono-saccharide modification occurring in the cytoplasm, nucleus and mitochondria. Recent advent of the mass spectrometry technology has enabled identification of abundant O-GlcNAc transferase (OGT) substrates in diverse biological processes, such as cell cycle progression, replication and DNA damage response. Herein we report the O-GlcNAcylation of Replication Protein A2 (RPA2), a component of the heterotrimeric RPA complex pivotal for DNA metabolism. We found that RPA2 interacts with OGT, and a topoisomerase II inhibitor, etoposide, diminishes the association. Using higher-energy collisional dissociation mass spectrometry, we mapped RPA2 O-GlcNAc sites to be Ser-4/Ser-8, which are well-known PIKK-dependent RPA2 phosphorylation sites involved in checkpoint activation upon replication stress. We further demonstrated that Ser-4/Ser-8 O-GlcNAcylation antagonizes phosphorylation and impairs downstream Chk1 activation. Moreover, RPA2 O-GlcNAcylation sustains H2AX phosphorylation upon etoposide treatment, and promotes inappropriate cell cycle progression, indicative of checkpoint defects. Our work not only unveils a new OGT substrate, but also underscores the distinct roles of OGT in replication versus replication stress.

Molecular and structural mechanisms of ZZ domain-mediated cargo selection by Nbr1.

In selective autophagy, cargo selectivity is determined by autophagy receptors. However, it remains scarcely understood how autophagy receptors recognize specific protein cargos. In the fission yeast Schizosaccharomyces pombe, a selective autophagy pathway termed Nbr1-mediated vacuolar targeting (NVT) employs Nbr1, an autophagy receptor conserved across eukaryotes including humans, to target cytosolic hydrolases into the vacuole. Here, we identify two new NVT cargos, the mannosidase Ams1 and the aminopeptidase Ape4, that bind competitively to the first ZZ domain of Nbr1 (Nbr1-ZZ1). High-resolution cryo-EM analyses reveal how a single ZZ domain recognizes two distinct protein cargos. Nbr1-ZZ1 not only recognizes the N-termini of cargos via a conserved acidic pocket, similar to other characterized ZZ domains, but also engages additional parts of cargos in a cargo-specific manner. Our findings unveil a single-domain bispecific mechanism of autophagy cargo recognition, elucidate its underlying structural basis, and expand the understanding of ZZ domain-mediated protein-protein interactions.

SUMO-Targeted DNA Translocase Rrp2 Protects the Genome from Top2-Induced DNA Damage.

The action of DNA topoisomerase II (Top2) creates transient DNA breaks that are normally concealed inside Top2-DNA covalent complexes. Top2 poisons, including ubiquitously present natural compounds and clinically used anti-cancer drugs, trap Top2-DNA complexes. Here, we show that cells actively prevent Top2 degradation to avoid the exposure of concealed DNA breaks. A genome-wide screen revealed that fission yeast cells lacking Rrp2, an Snf2-family DNA translocase, are strongly sensitive to Top2 poisons. Loss of Rrp2 enhances SUMOylation-dependent ubiquitination and degradation of Top2, which in turn increases DNA damage at sites where Top2-DNA complexes are trapped. Rrp2 possesses SUMO-binding ability and prevents excessive Top2 degradation by competing against the SUMO-targeted ubiquitin ligase (STUbL) for SUMO chain binding and by displacing SUMOylated Top2 from DNA. The budding yeast homolog of Rrp2, Uls1, plays a similar role, indicating that this genome protection mechanism is widely employed, a finding with implications for cancer treatment.

Structural basis of Tom20 and Tom22 cytosolic domains as the human TOM complex receptors.

Mitochondrial preproteins synthesized in cytosol are imported into mitochondria by a multisubunit translocase of the outer membrane (TOM) complex. Functioned as the receptor, the TOM complex components, Tom 20, Tom22, and Tom70, recognize the presequence and further guide the protein translocation. Their deficiency has been linked with neurodegenerative diseases and cardiac pathology. Although several structures of the TOM complex have been reported by cryoelectron microscopy (cryo-EM), how Tom22 and Tom20 function as TOM receptors remains elusive. Here we determined the structure of TOM core complex at 2.53 Å and captured the structure of the TOM complex containing Tom22 and Tom20 cytosolic domains at 3.74 Å. Structural analysis indicates that Tom20 and Tom22 share a similar three-helix bundle structural feature in the cytosolic domain. Further structure-guided biochemical analysis reveals that the Tom22 cytosolic domain is responsible for binding to the presequence, and the helix H1 is critical for this binding. Altogether, our results provide insights into the functional mechanism of the TOM complex recognizing and transferring preproteins across the mitochondrial membrane.

The Yun/Prohibitin complex regulates adult Drosophila intestinal stem cell proliferation through the transcription factor E2F1.

Stem cells constantly divide and differentiate to maintain adult tissue homeostasis, and uncontrolled stem cell proliferation leads to severe diseases such as cancer. How stem cell proliferation is precisely controlled remains poorly understood. Here, from an RNA interference (RNAi) screen in adult Drosophila intestinal stem cells (ISCs), we identify a factor, Yun, required for proliferation of normal and transformed ISCs. Yun is mainly expressed in progenitors; our genetic and biochemical evidence suggest that it acts as a scaffold to stabilize the Prohibitin (PHB) complex previously implicated in various cellular and developmental processes and diseases. We demonstrate that the Yun/PHB complex is regulated by and acts downstream of EGFR/MAPK signaling. Importantly, the Yun/PHB complex interacts with and positively affects the levels of the transcription factor E2F1 to regulate ISC proliferation. In addition, we find that the role of the PHB complex in cell proliferation is evolutionarily conserved. Thus, our study uncovers a Yun/PHB-E2F1 regulatory axis in stem cell proliferation.

Sphingolipid profiling reveals differential functions of sphingolipid biosynthesis isozymes of Caenorhabditis elegans.

Multiple isozymes are encoded in the Caenorhabditis elegans genome for the various sphingolipid biosynthesis reactions, but the contributions of individual isozymes are characterized only in part. We developed a simple but effective reversed-phase liquid chromatography-tandem mass spectrometry (RPLC-MS/MS) method that enables simultaneous identification and quantification of ceramides (Cer), glucosylceramides (GlcCer), and sphingomyelins (SM) from the same MS run. Validating this sphingolipid profiling method, we show that nearly all 47 quantifiable sphingolipid species found in young adult worms were reduced upon RNA interference (RNAi) of sptl-1 or elo-5, which are both required for synthesis of the id17:1 sphingoid base. We also confirm that HYL-1 and HYL-2, but not LAGR-1, constitute the major ceramide synthase activity with different preference for fatty acid substrates, and that CGT-3, but not CGT-1 and CGT-2, plays a major role in producing GlcCers. Deletion of sms-5 hardly affected SM levels. RNAi of sms-1, sms-2, and sms-3 all lowered the abundance of certain SMs with an odd-numbered N-acyl chains (mostly C21 and C23, with or without hydroxylation). Unexpectedly, sms-2 RNAi and sms-3 RNAi elevated a subset of SM species containing even-numbered N-acyls. This suggests that sphingolipids containing even-numbered N-acyls could be regulated separately, sometimes in opposite directions, from those containing odd-numbered N-acyls, which are presumably monomethyl branched chain fatty acyls. We also find that ceramide levels are kept in balance with those of GlcCers and SMs. These findings underscore the effectiveness of this RPLC-MS/MS method in studies of C. elegans sphingolipid biology.

Test-Time Training for Deep MS/MS Spectrum Prediction Improves Peptide Identification.

In bottom-up proteomics, peptide-spectrum matching is critical for peptide and protein identification. Recently, deep learning models have been used to predict tandem mass spectra of peptides, enabling the calculation of similarity scores between the predicted and experimental spectra for peptide-spectrum matching. These models follow the supervised learning paradigm, which trains a general model using paired peptides and spectra from standard data sets and directly employs the model on experimental data. However, this approach can lead to inaccurate predictions due to differences between the training data and the experimental data, such as sample types, enzyme specificity, and instrument calibration. To tackle this problem, we developed a test-time training paradigm that adapts the pretrained model to generate experimental data-specific models, namely, PepT3. PepT3 yields a 10-40% increase in peptide identification depending on the variability in training and experimental data. Intriguingly, when applied to a patient-derived immunopeptidomic sample, PepT3 increases the identification of tumor-specific immunopeptide candidates by 60%. Two-thirds of the newly identified candidates are predicted to bind to the patient's human leukocyte antigen isoforms. To facilitate access of the model and all the results, we have archived all the intermediate files in Zenodo.org with identifier 8231084.

Polo-like kinase 1 (PLK1) O-GlcNAcylation is essential for dividing mammalian cells and inhibits uterine carcinoma.

The O-linked ß-N-acetylglucosamine (O-GlcNAc) transferase (OGT) mediates intracellular O-GlcNAcylation modification. O-GlcNAcylation occurs on Ser/Thr residues and is important for numerous physiological processes. OGT is essential for dividing mammalian cells and is involved in many human diseases; however, many of its fundamental substrates during cell division remain unknown. Here, we focus on the effect of OGT on polo-like kinase 1 (PLK1), a mitotic master kinase that governs DNA replication, mitotic entry, chromosome segregation, and mitotic exit. We show that PLK1 interacts with OGT and is O-GlcNAcylated. By utilizing stepped collisional energy/higher-energy collisional dissociation mass spectrometry, we found a peptide fragment of PLK1 that is modified by O-GlcNAc. Further mutation analysis of PLK1 shows that the T291A mutant decreases O-GlcNAcylation. Interestingly, T291N is a uterine carcinoma mutant in The Cancer Genome Atlas. Our biochemical assays demonstrate that T291A and T291N both increase PLK1 stability. Using stable H2B-GFP cells, we found that PLK1-T291A and PLK1-T291N mutants display chromosome segregation defects and result in misaligned and lagging chromosomes. In mouse xenograft models, we demonstrate that the O-GlcNAc-deficient PLK1-T291A and PLK1-T291N mutants enhance uterine carcinoma in animals. Hence, we propose that OGT partially exerts its mitotic function through O-GlcNAcylation of PLK1, which might be one mechanism by which elevated levels of O-GlcNAc promote tumorigenesis.

O-GlcNAcylation promotes the cytosolic localization of the m6A reader YTHDF1 and colorectal cancer tumorigenesis.

O-linked GlcNAc (O-GlcNAc) is an emerging post-translation modification that couples metabolism with cellular signal transduction by crosstalk with phosphorylation and ubiquitination to orchestrate various biological processes. The mechanisms underlying the involvement of O-GlcNAc modifications in N6-methyladenosine (m6A) regulation are not fully characterized. Herein, we show that O-GlcNAc modifies the m6A mRNA reader YTH domain family 1 (YTHDF1) and fine-tunes its nuclear translocation by the exportin protein Crm1. First, we present evidence that YTHDF1 interacts with the sole O-GlcNAc transferase (OGT). Second, we verified Ser196/Ser197/Ser198 as the YTHDF1 O-GlcNAcylation sites, as described in numerous chemoproteomic studies. Then we constructed the O-GlcNAc-deficient YTHDF1-S196A/S197F/S198A (AFA) mutant, which significantly attenuated O-GlcNAc signals. Moreover, we revealed that YTHDF1 is a nucleocytoplasmic protein, whose nuclear export is mediated by Crm1. Furthermore, O-GlcNAcylation increases the cytosolic portion of YTHDF1 by enhancing binding with Crm1, thus upregulating downstream target (e.g. c-Myc) expression. Molecular dynamics simulations suggest that O-GlcNAcylation at S197 promotes the binding between the nuclear export signal motif and Crm1 through increasing hydrogen bonding. Mouse xenograft assays further demonstrate that YTHDF1-AFA mutants decreased the colon cancer mass and size via decreasing c-Myc expression. In sum, we found that YTHDF1 is a nucleocytoplasmic protein, whose cytosolic localization is dependent on O-GlcNAc modification. We propose that the OGT-YTHDF1-c-Myc axis underlies colorectal cancer tumorigenesis.