Analysis of endogenous NOTCH1 from POFUT1 S162L patient fibroblasts reveals the importance of the O-fucose modification on EGF12 in human development.

NOTCH1 is a transmembrane receptor interacting with membrane-tethered ligands on opposing cells that mediate the direct cell-cell interaction necessary for many cell fate decisions. Protein O-fucosyltransferase 1 (POFUT1) adds O-fucose to Epidermal Growth Factor (EGF)-like repeats in the NOTCH1 extracellular domain, which is required for trafficking and signaling activation. We previously showed that POFUT1 S162L caused a 90% loss of POFUT1 activity and global developmental defects in a patient; however, the mechanism by which POFUT1 contributes to these symptoms is still unclear. Compared to controls, POFUT1 S162L patient fibroblast cells had an equivalent amount of NOTCH1 on the cell surface but showed a 60% reduction of DLL1 ligand binding and a 70% reduction in JAG1 ligand binding. To determine if the reduction of O-fucose on NOTCH1 in POFUT1 S162L patient fibroblasts was the cause of these effects, we immunopurified endogenous NOTCH1 from control and patient fibroblasts and analyzed O-fucosylation using mass spectral glycoproteomics methods. NOTCH1 EGF8 to EGF12 comprise the ligand binding domain, and O-fucose on EGF8 and EGF12 physically interact with ligands to enhance affinity. Glycoproteomics of NOTCH1 from POFUT1 S162L patient fibroblasts showed WT fucosylation levels at all sites analyzed except for a large decrease at EGF9 and the complete absence of O-fucose at EGF12. Since the loss of O-fucose on EGF12 is known to have significant effects on NOTCH1 activity, this may explain the symptoms observed in the POFUT1 S162L patient.

Endothelial POFUT1 controls injury-induced liver fibrosis by repressing fibrinogen synthesis.

BACKGROUND & AIMS: NOTCH signaling in liver sinusoidal endothelial cells (LSECs) regulates liver fibrosis, a pathological feature of chronic liver diseases. POFUT1 is an essential regulator of NOTCH signaling. Here, we investigated the role of LSEC-expressed POFUT1 in liver fibrosis. METHODS: Endothelial-specific Pofut1 knockout mice were generated and experimental liver fibrosis was induced by chronic carbon tetrachloride exposure or common bile duct ligation. Liver samples were assessed by ELISA, histology, electron microscopy, immunostaining and RNA in situ hybridization. LSECs and hepatic stellate cells (HSCs) were isolated for gene expression analysis by RNA sequencing, qPCR, and western blotting. Signaling crosstalk between LSECs and HSCs was investigated by treating HSCs with supernatant from LSEC cultures. Liver single-cell RNA sequencing datasets from patients with cirrhosis and healthy individuals were analyzed to evaluate the clinical relevance of gene expression changes observed in mouse studies. RESULTS: POFUT1 loss promoted injury-induced LSEC capillarization and HSC activation, leading to aggravated liver fibrosis. RNA sequencing analysis revealed that POFUT1 deficiency upregulated fibrinogen expression in LSECs. Consistently, fibrinogen was elevated in LSECs of patients with cirrhosis. HSCs treated with supernatant from LSECs of Pofut1 null mice showed exacerbated activation compared to those treated with supernatant from control LSECs, and this effect was attenuated by knockdown of fibrinogen or by pharmacological inhibition of fibrinogen receptor signaling, altogether suggesting that LSEC-derived fibrinogen induced the activation of HSCs. Mechanistically, POFUT1 loss augmented fibrinogen expression by enhancing NOTCH/HES1/STAT3 signaling. CONCLUSIONS: Endothelial POFUT1 prevents injury-induced liver fibrosis by repressing the expression of fibrinogen, which functions as a profibrotic paracrine signal to activate HSCs. Therapies targeting the POFUT1/fibrinogen axis offer a promising strategy for the prevention and treatment of fibrotic liver diseases. IMPACT AND IMPLICATIONS: Paracrine signals produced by liver vasculature play a major role in the development of liver fibrosis, which is a pathological hallmark of most liver diseases. Identifying those paracrine signals is clinically relevant in that they may serve as therapeutic targets. In this study, we discovered that genetic deletion of Pofut1 aggravated experimental liver fibrosis in mouse models. Moreover, fibrinogen was identified as a downstream target repressed by Pofut1 in liver endothelial cells and functioned as a novel paracrine signal that drove liver fibrosis. In addition, fibrinogen was found to be relevant to cirrhosis and may serve as a potential therapeutic target for this devastating human disease.

Potential role of POFUT1 as a prognostic predictor in low-grade gliomas: Immune microenvironment insights from a pan-cancer analysis.

The POFUT1 gene, known to be up-regulated in various tumor tissues and associated with tumor biology, has yet to be explored for its potential role in immune response regulation and tumor immune microenvironment. The normalized pan-cancer dataset (TCGA Pan-Cancer) was downloaded from the UCSC database, followed by analysis of POFUT1 expression in various tumors and functional enrichment analysis. The correlation between POFUT1 expression levels and patient prognosis was assessed. GSEA of POFUT1 based on low-grade glioma (LGG) samples and immune infiltration analyses of LGG and glioblastoma (GBM) were conducted. The correlation between POFUT1 expression levels and infiltration levels of 22 immune cells in LGG and GBM was examined, as well as the correlation between immune cell infiltration levels and LGG patient prognosis. Additionally, the relationship between POFUT1 expression levels and characteristic gene expression of identified immune cells was evaluated. Lastly, external dataset validation was performed using the integrated CGGA dataset. Significant differences were observed in POFUT1 expression levels across 20 tumor types. High POFUT1 expression correlated with poor prognosis in GBMLGG, and LGG patients. Enrichment analysis and GSEA of POFUT1 in LGG demonstrated involvement in tumor-related and immune-related pathways. A positive correlation was identified between POFUT1 expression levels and infiltration levels of resting memory CD4+ T cells, as well as M2 macrophages or M2-like TAMs in the LGG immune microenvironment, potentially contributing to poor prognosis. External dataset validation revealed a positive correlation between M2 macrophages or M2-like TAMs and POFUT1 expression levels in LGG, and a negative correlation with LGG patient prognosis. POFUT1's negative impact on LGG prognosis may result from its influence on M2 macrophage and M2-like TAM infiltration levels within the immune microenvironment. This suggests its potential as a prognostic predictor and therapeutic target for LGG.

POFUT1-mediated O-fucosylation of glycoproteins expressed in the baculovirus Sf9 insect cell expression system.

The baculovirus-insect cell expression system allows addition of O-fucose to EGF-like domains of glycoproteins, following the action of the protein O-fucosyltransferase 1 named POFUT1. In this study, recombinant Spodoptera frugiperda POFUT1 from baculovirus-infected Sf9 cells was compared to recombinant Mus musculus POFUT1 produced by CHO cells. Contrary to recombinant murine POFUT1 carrying two hybrid and/or complex type N-glycans, Spodoptera frugiperda POFUT1 exhibited paucimannose N-glycans, at least on its highly evolutionary conserved across Metazoa NRT site. The abilities of both recombinant enzymes to add in vitro O -fucose to EGF-like domains of three different recombinant mammalian glycoproteins were then explored. In vitro POFUT1-mediated O-fucosylation experiments, followed by click chemistry and blot analyses, showed that Spodoptera frugiperda POFUT1 was able to add O-fucose to mouse NOTCH1 EGF-like 26 and WIF1 EGF-like 3 domains, similarly to the murine counterpart. As proved by mass spectrometry, full-length human WNT Inhibitor Factor 1 expressed by Sf9 cells was also modified with O-fucose. However, Spodoptera frugiperda POFUT1 was unable to modify the single EGF-like domain of mouse PAMR1 with O-fucose, contrary to murine POFUT1. Absence of orthologous proteins such as PAMR1 in insects may explain the enzyme's difficulty in adding O-fucose to a domain that it never encounters naturally.

Overexpression of POFUT1 promotes malignant phenotype and mediates perineural invasion in head and neck squamous cell carcinoma.

Head and neck squamous cell carcinoma (HNSCC) is one of the most aggressive neoplasms, which requires more effective prevention and treatment modalities. Previous studies found that protein O-fucosyltransferase 1 (POFUT1) upregulation promotes carcinogenesis, although the potential roles, underlying molecular mechanisms, and biological implications of POFUT1 in HNSCC were not investigated. In this study, in silico analyses referred POFUT1 as a potential oncogene in HNSCC. Further analysis of tumor and normal tissue samples as well as HNSCC cells with quantitative real-time polymerase chain reaction, Western blot analysis, and immunohistochemistry showed significant overexpression of POFUT1 in HNSCC clinical tumor tissue specimens and cell lines compared to corresponding controls. In vitro investigations revealed that overexpression of POFUT1 promoted phenotypes associated with cancer aggressiveness and its knockdown in HNSCC cells suppressed those phenotypes. Further xenograft experiments demonstrated that POFUT1 is an oncogene in vivo for HNSCC. Immunohistochemical analysis with human clinical samples and cancer cell-dorsal root ganglion ex-vivo coculture model showed that deregulation of POFUT1 is involved in the perineural invasion of HNSCC cells. These results suggest POFUT1 expression as a potential prognostic marker for patients with head and neck cancer and highlight its potential as a target for HNSCC therapy, although more molecular clues are needed to better define the functions of POFUT1 related to HNSCC carcinogenesis.

Preliminary study on the mechanism of POFUT1 in colorectal cancer.

To analyse the effect of POFUT1 (Protein O-Fucosyltransferase 1) on the proliferation, migration and apoptosis of colorectal cancer (CRC) cells and to explore its potential mechanism. The effects of POFUT1 silencing in vitro on the proliferation, migration, and apoptosis of CRC cells were investigated using the SW480 and RKO cell lines. The effect of POFUT1 expression on cell phenotype was detected by cell proliferation assay (CCK8), colony formation assay, flow cytometry, wound healing assay, transwell assay, cell apoptosis assay, etc. In vitro, silencing of POFUT1 resulted in decreased proliferation, cell cycle arrest, reduced migration and increased apoptosis of CRC cells. In CRC cells, POFUT1 plays a tumour-promoting role by promoting cell proliferation and migration and inhibiting apoptosis.

O-Fucosylation of BMP1 promotes endometrial decidualization by activating BMP/Smad signaling pathway.

Endometrial decidualization is critical to successful uterine receptivity and embryo implantation. Dysfunction of decidualization is associated with some pregnancy-related disorders, including miscarriage. Protein glycosylation is involved in many physiological and pathological processes. Protein O-fucosyltransferase 1 (poFUT1) is a key enzyme responsible for O-fucosylation biosynthesis on glycoproteins. Bone morphogenetic protein 1 (BMP1) is an essential glycoprotein in reproduction. However, the role and molecular mechanism of fucosylated BMP1 in endometrial stromal cell decidualization are still unknown. In the current study, we found that BMP1 contains a potential O-fucosylation site. Moreover, poFUT1 and BMP1 levels in the secretory phase are higher than those in the proliferative phase, and the highest level was observed in the human uterine tissues of early pregnancy, while a decrease of poFUT1 and BMP1 in the decidua was observed in miscarriage patients. Using human endometrial stromal cells (hESCs), we demonstrated that O-fucosylation of BMP1 was elevated after induced decidualization. Moreover, the increase of BMP1 O-fucosylation by poFUT1 promoted BMP1 secretion to the extracellular matrix, and more actively binds to CHRD. The binding of BMP1 and CHRD further released BMP4 originally bound to CHRD, and activated BMP/Smad signaling pathway, thereby accelerating the decidualization of human endometrial stromal cells. In summary, these results suggest that BMP1 O-fucosylation by poFUT1 could be a potential diagnostic and therapeutic target to predict miscarriage in early pregnancy examinations.

Role of poFUT1 and O-fucosylation in placental angiogenesis†.

Trophoblast cells are critical to placental angiogenesis in the first trimester of pregnancy. Dysfunction of trophoblast leads to defective vascular remodeling and impaired angiogenesis, which is believed as the major cause of placental insufficiency and pregnancy failure. Protein O-fucosyltransferase 1 (poFUT1) is mainly responsible for O-fucosylated glycan biosynthesis on glycoproteins, and poFUT1 deficiency causes embryonic lethality in mice. However, the expression and function of poFUT1 in trophoblast-mediated human placental vessel formation remain unclear. In the current study, we showed that fewer blood vessels were observed in the villi and decidua of miscarriage patients than in normal pregnancy women. The expression of poFUT1 was decreased in the trophoblast cells of miscarriage patients compared with normal pregnancy women. Employing HTR/SVneo cells and an in vivo chorioallantoic membrane assay, we demonstrated that poFUT1 promoted the proliferation, migration ability, and angiogenesis potential of trophoblast cells. The results also indicated that poFUT1 upregulated O-fucosylation on uPA, facilitated the binding of uPA and uPAR, activated the RhoA signaling pathway, and further enhanced the angiogenic capacity of trophoblast cells. Our study provides new evidence for a relationship between poFUT1/O-fucosylation and placental angiogenesis. These findings may provide potential diagnostic biomarkers and targeted therapies for miscarriage patients.

1H, 15N, 13C backbone and sidechain resonance assignments and secondary structure of mouse NOTCH1 EGF27.

NOTCH1 is a transmembrane receptor in metazoans that is linked to a variety of disorders. The receptor contains an extracellular domain (ECD) with 36 tandem epidermal growth factor-like (EGF) repeats. The ECD is responsible for intercellular signaling via protein-ligand interactions with neighboring cells. Each EGF repeat consists of approximately 40 amino acids and 3 conserved disulfide bonds. The Abruptex region (EGF24-29) is critical for NOTCH1 signaling and is known for its missense mutations. Certain EGF repeats are modified with the addition of O-linked glycans and many have calcium binding sites, which give each EGF repeat a unique function. It has been shown that the loss of the O-fucose site of EGF27 alters NOTCH1 activity. To investigate the role of glycosylation in the NOTCH1 signaling pathway, nuclear magnetic resonance spectroscopy has been employed to study the structures of EGF27 and its glycoforms. Here, we report the backbone and sidechain 1H, 15N, and 13C-resonance assignments of the unmodified EGF27 protein and the predicted secondary structure derived from the assigned chemical shifts.

Cancer-associated Notch receptor variants lead to O-fucosylation defects that deregulate Notch signaling.

NOTCH1 is a transmembrane receptor that initiates a signaling pathway involved in embryonic development of adult tissue homeostasis. The extracellular domain of NOTCH1 is composed largely of epidermal growth factor-like repeats (EGFs), many of which can be O-fucosylated at a specific consensus sequence by protein O-fucosyltransferase 1 (POFUT1). O-fucosylation of NOTCH1 is necessary for its function. The Notch pathway is deregulated in many cancers, and alteration of POFUT1 has been reported in several cancers, but further investigation is needed to assess whether there is deregulation of the Notch pathway associated with mutations that affect O-fucosylation in cancers. Using Biomuta and COSMIC databases, we selected nine NOTCH1 variants that could cause a change in O-fucosylation of key EGFs. Mass spectral glycoproteomic site mapping was used to identify alterations in O-fucosylation of EGFs containing the mutations. Cell-based NOTCH-1 signaling assays, ligand-binding assays, and cellsurface analysis were used to determine the effect of each mutation on Notch activation. Two variants led to a gain of function (GOF), six to a loss of function (LOF), and one had minimal effects. Most GOF and LOF were associated with a change in O-fucosylation. Finally, by comparing our results with known NOTCH1 alterations in cancers from which our mutations originated, we were able to establish a correlation between our results and the known GOF or LOF of NOTCH1 in these cancers. This study shows that point mutations in N1 can lead to alterations in O-fucosylation that deregulate the Notch pathway and be associated with cancer processes.

POFUT1 acts as a tumor promoter in glioblastoma by enhancing the activation of Notch signaling.

Dysregulation of protein O-fucosyl transferase 1 (POFUT1) contributes to the occurrence and progression of multiple cancers. However, whether POFUT1 has a relationship with the pathogenesis of glioblastoma (GBM) is unknown. This work was aimed at evaluating the detailed relevance of POFUT1 in GBM. Here, we demonstrated high levels of POFUT1 in GBM tissue and elucidated that GBM patients with high levels of POFUT1 had a shorter survival rate than those with low levels of POFUT1. POFUT1 knockdown in GBM cells markedly downregulated the ability to proliferate and invade, while overexpression of POFUT1 potentiated the proliferative and invasive ability of GBM cells. Further mechanistic studies indicated that silencing POFUT1 prohibited the activation of Notch signaling, leading to a reduction in the expression of HES1 and HEY1. On the contrary, overexpression of POFUT1 enhanced the activation of Notch signaling. Notably, inhibition of Notch signaling markedly reversed POFUT1-overexpression-induced tumor promotion effects in GBM cells. In addition, POFUT1 silencing markedly repressed the potential of GBM cells to form tumors in vivo. In conclusion, the data of this work indicates that POFUT1 serves a tumor promotion role in GBM by enhancing the activation of Notch signaling. This study underlines the potential role of the POFUT1/Notch axis in GBM progression and proposes POFUT1 as a promising anticancer target for GBM.

Diseases related to Notch glycosylation.

The Notch receptors are a family of transmembrane proteins that mediate direct cell-cell interactions and control numerous cell-fate specifications in humans. The extracellular domains of mammalian Notch proteins contain 29-36 tandem epidermal growth factor-like (EGF) repeats, most of which have O-linked glycan modifications: O-glucose added by POGLUT1, O-fucose added by POFUT1 and elongated by Fringe enzymes, and O-GlcNAc added by EOGT. The extracellular domain is also N-glycosylated. Mutations in the glycosyltransferases modifying Notch have been identified in several diseases, including Dowling-Degos Disease (haploinsufficiency of POFUT1 or POGLUT1), a form of limb-girdle muscular dystrophy (autosomal recessive mutations in POGLUT1), Spondylocostal Dysostosis 3 (autosomal recessive mutations in LFNG), Adams-Oliver syndrome (autosomal recessive mutations in EOGT), and some cancers (amplification, gain or loss-of-function of POFUT1, Fringe enzymes, POGLUT1, MGAT3). Here we review the characteristics of these diseases and potential molecular mechanisms.

The single EGF-like domain of mouse PAMR1 is modified by O-Glucose, O-Fucose and O-GlcNAc.

Epidermal growth factor-like domains (EGF-LDs) of membrane and secreted proteins can be modified by N-glycans and/or potentially elongated O-linked monosaccharides such as O-glucose (O-Glc) found at two positions (O-Glc 1 and O-Glc2), O-fucose (O-Fuc) and O-N-acetylglucosamine (O-GlcNAc). The presence of three O-linked sugars within the same EGF-LD, such as in EGF-LD 20 of NOTCH1, has rarely been evidenced. We searched in KEGG GENES database to list mouse and human proteins with an EGF-LD sequence including one, two, three or four potential O-glycosylation consensus sites. Among the 129 murine retrieved proteins, most had predicted O-fucosylation and/or O-GlcNAcylation sites. Around 68% of EGF-LDs were subjected to only one O-linked sugar modification and near 5% to three modifications. Among these latter, we focused on the peptidase domain-containing protein associated with muscle regeneration 1 (PAMR1), having only one EGF-LD. To test the ability of this domain to be glycosylated, a correctly folded EGF-LD was produced in Escherichia coli periplasm, purified and subjected to in vitro incubations with the recombinant O-glycosyltransferases POGLUT1, POFUT1 and EOGT, adding O-Glc1, O-Fuc and O-GlcNAc, respectively. Using click chemistry and mass spectrometry, isolated PAMR1 EGF-LD was demonstrated to be modified by the three O-linked sugars. Their presence was individually confirmed on EGF-LD of full-length mouse recombinant PAMR1, with at least some molecules modified by both O-Glc1 and O-Fuc. Overall, these results are consistent with the presence of a triple O-glycosylated EGF-LD in mouse PAMR1.

Mouse WIF1 Is Only Modified with O-Fucose in Its EGF-like Domain III Despite Two Evolutionarily Conserved Consensus Sites.

The Wnt Inhibitory Factor 1 (Wif1), known to inhibit Wnt signaling pathways, is composed of a WIF domain and five EGF-like domains (EGF-LDs) involved in protein interactions. Despite the presence of a potential O-fucosylation site in its EGF-LDs III and V, the O-fucose sites occupancy has never been demonstrated for WIF1. In this study, a phylogenetic analysis on the distribution, conservation and evolution of Wif1 proteins was performed, as well as biochemical approaches focusing on O-fucosylation sites occupancy of recombinant mouse WIF1. In the monophyletic group of gnathostomes, we showed that the consensus sequence for O-fucose modification by Pofut1 is highly conserved in Wif1 EGF-LD III while it was more divergent in EGF-LD V. Using click chemistry and mass spectrometry, we demonstrated that mouse WIF1 was only modified with a non-extended O-fucose on its EGF-LD III. In addition, a decreased amount of mouse WIF1 in the secretome of CHO cells was observed when the O-fucosylation site in EGF-LD III was mutated. Based on sequence comparison and automated protein modeling, we suggest that the absence of O-fucose on EGF-LD V of WIF1 in mouse and probably in most gnathostomes, could be related to EGF-LD V inability to interact with POFUT1.

Association Analysis between SPP1, POFUT1 and PRLR Gene Variation and Milk Yield, Composition and Coagulation Traits in Sarda Sheep.

Many studies focus on the identification of genomic regions that undergo selective processes, where evidence of selection is revealed and positional candidate genes are identified. The aim of the research was to evaluate the association between positional candidate genes, namely secreted phosphoprotein 1 (SPP1, sheep chromosome Ovis aries OAR6, 36.651-36.658 Mb), protein O-fucosyltransferase 1 (POFUT1, OAR13, 61.006-61.027 Mb) and prolactin receptor (PRLR, OAR16, 38.969-39.028 Mb) with milk yield, composition and coagulation traits. Eight single nucleotide polymorphisms (SNPs) mapping to the three genes were genotyped in 380 Sarda dairy sheep. Statistical analysis revealed an association between SNP rs161844011 at SPP1 (chromosome position Oar_v3 OAR6:36651870, gene region exon 7) and somatic cell score, while POFUT1 SNP rs424501869 (OAR13:61007495, intron 1) was associated with curd firmness both 45 and 60 min after rennet addition (p = 0.015 and p = 0.007, respectively). SNP rs400874750 at PRLR gene (OAR16:39004070, intron 2) had a significant association with lactose content (p = 0.020), somatic cell score (p = 0.038), rennet coagulation time (p = 0.018) and curd firming time (p = 0.047). The outcome of this research confirmed predictions based on genomic studies, producing new information regarding the SPP1, POFUT1 and PRLR genes, which may be useful for future breeding schemes.

POFUT1 is dispensable for structure, function and survival of mouse podocytes.

Pofut1 gene encodes a O-fucosyltransferase that adds fucose to the serine/threonine residue in the sequence of C2XXXX(S/T)C3 of EGF-like domain in a protein. O-fucosylation has been shown to be required for some EGF-like domain-containing proteins to function, e.g., Notch1, and POFUT1 deficiency could affect cellular function and cause diseases. Pofut1 is ubiquitously expressed, but its essentiality for most cell types is not known. In the present study, we examined the consequence of Pofut1 gene abrogation in mouse podocytes using Cre-loxP system, and found that the conditional knockout mice were indistinguishable from wild-type controls in urinary protein level, glomerular morphology, podocyte foot process ultrastructure, podocyte marker expression and podocyte numbers. These results indicated that POFUT1 is not essential for podocyte structure, function and survival in mice. To understand why POFUT1 is dispensable for podocytes, we searched mouse podocyte essential gene candidates (as determined by single-cell RNA-seq) and found only two POFUT1 substrates, NOTCH2 and tPA. It has been shown that abrogation of these genes does not cause podocyte injury, explaining dispensability of POFUT1 for mouse podocytes and demonstrating a feasibility to predict POFUT1 essentiality for a given cell type. At present, most mouse cell types have been subject to single-cell RNA-seq, making essential gene prediction and thus POFUT1 requirement prediction possible for the cell types.

Functional Characterization of POFUT1 Variants Associated with Colorectal Cancer.

BACKGROUND: Protein O-fucosyltransferase 1 (POFUT1) overexpression, which is observed in many cancers such as colorectal cancer (CRC), leads to a NOTCH signaling dysregulation associated with the tumoral process. In rare CRC cases, with no POFUT1 overexpression, seven missense mutations were found in human POFUT1. METHODS: Recombinant secreted forms of human WT POFUT1 and its seven mutated counterparts were produced and purified. Their O-fucosyltransferase activities were assayed in vitro using a chemo-enzymatic approach with azido-labeled GDP-fucose as a donor substrate and NOTCH1 EGF-LD26, produced in E. coli periplasm, as a relevant acceptor substrate. Targeted mass spectrometry (MS) was carried out to quantify the O-fucosyltransferase ability of all POFUT1 proteins. FINDINGS: MS analyses showed a significantly higher O-fucosyltransferase activity of six POFUT1 variants (R43H, Y73C, T115A, I343V, D348N, and R364W) compared to WT POFUT1. INTERPRETATION: This study provides insights on the possible involvement of these seven missense mutations in colorectal tumors. The hyperactive forms could lead to an increased O-fucosylation of POFUT1 protein targets such as NOTCH receptors in CRC patients, thereby leading to a NOTCH signaling dysregulation. It is the first demonstration of gain-of-function mutations for this crucial glycosyltransferase, modulating NOTCH activity, as well as that of other potential glycoproteins.

Molecular characterization of colorectal adenomas reveals POFUT1 as a candidate driver of tumor progression.

Removal of colorectal adenomas is an effective strategy to reduce colorectal cancer (CRC) mortality rates. However, as only a minority of adenomas progress to cancer, such strategies may lead to overtreatment. The present study aimed to characterize adenomas by in-depth molecular profiling, to obtain insights into altered biology associated with the colorectal adenoma-to-carcinoma progression. We obtained low-coverage whole genome sequencing, RNA sequencing and tandem mass spectrometry data for 30 CRCs, 30 adenomas and 18 normal adjacent colon samples. These data were used for DNA copy number aberrations profiling, differential expression, gene set enrichment and gene-dosage effect analysis. Protein expression was independently validated by immunohistochemistry on tissue microarrays and in patient-derived colorectal adenoma organoids. Stroma percentage was determined by digital image analysis of tissue sections. Twenty-four out of 30 adenomas could be unambiguously classified as high risk (n = 9) or low risk (n = 15) of progressing to cancer, based on DNA copy number profiles. Biological processes more prevalent in high-risk than low-risk adenomas were related to proliferation, tumor microenvironment and Notch, Wnt, PI3K/AKT/mTOR and Hedgehog signaling, while metabolic processes and protein secretion were enriched in low-risk adenomas. DNA copy number driven gene-dosage effect in high-risk adenomas and cancers was observed for POFUT1, RPRD1B and EIF6. Increased POFUT1 expression in high-risk adenomas was validated in tissue samples and organoids. High POFUT1 expression was also associated with Notch signaling enrichment and with decreased goblet cells differentiation. In-depth molecular characterization of colorectal adenomas revealed POFUT1 and Notch signaling as potential drivers of tumor progression.

Glycosylation of Specific Notch EGF Repeats by O-Fut1 and Fringe Regulates Notch Signaling in Drosophila.

Fringe glycosyltransferases differentially modulate the binding of Notch receptors to Delta/DLL versus Serrate/Jagged ligands by adding GlcNAc to O-linked fucose on Notch epidermal growth factor-like (EGF) repeats. Although Notch has 22 O-fucosylation sites, the biologically relevant sites affecting Notch activity during animal development in vivo in the presence or absence of Fringe are not known. Using a variety of assays, we find important roles in Drosophila Notch signaling for GlcNAc-fucose-O glycans on three sites: EGF8, EGF9, and EGF12. O-Fucose monosaccharide on EGF12 (in the absence of Fringe) is essential for Delta-mediated lateral inhibition in embryos. However, wing vein development depends on the addition of GlcNAc to EGF8 and EGF12 by Fringe, with a minor contribution from EGF9. Fringe modifications of EGF8 and EGF12 together prevent Notch from cis-inhibiting Serrate, thereby promoting normal wing margin formation. Our work shows the combinatorial and context-dependent roles of GlcNAc-fucose-O glycans on these sites in Drosophila Notch-ligand interactions.