NEU4-mediated desialylation ignites the oncogenic receptors for the dissemination of ovarian carcinoma

NEU4-mediated desialylation ignites the oncogenic receptors for the dissemination of ovarian carcinoma | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article NEU4-mediated desialylation ignites the oncogenic receptors for the dissemination of ovarian carcinoma Long Wang, Jie Shi, Rui Zhou, Shuo Wang, Yuxin Liu, Baorui Tian, and 12 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3772327/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 14 Oct, 2024 Read the published version in Oncogene → Version 1 posted 9 You are reading this latest preprint version Abstract Glycosylation profoundly influences the interactions between cancer cells and microenvironmental stromal cells during the peritoneal disseminated metastasis of ovarian carcinoma (OC), which is the major cause of cancer-related death. Although the characteristic cancer glycoconjugates are widely used as biomarkers for cancer diagnosis, our knowledge about cancer glycome remains quite fragmented due to the technique limitations in analyzing glycan chains with tremendous structural and functional heterogeneity. Given the dysregulated cancer glycome is defined by the altered glycosylation machinery, here we performed a systematic loss-of-function screen on 498 genes involved in glycosylation for key regulators of OC dissemination. We identified neuraminidase 4 (NEU4), an enzyme capable of hydrolyzing terminal sialic acid from glycoconjugates, as a vital peritoneal dissemination-promoting modifier of OC glycome. In human patients with high-grade serous OC (HGSOC), increased NEU4 was detected in the disseminated OC cells when compared with that in the primary tumor cells, which significantly correlated with the worse survival. Among three alternative splice-generated isoforms of human NEU4, we revealed that only the plasma membrane-localized NEU4 isoform 2 (NEU4-iso2) and intracellular isoform 3 promoted the peritoneal dissemination of OC by enhancing the cell motility and epithelial-mesenchymal transition. We also identified NEU4-iso2-regulated cell surface glycoproteome and found that NEU4-iso2 desialylated the epithelial growth factor receptor (EGFR), in particular at N 196 residue, for the hyperactivation of EGFR and its downstream tumor-promoting signaling cascades. Our results provide new insights into how the OC glycome is dysregulated during OC progression and reveals a functionally important glycosite on EGFR for its abnormal activation in cancer. Biological sciences/Cancer/Gynaecological cancer/Ovarian cancer Biological sciences/Cell biology/Glycobiology Ovarian carcinoma NEU4 sialylation transcoelomic metastasis EGFR Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Significance Statement The characteristic cancer glycoconjugates are widely used as biomarkers for cancer diagnosis. However, our knowledge about cancer glycome remains quite fragmented due to the technique limitations in analyzing glycan chains with tremendous structural and functional heterogeneity. We identified neuraminidase 4 (NEU4), that hydrolyzes terminal sialic acid from glycoconjugates, as a vital peritoneal dissemination-promoting modifier of OC glycome. In addition, we for the first time revealed that the plasma membrane-localized NEU4 isoform 2 desialylated the epithelial growth factor receptor (EGFR), in particular at N196 residue, for the hyperactivation of EGFR and its downstream tumor-promoting signaling cascades. Our results provide new insights into how the OC glycome is dysregulated during OC progression and reveals a functionally important glycosite on EGFR for its abnormal activation in cancer. Introduction Peritoneal disseminated metastasis (PDM), also known as transcoelomic metastasis, is a major cause of the death of epithelial ovarian carcinoma (EOC) patients ( 1 , 2 ). The peritoneal disseminated metastasis of EOC cells involves multiple steps, i.e., the detachment of cancer cells from the primary tumor, the peritoneal migration of the free cancer cells, the adhesion of disseminated cancer cells to the mesothelial cells lining the peritoneal cavities and intra-abdominal organs or the underlying extracellular matrix (ECM), and the colonization of the disseminated cancer cells to form the secondary tumor nodules ( 2 , 3 ). Complex interactions between cancer cells and the peritoneal stromal cells or ECMs occur throughout the whole dissemination cascade, in which aberrant glycosylation has long been realized to play essential roles and hence to be a common signature of malignancies ( 4 , 5 ). Characteristic alterations of the N-glycosylation, GalNAc-type O-glycosylation and O-Xyl glycosaminoglycans have been found in cancers ( 5 , 6 ). For example, distinct changes in the branching of N-glycans and core fucosylation and increased α2-6-sialic acid capping have been linked to the altered adhesion, migration and epithelial-mesenchymal transition (EMT) capacities. The upregulation of sialyl-Lewis x and sialyl-Lewis a glycan antigens in cancer cells enhances their binding on the endothelial cells through selectins and thus leads to their extravasation and homing to the metastatic loci ( 7 , 8 ). And altered sialylation pattern also contributes to the immune evasion of tumor cells ( 9 ). Immature, truncated O-glycan structure has also been identified to be a feature of many cancers, e.g., Tn and STn antigens that impact the interaction between cancer cells and macrophages to regulate tumor immune microenvironment ( 10 , 11 ). A recent integrated analysis of the proteomics and glycoproteomics in human high-grade serous OC (HGSOC) reveals tumor-specific changes of glycoproteins and glycosites ( 12 , 13 ). Compared to the adjacent non-tumor cells, HGSOC cells contain more abundant high-mannose glycans, but less fucosylated and sialylated glycans ( 12 ). However, the signature glycosylation patterns and their functional consequences in tumor progression remain largely unknown due to the too limited glycoproteomic technologies to characterize the glycan chains with huge complexity and heterogeneity that are not encoded directly in the genome ( 6 ). The dysregulated glycome in cancer is largely due to the altered glycosylation machinery that is composed of glycosyltransferases, transporters, chaperones and glycan modification and degradation enzymes including glycosidases, which are encoded by ~ 700 genes in the human genome ( 6 ). For example, fucosyltransferase 3 or 6 (FUT3, FUT6) is upregulated in cancer cells to increase the capping of N-acetyllactosamine (LacNAc)-terminated oligosaccharides by sialyl-Lewis x antigen, resulting in enhanced E-selectin binding in the vascular niche and hence promotes bone metastasis ( 7 ). Aberrant sialylation caused by altered expressions of sialyltransferases and sialidases (also known as neuraminidases) has been observed in many cancer types including OC, colon cancer, renal cancer and prostate cancer ( 14 , 15 ). Increased expression of neuraminidase 3 (NEU3) was reported in colon and prostate cancers to promote cancer progression by enhancing EGF-EGFR signaling cascades ( 16 , 17 ). On the contrary, NEU1 functions as a tumor suppressor with decreased expression in cancer cells and negatively regulates metastasis via modulating integrin β4-mediated signaling ( 18 ). Due to the tremendous structural complexity and functional diversity of glycans and current technical limitations in the sequencing and manipulation of the glycan chains, our knowledge about the cancer glycome remains fragmented. However, new gene editing tools allow more comprehensive dissection of the functions of cancer glycome. In this study, we performed a systematic loss-of-function screen on 498 genes involved in the glycosylation machinery using CRISPR/Cas9 knockout library for key regulators of OC dissemination. We identified neuraminidase 4 (NEU4), an enzyme capable of removing terminal sialic acid residue from all types of sialylated glycoconjugates including glycoproteins, oligosaccharides and gangliosides ( 14 ), as a novel peritoneal dissemination-promoting factor for OC. Given the existence of three isoforms of human NEU4 generated by alternative splicing, we further investigated their roles in the progression of OC and revealed that the plasma membrane-localized NEU4 isoform 2 was able to desialylate EGFR for its hyperactivation, which contributed to the peritoneal dissemination of OC. Our results provide new evidences that the alternative splice-generated isoforms of glycosylation-related enzymes, which have different subcellular distribution and functions, contribute to the vast heterogeneity of cancer glycome and greatly impact the cancer progression. Results Systematic gene knockout screen identifies NEU4 as a potential dissemination-promoting gene in OC To systematically identify key glycosylation-related genes involved in the peritoneal dissemination of OC, we selected 2,984 lentivirus-based sgRNA plasmids against 498 genes (4 ~ 6 sgRNAs per gene) involved in the protein glycosylation modification and 1,000 non-targeting control sgRNA plasmids from the genome-wide CRISPR/Cas9 library (GeCKO-v2.0) made by Feng Zhang’s laboratory ( 19 , 20 ) ( Supplementary Table 1 ), which were subsequently transfected into human OC cells SK-OV-3 to generate heterogeneous cell population with deficiency of different glycosylation gene in each cell. These cells were orthotopically inoculated into the ovarian bursas of NOD-SCID mice and the peritoneal disseminated OC cells that formed the new tumor nodules on the peritoneal wall were dissected and expanded in vitro for the next two rounds of selections for OC cells with high dissemination capacity ( Fig. 1A ). The genomic DNAs from the primary OC xenografts in the first-round selection and the disseminated OC cells in the third-round selection were isolated for high throughput DNA sequencing to identify enriched sgRNAs and target genes in these cell populations ( Fig. 1B, Supplementary Table 1 ). By using the MAGeCK algorithm to analyze the enrichment of identified sgRNAs targeting 143 glycosylation-related genes, we got eight genes with two sgRNAs enriched in the primary OC xenograft cells ( Fig. 1B, Supplementary Table 2 ), suggesting their potential roles in promoting peritoneal dissemination of OC. We further investigated the correlation of these candidate dissemination-promoting genes with the overall survivals of 373 OC patients in The Cancer Genome Atlas (TCGA) database since the peritoneal dissemination of OC cells was the leading cause of the death of OC patients ( 21 ) ( Supplementary Table 3 ). Among these eight genes, only the higher expressions of NEU4, SPTA1 and ANKRD29 significantly correlate with the poorer survivals of OC patients ( Fig. 1C, Supplementary Fig. 1, Supplementary Table 3 ). Given that NEU4 is an enzyme responsible for the trimming of sialic acid from glycosylated proteins and the significant roles of sialic acid in the cell-cell and cell-environment communications ( 22 ), here we focused on the potential roles of NEU4 in OC dissemination. To investigate the clinical relevance of NEU4 with OC dissemination, we collected the primary tumor tissues from eight HGSOC patients with paired peritoneal disseminated OC tissues and performed quantitative RT-PCR analysis of all NEU4 transcripts ( Fig. 1D ). A statistically significant increase of NEU4 transcription was observed in five disseminated OC patient samples while decreased NEU4 transcription was observed in other three disseminated OC tissues ( Fig. 1E ). However, when analyzed by western blot and immunohistochemistry, a dramatically elevated protein level of NEU4 was observed in most disseminated OC tissues ( Fig. 1F and 1G , respectively), suggesting there probably existed post-transcriptional regulation of NEU4 expression during OC dissemination. There are several transcript variants of NEU4 that generate three isoforms with difference in the N-terminal sequences. Compared to the NEU4 isoform 1 and 2, the isoform 3 is shorter due to the lack of an N-terminal signal peptide, in which isoform 2 has an additional amino acid residue than isoform 1 ( Fig. 1D ). In addition to the localizations of NEU4 isoform 1 and isoform 3 in the lysosome and intracellular membranes as reported ( 23 – 25 ) ( Fig. 1H, Supplementary Fig. 2A and 2B ), IF staining of V5-tagged NEU4 isoform 2 showed the distinct plasma membrane localization in Caov-3 and SK-OV-3 cells ( Fig. 1H, Supplementary Fig. 2A ), which was further confirmed by the cell fractionation analysis ( Supplementary Fig. 2C and 2D ). The one-amino-acid-longer signal peptide of isoform 2, when attached to the enhanced green fluorescent protein (EGFP), is robust enough to localize the EGFP proteins on the plasma membrane when compared with that of the isoform 1 ( Supplementary Fig. 2E ), suggesting that the isoform 2 of NEU4 may modify the glycoproteins on the plasma membrane and directly affect the signal recognition and cell-cell communications during the dissemination of OC cells. We further investigated the transcriptional change of NEU4 isoform 2 in human OC specimen by a specific primer set ( Fig. 1D ). Among eight paired human OC tissues, six disseminated OC tissues showed increased expression of NEU4 isoform 2, in which three, i.e., patient P2, P3 and P5, showed even higher fold of increase than that of the total NEU4 isoforms, indicating the potential functional significance of isoform 2 in OC dissemination ( Fig. 1E and 1I ). However, the other three showed similar or a bit less increase of NEU4 isoform 2 compared with the change of total NEU4 isoforms ( Fig. 1E and 1I ), suggesting the possible functional significance of other isoforms of NEU4. Figure 1. High-throughput CRISPR/Cas9 screen identified NEU4 as a key regulator of OC peritoneal dissemination. (A) The schematic diagram of CRISPR/Cas9 library-based screen of key glycosylation-related genes regulating peritoneal dissemination in the orthotopic murine model of OC. (B) Scatterplot showing the enrichment of specific sgRNAs in primary (Pri) or peritoneal disseminated (Diss) OC cells. (C) Kaplan-Meier survival curve to show the overall survival (OS) of OC patients with different NEU4 expression (n = 373) from the TCGA database. (D) The different amino acid sequences of three NEU4 isoforms and the sites of specific qRT-PCR primer sets. The Primer 1 was designed at the common encoding regions of NEU4 isoforms. The additional N-terminal amino acid residues of isoform 1 (iso1) and 2 (iso2) were presented with the differential residue of iso2 highlighted, where the specific forward primer of iso2 bound (Primer 2). (E) qRT-PCR analysis of the total NEU4 transcripts in the primary and the paired peritoneal disseminated OC tissues in each OC patients (the left bargraph, data are plotted as means ± SEM from three independent measurements, **p < 0.01, ***p < 0.001, ns-not significant, by unpaired Student’s t-test) and the statistical analysis of NEU4 transcripts in eight OC patients ( the right bargraph, ***p < 0.001, by paired Student’s t-test). (F) Western blot analysis of NEU4 in the primary and the paired disseminated OC tissues from eight OC patients and the quantification results (n = 8, **p < 0.01, by paired Student’s t-test). (G) Representative images of the immunohistochemical staining of NEU4 in the primary and disseminated OC tissues from an OC patient and the H-score quantification results (n = 8, ***p < 0.001, by paired Student’s t-test). (H) Immunofluorescent staining of v5-tagged NEU4 isoforms ectopically expressed in Caov-3 cells. NEU4 isoforms are stained by anti-v5 antibody in green, while the actin filaments (by phalloidin) and lysosome marker LAMP1 in red. (I) qRT-PCR analysis of NEU4 isoform 2 (NEU4-iso2) in the primary and paired disseminated OC tissues in each OC patient (left bargraph, means ± SEM from three independent measurements, *p < 0.05, **p < 0.01, ***p < 0.001, by unpaired Student’s t-test) and the statistical analysis of NEU4-iso2 expression in eight OC patients (n = 8, ***p < 0.001, by paired Student’s t-test). NEU4, especially the isoform 2 and 3, is essential for the peritoneal dissemination of OC cells in mice To investigate the role of NEU4 in the progression of OC, we selected a mouse HGSOC cell line ID8 with high NEU4 expression and a human HGSOC cell line Caov-3 with relatively low NEU4 expression to establish different murine models of HGSOC ( Fig. 2A ) ( 26 ). In the syngeneic murine model of HGSOC, we knocked down Neu4 effectively by two small hairpin RNAs (shRNAs) in ID8 cells ( Fig. 2B ) and orthotopically transplanted these cells into the ovarian bursas of C57BL/6 mice ( Fig. 2C ). Either shRNA targeting Neu4 significantly inhibited the formation of malignant ascites ( Fig. 2D ) and reduced the disseminated OC lesions in the abdominal cavity ( Fig. 2E ). To further investigate the role of NEU4 in the peritoneal seeding and colonization of disseminated OC cells that involved complex cell-cell and cell-matrix communications, we performed intraperitoneal injection of ID8 cells expressing the luciferase reporter gene into C57BL/6 mice ( Fig. 2F ). Live imaging showed that knocking down Neu4 greatly reduced the formation of cancer lesions in the abdominal cavity ( Fig. 2G ) and, as the cancer developed, ID8 cells with silenced Neu4 expression caused much less formation of malignant ascites ( Fig. 2H ) and cancer nodules ( Fig. 2I ). Mouse Neu4 has only two isoforms reported. The shorter one is analogous to the human NEU4 isoform 3, while the longer one has an additional N-terminal signal peptide that is longer than those of the human NEU4 isoform 1 and 2 ( Supplementary Fig. 2E and 2F ) and displays both the cytoplasmic and plasma membrane localizations ( Supplementary Fig. 2F ) and may be functionally equivalent to human NEU4 isoform 1 and 2. To clarify the roles of different isoforms of human NEU4 in the dissemination of OC, we ectopically expressed these isoforms of NEU4 in human HGSOC cells Caov-3 that have very low level of endogenous NEU4 expression ( Fig. 2A and 2J ) and orthotopically transplanted these cells into the ovarian bursas of immunodeficient NOD-SCID mice ( Fig. 2K ). Although all the isoforms of NEU4 did not affect the growth of primary tumor xenografts ( Fig. 2L ), the isoform 2 and 3 of NEU4 dramatically promoted the formation of peritoneal cancer nodules ( Fig. 2M ). These results consistently suggest that the NEU4, in particular the isoform 2 and 3, is a potent dissemination-promoting factor of OC. The isoform 3, also known as NEU4-short that localizes in the endoplasmic reticulum where the post-translational glycosylation occurs ( 25 ), may regulate the sialylation during the biosynthesis of glycan chains. And the cell-surface localized NEU4 isoform 2 may modify the mature glycan chains of the cell surface proteins or environmental glycans during the peritoneal dissemination of OC cells. Figure 2. NEU4 promotes the peritoneal dissemination of OC cells in mice. (A) Western blot analysis of NEU4 expression in indicated OC cell lines. (B) Western blot analysis of the knock-down efficiencies of shRNAs targeting mouse Neu4 (shNeu4) in ID8 cells. The shRNA targeting bacteria lacZ (shLacZ) was used as a negative control. (C) The schematic diagram of the syngeneic mouse model of OC by the intrabursal injection of ID8 cells into the C57BL/6 mice. (D) Representative images of C57BL/6 mice 75 days after the intrabursal injection of indicated ID8 cells and the quantification of the ascites (n = 3 ~ 6, *p < 0.05, ***p < 0.001, by unpaired Student’s t-test). (E) Representative images of peritoneal disseminated tumor nodules (indicated by white arrows) and the quantification results (n = 3 ~ 6, **p < 0.01, ***p < 0.001, by unpaired Student’s t-test). (F) The schematic diagram of the experimental dissemination mouse model of OC by the intraperitoneal (i.p.) injection of the luciferase-transfected ID8 cells into C57BL/6 mice. (G-I) Representative images of the peritoneal dissemination detected by live imaging ( G , left panel), ascites formation ( H , left panel) and disseminated tumor nodules ( I , left panel, indicated by white arrows) and the quantification results (n = 4 ~ 6, *p < 0.05 by unpaired Student’s t-test). (J) Western blot analysis of the V5-tagged NEU4 isoforms (iso1 ~ iso3) stably transfected into the Caov-3 cells and empty vector-transfected control (Vec) cells. (K) The schematic diagram of the murine xenograft model of OC by the intrabursal injection of human Caov-3 cells into NOD-SCID mice. (L) Images of Caov-3 tumor xenografts dissected from the primary tumor sites 120 days after the intrabursal injection (left panel) and the quantification results of the weights of xenografts (n = 4 ~ 5 per group, ns-not significant, by unpaired Student’s t-test). (M) Representative images of the peritoneal dissemination of cancer cells (indicated by white arrows) in mice inoculated with indicated Caov-3 cells and the quantification results (n = 4 ~ 5, **p < 0.01, *p < 0.05, ns-not significant, by unpaired Student’s t-test). NEU4 isoform 2 enhances the motility and EMT of OC cells To explore how NEU4 promotes the disseminated metastasis of OC cells, we first examined its roles in the regulation of cell proliferation. Knocking down Neu4 in mouse ID8 cells did not affect their proliferation and clone formation capacity when cultured in vitro ( Supplementary Fig. 3A and 3B ). And the ectopic expression of either isoform of NEU4 did not affect the proliferation of the in vitro cultured human HGSOC cell Caov-3 ( Supplementary Fig. 3C and 3D ) and another type of OC cell SK-OV-3 ( Supplementary Fig. 3E and 3F ) as well. We further investigated the roles of NEU4 in regulating cell mobility. Silencing Neu4 significantly attenuated the mobility and invasion of mouse ID8 cells shown by transwell cell migration assay ( Fig. 3A ) and the dynamic high-content imaging and analysis of the track and speed of each cell ( Fig. 3B ). Given the essential role of the EMT in the peritoneal dissemination of OC cells ( 27 ), we also examined the EMT markers in Neu4-silenced ID8 cells. As shown in Fig. 3C , either shRNA against Neu4, especially shNeu4 #3, was able to reduce the expression of mesenchymal marker genes, i.e., N-Cadherin (N-Cad), Vimentin (VIM) and Snail, while enhanced the expression of epithelial maker gene E-Cadherin (E-Cad), suggesting the potential role of Neu4 in maintaining the mesenchymal phenotype of OC cells. To further investigate the roles of different isoforms of NEU4 in the regulation of OC cell motility, we performed the transwell cell migration assay and the high-content imaging and analysis of single cell motility of human HGSOC cell Caov-3, in which the ectopically expressed different isoforms of NEU4 supplemented the endogenous NEU4 ( Fig. 3D ). The isoform 2 and isoform 3, but not isoform 1, significantly enhanced the migration and invasion ( Fig. 3E and 3F ) and promoted the EMT markers changing toward the mesenchymal phenotype ( Fig. 3D ), which were also observed in SK-OV-3 cells ( Fig. 3G and 3H ). Taken together, NEU4, especially the isoforms 2 and 3, is a potent EMT-promoting factor for OC cells. Figure 3. NEU4 promotes the motility and epithelial mesenchymal transition (EMT) of OC cells. (A) Transwell cell migration and invasion assays of ID8 cells transfected with shNeu4 or shLacZ as control. Data are shown as means ± SEM from three independent experiments (***P < 0.001, by unpaired Student’s t-test). (B) Dynamic imaging analysis of ID8 cell migration by the high-content cell imaging and analysis system. Representative images of single-cell ultimate displacement are shown in the left panels. The average speed of the cells and the mean square displacement (MSD) are shown as means ± SEM (**P < 0.01 by unpaired Student’s t-test). (C) Western blot analysis of EMT markers in ID8 cells transfected with indicated shRNAs. (D) Western blot analysis of the Caov-3 cells, in which all the endogenous NEU4 was replaced with ectopically expressed isoforms of NEU4. (E) Transwell cell migration and invasion assays of Caov-3 cells with reconstituted NEU4 isoforms (iso1 ~ iso3) or empty vector (Vec) as control (means ± SEM from three independent experiments, **P < 0.01, ***P < 0.001, ns-not significant, by unpaired Student’s t-test). (F) Dynamic imaging analysis of Caov-3 cell migration as described in (B) (means ± SEM from three independent experiments, **P < 0.01, ***P < 0.001, ns-not significant, by unpaired Student’s t-test). (G) The transwell cell migration and invasion assays of SK-OV-3 cells with the overexpression of NEU4 isoforms (means ± SEM from three independent experiments, ***P < 0.001, ns-not significant, by unpaired Student’s t-test). (H) Western blot analysis of EMT markers in SK-OV-3 cells. NEU4 attenuates the cell surface sialylation of OC Given NEU4 has been reported to be a sialidase that removes the terminal sialic acid residue from a broad spectrum of glycoconjugates ( 14 ), we investigated the activities of different NEU4 isoforms in modifying the cell surface sialylation. We utilized an analog of the natural sialic acid precursor N-acetylmannosamine (ManNAc), i.e., the tetra acetyl-N-azidoacetylmannosamine (AC 4 ManNAz), to metabolically label the sialylated glycoproteins and glycolipids with the fluorescent dye or biotin ( Fig. 4A ). Knocking down Neu4 in mouse ID8 cells greatly enhanced the whole sialylation levels when examined by confocal microscopy ( Fig. 4B ) and by western blot analysis of the whole cell lysates ( Fig. 4C ). In human Caov-3 and SK-OV-3 cells, ectopic expression of the isoform 2 or 3, but not isoform 1, dramatically decreased the cell surface sialylation level ( Fig. 4D ) and the whole sialylation level ( Fig. 4E and Supplementary Fig. 4A ). We further analyzed the correlation between the cellular sialylation level and NEU4 expression in the orthotopic murine models of OC established by the intrabursal transplantation of Caov-3 cells or ID8 cells into the ovarian bursas of mice. In the syngeneic murine model of OC by the intrabursal injection of ID8 cells, we collected the OC tissues from the primary OC allografts and detected the sialylation by lectins SNA ( Sambucus Nigra Lectin ), specifically recognizing α2, 6-linked sialic acid, and MAL II ( Maackia Amurensis Lectin ) specifically for α2, 3-linked sialic acid. We found that silencing Neu4 dramatically increased the protein sialylation ( Fig. 4F ). Consistently, in the Caov-3 xenograft murine model of OC, we found elevated expression of NEU4 in disseminated OC cells ( Fig. 4G ), which showed lower level of sialylation when compared with that of the primary OC tissues by either lectin blot ( Fig. 4G ) or the lectin histochemistry ( Fig. 4H ) analyses. In addition, the ectopic expression of isoform 2 or 3 of NEU4, but not isoform 1, significantly reduced the sialylation in Caov-3-derived OC xenografts ( Fig. 4I ). In eight human HGSOC tumor tissues, we also found that the disseminated OC cells with high NEU4 expression ( Fig. 1E-G ) exhibited lower sialylation level when compared with that in the primary tumor cells which had low NEU4 expression ( Fig. 4J ). Figure 4. NEU4 significantly decreases the sialylation level in OC. (A) The schematic diagram of metabolic labeling of the sialylated molecules with click chemistry. (B, C) Representative images of fluorescent labeling ( B ) and western blot (WB) analysis ( C ) of the sialylation in ID8 cells transfected with indicated shRNAs. The quantification results are shown as means ± SEM from three independent experiments (***P < 0.001, **P < 0.01, by unpaired Student’s t-test). (D, E) Representative images of fluorescent labeling ( D ) and WB analysis ( E ) of the sialylation in Caov-3 cells transfected with NEU4 isoforms or empty vector (Vec) as a control. The quantification results are shown as means ± SEM from three independent experiments (**P < 0.01, ns-not significant, by unpaired Student’s t-test). (F) Lectin blot analysis of the tumor allografts in the murine model of OC established by the intrabursal injection of ID8 cells transfected with Neu4 shRNAs. SNA and MAL II were used to detect the α2, 6- and α2, 3- sialylation, respectively. (G, H) Lectin blot ( G ) and Lectin histochemistry ( H ) analyses of the primary (Pri) and disseminated (Diss) tumor tissues in the orthotopic murine model of OC established by the intrabursal injection of Caov-3 cells (n = 4 ~ 5, **p < 0.01, ***p < 0.001, by unpaired Student’s t-test). (I) Lectin histochemistry analysis of the primary tumors tissues dissected from the orthotopic murine model of OC by the intrabursal injection of Caov-3 cells transfected with indicated NEU4 isoforms (n = 4 ~ 5, *p < 0.05, **p < 0.01, ***p < 0.001, by unpaired Student’ s t-test). (J) Lectin histochemistry analysis of the primary and paired disseminated tumor tissues dissected from human HGSOC patients (n = 8, **p < 0.01, ***p < 0.001, by paired Student’ s t-test). The analysis of the cell surface glycoproteome affected by NEU4 isoform 2 in OC cells Given the essential roles of terminal sialylation in the regulation of the cell-matrix and cell-cell communications, we hypothesized that the plasma membrane-localized NEU4 isoform 2 might affect the cell surface sialylation and hence the essential transmembrane signaling for the dissemination of OC cells. To identify the NEU4-regulated glycoproteome, we metabolically labeled OC cells with AC 4 ManNAz that allowed the subsequent conjugation of biotin on the sialylated proteins and their affinity purification by streptavidin-conjugated resins ( Fig. 5A ). We next performed SDS-PAGE to separate these proteins and selected the membrane proteins whose sialylation were able to be reduced by NEU4 isoform2 in Caov-3 and increased by silencing Neu4 in ID8 cells ( Fig. 5B ) for mass spectrometry identification ( Fig. 5C ). Among 2,014 proteins with over 3 unique peptides identified, we identified 615 plasma membrane proteins with different biological functions ranging from cell-cell adhesion, protein-binding modulator to transmembrane signal receptors ( Fig. 5D, Supplementary Table 4 ). We were specifically interested in the 21 transmembrane receptors that were able to transduce signals in multiple biological processes, such as angiogenesis, cell adherence and growth-stimulating pathways, that were essential for the dissemination, survival and colonization of cancer cells ( Fig. 5E ). These results suggested that NEU4 might modulate the sialylation of a variety of signaling receptors essential for a broad range of biological processes during the dissemination of OC cells. Figure 5. The identification of NEU4-regulated cell surface glycoproteins in OC cells. (A) The schematic diagram of the purification and identification of membrane proteins whose sialylation are affected by NEU4. (B, C) The membrane associated sialylated proteins in NEU4-iso2-overexpressed Caov-3 cells or Neu4-silenced ID8 cells were resolved by SDS-PAGE and stained by the coomassie blue to show NEU4-affected protein bands ( B ), which were cut out for mass spectrometry (MS) identification and gene ontology analysis ( C ). (D) The Reactome pathway enrichment analysis of NEU4-regulated candidate proteins by Metascape tool. (E) The signaling receptors that might be regulated by NEU4 iso2. NEU4 desialylates EGFR to enhance its activation in OC To get insights into the functional significance of NEU4-mediated desialylation in the regulation of the oncogenic signaling, we selected one important receptor EGFR that was highly activated during the dissemination of many cancers including OC ( 28 ). Ectopic expression of NEU4 isoforms 2 and 3, but not isoform 1, greatly attenuated the sialylation of EGFR, resulting in the hyper activation of EGFR shown by its autophosphorylation at tyrosine 1068 in Caov-3 cells ( Fig. 6A ) and SK-OV-3 cells ( Supplementary Fig. 4B ). In addition, the NEU4 isoform 2-triggered hyperphosphorylation of EGFR induced more activated downstream FAK, ERK and AKT signaling cascades in both cell lines ( Fig. 6B and Supplementary Fig. 4C ). In ID8 cells, silencing Neu4 inhibited the activation of EGFR and the downstream signaling cascades ( Fig. 6C ). In human OC tissues, we also found that the elevated NEU4 protein level tightly correlated with the autophosphorylation level of EGFR, but not with the total EGFR protein level ( Fig. 6D and 6E ), suggesting that NEU4-mediated desialylation of EGFR only affected its activation, but not its stability. Consistently, the tight positive correlations between NEU4 and the activation of EGFR downstream signaling molecules were also observed in human OC tissues ( Fig. 6F ). In Caov-3 and SK-OV-3 cells, we silenced the endogenous EGFR by 3’ UTR-specific shRNAs and reconstituted the cells with glycosylation-deficient EGFR with N175Q mutation, N196Q mutation or double mutations, given these two glycosites were getting much closer when EGFR dimerized by its ligand and the negatively charged sialic acid might interfere with the dimerization ( Fig. 6G and Supplementary Fig. 4D-F ). These mutations, especially N196Q and double mutations, greatly abolished the sialylation of EGFR and caused its hyper autophosphorylation ( Fig. 6H and Supplementary Fig. 4G ), indicating that sialylation of EGFR at N196 and N175 was able to inhibit the activation of EGFR. Figure 6. NEU4 desialylates the epithelial growth factor receptor (EGFR) to enhance its activation in OC. (A) Western blot and Lectin blot analyses of the immunoprecipitated (IP) EGFR from epithelial growth factor (EGF)-treated Caov-3 cells with ectopically expressed NEU4 isoforms. The protein samples were denatured using the loading buffers with (top panel) or without (bottom panel) β-mercaptoethanol (β-ME). (B) Western blot analysis of the EGFR downstream FAK, ERK and AKT signaling cascades in indicated Caov-3 cells treated with or without 20 ng/mL of EGF for 1 h. (C) Western blot analysis of the EGFR downstream signaling cascades in shNeu4- or shLacZ-transfected ID8 cells after 10 min-treatment with or without 5 ng/mL EGF. (D, E) Immunofluorescent staining of the primary and disseminated human OC tissues for the correlation analysis of NEU4 with total EGFR proteins ( D ) or with phosphorylated EGFR (p-EGFR) ( E ) (n = 10, by the spearman rank correlation test). (F) IHC analysis of the human OC tissues for the correlation between NEU4 with the activation of EGFR downstream FAK, ERK and AKT signaling molecules. (n = 10, by the spearman rank correlation test). (G) Western blot analysis of the knock-down efficiencies of endogenous EGFR-specific shRNAs targeting the 3’ UTR region (left panel) and the ectopically expressed EGFR mutants (right panel) in Caov-3 cells. (H) Western blot and Lectin blot analyses of immunoprecipitated EGFR in Caov-3 cells, whose endogenous EGFR was substituted with indicated EGFR mutants, for EGFR phosphorylation and sialylation. NEU4 isoform2 promotes the dissemination of OC cells partly through desialylating EGFR To evaluate the biological significance of NEU4-regulated desialylation of EGFR in the dissemination of OC, we inhibited the EGFR signaling by either RNAi ( Fig. 7A and 7B ) or by the specific EGFR inhibitor Osimertinib ( Fig. 7C and 7D ). Once the EGFR was blocked, NEU4 isoform 2 was no longer able to effectively enhance the mobility and invasion of Caov-3 cells ( Fig. 7A and 7C ). Consistently, NEU4 isoform2 did not promote the EMT of Caov-3 cells whose EGFR was silenced by shRNA ( Fig. 7B ) or inhibited by Osimertinib ( Fig. 7D ). We further explored the importance of EGFR in NEU4-regulated dissemination in Caov-3 cells whose endogenous EGFR was substituted with N196-glycosylation-deficient EGFR (N196Q) or wild type (WT) EGFR ( Fig. 6H ). In N196-glycosylation-deficient Caov-3 cells, both the EMT process ( Fig. 7E ) the EGFR downstream signaling cascades ( Fig. 7F ) were dramatically enhanced, where NEU4 isoform 2 could only slightly enhance some mesenchymal markers, such as N-cadherin and Snail ( Fig. 7E ), and EGFR downstream signaling cascades, such as ERK and AKT ( Fig. 7F ), suggesting that EGFR was an essential NEU4 downstream molecule that mediated the dissemination-promoting roles of NEU4 in OC. In conclusion, we identified NEU4-iso2 and NEU4-iso3 as new glycome regulators that promoted the peritoneal disseminated metastasis of OC. In addition, we revealed that NEU4-iso2 desialylated EGFR that resulted in the hyperphosphorylation of EGFR and its downstream signaling cascades to promote the peritoneal disseminated metastasis of OC ( Fig. 7G) . Figure 7. NEU4 isoform2 enhances the motility and EMT of OC cells partly through desialylating EGFR. (A) The transwell cell migration and invasion assays of Caov-3 cells transfected with NEU4 isoform 2 (NEU4-iso2) or empty vector (Vec) control when EGFR was silenced (shEGFR) or not (shLacZ) (ns-not significant, *P < 0.05, by unpaired Student’s t-test). (B) Western blot analysis of EMT markers in Caov-3 cells transfected with indicated plasmids. (C) The transwell cell migration and invasion assays of Caov-3 cells transfected with NEU4-iso2 or Vec with (+) or without (-) the treatment of 5 µM EGFR inhibitor Osimertinib (Osi). The quantification results in ( A ) and ( C ) are shown as means ± SEM from three independent experiments (ns-not significant, *P < 0.05, by unpaired Student’s t-test). (D) Western blot analysis of EMT markers in Caov-3 cells transfected with indicated plasmids and treated with or without 5 µM Osi. (E, F) Western blot analysis of EMT markers ( E ) and EGFR downstream signaling cascades ( F ) in Caov-3 cells whose endogenous EGFR was silenced by 3’ UTR-specific shRNA and substituted with the ectopically expressed wild type (EGFR WT ) or N196-glycosylation-deficient EGFR mutant (EGFR N196Q ). (G) The Schematic diagram summarizing the role of NEU4 in the desialylation of EGFR to remove the negative charges to promote the dimerization and hyperactivation of EGFR and the downstream dissemination-promoting signaling cascades. Discussion The roles of NEU4 in malignancies are variable in different cancer types. In colon cancer and hepatocellular carcinoma (HCC), NEU4 has been reported to be a tumor suppressor. NEU4 hydrolyzes the sialyl-Lewis antigens on colon cancer cells and results in attenuated E-selectin-mediated cell adhesion, motility and growth ( 29 ). And NEU4 expression is downregulated by hypoxia stimuli. They also observed the presence of some NEU4 proteins on the cell surface with unknown biologic functions ( 29 ). Significant down-regulation of NEU4 was also found in highly metastatic hepatocellular carcinoma (HCC), correlating with high grades and poor outcomes ( 30 ). NEU4 inhibits the motility and metastasis of HCC cells partially through removing the α2,3-linked sialic acid residues on CD44 to enhance its binding on the hyaluronic acid. On the contrary, the tumor-promoting function of NEU4 in neuroblastoma (NB) was also reported( 31 ). NEU4 promotes the proliferation, survival and stemness of NB cells through the hyperactivation of the Wnt/β-catenin signaling pathway particularly by NEU4 isoform 1 (also known as NEU4 long or NEU4L) ( 32 ). The functional diversity of NEU4 in malignancies may be caused by the presence of multiple, alternative splicing-generated isoforms of NEU4, which have different subcellular localizations. Previous studies of NEU4 are mostly focused on the mitochondrial- or lysosome-localized isoform 1 (NEU4L)( 23 , 24 ), while the roles of endoplasmic reticulum (ER)-localized isoform 3 (also known as NEU4S)( 25 ) and plasma membrane-localized isoform 2 of NEU4 are largely unexplored. Here, we identified both the isoform 2 and isoform 3 of NEU4 were key players in shaping the OC glylcome favoring the peritoneal dissemination. The ER-localized isoform 3 probably regulates the sialylation in the biosynthesis step of glycan chains and intrinsically influence the dissemination capacity of OC cells, while the cell-surface localized isoform 2 may instantly modify the glycan chains of the signaling receptors to adjust their reactivities in response to the environmental cues during the peritoneal dissemination of OC cells. It keeps an open question how different environmental stimuli regulate the alternative splicing of NEU4 in the disseminating OC cells to generate the isoform 2 and 3. In this study, we also identified membrane proteins whose sialylation was able to be modified by NEU4 in OC cells. To investigate the functional significance of NEU4-mediated desialylation, we chosed one tyrosine kinase receptor EGFR, whose amplification and overexpression were found in some epithelial OC patients and correlated with poorer clinical outcomes ( 33 ). We showed that NEU4-mediated desialylation at N196 and N175 greatly contributed to the phosphorylation and activation of EGFR, possibly by eliminating the repulsive force of negatively charged terminal sialic acid on N196 and N175-glycan chains that opposed the dimerization of EGFR. However, when substituted with glycosylation-deficient EGFR, NEU4 isoform 2 was still able to further enhance the activation of ERK and AKT ( Fig. 7F ), indicating that NEU4 regulated more oncogenic signaling receptors. In addition to EGFR, we also identified 20 transmembrane signaling receptors including insulin-like growth factor 1 receptor (IGF1R), ephrin type A receptor 2 (EphA2), epithelial discoidin domain containing receptor 1 and 2 (DDR1 and DDR2) ( Supplementary Table 4 ). The involvement of IGF1/IGF1R signaling in the invasion, angiogenesis and cell survival of OC cells by the activation the downstream MAPK/ERK and PI3K/AKT pathways has been reported( 34 ). As collagen-activated receptor tyrosine kinases, DDR1 and DDR2 regulate cell differentiation, proliferation, adhesion, migration, invasion, and matrix remodeling, whose dysregulation has also been found in ovarian cancer ( 35 , 36 ). And the ligand-independent EphA2 signaling by S897 phosphorylation has been implicated in the adaptive chemotherapy resistance of OC during dissemination ( 37 ). Therefore, the NEU4 might affect all these oncogenic signaling cascades by modifying the glycosylation of the key receptors. Materials and Methods Patient samples Eight paired OC patient samples (including primary tumors and disseminated tumor nodules) were obtained from Tianjin Center Hospital of Gynecology Obstetrics from December 2017 to December 2019. Half of each sample was kept in 4% paraformaldehyde for Hematoxylin-eosin (H&E) and immunohistochemistry (IHC) staining, the other half was lysed in TRIeasy™ LS Total RNA Extraction Reagent (Yeasen Biotechnology, Shanghai, China) for the extraction of protein and RNA. The human patient sample-based studies were performed in accordance with the ethics guidelines of the committee of Nankai University and Tianjin Center Hospital of Gynecology Obstetrics (2018KY032). All OC patients from Tianjin Center Hospital of Gynecology Obstetrics provided informed consents. Cell culture Caov-3 cell lines was purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). HEK 293T and SK-OV-3 cell lines were purchased from American Type Culture Collection (ATCC). ID8 cells were purchased from Merck (Darmstadt, Germany). The identities of all cell lines were confirmed by STR analysis. And all cells were confirmed to be mycoplasma negative by using the MycAway™ -Color One-Step Mycoplasma Detection Kit (Yeasen Biotechnology). The cells were maintained in an atmosphere of 5% CO 2 at 37°C in recommended medium. HEK 293T, Caov-3 and ID8 cells were cultured in Dulbecco’s Minimum Essential Medium (DMEM) (Biological Industries, Kibbutz Beit-Haemek, Isreal) supplemented with 10% of heat-inactivated fetal bovine serum (FBS) (Biological Industries), 2 mM of L-glutamine, 100 U/mL of penicillin and 100 µg/mL of streptomycin (Thermo-Fisher Scientific, Waltham, MA, USA). SK-OV-3 cells were cultured with McCoy's 5A Medium Modified. Plasmid construction and stable cell line establishment Three transcript variants of human NEU4 isoforms were cloned by RT-PCR using cDNA generated from total RNAs of Caov-3 cells. The coding sequences of NEU4 isoforms were cloned into the lentiviral plasmid pLV-EF1α-MCS-IRES-Bsd (Biosettia Inc., San Diego, CA, USA), with the V5 tag-encoding sequence was added at the C-terminal just before the stop codon. The sequences of the primers are listed in the Supplementary Table 5 . The shRNA target sites were selected by using the Thermo-Fisher RNAi Designer ( http://rnaidesigner.thermofisher.com/rnaiexpress/ ) and the shRNA templates were inserted into the lentiviral plasmid pLV-H1-EF1α-puro (Biosettia Inc.). The sequences of the shRNA templates are listed in the Supplementary Table 5 . All plasmids were further confirmed by DNA sequencing (Sangon, Beijing, China). The lentiviruses carrying these plasmids were produced in Lenti-293T cells (Biosettia Inc.) and were used to infect the OC cells at the multiplicity of infection (MOI) of 1.0 before the selection with 2.5-5.0 µg/mL of blasticidin (Sigma-Aldrich, Darmstadt, Germany) for the cells with the stable ectopic expression of target genes or 5 µg/mL of puromycin (Sigma-Aldrich) for the cells with stable knock-down of target genes. After 2-week-selection, the expression levels of target genes in the polyclonal cell populations were verified by western blot and quantitative RT-PCR analyses before the in vitro studies or inoculation into the mice. Protein extraction and western blot Cells were lysed by RIPA buffer [50 mM Tris-HCl pH 7.4, 150 mM NaCl, 1% (v/v) Triton X-100, 1% (w/v) sodium deoxycholate, 0.1% (w/v) SDS with protease inhibitor cocktail and phosphatase inhibitor cocktail (Sigma-Aldrich). The membrane proteins were extracted by using the Membrane Protein Extraction Kit (Bestbio, Nanjing, Jiangsu, China). The protein concentration was then measured by using the Pierce TM BCA Protein Assay Kit (Thermo-Fisher Scientific). Equal amount of proteins was resolved by SDS-PAGE (10%). The proteins in the gel were transferred onto the PVDF membrane and incubated with 5% defatted milk for 1 hour (h). The membrane was incubated with primary antibodies overnight at 4°C followed by the incubation with secondary antibodies (Proteintech, Wuhan, Hubei, China) for 1 h at room temperature. The primary antibodies used in this study are listed in the Supplementary Table 2 . Quantitative Real-time PCR The total RNAs were extracted from cells by using the TRIeasy™ LS Total RNA Extraction Reagent and reverse-transcribed into cDNA using M-MLV reverse transcriptase (Promega, Madison, WI, USA) according to the manufacturer’s protocol. Quantitative real time PCR analysis was performed by using SYBR Green SuperMix (Yeasen) on the Applied Biosystems-StepOnePlus Real-Time PCR system (Thermo-Fisher Scientific). The PCR program was as follows: initial denaturation at 95°C for 6 minutes (min), followed by 45 cycles of 95°C for 30 seconds (s) and 60°C for 45 s. And the relative expression of gene was calculated using the 2 −ΔΔCt method with β-actin gene as the normalization control. The primer sequences are listed in the Supplementary Table 5 . Hematoxylin and Eosin (H&E) and Immunohistochemistry (IHC) staining The tissue samples were fixed with 4% (w/v) paraformaldehyde (Sigma-Aldrich) and dehydrated before being put into wax cylinder overnight for paraffin wax embedding. After consecutive sectioning, tissue sections of 5-µm thickness were stained with hematoxylin and eosin (OriGene, Rockville, MD, USA) for pathological analysis. For IHC analysis, sections were treated with 3% (v/v) hydrogen peroxide for quenching the endogenous peroxidase and heated for 1 h for the retrieval of antigens. After blocked with 5% (v/v) goat serum, the sections were incubated with primary antibodies, biotin-conjugated secondary antibodies (Vector Laboratories, Newark, CA, USA), and streptavidin-HRP (Vector Laboratories). Finally, DAB substrate (Zsgb-Bio, Beijing, China) was dropped onto sections for less than 2 min and hematoxylin was used to stain the cell nuclei. The primary antibodies used in this study are listed in the Supplementary Table 6 . The H score was calculated by multiplying positively stained area (P) with staining intensity (I), where the degrees for P were 0–4 (0, < 5%; 1, 5–25%; 2, 25–50%;3, 50–75%; 4, 75–100%) and the degrees for I were 0–3 (0, none; 1, weak; 2, moderate; 3, strong). Immunofluorescent (IF) staining Cells were seeded on glass coverslips put in 24-well plate. The cells were washed once with cold phosphate buffered saline (PBS) and fixed by 4% paraformaldehyde for 10 min. Tissue sections were blocked with 5% goat serum before the sequential incubation with primary antibodies, secondary antibodies conjugated with Alexa Fluor-488 and Alexa Fluor-594 (Thermo-Fisher Scientific), and 4’,6-diamidino-2-phenylindole (DAPI) (Sigma-Aldrich) for the nuclear counterstain. Images were taken by using an Olympus FV1000 confocal microscope (Tokyo, Japan). The primary antibodies used in this study are listed in the Supplementary Table 6 . Cell migration and invasion assay The migration and invasion assays were carried with the Corning® Transwell® chambers with 8-µm pore size (Merck). For invasion assay, the transwell chambers were additionally coated with Matrigel™ basement membrane matrix (BD Biosciences, San Jose, CA, USA). Ten thousand of ID8 cells or 5×10 4 Caov-3 cells were mixed with FBS-free medium and added into top chambers. 600 µL medium complemented with 10% FBS was added to the lower chamber. To analyze the EGFR signaling, cells were seeded in the top chambers with 5 µM Osimertinib (MedChemExpress, Shanghai, China) added in FBS-free medium. After 6 h for ID8 or 18 h for Caov-3, cells migrated to the bottom surface of the transwell membranes were fixed in 4% paraformaldehyde and stained with crystal violet (Beyotime, Shanghai, China). Cell proliferation assay 1 × 10 5 ID8 cells or 5 × 10 5 Caov-3 cells were seeded in 6-well plates and maintained at 37°C in the incubator. Cells were detached by trypsinization with 0.25% trypsin–0.02% ethylenediaminetetraacetic acid (EDTA) (Biological Industries) and counted every 24 h by using Countess 2 Automated Cell Counter (Thermo-Fisher Scientific) until the cells reached full confluency. The cell number at each time point was the average number from three wells. Colony formation assay 1× 10 3 ID8 cells or 5× 10 3 Caov-3 cells were seeded in 6-well plate and maintained at 37°C in incubator until colonies could be identified. The colonies on the plates were fixed in 4% paraformaldehyde for 1 h at 4°C, then stained with crystal violet (C0121, Beyotime). The number of colonies was counted to evaluate the proliferation of cells. Lectin Blot and Lectin histochemistry Total proteins were extracted from cells by using RIPA buffer and quantified by using the Pierce TM BCA Protein Assay Kit. The proteins were then resolved by SDS-PAGE and transferred on to PVDF membranes. After the PVDF membranes were blocked with 2.5% (w/g) oxidized bovine serum albumin (oBSA) for 4 h, the proteins with α2,6- or α2,3- sialylation modification were detected by the incubation with biotinylated Sambucus Nigra Lectin (SNA) or Maackia Amurensis Lectin (MAL II) (Vector Laboratories) for 1 h, respectively. Sialylated proteins were visualized by the incubation with HRP-conjugated streptavidin (Thermo-Fisher Scientific) for 1 h followed by ECL enhanced chemiluminescence reagents (EpiZyme Scientific, Shanghai, China). The Coomassie Brilliant Blue (CBB) staining of the membranes were used to estimate the total protein amounts. oBSA was prepared as follows: BSA (Solarbio, Beijing, China) was dissolved in 10 mM sodium metaperiodate (Aladdin, Shanghai, China) and maintained at 4 ˚C for 1 h before the dialysis against TBS-T buffer for 16 h at 4˚C to remove the sodium metaperiodate. All lectins were dissolved in 0.5% oBSA. For lectin histochemistry, 5-µm sections were blocked with 2.5% oBSA for 3 h and incubated with biotinylated SNA or MAL II overnight. After washing with TBST-T buffer for three times, sections were incubated with HRP-conjugated streptavidin for 1 h and visualized with 3,3’-diaminobenzidine (DAB) substrate (ZSGB-BIO, Beijing, China). The quantification was performed by the H score method calculated as described above in IHC segment. The murine models of OC All animal studies and procedures were approved by the Nankai University Animal Care and Use Committee. For the orthotopic murine model of OC, 5 × 10 6 human Caov-3 cells (stably transfected with NEU4 isoforms or empty vector as the control) or murine ID8 cells (stably transfected with shNeu4 or shLacZ as a control) were suspended in 20 µL PBS and implanted into the surgically exposed right ovarian bursa of anaesthetized 6-week-old, female NOD-SCID mice or C57BL/6 mice (SPF Biotechnology, Beijing, China) by intrabursal injection. 75 days after ID8 injection or 120 days after Caov-3 injection, mice were sacrificed for OC development analysis. For the experimental dissemination murine model of OC, 1 × 10 7 ID8 cells were suspended in 100 µL PBS and implanted into the peritoneal cavity of C57BL/6 mice by intraperitoneal injection. Then mice were sacrificed at day 60 after the injection. The ascites, primary tumors and peritoneal disseminated tumor nodules were collected. The CRISPR-Cas9 library screen in the murine model of OC The CRISPR/Cas9 library containing 2,984 lentivirus-based sgRNA plasmids against 498 protein glycosylation–related genes (4 ~ 6 sgRNAs per gene) and 1,000 non-targeting control sgRNA plasmids were selected from the genome-wide CRISPR/Cas9 library (GeCKO-v2.0) made by Feng Zhang’s laboratory ( 19 , 20 ) ( Supplementary Table 3 ). The amplification and lentivirus preparation of CRIPSR-Cas9 library were performed as described by Feng Zhang. For the in vivo screen, a total of 1×10 7 of SK-OV-3 cells were infected in the presence of 8 µg/mL polybrene (Sigma-Aldrich) with the lentiviral particles at a multiplicity of infection of 0.3 to avoid multiple gene knock outs in one cell. The infected cells were selected by using 5 µg/mL puromycin (Sigma-Aldrich) after 72 h. After 14-day in vitro selection and expansion, a total of 2×10 6 of SK-OV-3 cells were suspended in 20 µL PBS and implanted into the surgically exposed right ovarian bursa of anaesthetized 6-week-old, female NOD-SCID mice (SPF Biotechnology, Beijing, China) by intrabursal injection (n = 3). Around 40 days post-implantation, the tumor tissues from the primary sites and peritoneal disseminated sites (including the tumor nodules on the abdominal wall, the out surfaces of the organs in the peritoneal cavity, and the malignant ascites) were dissected from all the mice. The tumor tissues were then minced and further dissociated with type I collagenase (Sigma-Aldrich) for in vitro culture and expansion in the presence of 5 µg/mL puromycin for two weeks. The expanded cells from all the disseminated sites of different mice were pooled for the second round of intrabursal injection into three NOD-SCID mice (2×10 6 SK-OV-3 cells per mouse). Around 30 d post-implantation, we sacrificed the mice and collected tumor cells for in vitro expansion and the third round of in vivo screen. After three rounds of screen, the pooled peritoneal disseminated tumor cells from the mice in third round screen and the pooled cells from the primary tumor sites in the first-round screen were collected separately for genomic DNA extraction, followed by PCR amplification of the sgRNA-coding region and deep DNA sequencing analysis as described by Feng Zhang ( 19 ). Live imaging of the OC progression 50 days after the intraperitoneal injection of ID8 cells into the peritoneal cavity, C57BL/6 mice were anaesthetized and 100 µL D-luciferin (Sigma-Aldrich) was injected intraperitoneally. Bioluminescence images was acquired by using the IVIS Lumina II Imager (Caliper Life Sciences, Hopkinton, MA, USA). Metabolic labeling of sialylated proteins by AC 4 ManNAz Cells were cultured in indicated medium with 50 µM of tetra acetyl-N-azidoacetylmannosamine (AC 4 ManNAz) for 72 h at 37°C. For the detection of sialylated-proteins, cells were lysed with RIPA buffer and the protein concentrations were adjusted to 5 µg/µL. 20 µL of the cell lysates (~ 100 µg of total proteins) were incubated with 2 µL of 5 mM EZ-Link® Phosphine-PEG3-Biotin (Thermo-Fisher Scientific) at 37°C for 2–4 h. After the addition of SDS loading buffer and heat denaturation, the proteins were resolved by SDS-PAGE and transferred onto PVDF membranes. The membranes were blocked with 5% defatted milk for 1 h at room temperature and incubated with HRP-conjugated streptavidin (Thermo-Fisher Scientific). For the fluorescent labeling of sialylated-proteins in cells, cells were cultured in the medium containing 50 µM of DBCO-Cy5 (Sigma-Aldrich) for 1 h at 37°C after the 72-hour incubation with AC 4 ManNAz. Identification of sialylated proteins After the metabolic labeling of sialylated proteins in cultured cells with AC 4 ManNAz as described above, the membrane proteins were extracted by using the Mem-PER Plus Membrane Protein Extraction Kit (Thermo-Fisher Scientific) and the concentration was adjusted to 5 µg/µL. The membrane proteins were then incubated with Phosphine-PEG3-Biotin for 4 h to conjugate the biotin. Subsequently, the pre-washed streptavidin agarose beads (Yeasen Biotechnology) were added to the conjugates and incubated overnight at 4°C with rotation. After washed with the wash buffer (150 mM NaCl, 20 mM NaH 2 PO 4 , pH 7.4) three times, the biotin-conjugated membrane proteins were eluted with the elution buffer (0.1M glycine-HCl, pH 2.8) and resolved by SDS-PAGE. The protein bands were visualized by Coomassie blue staining before the identification by mass spectrometry analysis (LC-MS/MS) (BGI, Beijing, China). Immunoprecipitation Cell lysates were incubated with the EGFR antibody ( Supplementary Table 6 ) for 4 h before the addition of pre-washed protein G agarose beads (CWBIO, Beijing, China) overnight at 4°C with rotation. After three washes, the proteins bond to the beads was eluted by boiling in 1× SDS loading buffer and resolved by SDS-PAGE for western blot to analyze the sialylation of EGFR. Statistical analysis Prism 8.0 software was used for statistical analysis. Quantitative data were presented as means ± SEM, and the differences between the groups were analyzed using the Student’s t-test. Two-way ANOVA test was used to analyze the continuous variables. Pearson correlation analysis was performed to examine the relation between the level of p-EGFR and the NEU4 expression. Differences are considered statistically significant at *p < 0.05; **p < 0.01; ***p < 0.001; ns means not significant. Declarations Author Contributions: J.S., J. L., Y. S., P. Q., L. W. and S. Y. conceived the project. J.S., R. Z., Y. L., B.T., and S.W. performed experiments. Y. L., Y. C, J. Y., T.H., Y. M, S.W., and X. S. helped with the design of experiments. J.S. performed the data analysis. Y. S. supervised the study. J.S., J. L., L. W., and Y. S. wrote the manuscript with comments from all authors. Competing Interest Statement: The authors declare no potential conflict of interests. Acknowledgements This work was supported by the grants from the National Natural Science Foundation of China [32070752, 32200641] and Nankai University Undergraduate Innovation and Entrepreneurship Project [202110055083]. Data Availability The data generated in this study are available upon request from the corresponding authors. References Bast RC, Jr., Hennessy B, & Mills GB (2009) The biology of ovarian cancer: new opportunities for translation. Nat Rev Cancer 9(6):415–428. Cortes-Guiral D, et al. (2021) Primary and metastatic peritoneal surface malignancies. Nat Rev Dis Primers 7(1):91. Tan DS, Agarwal R, & Kaye SB (2006) Mechanisms of transcoelomic metastasis in ovarian cancer. Lancet Oncol 7(11):925–934. Mereiter S, Balmana M, Campos D, Gomes J, & Reis CA (2019) Glycosylation in the Era of Cancer-Targeted Therapy: Where Are We Heading? Cancer Cell 36(1):6–16. Christiansen MN, et al. (2014) Cell surface protein glycosylation in cancer. Proteomics 14(4–5):525–546. Schjoldager KT, Narimatsu Y, Joshi HJ, & Clausen H (2020) Global view of human protein glycosylation pathways and functions. Nat Rev Mol Cell Biol 21(12):729–749. Esposito M, et al. (2019) Bone vascular niche E-selectin induces mesenchymal-epithelial transition and Wnt activation in cancer cells to promote bone metastasis. Nat Cell Biol 21(5):627–639. Barthel SR, et al. (2009) Alpha 1,3 fucosyltransferases are master regulators of prostate cancer cell trafficking. Proc Natl Acad Sci U S A 106(46):19491–19496. RodrIguez E, Schetters STT, & van Kooyk Y (2018) The tumour glyco-code as a novel immune checkpoint for immunotherapy. Nat Rev Immunol 18(3):204–211. Marcos NT, et al. (2004) Role of the human ST6GalNAc-I and ST6GalNAc-II in the synthesis of the cancer-associated sialyl-Tn antigen. Cancer Res 64(19):7050–7057. van de Wall S, Santegoets KCM, van Houtum EJH, Bull C, & Adema GJ (2020) Sialoglycans and Siglecs Can Shape the Tumor Immune Microenvironment. Trends Immunol 41(4):274–285. Hu Y, et al. (2020) Integrated Proteomic and Glycoproteomic Characterization of Human High-Grade Serous Ovarian Carcinoma. Cell Rep 33(3):108276. Pan J, et al. (2020) Glycoproteomics-based signatures for tumor subtyping and clinical outcome prediction of high-grade serous ovarian cancer. Nat Commun 11(1):6139. Miyagi T, Takahashi K, Hata K, Shiozaki K, & Yamaguchi K (2012) Sialidase significance for cancer progression. Glycoconj J 29(8–9):567–577. Pshezhetsky AV & Ashmarina LI (2013) Desialylation of surface receptors as a new dimension in cell signaling. Biochemistry (Mosc) 78(7):736–745. Kakugawa Y, et al. (2002) Up-regulation of plasma membrane-associated ganglioside sialidase (Neu3) in human colon cancer and its involvement in apoptosis suppression. Proc Natl Acad Sci U S A 99(16):10718–10723. Miyagi T, Wada T, & Yamaguchi K (2008) Roles of plasma membrane-associated sialidase NEU3 in human cancers. Biochim Biophys Acta 1780(3):532–537. Uemura T, et al. (2009) Contribution of sialidase NEU1 to suppression of metastasis of human colon cancer cells through desialylation of integrin beta4. Oncogene 28(9):1218–1229. Sanjana NE, Shalem O, & Zhang F (2014) Improved vectors and genome-wide libraries for CRISPR screening. Nat Methods 11(8):783–784. Shalem O, et al. (2014) Genome-scale CRISPR-Cas9 knockout screening in human cells. Science 343(6166):84–87. Berek JS, Renz M, Kehoe S, Kumar L, & Friedlander M (2021) Cancer of the ovary, fallopian tube, and peritoneum: 2021 update. Int J Gynaecol Obstet 155 Suppl 1:61–85. Varki A (2008) Sialic acids in human health and disease. Trends Mol Med 14(8):351–360. Seyrantepe V, et al. (2004) Neu4, a novel human lysosomal lumen sialidase, confers normal phenotype to sialidosis and galactosialidosis cells. J Biol Chem 279(35):37021–37029. Yamaguchi K, et al. (2005) Evidence for mitochondrial localization of a novel human sialidase (NEU4). Biochem J 390(Pt 1):85–93. Bigi A, et al. (2010) Human sialidase NEU4 long and short are extrinsic proteins bound to outer mitochondrial membrane and the endoplasmic reticulum, respectively. Glycobiology 20(2):148–157. Domcke S, Sinha R, Levine DA, Sander C, & Schultz N (2013) Evaluating cell lines as tumour models by comparison of genomic profiles. Nat Commun 4:2126. Vergara D, et al. (2010) Epithelial-mesenchymal transition in ovarian cancer. Cancer Lett 291(1):59–66. Sheng Q & Liu J (2011) The therapeutic potential of targeting the EGFR family in epithelial ovarian cancer. Br J Cancer 104(8):1241–1245. Shiozaki K, Yamaguchi K, Takahashi K, Moriya S, & Miyagi T (2011) Regulation of sialyl Lewis antigen expression in colon cancer cells by sialidase NEU4. J Biol Chem 286(24):21052–21061. Zhang X, et al. (2021) NEU4 inhibits motility of HCC cells by cleaving sialic acids on CD44. Oncogene 40(35):5427–5440. Silvestri I, et al. (2014) Sialidase NEU4 is involved in glioblastoma stem cell survival. Cell Death Dis 5:e1381. Tringali C, et al. (2012) NEU4L sialidase overexpression promotes beta-catenin signaling in neuroblastoma cells, enhancing stem-like malignant cell growth. Int J Cancer 131(8):1768–1778. Siwak DR, et al. (2010) Targeting the epidermal growth factor receptor in epithelial ovarian cancer: current knowledge and future challenges. J Oncol 2010:568938. Shao M, et al. (2012) Targeting the Insulin Growth Factor and the Vascular Endothelial Growth Factor Pathways in Ovarian Cancer. Molecular Cancer Therapeutics 11(7):1576–1586. Bali; James G. Kench; Lyndal S. Edwards; Patricia M. Vanden Bergh; Neville F. Hacker; Robert L. Sutherland; Philippa M. O’Brien VAH-SMG-GSMHJSRASMJDMHLHKA (2004) . Clin Cancer Res 10(13): 4427–4436. Quan J, Yahata T, Adachi S, Yoshihara K, & Tanaka K (2011) Identification of Receptor Tyrosine Kinase, Discoidin Domain Receptor 1 (DDR1), as a Potential Biomarker for Serous Ovarian Cancer. International Journal of Molecular Sciences 12(2):971–982. Moyano-Galceran L, et al. (2020) Adaptive RSK-EphA2-GPRC5A signaling switch triggers chemotherapy resistance in ovarian cancer. EMBO Mol Med 12(4):e11177. Additional Declarations There is NO conflict of interest to disclose. Supplementary Files FigS1.tif FigS2.tif FigS3.tif FigS4.tif SupplementaryTable1.xlsx SupplementaryTable2.xlsx SupplementaryTable3.xlsx SupplementaryTable4.xlsx SupplementaryTable5.xlsx SupplementaryTable6.xlsx Cite Share Download PDF Status: Published Journal Publication published 14 Oct, 2024 Read the published version in Oncogene → Version 1 posted Editorial decision: revise 29 Jan, 2024 Review # 2 received at journal 23 Jan, 2024 Reviewer # 2 agreed at journal 30 Dec, 2023 Review # 1 received at journal 29 Dec, 2023 Reviewer # 1 agreed at journal 28 Dec, 2023 Reviewers invited by journal 28 Dec, 2023 Submission checks completed at journal 19 Dec, 2023 Editor assigned by journal 18 Dec, 2023 First submitted to journal 18 Dec, 2023 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3772327","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":264079205,"identity":"6beca94f-57cb-439f-aa67-4e80e265d069","order_by":0,"name":"Long Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3UlEQVRIie3NMQrCMBSA4VcEu+QAOohXiBSqIOJV8hDsUsTRwSFTJ8ULiGcICIJbIKBLSlcHB13EwaEnECOCY4ybYP4hvAfvIwA+3692mpiHAEiAiiNh+muC2YuAE6H7XF1xdWy2iTpLmHSRh7m0Ez0adnBzaW3nGZWgE+RkxKwklmlMcaMCUQCVQaaQ1wi1k+JmyFL1RRGWMri7kEManZArFPnM/MIdSP9wi4Ht1EBoPZZsl0QZSe2kvkijspyqntDJ2gzdxiLUdmKq1t4je66f7k2V0uHI5/P5/rkHh1lPKlKUW/MAAAAASUVORK5CYII=","orcid":"","institution":"Nankai university","correspondingAuthor":true,"prefix":"","firstName":"Long","middleName":"","lastName":"Wang","suffix":""},{"id":264079206,"identity":"981e7c17-c3d7-45d3-8850-f0aeffeab770","order_by":1,"name":"Jie Shi","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Jie","middleName":"","lastName":"Shi","suffix":""},{"id":264079207,"identity":"0841680e-e4ac-4842-b9a6-474033f7d169","order_by":2,"name":"Rui Zhou","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Rui","middleName":"","lastName":"Zhou","suffix":""},{"id":264079208,"identity":"a3f81e13-21dc-45d3-b1dc-46b6081523c3","order_by":3,"name":"Shuo Wang","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Shuo","middleName":"","lastName":"Wang","suffix":""},{"id":264079209,"identity":"f8085631-21bc-42bf-81a4-2a66910fad70","order_by":4,"name":"Yuxin Liu","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Yuxin","middleName":"","lastName":"Liu","suffix":""},{"id":264079210,"identity":"e78d7f26-6afa-4d3a-ba1e-65901afb1095","order_by":5,"name":"Baorui Tian","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Baorui","middleName":"","lastName":"Tian","suffix":""},{"id":264079211,"identity":"44abf8ce-6a6a-4b44-bcee-c53ea0f85884","order_by":6,"name":"Yanhua Liu","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Yanhua","middleName":"","lastName":"Liu","suffix":""},{"id":264079212,"identity":"aceda8da-28e9-44b0-821d-0f9a761b1558","order_by":7,"name":"Yanan Chen","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Yanan","middleName":"","lastName":"Chen","suffix":""},{"id":264079213,"identity":"048c1914-bc7a-49bd-aa32-451b8de275b1","order_by":8,"name":"Taoyu Hu","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Taoyu","middleName":"","lastName":"Hu","suffix":""},{"id":264079214,"identity":"6ffeaa4b-e34d-41f7-a0ba-58cbc8324451","order_by":9,"name":"Yuhao Mu","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Yuhao","middleName":"","lastName":"Mu","suffix":""},{"id":264079215,"identity":"9123c420-abb0-4d8f-aff7-f0544c7c149b","order_by":10,"name":"Shufan Wang","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Shufan","middleName":"","lastName":"Wang","suffix":""},{"id":264079216,"identity":"f2a4c506-289e-47e1-9019-cba1bfd5bd88","order_by":11,"name":"Xintao Shao","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Xintao","middleName":"","lastName":"Shao","suffix":""},{"id":264079217,"identity":"984d0ad7-4400-4d6c-9359-860fb1a5af03","order_by":12,"name":"Jie Yan","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Jie","middleName":"","lastName":"Yan","suffix":""},{"id":264079218,"identity":"2ef909b4-a505-4fbc-b62d-12a4ede04edb","order_by":13,"name":"Pengpeng Qu","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Pengpeng","middleName":"","lastName":"Qu","suffix":""},{"id":264079219,"identity":"a1e5177b-9727-47f7-843b-bed31ffd14ef","order_by":14,"name":"Ding Wei","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Ding","middleName":"","lastName":"Wei","suffix":""},{"id":264079220,"identity":"a5c0681d-2852-4dfa-b05a-2ff6f9af8971","order_by":15,"name":"Shuang Yang","email":"","orcid":"https://orcid.org/0000-0002-4779-8553","institution":"Medical College of Nankai University","correspondingAuthor":false,"prefix":"","firstName":"Shuang","middleName":"","lastName":"Yang","suffix":""},{"id":264079221,"identity":"3ea3149e-7c31-42a5-82a6-70cf18ca60da","order_by":16,"name":"Yi Shi","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Yi","middleName":"","lastName":"Shi","suffix":""},{"id":264079222,"identity":"e06e9366-955d-4200-9c02-3062d86f3645","order_by":17,"name":"Jia Li","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Jia","middleName":"","lastName":"Li","suffix":""}],"badges":[],"createdAt":"2023-12-18 14:07:22","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3772327/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3772327/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41388-024-03187-x","type":"published","date":"2024-10-14T04:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":49093678,"identity":"25b174e3-00b4-46e4-83ce-2890e44aea42","added_by":"auto","created_at":"2024-01-03 02:17:50","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":586728,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eHigh-throughput CRISPR/Cas9 screen identified NEU4 as a key regulator of OC peritoneal dissemination.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A) \u003c/strong\u003eThe schematic diagram of CRISPR/Cas9 library-based screen of key glycosylation-related genes regulating peritoneal dissemination in the orthotopic murine model of OC. \u003cstrong\u003e(B)\u003c/strong\u003e Scatterplot showing the enrichment of specific sgRNAs in primary (Pri) or peritoneal disseminated (Diss) OC cells.\u003cstrong\u003e (C)\u003c/strong\u003e Kaplan-Meier survival curve to show the overall survival (OS) of OC patients with different NEU4 expression (n = 373) from the TCGA database. \u003cstrong\u003e(D)\u003c/strong\u003eThe different amino acid sequences of three NEU4 isoforms and the sites of specific qRT-PCR primer sets. The Primer 1 was designed at the common encoding regions of NEU4 isoforms. The additional N-terminal amino acid residues of isoform 1 (iso1) and 2 (iso2) were presented with the differential residue of iso2 highlighted, where the specific forward primer of iso2 bound (Primer 2). \u003cstrong\u003e(E)\u003c/strong\u003e qRT-PCR analysis of the total NEU4 transcripts in the primary and the paired peritoneal disseminated OC tissues in each OC patients (the left bargraph, data are plotted as means ± SEM from three independent measurements, **p\u0026lt;0.01, ***p\u0026lt;0.001, ns-not significant, by unpaired Student’s t-test) and the statistical analysis of NEU4 transcripts in eight OC patients ( the right bargraph, ***p\u0026lt;0.001, by paired Student’s t-test). \u003cstrong\u003e(F)\u003c/strong\u003eWestern blot analysis of NEU4 in the primary and the paired disseminated OC tissues from eight OC patients and the quantification results (n=8, **p\u0026lt;0.01, by paired Student’s t-test).\u003cstrong\u003e(G)\u003c/strong\u003e Representative images of the immunohistochemical staining of NEU4 in the primary and disseminated OC tissues from an OC patient and the H-score quantification results (n=8, ***p\u0026lt;0.001, by paired Student’s t-test). \u003cstrong\u003e(H) \u003c/strong\u003eImmunofluorescent staining of v5-tagged NEU4 isoforms ectopically expressed in Caov-3 cells. NEU4 isoforms are stained by anti-v5 antibody in green, while the actin filaments (by phalloidin) and lysosome marker LAMP1 in red. \u003cstrong\u003e(I)\u003c/strong\u003e qRT-PCR analysis of NEU4 isoform 2 (NEU4-iso2) in the primary and paired disseminated OC tissues in each OC patient (left bargraph, means ± SEM from three independent measurements, *p\u0026lt;0.05, **p\u0026lt;0.01, ***p\u0026lt;0.001, by unpaired Student’s t-test) and the statistical analysis of NEU4-iso2 expression in eight OC patients (n=8, ***p\u0026lt;0.001, by paired Student’s t-test).\u003c/p\u003e","description":"","filename":"Fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/4e4c467013c4c1914088a43f.png"},{"id":49092930,"identity":"3dce37bc-abca-4d41-8c86-57f7117a39ca","added_by":"auto","created_at":"2024-01-03 02:09:50","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":702294,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eNEU4 promotes the peritoneal dissemination of OC cells in mice. (A)\u003c/strong\u003e Western blot analysis of NEU4 expression in indicated OC cell lines. \u003cstrong\u003e(B) \u003c/strong\u003eWestern blot analysis of the knock-down efficiencies of shRNAs targeting mouse Neu4 (shNeu4) in ID8 cells. The shRNA targeting bacteria lacZ (shLacZ) was used as a negative control.\u003cstrong\u003e(C)\u003c/strong\u003e The schematic diagram of the syngeneic mouse model of OC by the intrabursal injection of ID8 cells into the C57BL/6 mice. \u003cstrong\u003e(D)\u003c/strong\u003e Representative images of C57BL/6 mice 75 days after the intrabursal injection of indicated ID8 cells and the quantification of the ascites (n=3~6, *p\u0026lt;0.05, ***p\u0026lt;0.001, by unpaired Student’s t-test). \u003cstrong\u003e(E)\u003c/strong\u003e Representative images of peritoneal disseminated tumor nodules (indicated by white arrows) and the quantification results (n=3~6, **p\u0026lt;0.01, ***p\u0026lt;0.001, by unpaired Student’s t-test). \u003cstrong\u003e(F) \u003c/strong\u003eThe schematic diagram of the experimental dissemination mouse model of OC by the intraperitoneal (i.p.) injection of the luciferase-transfected ID8 cells into C57BL/6 mice. \u003cstrong\u003e(G-I)\u003c/strong\u003e Representative images of the peritoneal dissemination detected by live imaging (\u003cstrong\u003eG\u003c/strong\u003e, left panel), ascites formation (\u003cstrong\u003eH\u003c/strong\u003e, left panel) and disseminated tumor nodules (\u003cstrong\u003eI\u003c/strong\u003e, left panel, indicated by white arrows) and the quantification results (n=4~6, *p\u0026lt;0.05 by unpaired Student’s t-test).\u003cstrong\u003e(J)\u003c/strong\u003e Western blot analysis of the V5-tagged NEU4 isoforms (iso1~iso3) stably transfected into the Caov-3 cells and empty vector-transfected control (Vec) cells. \u003cstrong\u003e(K)\u003c/strong\u003e The schematic diagram of the murine xenograft model of OC by the intrabursal injection of human Caov-3 cells into NOD-SCID mice. \u003cstrong\u003e(L)\u003c/strong\u003e Images of Caov-3 tumor xenografts dissected from the primary tumor sites 120 days after the intrabursal injection (left panel) and the quantification results of the weights of xenografts (n=4~5 per group, ns-not significant, by unpaired Student’s t-test). \u003cstrong\u003e(M)\u003c/strong\u003eRepresentative images of the peritoneal dissemination of cancer cells (indicated by white arrows) in mice inoculated with indicated Caov-3 cells and the quantification results (n=4~5, **p\u0026lt;0.01, *p\u0026lt;0.05, ns-not significant, by unpaired Student’s t-test).\u003c/p\u003e","description":"","filename":"Fig2.png","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/15ef50585d99cfbd101d871e.png"},{"id":49092926,"identity":"120f89c1-8e1c-4277-af6f-d308176c55bf","added_by":"auto","created_at":"2024-01-03 02:09:50","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":656198,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eNEU4 promotes the motility and epithelial mesenchymal transition (EMT) of OC cells. (A)\u003c/strong\u003eTranswell cell migration and invasion assays of ID8 cells transfected with shNeu4 or shLacZ as control. Data are shown as means ± SEM from three independent experiments (***P\u0026lt;0.001, by unpaired Student’s t-test). \u003cstrong\u003e(B)\u003c/strong\u003e Dynamic imaging analysis of ID8 cell migration by the high-content cell imaging and analysis system. Representative images of single-cell ultimate displacement are shown in the left panels. The average speed of the cells and the mean square displacement (MSD) are shown as means ± SEM (**P \u0026lt; 0.01 by unpaired Student’s t-test). \u003cstrong\u003e(C)\u003c/strong\u003e Western blot analysis of EMT markers in ID8 cells transfected with indicated shRNAs. \u003cstrong\u003e(D)\u003c/strong\u003e Western blot analysis of the Caov-3 cells, in which all the endogenous NEU4 was replaced with ectopically expressed isoforms of NEU4.\u003cstrong\u003e (E) \u003c/strong\u003eTranswell cell migration and invasion assays of Caov-3 cells with reconstituted NEU4 isoforms (iso1~iso3) or empty vector (Vec) as control (means ± SEM from three independent experiments, **P \u0026lt; 0.01, ***P \u0026lt; 0.001，ns-not significant, by unpaired Student’s t-test). \u003cstrong\u003e(F)\u003c/strong\u003e Dynamic imaging analysis of Caov-3 cell migration as described in (B) (means ± SEM from three independent experiments, **P \u0026lt; 0.01, ***P \u0026lt; 0.001, ns-not significant, by unpaired Student’s t-test).\u003cstrong\u003e(G)\u003c/strong\u003e The transwell cell migration and invasion assays of SK-OV-3 cells with the overexpression of NEU4 isoforms (means ± SEM from three independent experiments, ***P \u0026lt; 0.001, ns-not significant, by unpaired Student’s t-test). \u003cstrong\u003e(H)\u003c/strong\u003e Western blot analysis of EMT markers in SK-OV-3 cells.\u003c/p\u003e","description":"","filename":"Fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/9937386ee04839b8cfea2473.png"},{"id":49094074,"identity":"07bf4bfc-6c6f-4597-8ed1-c39aeb760031","added_by":"auto","created_at":"2024-01-03 02:33:50","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1115526,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eNEU4 significantly decreases the sialylation level in OC. (A)\u003c/strong\u003e The schematic diagram of metabolic labeling of the sialylated molecules with click chemistry. \u003cstrong\u003e(B, C)\u003c/strong\u003e Representative images of fluorescent labeling (\u003cstrong\u003eB\u003c/strong\u003e) and western blot (WB) analysis (\u003cstrong\u003eC\u003c/strong\u003e) of the sialylation in ID8 cells transfected with indicated shRNAs. The quantification results are shown as means±SEM from three independent experiments (***P \u0026lt; 0.001, **P \u0026lt; 0.01, by unpaired Student’s t-test). \u003cstrong\u003e(D, E)\u003c/strong\u003eRepresentative images of fluorescent labeling (\u003cstrong\u003eD\u003c/strong\u003e) and WB analysis (\u003cstrong\u003eE\u003c/strong\u003e) of the sialylation in Caov-3 cells transfected with NEU4 isoforms or empty vector (Vec) as a control. The quantification results are shown as means±SEM from three independent experiments (**P \u0026lt; 0.01, ns-not significant, by unpaired Student’s t-test). \u003cstrong\u003e(F) \u003c/strong\u003eLectin blot analysis of the tumor allografts in the murine model of OC established by the intrabursal injection of ID8 cells transfected with Neu4 shRNAs. SNA and MAL II were used to detect the α2, 6- and α2, 3- sialylation, respectively. \u003cstrong\u003e(G, H)\u003c/strong\u003e Lectin blot (\u003cstrong\u003eG\u003c/strong\u003e) and Lectin histochemistry (\u003cstrong\u003eH\u003c/strong\u003e) analyses of the primary (Pri) and disseminated (Diss) tumor tissues in the orthotopic murine model of OC established by the intrabursal injection of Caov-3 cells (n=4~5, **p\u0026lt;0.01, ***p\u0026lt;0.001, by unpaired Student’s t-test).\u003cstrong\u003e (I)\u003c/strong\u003e Lectin histochemistry analysis of the primary tumors tissues dissected from the orthotopic murine model of OC by the intrabursal injection of Caov-3 cells transfected with indicated NEU4 isoforms (n=4~5, *p\u0026lt;0.05, **p\u0026lt;0.01, ***p\u0026lt;0.001，by unpaired Student’ s t-test). \u003cstrong\u003e(J) \u003c/strong\u003eLectin histochemistry analysis of the primary and paired disseminated tumor tissues dissected from human HGSOC patients (n=8，**p\u0026lt;0.01, ***p\u0026lt;0.001, by paired Student’ s t-test).\u003c/p\u003e","description":"","filename":"Fig4.png","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/aa6f636214d471071e4575f5.png"},{"id":49093857,"identity":"48a5beb1-a1f1-42cd-b8c7-625713f18924","added_by":"auto","created_at":"2024-01-03 02:25:50","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":195113,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe identification of NEU4-regulated cell surface glycoproteins in OC cells. (A) \u003c/strong\u003eThe schematic diagram of the purification and identification of membrane proteins whose sialylation are affected by NEU4. \u003cstrong\u003e(B, C) \u003c/strong\u003eThe membrane associated sialylated proteins in NEU4-iso2-overexpressed Caov-3 cells or Neu4-silenced ID8 cells were resolved by SDS-PAGE and stained by the coomassie blue to show NEU4-affected protein bands (\u003cstrong\u003eB\u003c/strong\u003e), which were cut out for mass spectrometry (MS) identification and gene ontology analysis (\u003cstrong\u003eC\u003c/strong\u003e). \u003cstrong\u003e(D)\u003c/strong\u003e The Reactome pathway enrichment analysis of NEU4-regulated candidate proteins by Metascape tool. \u003cstrong\u003e(E)\u003c/strong\u003eThe signaling receptors that might be regulated by NEU4 iso2.\u003c/p\u003e","description":"","filename":"Fig5.png","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/0ea442b64505645e9a33a8ae.png"},{"id":49093680,"identity":"9d69e328-1b03-499a-a6a7-ba7b2794c20f","added_by":"auto","created_at":"2024-01-03 02:17:51","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":705990,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eNEU4 desialylates the epithelial growth factor receptor (EGFR) to enhance its activation in OC. (A)\u003c/strong\u003eWestern blot and Lectin blot analyses of the immunoprecipitated (IP) EGFR from epithelial growth factor (EGF)-treated Caov-3 cells with ectopically expressed NEU4 isoforms. The protein samples were denatured using the loading buffers with (top panel) or without (bottom panel) β-mercaptoethanol (β-ME). \u003cstrong\u003e(B)\u003c/strong\u003e Western blot analysis of the EGFR downstream FAK, ERK and AKT signaling cascades in indicated Caov-3 cells treated with or without 20 ng/mL of EGF for 1 h. \u003cstrong\u003e(C)\u003c/strong\u003e Western blot analysis of the EGFR downstream signaling cascades in shNeu4- or shLacZ-transfected ID8 cells after 10 min-treatment with or without 5 ng/mL EGF. \u003cstrong\u003e(D, E)\u003c/strong\u003e Immunofluorescent staining of the primary and disseminated human OC tissues for the correlation analysis of NEU4 with total EGFR proteins (\u003cstrong\u003eD\u003c/strong\u003e) or with phosphorylated EGFR (p-EGFR) (\u003cstrong\u003eE\u003c/strong\u003e) (n=10, by the spearman rank correlation test). \u003cstrong\u003e(F)\u003c/strong\u003e IHC analysis of the human OC tissues for the correlation between NEU4 with the activation of EGFR downstream FAK, ERK and AKT signaling molecules. (n=10, by the spearman rank correlation test). \u003cstrong\u003e(G)\u003c/strong\u003e Western blot analysis of the knock-down efficiencies of endogenous EGFR-specific shRNAs targeting the 3’ UTR region (left panel) and the ectopically expressed EGFR mutants (right panel) in Caov-3 cells.\u003cstrong\u003e (H)\u003c/strong\u003e Western blot and Lectin blot analyses of immunoprecipitated EGFR in Caov-3 cells, whose endogenous EGFR was substituted with indicated EGFR mutants, for EGFR phosphorylation and sialylation.\u003c/p\u003e","description":"","filename":"Fig6.png","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/90a7251a5cb851796e82b769.png"},{"id":49093859,"identity":"dbee7b14-cf97-40bd-b5d4-7cfb2765d68c","added_by":"auto","created_at":"2024-01-03 02:25:50","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":2253989,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eNEU4 isoform2 enhances the motility and EMT of OC cells partly through desialylating EGFR. (A)\u003c/strong\u003e The transwell cell migration and invasion assays of Caov-3 cells transfected with NEU4 isoform 2 (NEU4-iso2) or empty vector (Vec) control when EGFR was silenced (shEGFR) or not (shLacZ) (ns-not significant, *P\u0026lt;0.05, by unpaired Student’s t-test). \u003cstrong\u003e(B)\u003c/strong\u003e Western blot analysis of EMT markers in Caov-3 cells transfected with indicated plasmids. \u003cstrong\u003e(C)\u003c/strong\u003e The transwell cell migration and invasion assays of Caov-3 cells transfected with NEU4-iso2 or Vec with (+) or without (-) the treatment of 5 μM EGFR inhibitor Osimertinib (Osi). The quantification results in (\u003cstrong\u003eA\u003c/strong\u003e) and (\u003cstrong\u003eC\u003c/strong\u003e) are shown as means±SEM from three independent experiments (ns-not significant, *P\u0026lt;0.05, by unpaired Student’s t-test). \u003cstrong\u003e(D) \u003c/strong\u003eWestern blot analysis of EMT markers in Caov-3 cells transfected with indicated plasmids and treated with or without 5 μM Osi.\u003cstrong\u003e (E, F)\u003c/strong\u003e Western blot analysis of EMT markers (\u003cstrong\u003eE\u003c/strong\u003e) and EGFR downstream signaling cascades (\u003cstrong\u003eF\u003c/strong\u003e) in Caov-3 cells whose endogenous EGFR was silenced by 3’ UTR-specific shRNA and substituted with the ectopically expressed wild type (EGFR\u003csup\u003eWT\u003c/sup\u003e) or N196-glycosylation-deficient EGFR mutant (EGFR\u003csup\u003eN196Q\u003c/sup\u003e). \u003cstrong\u003e(G)\u003c/strong\u003e The Schematic diagram summarizing the role of NEU4 in the desialylation of EGFR to remove the negative charges to promote the dimerization and hyperactivation of EGFR and the downstream dissemination-promoting signaling cascades.\u003c/p\u003e","description":"","filename":"Fig7.png","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/04cd7899522ce4f6bfc57f09.png"},{"id":66646945,"identity":"84306d66-03a4-46e5-95f8-1c7cb88ba2c2","added_by":"auto","created_at":"2024-10-15 07:05:58","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":7703243,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/26b40c20-fdb3-4106-9eb8-225a48d55aa6.pdf"},{"id":49092939,"identity":"f73a83aa-fd66-4d32-b33f-95329a673650","added_by":"auto","created_at":"2024-01-03 02:09:51","extension":"tif","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":5985708,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"FigS1.tif","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/d863a9dc2e8edf4cb0c4d657.tif"},{"id":49093685,"identity":"572dc31d-8655-4fb6-b4f2-1e7e6378eb2d","added_by":"auto","created_at":"2024-01-03 02:17:51","extension":"tif","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":8231768,"visible":true,"origin":"","legend":"","description":"","filename":"FigS2.tif","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/71397d8a7b361aae8c80e736.tif"},{"id":49092940,"identity":"6bfe729e-bc7c-4c0d-b442-8c64ea5539c0","added_by":"auto","created_at":"2024-01-03 02:09:51","extension":"tif","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":6004376,"visible":true,"origin":"","legend":"","description":"","filename":"FigS3.tif","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/7110788d5027e3754a9c7b0a.tif"},{"id":49093684,"identity":"9fd0191b-0f87-4caf-b72c-c1fa57adeb4c","added_by":"auto","created_at":"2024-01-03 02:17:51","extension":"tif","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":6404744,"visible":true,"origin":"","legend":"","description":"","filename":"FigS4.tif","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/bb3019f6ef51eaa8f510ddf1.tif"},{"id":49092937,"identity":"c9c270a1-0fe8-4a08-ac6e-71686a788db0","added_by":"auto","created_at":"2024-01-03 02:09:51","extension":"xlsx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":210048,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/09b9a25fa6ebef24af18b94f.xlsx"},{"id":49092931,"identity":"c199c270-2d48-45b6-a9c2-648c46248d5d","added_by":"auto","created_at":"2024-01-03 02:09:51","extension":"xlsx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":20562,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/ae4c1bdbd62a83659bc19701.xlsx"},{"id":49093860,"identity":"6695f378-7a77-44ba-a6e1-6a95a056e943","added_by":"auto","created_at":"2024-01-03 02:25:51","extension":"xlsx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":36757,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable3.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/b8d28a997bca01bbc6d35561.xlsx"},{"id":49092943,"identity":"98c71646-3adf-431f-bfb2-5b8a2c521ad5","added_by":"auto","created_at":"2024-01-03 02:09:51","extension":"xlsx","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":588160,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable4.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/fb79126aa17055e93b82fb48.xlsx"},{"id":49092935,"identity":"34fb0aa4-9232-43fa-8ae0-fc6bd406f315","added_by":"auto","created_at":"2024-01-03 02:09:51","extension":"xlsx","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":10764,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable5.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/a97537e87952da791871dfcf.xlsx"},{"id":49093682,"identity":"ea20e2b6-caa3-423c-bff8-cf423cf5460f","added_by":"auto","created_at":"2024-01-03 02:17:51","extension":"xlsx","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":11048,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable6.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-3772327/v1/caea38f891b7c711cebd85a0.xlsx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e conflict of interest to disclose.","formattedTitle":"NEU4-mediated desialylation ignites the oncogenic receptors for the dissemination of ovarian carcinoma","fulltext":[{"header":"Significance Statement","content":"\u003cp\u003eThe characteristic cancer glycoconjugates are widely used as biomarkers for cancer diagnosis. However, our knowledge about cancer glycome remains quite fragmented due to the technique limitations in analyzing glycan chains with tremendous structural and functional heterogeneity. We identified neuraminidase 4 (NEU4), that hydrolyzes terminal sialic acid from glycoconjugates, as a vital peritoneal dissemination-promoting modifier of OC glycome. In addition, we for the first time revealed that the plasma membrane-localized NEU4 isoform 2 desialylated the epithelial growth factor receptor (EGFR), in particular at N196 residue, for the hyperactivation of EGFR and its downstream tumor-promoting signaling cascades. Our results provide new insights into how the OC glycome is dysregulated during OC progression and reveals a functionally important glycosite on EGFR for its abnormal activation in cancer.\u003c/p\u003e"},{"header":"Introduction","content":"\u003cp\u003ePeritoneal disseminated metastasis (PDM), also known as transcoelomic metastasis, is a major cause of the death of epithelial ovarian carcinoma (EOC) patients (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). The peritoneal disseminated metastasis of EOC cells involves multiple steps, i.e., the detachment of cancer cells from the primary tumor, the peritoneal migration of the free cancer cells, the adhesion of disseminated cancer cells to the mesothelial cells lining the peritoneal cavities and intra-abdominal organs or the underlying extracellular matrix (ECM), and the colonization of the disseminated cancer cells to form the secondary tumor nodules (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). Complex interactions between cancer cells and the peritoneal stromal cells or ECMs occur throughout the whole dissemination cascade, in which aberrant glycosylation has long been realized to play essential roles and hence to be a common signature of malignancies (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). Characteristic alterations of the N-glycosylation, GalNAc-type O-glycosylation and O-Xyl glycosaminoglycans have been found in cancers (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). For example, distinct changes in the branching of N-glycans and core fucosylation and increased α2-6-sialic acid capping have been linked to the altered adhesion, migration and epithelial-mesenchymal transition (EMT) capacities. The upregulation of sialyl-Lewis\u003csup\u003ex\u003c/sup\u003e and sialyl-Lewis\u003csup\u003ea\u003c/sup\u003e glycan antigens in cancer cells enhances their binding on the endothelial cells through selectins and thus leads to their extravasation and homing to the metastatic loci (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). And altered sialylation pattern also contributes to the immune evasion of tumor cells (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). Immature, truncated O-glycan structure has also been identified to be a feature of many cancers, e.g., Tn and STn antigens that impact the interaction between cancer cells and macrophages to regulate tumor immune microenvironment (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). A recent integrated analysis of the proteomics and glycoproteomics in human high-grade serous OC (HGSOC) reveals tumor-specific changes of glycoproteins and glycosites (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e). Compared to the adjacent non-tumor cells, HGSOC cells contain more abundant high-mannose glycans, but less fucosylated and sialylated glycans (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). However, the signature glycosylation patterns and their functional consequences in tumor progression remain largely unknown due to the too limited glycoproteomic technologies to characterize the glycan chains with huge complexity and heterogeneity that are not encoded directly in the genome (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe dysregulated glycome in cancer is largely due to the altered glycosylation machinery that is composed of glycosyltransferases, transporters, chaperones and glycan modification and degradation enzymes including glycosidases, which are encoded by ~\u0026thinsp;700 genes in the human genome (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). For example, fucosyltransferase 3 or 6 (FUT3, FUT6) is upregulated in cancer cells to increase the capping of N-acetyllactosamine (LacNAc)-terminated oligosaccharides by sialyl-Lewis\u003csup\u003ex\u003c/sup\u003e antigen, resulting in enhanced E-selectin binding in the vascular niche and hence promotes bone metastasis (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). Aberrant sialylation caused by altered expressions of sialyltransferases and sialidases (also known as neuraminidases) has been observed in many cancer types including OC, colon cancer, renal cancer and prostate cancer (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). Increased expression of neuraminidase 3 (NEU3) was reported in colon and prostate cancers to promote cancer progression by enhancing EGF-EGFR signaling cascades (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). On the contrary, NEU1 functions as a tumor suppressor with decreased expression in cancer cells and negatively regulates metastasis via modulating integrin β4-mediated signaling (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDue to the tremendous structural complexity and functional diversity of glycans and current technical limitations in the sequencing and manipulation of the glycan chains, our knowledge about the cancer glycome remains fragmented. However, new gene editing tools allow more comprehensive dissection of the functions of cancer glycome. In this study, we performed a systematic loss-of-function screen on 498 genes involved in the glycosylation machinery using CRISPR/Cas9 knockout library for key regulators of OC dissemination. We identified neuraminidase 4 (NEU4), an enzyme capable of removing terminal sialic acid residue from all types of sialylated glycoconjugates including glycoproteins, oligosaccharides and gangliosides (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e), as a novel peritoneal dissemination-promoting factor for OC. Given the existence of three isoforms of human NEU4 generated by alternative splicing, we further investigated their roles in the progression of OC and revealed that the plasma membrane-localized NEU4 isoform 2 was able to desialylate EGFR for its hyperactivation, which contributed to the peritoneal dissemination of OC. Our results provide new evidences that the alternative splice-generated isoforms of glycosylation-related enzymes, which have different subcellular distribution and functions, contribute to the vast heterogeneity of cancer glycome and greatly impact the cancer progression.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eSystematic gene knockout screen identifies NEU4 as a potential dissemination-promoting gene in OC\u003c/h2\u003e \u003cp\u003eTo systematically identify key glycosylation-related genes involved in the peritoneal dissemination of OC, we selected 2,984 lentivirus-based sgRNA plasmids against 498 genes (4\u0026thinsp;~\u0026thinsp;6 sgRNAs per gene) involved in the protein glycosylation modification and 1,000 non-targeting control sgRNA plasmids from the genome-wide CRISPR/Cas9 library (GeCKO-v2.0) made by Feng Zhang\u0026rsquo;s laboratory (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e) (\u003cb\u003eSupplementary Table\u0026nbsp;1\u003c/b\u003e), which were subsequently transfected into human OC cells SK-OV-3 to generate heterogeneous cell population with deficiency of different glycosylation gene in each cell. These cells were orthotopically inoculated into the ovarian bursas of NOD-SCID mice and the peritoneal disseminated OC cells that formed the new tumor nodules on the peritoneal wall were dissected and expanded \u003cem\u003ein vitro\u003c/em\u003e for the next two rounds of selections for OC cells with high dissemination capacity (\u003cb\u003eFig.\u0026nbsp;1A\u003c/b\u003e). The genomic DNAs from the primary OC xenografts in the first-round selection and the disseminated OC cells in the third-round selection were isolated for high throughput DNA sequencing to identify enriched sgRNAs and target genes in these cell populations (\u003cb\u003eFig.\u0026nbsp;1B, Supplementary Table\u0026nbsp;1\u003c/b\u003e). By using the MAGeCK algorithm to analyze the enrichment of identified sgRNAs targeting 143 glycosylation-related genes, we got eight genes with two sgRNAs enriched in the primary OC xenograft cells (\u003cb\u003eFig.\u0026nbsp;1B, Supplementary Table\u0026nbsp;2\u003c/b\u003e), suggesting their potential roles in promoting peritoneal dissemination of OC. We further investigated the correlation of these candidate dissemination-promoting genes with the overall survivals of 373 OC patients in The Cancer Genome Atlas (TCGA) database since the peritoneal dissemination of OC cells was the leading cause of the death of OC patients (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e) (\u003cb\u003eSupplementary Table\u0026nbsp;3\u003c/b\u003e). Among these eight genes, only the higher expressions of NEU4, SPTA1 and ANKRD29 significantly correlate with the poorer survivals of OC patients (\u003cb\u003eFig.\u0026nbsp;1C, Supplementary Fig.\u0026nbsp;1, Supplementary Table\u0026nbsp;3\u003c/b\u003e). Given that NEU4 is an enzyme responsible for the trimming of sialic acid from glycosylated proteins and the significant roles of sialic acid in the cell-cell and cell-environment communications (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e), here we focused on the potential roles of NEU4 in OC dissemination.\u003c/p\u003e \u003cp\u003eTo investigate the clinical relevance of NEU4 with OC dissemination, we collected the primary tumor tissues from eight HGSOC patients with paired peritoneal disseminated OC tissues and performed quantitative RT-PCR analysis of all NEU4 transcripts (\u003cb\u003eFig.\u0026nbsp;1D\u003c/b\u003e). A statistically significant increase of NEU4 transcription was observed in five disseminated OC patient samples while decreased NEU4 transcription was observed in other three disseminated OC tissues (\u003cb\u003eFig.\u0026nbsp;1E\u003c/b\u003e). However, when analyzed by western blot and immunohistochemistry, a dramatically elevated protein level of NEU4 was observed in most disseminated OC tissues (\u003cb\u003eFig.\u0026nbsp;1F and 1G\u003c/b\u003e, respectively), suggesting there probably existed post-transcriptional regulation of NEU4 expression during OC dissemination.\u003c/p\u003e \u003cp\u003eThere are several transcript variants of NEU4 that generate three isoforms with difference in the N-terminal sequences. Compared to the NEU4 isoform 1 and 2, the isoform 3 is shorter due to the lack of an N-terminal signal peptide, in which isoform 2 has an additional amino acid residue than isoform 1 (\u003cb\u003eFig.\u0026nbsp;1D\u003c/b\u003e). In addition to the localizations of NEU4 isoform 1 and isoform 3 in the lysosome and intracellular membranes as reported (\u003cspan additionalcitationids=\"CR24\" citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e) (\u003cb\u003eFig.\u0026nbsp;1H, Supplementary Fig.\u0026nbsp;2A and 2B\u003c/b\u003e), IF staining of V5-tagged NEU4 isoform 2 showed the distinct plasma membrane localization in Caov-3 and SK-OV-3 cells (\u003cb\u003eFig.\u0026nbsp;1H, Supplementary Fig.\u0026nbsp;2A\u003c/b\u003e), which was further confirmed by the cell fractionation analysis (\u003cb\u003eSupplementary Fig.\u0026nbsp;2C and 2D\u003c/b\u003e). The one-amino-acid-longer signal peptide of isoform 2, when attached to the enhanced green fluorescent protein (EGFP), is robust enough to localize the EGFP proteins on the plasma membrane when compared with that of the isoform 1 (\u003cb\u003eSupplementary Fig.\u0026nbsp;2E\u003c/b\u003e), suggesting that the isoform 2 of NEU4 may modify the glycoproteins on the plasma membrane and directly affect the signal recognition and cell-cell communications during the dissemination of OC cells. We further investigated the transcriptional change of NEU4 isoform 2 in human OC specimen by a specific primer set (\u003cb\u003eFig.\u0026nbsp;1D\u003c/b\u003e). Among eight paired human OC tissues, six disseminated OC tissues showed increased expression of NEU4 isoform 2, in which three, i.e., patient P2, P3 and P5, showed even higher fold of increase than that of the total NEU4 isoforms, indicating the potential functional significance of isoform 2 in OC dissemination (\u003cb\u003eFig.\u0026nbsp;1E and 1I\u003c/b\u003e). However, the other three showed similar or a bit less increase of NEU4 isoform 2 compared with the change of total NEU4 isoforms (\u003cb\u003eFig.\u0026nbsp;1E and 1I\u003c/b\u003e), suggesting the possible functional significance of other isoforms of NEU4.\u003c/p\u003e \u003cp\u003e \u003cb\u003eFigure 1. High-throughput CRISPR/Cas9 screen identified NEU4 as a key regulator of OC peritoneal dissemination.\u003c/b\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003e(A)\u003c/b\u003e The schematic diagram of CRISPR/Cas9 library-based screen of key glycosylation-related genes regulating peritoneal dissemination in the orthotopic murine model of OC. \u003cb\u003e(B)\u003c/b\u003e Scatterplot showing the enrichment of specific sgRNAs in primary (Pri) or peritoneal disseminated (Diss) OC cells. \u003cb\u003e(C)\u003c/b\u003e Kaplan-Meier survival curve to show the overall survival (OS) of OC patients with different NEU4 expression (n\u0026thinsp;=\u0026thinsp;373) from the TCGA database. \u003cb\u003e(D)\u003c/b\u003e The different amino acid sequences of three NEU4 isoforms and the sites of specific qRT-PCR primer sets. The Primer 1 was designed at the common encoding regions of NEU4 isoforms. The additional N-terminal amino acid residues of isoform 1 (iso1) and 2 (iso2) were presented with the differential residue of iso2 highlighted, where the specific forward primer of iso2 bound (Primer 2). \u003cb\u003e(E)\u003c/b\u003e qRT-PCR analysis of the total NEU4 transcripts in the primary and the paired peritoneal disseminated OC tissues in each OC patients (the left bargraph, data are plotted as means\u0026thinsp;\u0026plusmn;\u0026thinsp;SEM from three independent measurements, **p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, ***p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ns-not significant, by unpaired Student\u0026rsquo;s t-test) and the statistical analysis of NEU4 transcripts in eight OC patients ( the right bargraph, ***p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, by paired Student\u0026rsquo;s t-test). \u003cb\u003e(F)\u003c/b\u003e Western blot analysis of NEU4 in the primary and the paired disseminated OC tissues from eight OC patients and the quantification results (n\u0026thinsp;=\u0026thinsp;8, **p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, by paired Student\u0026rsquo;s t-test). \u003cb\u003e(G)\u003c/b\u003e Representative images of the immunohistochemical staining of NEU4 in the primary and disseminated OC tissues from an OC patient and the H-score quantification results (n\u0026thinsp;=\u0026thinsp;8, ***p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, by paired Student\u0026rsquo;s t-test). \u003cb\u003e(H)\u003c/b\u003e Immunofluorescent staining of v5-tagged NEU4 isoforms ectopically expressed in Caov-3 cells. NEU4 isoforms are stained by anti-v5 antibody in green, while the actin filaments (by phalloidin) and lysosome marker LAMP1 in red. \u003cb\u003e(I)\u003c/b\u003e qRT-PCR analysis of NEU4 isoform 2 (NEU4-iso2) in the primary and paired disseminated OC tissues in each OC patient (left bargraph, means\u0026thinsp;\u0026plusmn;\u0026thinsp;SEM from three independent measurements, *p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, **p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, ***p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, by unpaired Student\u0026rsquo;s t-test) and the statistical analysis of NEU4-iso2 expression in eight OC patients (n\u0026thinsp;=\u0026thinsp;8, ***p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, by paired Student\u0026rsquo;s t-test).\u003c/p\u003e \u003cp\u003e \u003cb\u003eNEU4, especially the isoform 2 and 3, is essential for the peritoneal dissemination of OC cells in mice\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTo investigate the role of NEU4 in the progression of OC, we selected a mouse HGSOC cell line ID8 with high NEU4 expression and a human HGSOC cell line Caov-3 with relatively low NEU4 expression to establish different murine models of HGSOC (\u003cb\u003eFig.\u0026nbsp;2A\u003c/b\u003e) (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e). In the syngeneic murine model of HGSOC, we knocked down \u003cem\u003eNeu4\u003c/em\u003e effectively by two small hairpin RNAs (shRNAs) in ID8 cells (\u003cb\u003eFig.\u0026nbsp;2B\u003c/b\u003e) and orthotopically transplanted these cells into the ovarian bursas of C57BL/6 mice (\u003cb\u003eFig.\u0026nbsp;2C\u003c/b\u003e). Either shRNA targeting \u003cem\u003eNeu4\u003c/em\u003e significantly inhibited the formation of malignant ascites (\u003cb\u003eFig.\u0026nbsp;2D\u003c/b\u003e) and reduced the disseminated OC lesions in the abdominal cavity (\u003cb\u003eFig.\u0026nbsp;2E\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eTo further investigate the role of NEU4 in the peritoneal seeding and colonization of disseminated OC cells that involved complex cell-cell and cell-matrix communications, we performed intraperitoneal injection of ID8 cells expressing the luciferase reporter gene into C57BL/6 mice (\u003cb\u003eFig.\u0026nbsp;2F\u003c/b\u003e). Live imaging showed that knocking down \u003cem\u003eNeu4\u003c/em\u003e greatly reduced the formation of cancer lesions in the abdominal cavity (\u003cb\u003eFig.\u0026nbsp;2G\u003c/b\u003e) and, as the cancer developed, ID8 cells with silenced \u003cem\u003eNeu4\u003c/em\u003e expression caused much less formation of malignant ascites (\u003cb\u003eFig.\u0026nbsp;2H\u003c/b\u003e) and cancer nodules (\u003cb\u003eFig.\u0026nbsp;2I\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eMouse Neu4 has only two isoforms reported. The shorter one is analogous to the human NEU4 isoform 3, while the longer one has an additional N-terminal signal peptide that is longer than those of the human NEU4 isoform 1 and 2 (\u003cb\u003eSupplementary Fig.\u0026nbsp;2E and 2F\u003c/b\u003e) and displays both the cytoplasmic and plasma membrane localizations (\u003cb\u003eSupplementary Fig.\u0026nbsp;2F\u003c/b\u003e) and may be functionally equivalent to human NEU4 isoform 1 and 2. To clarify the roles of different isoforms of human NEU4 in the dissemination of OC, we ectopically expressed these isoforms of NEU4 in human HGSOC cells Caov-3 that have very low level of endogenous NEU4 expression (\u003cb\u003eFig.\u0026nbsp;2A and 2J\u003c/b\u003e) and orthotopically transplanted these cells into the ovarian bursas of immunodeficient NOD-SCID mice (\u003cb\u003eFig.\u0026nbsp;2K\u003c/b\u003e). Although all the isoforms of NEU4 did not affect the growth of primary tumor xenografts (\u003cb\u003eFig.\u0026nbsp;2L\u003c/b\u003e), the isoform 2 and 3 of NEU4 dramatically promoted the formation of peritoneal cancer nodules (\u003cb\u003eFig.\u0026nbsp;2M\u003c/b\u003e). These results consistently suggest that the NEU4, in particular the isoform 2 and 3, is a potent dissemination-promoting factor of OC. The isoform 3, also known as NEU4-short that localizes in the endoplasmic reticulum where the post-translational glycosylation occurs (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e), may regulate the sialylation during the biosynthesis of glycan chains. And the cell-surface localized NEU4 isoform 2 may modify the mature glycan chains of the cell surface proteins or environmental glycans during the peritoneal dissemination of OC cells.\u003c/p\u003e \u003cp\u003e \u003cb\u003eFigure 2. NEU4 promotes the peritoneal dissemination of OC cells in mice. (A)\u003c/b\u003e Western blot analysis of NEU4 expression in indicated OC cell lines. \u003cb\u003e(B)\u003c/b\u003e Western blot analysis of the knock-down efficiencies of shRNAs targeting mouse Neu4 (shNeu4) in ID8 cells. The shRNA targeting bacteria lacZ (shLacZ) was used as a negative control. \u003cb\u003e(C)\u003c/b\u003e The schematic diagram of the syngeneic mouse model of OC by the intrabursal injection of ID8 cells into the C57BL/6 mice. \u003cb\u003e(D)\u003c/b\u003e Representative images of C57BL/6 mice 75 days after the intrabursal injection of indicated ID8 cells and the quantification of the ascites (n\u0026thinsp;=\u0026thinsp;3\u0026thinsp;~\u0026thinsp;6, *p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, ***p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, by unpaired Student\u0026rsquo;s t-test). \u003cb\u003e(E)\u003c/b\u003e Representative images of peritoneal disseminated tumor nodules (indicated by white arrows) and the quantification results (n\u0026thinsp;=\u0026thinsp;3\u0026thinsp;~\u0026thinsp;6, **p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, ***p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, by unpaired Student\u0026rsquo;s t-test). \u003cb\u003e(F)\u003c/b\u003e The schematic diagram of the experimental dissemination mouse model of OC by the intraperitoneal (i.p.) injection of the luciferase-transfected ID8 cells into C57BL/6 mice. \u003cb\u003e(G-I)\u003c/b\u003e Representative images of the peritoneal dissemination detected by live imaging (\u003cb\u003eG\u003c/b\u003e, left panel), ascites formation (\u003cb\u003eH\u003c/b\u003e, left panel) and disseminated tumor nodules (\u003cb\u003eI\u003c/b\u003e, left panel, indicated by white arrows) and the quantification results (n\u0026thinsp;=\u0026thinsp;4\u0026thinsp;~\u0026thinsp;6, *p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 by unpaired Student\u0026rsquo;s t-test). \u003cb\u003e(J)\u003c/b\u003e Western blot analysis of the V5-tagged NEU4 isoforms (iso1\u0026thinsp;~\u0026thinsp;iso3) stably transfected into the Caov-3 cells and empty vector-transfected control (Vec) cells. \u003cb\u003e(K)\u003c/b\u003e The schematic diagram of the murine xenograft model of OC by the intrabursal injection of human Caov-3 cells into NOD-SCID mice. \u003cb\u003e(L)\u003c/b\u003e Images of Caov-3 tumor xenografts dissected from the primary tumor sites 120 days after the intrabursal injection (left panel) and the quantification results of the weights of xenografts (n\u0026thinsp;=\u0026thinsp;4\u0026thinsp;~\u0026thinsp;5 per group, ns-not significant, by unpaired Student\u0026rsquo;s t-test). \u003cb\u003e(M)\u003c/b\u003e Representative images of the peritoneal dissemination of cancer cells (indicated by white arrows) in mice inoculated with indicated Caov-3 cells and the quantification results (n\u0026thinsp;=\u0026thinsp;4\u0026thinsp;~\u0026thinsp;5, **p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, *p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, ns-not significant, by unpaired Student\u0026rsquo;s t-test).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eNEU4 isoform 2 enhances the motility and EMT of OC cells\u003c/h2\u003e \u003cp\u003eTo explore how NEU4 promotes the disseminated metastasis of OC cells, we first examined its roles in the regulation of cell proliferation. Knocking down Neu4 in mouse ID8 cells did not affect their proliferation and clone formation capacity when cultured \u003cem\u003ein vitro\u003c/em\u003e (\u003cb\u003eSupplementary Fig.\u0026nbsp;3A and 3B\u003c/b\u003e). And the ectopic expression of either isoform of NEU4 did not affect the proliferation of the \u003cem\u003ein vitro\u003c/em\u003e cultured human HGSOC cell Caov-3 (\u003cb\u003eSupplementary Fig.\u0026nbsp;3C and 3D\u003c/b\u003e) and another type of OC cell SK-OV-3 (\u003cb\u003eSupplementary Fig.\u0026nbsp;3E and 3F\u003c/b\u003e) as well.\u003c/p\u003e \u003cp\u003eWe further investigated the roles of NEU4 in regulating cell mobility. Silencing Neu4 significantly attenuated the mobility and invasion of mouse ID8 cells shown by transwell cell migration assay (\u003cb\u003eFig.\u0026nbsp;3A\u003c/b\u003e) and the dynamic high-content imaging and analysis of the track and speed of each cell (\u003cb\u003eFig.\u0026nbsp;3B\u003c/b\u003e). Given the essential role of the EMT in the peritoneal dissemination of OC cells (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e), we also examined the EMT markers in Neu4-silenced ID8 cells. As shown in \u003cb\u003eFig.\u0026nbsp;3C\u003c/b\u003e, either shRNA against Neu4, especially shNeu4 #3, was able to reduce the expression of mesenchymal marker genes, i.e., N-Cadherin (N-Cad), Vimentin (VIM) and Snail, while enhanced the expression of epithelial maker gene E-Cadherin (E-Cad), suggesting the potential role of Neu4 in maintaining the mesenchymal phenotype of OC cells.\u003c/p\u003e \u003cp\u003eTo further investigate the roles of different isoforms of NEU4 in the regulation of OC cell motility, we performed the transwell cell migration assay and the high-content imaging and analysis of single cell motility of human HGSOC cell Caov-3, in which the ectopically expressed different isoforms of NEU4 supplemented the endogenous NEU4 (\u003cb\u003eFig.\u0026nbsp;3D\u003c/b\u003e). The isoform 2 and isoform 3, but not isoform 1, significantly enhanced the migration and invasion (\u003cb\u003eFig.\u0026nbsp;3E and 3F\u003c/b\u003e) and promoted the EMT markers changing toward the mesenchymal phenotype (\u003cb\u003eFig.\u0026nbsp;3D\u003c/b\u003e), which were also observed in SK-OV-3 cells (\u003cb\u003eFig.\u0026nbsp;3G and 3H\u003c/b\u003e). Taken together, NEU4, especially the isoforms 2 and 3, is a potent EMT-promoting factor for OC cells.\u003c/p\u003e \u003cp\u003e \u003cb\u003eFigure 3. NEU4 promotes the motility and epithelial mesenchymal transition (EMT) of OC cells. (A)\u003c/b\u003e Transwell cell migration and invasion assays of ID8 cells transfected with shNeu4 or shLacZ as control. Data are shown as means\u0026thinsp;\u0026plusmn;\u0026thinsp;SEM from three independent experiments (***P\u0026thinsp;\u0026lt;\u0026thinsp;0.001, by unpaired Student\u0026rsquo;s t-test). \u003cb\u003e(B)\u003c/b\u003e Dynamic imaging analysis of ID8 cell migration by the high-content cell imaging and analysis system. Representative images of single-cell ultimate displacement are shown in the left panels. The average speed of the cells and the mean square displacement (MSD) are shown as means\u0026thinsp;\u0026plusmn;\u0026thinsp;SEM (**P\u0026thinsp;\u0026lt;\u0026thinsp;0.01 by unpaired Student\u0026rsquo;s t-test). \u003cb\u003e(C)\u003c/b\u003e Western blot analysis of EMT markers in ID8 cells transfected with indicated shRNAs. \u003cb\u003e(D)\u003c/b\u003e Western blot analysis of the Caov-3 cells, in which all the endogenous NEU4 was replaced with ectopically expressed isoforms of NEU4. \u003cb\u003e(E)\u003c/b\u003e Transwell cell migration and invasion assays of Caov-3 cells with reconstituted NEU4 isoforms (iso1\u0026thinsp;~\u0026thinsp;iso3) or empty vector (Vec) as control (means\u0026thinsp;\u0026plusmn;\u0026thinsp;SEM from three independent experiments, **P\u0026thinsp;\u0026lt;\u0026thinsp;0.01, ***P\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ns-not significant, by unpaired Student\u0026rsquo;s t-test). \u003cb\u003e(F)\u003c/b\u003e Dynamic imaging analysis of Caov-3 cell migration as described in (B) (means\u0026thinsp;\u0026plusmn;\u0026thinsp;SEM from three independent experiments, **P\u0026thinsp;\u0026lt;\u0026thinsp;0.01, ***P\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ns-not significant, by unpaired Student\u0026rsquo;s t-test). \u003cb\u003e(G)\u003c/b\u003e The transwell cell migration and invasion assays of SK-OV-3 cells with the overexpression of NEU4 isoforms (means\u0026thinsp;\u0026plusmn;\u0026thinsp;SEM from three independent experiments, ***P\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ns-not significant, by unpaired Student\u0026rsquo;s t-test). \u003cb\u003e(H)\u003c/b\u003e Western blot analysis of EMT markers in SK-OV-3 cells.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eNEU4 attenuates the cell surface sialylation of OC\u003c/h2\u003e \u003cp\u003eGiven NEU4 has been reported to be a sialidase that removes the terminal sialic acid residue from a broad spectrum of glycoconjugates (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e), we investigated the activities of different NEU4 isoforms in modifying the cell surface sialylation. We utilized an analog of the natural sialic acid precursor N-acetylmannosamine (ManNAc), i.e., the tetra acetyl-N-azidoacetylmannosamine (AC\u003csub\u003e4\u003c/sub\u003eManNAz), to metabolically label the sialylated glycoproteins and glycolipids with the fluorescent dye or biotin (\u003cb\u003eFig.\u0026nbsp;4A\u003c/b\u003e). Knocking down Neu4 in mouse ID8 cells greatly enhanced the whole sialylation levels when examined by confocal microscopy (\u003cb\u003eFig.\u0026nbsp;4B\u003c/b\u003e) and by western blot analysis of the whole cell lysates (\u003cb\u003eFig.\u0026nbsp;4C\u003c/b\u003e). In human Caov-3 and SK-OV-3 cells, ectopic expression of the isoform 2 or 3, but not isoform 1, dramatically decreased the cell surface sialylation level (\u003cb\u003eFig.\u0026nbsp;4D\u003c/b\u003e) and the whole sialylation level (\u003cb\u003eFig.\u0026nbsp;4E\u003c/b\u003e and \u003cb\u003eSupplementary Fig.\u0026nbsp;4A\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eWe further analyzed the correlation between the cellular sialylation level and NEU4 expression in the orthotopic murine models of OC established by the intrabursal transplantation of Caov-3 cells or ID8 cells into the ovarian bursas of mice. In the syngeneic murine model of OC by the intrabursal injection of ID8 cells, we collected the OC tissues from the primary OC allografts and detected the sialylation by lectins SNA (\u003cem\u003eSambucus Nigra Lectin\u003c/em\u003e), specifically recognizing α2, 6-linked sialic acid, and MAL II (\u003cem\u003eMaackia Amurensis Lectin\u003c/em\u003e) specifically for α2, 3-linked sialic acid. We found that silencing Neu4 dramatically increased the protein sialylation (\u003cb\u003eFig.\u0026nbsp;4F\u003c/b\u003e). Consistently, in the Caov-3 xenograft murine model of OC, we found elevated expression of NEU4 in disseminated OC cells (\u003cb\u003eFig.\u0026nbsp;4G\u003c/b\u003e), which showed lower level of sialylation when compared with that of the primary OC tissues by either lectin blot (\u003cb\u003eFig.\u0026nbsp;4G\u003c/b\u003e) or the lectin histochemistry (\u003cb\u003eFig.\u0026nbsp;4H\u003c/b\u003e) analyses. In addition, the ectopic expression of isoform 2 or 3 of NEU4, but not isoform 1, significantly reduced the sialylation in Caov-3-derived OC xenografts (\u003cb\u003eFig.\u0026nbsp;4I\u003c/b\u003e). In eight human HGSOC tumor tissues, we also found that the disseminated OC cells with high NEU4 expression (\u003cb\u003eFig.\u0026nbsp;1E-G\u003c/b\u003e) exhibited lower sialylation level when compared with that in the primary tumor cells which had low NEU4 expression (\u003cb\u003eFig.\u0026nbsp;4J\u003c/b\u003e).\u003c/p\u003e \u003cp\u003e \u003cb\u003eFigure 4. NEU4 significantly decreases the sialylation level in OC. (A)\u003c/b\u003e The schematic diagram of metabolic labeling of the sialylated molecules with click chemistry. \u003cb\u003e(B, C)\u003c/b\u003e Representative images of fluorescent labeling (\u003cb\u003eB\u003c/b\u003e) and western blot (WB) analysis (\u003cb\u003eC\u003c/b\u003e) of the sialylation in ID8 cells transfected with indicated shRNAs. The quantification results are shown as means\u0026thinsp;\u0026plusmn;\u0026thinsp;SEM from three independent experiments (***P\u0026thinsp;\u0026lt;\u0026thinsp;0.001, **P\u0026thinsp;\u0026lt;\u0026thinsp;0.01, by unpaired Student\u0026rsquo;s t-test). \u003cb\u003e(D, E)\u003c/b\u003e Representative images of fluorescent labeling (\u003cb\u003eD\u003c/b\u003e) and WB analysis (\u003cb\u003eE\u003c/b\u003e) of the sialylation in Caov-3 cells transfected with NEU4 isoforms or empty vector (Vec) as a control. The quantification results are shown as means\u0026thinsp;\u0026plusmn;\u0026thinsp;SEM from three independent experiments (**P\u0026thinsp;\u0026lt;\u0026thinsp;0.01, ns-not significant, by unpaired Student\u0026rsquo;s t-test). \u003cb\u003e(F)\u003c/b\u003e Lectin blot analysis of the tumor allografts in the murine model of OC established by the intrabursal injection of ID8 cells transfected with Neu4 shRNAs. SNA and MAL II were used to detect the α2, 6- and α2, 3- sialylation, respectively. \u003cb\u003e(G, H)\u003c/b\u003e Lectin blot (\u003cb\u003eG\u003c/b\u003e) and Lectin histochemistry (\u003cb\u003eH\u003c/b\u003e) analyses of the primary (Pri) and disseminated (Diss) tumor tissues in the orthotopic murine model of OC established by the intrabursal injection of Caov-3 cells (n\u0026thinsp;=\u0026thinsp;4\u0026thinsp;~\u0026thinsp;5, **p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, ***p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, by unpaired Student\u0026rsquo;s t-test). \u003cb\u003e(I)\u003c/b\u003e Lectin histochemistry analysis of the primary tumors tissues dissected from the orthotopic murine model of OC by the intrabursal injection of Caov-3 cells transfected with indicated NEU4 isoforms (n\u0026thinsp;=\u0026thinsp;4\u0026thinsp;~\u0026thinsp;5, *p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, **p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, ***p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, by unpaired Student\u0026rsquo; s t-test). \u003cb\u003e(J)\u003c/b\u003e Lectin histochemistry analysis of the primary and paired disseminated tumor tissues dissected from human HGSOC patients (n\u0026thinsp;=\u0026thinsp;8, **p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, ***p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, by paired Student\u0026rsquo; s t-test).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eThe analysis of the cell surface glycoproteome affected by NEU4 isoform 2 in OC cells\u003c/h2\u003e \u003cp\u003eGiven the essential roles of terminal sialylation in the regulation of the cell-matrix and cell-cell communications, we hypothesized that the plasma membrane-localized NEU4 isoform 2 might affect the cell surface sialylation and hence the essential transmembrane signaling for the dissemination of OC cells. To identify the NEU4-regulated glycoproteome, we metabolically labeled OC cells with AC\u003csub\u003e4\u003c/sub\u003eManNAz that allowed the subsequent conjugation of biotin on the sialylated proteins and their affinity purification by streptavidin-conjugated resins (\u003cb\u003eFig.\u0026nbsp;5A\u003c/b\u003e). We next performed SDS-PAGE to separate these proteins and selected the membrane proteins whose sialylation were able to be reduced by NEU4 isoform2 in Caov-3 and increased by silencing Neu4 in ID8 cells (\u003cb\u003eFig.\u0026nbsp;5B\u003c/b\u003e) for mass spectrometry identification (\u003cb\u003eFig.\u0026nbsp;5C\u003c/b\u003e). Among 2,014 proteins with over 3 unique peptides identified, we identified 615 plasma membrane proteins with different biological functions ranging from cell-cell adhesion, protein-binding modulator to transmembrane signal receptors (\u003cb\u003eFig.\u0026nbsp;5D, Supplementary Table\u0026nbsp;4\u003c/b\u003e). We were specifically interested in the 21 transmembrane receptors that were able to transduce signals in multiple biological processes, such as angiogenesis, cell adherence and growth-stimulating pathways, that were essential for the dissemination, survival and colonization of cancer cells (\u003cb\u003eFig.\u0026nbsp;5E\u003c/b\u003e). These results suggested that NEU4 might modulate the sialylation of a variety of signaling receptors essential for a broad range of biological processes during the dissemination of OC cells.\u003c/p\u003e \u003cp\u003e \u003cb\u003eFigure 5. The identification of NEU4-regulated cell surface glycoproteins in OC cells. (A)\u003c/b\u003e The schematic diagram of the purification and identification of membrane proteins whose sialylation are affected by NEU4. \u003cb\u003e(B, C)\u003c/b\u003e The membrane associated sialylated proteins in NEU4-iso2-overexpressed Caov-3 cells or Neu4-silenced ID8 cells were resolved by SDS-PAGE and stained by the coomassie blue to show NEU4-affected protein bands (\u003cb\u003eB\u003c/b\u003e), which were cut out for mass spectrometry (MS) identification and gene ontology analysis (\u003cb\u003eC\u003c/b\u003e). \u003cb\u003e(D)\u003c/b\u003e The Reactome pathway enrichment analysis of NEU4-regulated candidate proteins by Metascape tool. \u003cb\u003e(E)\u003c/b\u003e The signaling receptors that might be regulated by NEU4 iso2.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eNEU4 desialylates EGFR to enhance its activation in OC\u003c/h2\u003e \u003cp\u003eTo get insights into the functional significance of NEU4-mediated desialylation in the regulation of the oncogenic signaling, we selected one important receptor EGFR that was highly activated during the dissemination of many cancers including OC (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e). Ectopic expression of NEU4 isoforms 2 and 3, but not isoform 1, greatly attenuated the sialylation of EGFR, resulting in the hyper activation of EGFR shown by its autophosphorylation at tyrosine 1068 in Caov-3 cells (\u003cb\u003eFig.\u0026nbsp;6A\u003c/b\u003e) and SK-OV-3 cells (\u003cb\u003eSupplementary Fig.\u0026nbsp;4B\u003c/b\u003e). In addition, the NEU4 isoform 2-triggered hyperphosphorylation of EGFR induced more activated downstream FAK, ERK and AKT signaling cascades in both cell lines (\u003cb\u003eFig.\u0026nbsp;6B\u003c/b\u003e and \u003cb\u003eSupplementary Fig.\u0026nbsp;4C\u003c/b\u003e). In ID8 cells, silencing Neu4 inhibited the activation of EGFR and the downstream signaling cascades (\u003cb\u003eFig.\u0026nbsp;6C\u003c/b\u003e). In human OC tissues, we also found that the elevated NEU4 protein level tightly correlated with the autophosphorylation level of EGFR, but not with the total EGFR protein level (\u003cb\u003eFig.\u0026nbsp;6D and 6E\u003c/b\u003e), suggesting that NEU4-mediated desialylation of EGFR only affected its activation, but not its stability. Consistently, the tight positive correlations between NEU4 and the activation of EGFR downstream signaling molecules were also observed in human OC tissues (\u003cb\u003eFig.\u0026nbsp;6F\u003c/b\u003e). In Caov-3 and SK-OV-3 cells, we silenced the endogenous EGFR by 3\u0026rsquo; UTR-specific shRNAs and reconstituted the cells with glycosylation-deficient EGFR with N175Q mutation, N196Q mutation or double mutations, given these two glycosites were getting much closer when EGFR dimerized by its ligand and the negatively charged sialic acid might interfere with the dimerization (\u003cb\u003eFig.\u0026nbsp;6G\u003c/b\u003e and \u003cb\u003eSupplementary Fig.\u0026nbsp;4D-F\u003c/b\u003e). These mutations, especially N196Q and double mutations, greatly abolished the sialylation of EGFR and caused its hyper autophosphorylation (\u003cb\u003eFig.\u0026nbsp;6H\u003c/b\u003e and \u003cb\u003eSupplementary Fig.\u0026nbsp;4G\u003c/b\u003e), indicating that sialylation of EGFR at N196 and N175 was able to inhibit the activation of EGFR.\u003c/p\u003e \u003cp\u003e \u003cb\u003eFigure 6. NEU4 desialylates the epithelial growth factor receptor (EGFR) to enhance its activation in OC. (A)\u003c/b\u003e Western blot and Lectin blot analyses of the immunoprecipitated (IP) EGFR from epithelial growth factor (EGF)-treated Caov-3 cells with ectopically expressed NEU4 isoforms. The protein samples were denatured using the loading buffers with (top panel) or without (bottom panel) β-mercaptoethanol (β-ME). \u003cb\u003e(B)\u003c/b\u003e Western blot analysis of the EGFR downstream FAK, ERK and AKT signaling cascades in indicated Caov-3 cells treated with or without 20 ng/mL of EGF for 1 h. \u003cb\u003e(C)\u003c/b\u003e Western blot analysis of the EGFR downstream signaling cascades in shNeu4- or shLacZ-transfected ID8 cells after 10 min-treatment with or without 5 ng/mL EGF. \u003cb\u003e(D, E)\u003c/b\u003e Immunofluorescent staining of the primary and disseminated human OC tissues for the correlation analysis of NEU4 with total EGFR proteins (\u003cb\u003eD\u003c/b\u003e) or with phosphorylated EGFR (p-EGFR) (\u003cb\u003eE\u003c/b\u003e) (n\u0026thinsp;=\u0026thinsp;10, by the spearman rank correlation test). \u003cb\u003e(F)\u003c/b\u003e IHC analysis of the human OC tissues for the correlation between NEU4 with the activation of EGFR downstream FAK, ERK and AKT signaling molecules. (n\u0026thinsp;=\u0026thinsp;10, by the spearman rank correlation test). \u003cb\u003e(G)\u003c/b\u003e Western blot analysis of the knock-down efficiencies of endogenous EGFR-specific shRNAs targeting the 3\u0026rsquo; UTR region (left panel) and the ectopically expressed EGFR mutants (right panel) in Caov-3 cells. \u003cb\u003e(H)\u003c/b\u003e Western blot and Lectin blot analyses of immunoprecipitated EGFR in Caov-3 cells, whose endogenous EGFR was substituted with indicated EGFR mutants, for EGFR phosphorylation and sialylation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eNEU4 isoform2 promotes the dissemination of OC cells partly through desialylating EGFR\u003c/h2\u003e \u003cp\u003eTo evaluate the biological significance of NEU4-regulated desialylation of EGFR in the dissemination of OC, we inhibited the EGFR signaling by either RNAi (\u003cb\u003eFig.\u0026nbsp;7A and 7B\u003c/b\u003e) or by the specific EGFR inhibitor Osimertinib (\u003cb\u003eFig.\u0026nbsp;7C and 7D\u003c/b\u003e). Once the EGFR was blocked, NEU4 isoform 2 was no longer able to effectively enhance the mobility and invasion of Caov-3 cells (\u003cb\u003eFig.\u0026nbsp;7A and 7C\u003c/b\u003e). Consistently, NEU4 isoform2 did not promote the EMT of Caov-3 cells whose EGFR was silenced by shRNA (\u003cb\u003eFig.\u0026nbsp;7B\u003c/b\u003e) or inhibited by Osimertinib (\u003cb\u003eFig.\u0026nbsp;7D\u003c/b\u003e). We further explored the importance of EGFR in NEU4-regulated dissemination in Caov-3 cells whose endogenous EGFR was substituted with N196-glycosylation-deficient EGFR (N196Q) or wild type (WT) EGFR (\u003cb\u003eFig.\u0026nbsp;6H\u003c/b\u003e). In N196-glycosylation-deficient Caov-3 cells, both the EMT process (\u003cb\u003eFig.\u0026nbsp;7E\u003c/b\u003e) the EGFR downstream signaling cascades (\u003cb\u003eFig.\u0026nbsp;7F\u003c/b\u003e) were dramatically enhanced, where NEU4 isoform 2 could only slightly enhance some mesenchymal markers, such as N-cadherin and Snail (\u003cb\u003eFig.\u0026nbsp;7E\u003c/b\u003e), and EGFR downstream signaling cascades, such as ERK and AKT (\u003cb\u003eFig.\u0026nbsp;7F\u003c/b\u003e), suggesting that EGFR was an essential NEU4 downstream molecule that mediated the dissemination-promoting roles of NEU4 in OC.\u003c/p\u003e \u003cp\u003eIn conclusion, we identified NEU4-iso2 and NEU4-iso3 as new glycome regulators that promoted the peritoneal disseminated metastasis of OC. In addition, we revealed that NEU4-iso2 desialylated EGFR that resulted in the hyperphosphorylation of EGFR and its downstream signaling cascades to promote the peritoneal disseminated metastasis of OC (\u003cb\u003eFig.\u0026nbsp;7G)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003e \u003cb\u003eFigure 7. NEU4 isoform2 enhances the motility and EMT of OC cells partly through desialylating EGFR. (A)\u003c/b\u003e The transwell cell migration and invasion assays of Caov-3 cells transfected with NEU4 isoform 2 (NEU4-iso2) or empty vector (Vec) control when EGFR was silenced (shEGFR) or not (shLacZ) (ns-not significant, *P\u0026thinsp;\u0026lt;\u0026thinsp;0.05, by unpaired Student\u0026rsquo;s t-test). \u003cb\u003e(B)\u003c/b\u003e Western blot analysis of EMT markers in Caov-3 cells transfected with indicated plasmids. \u003cb\u003e(C)\u003c/b\u003e The transwell cell migration and invasion assays of Caov-3 cells transfected with NEU4-iso2 or Vec with (+) or without (-) the treatment of 5 \u0026micro;M EGFR inhibitor Osimertinib (Osi). The quantification results in (\u003cb\u003eA\u003c/b\u003e) and (\u003cb\u003eC\u003c/b\u003e) are shown as means\u0026thinsp;\u0026plusmn;\u0026thinsp;SEM from three independent experiments (ns-not significant, *P\u0026thinsp;\u0026lt;\u0026thinsp;0.05, by unpaired Student\u0026rsquo;s t-test). \u003cb\u003e(D)\u003c/b\u003e Western blot analysis of EMT markers in Caov-3 cells transfected with indicated plasmids and treated with or without 5 \u0026micro;M Osi. \u003cb\u003e(E, F)\u003c/b\u003e Western blot analysis of EMT markers (\u003cb\u003eE\u003c/b\u003e) and EGFR downstream signaling cascades (\u003cb\u003eF\u003c/b\u003e) in Caov-3 cells whose endogenous EGFR was silenced by 3\u0026rsquo; UTR-specific shRNA and substituted with the ectopically expressed wild type (EGFR\u003csup\u003eWT\u003c/sup\u003e) or N196-glycosylation-deficient EGFR mutant (EGFR\u003csup\u003eN196Q\u003c/sup\u003e). \u003cb\u003e(G)\u003c/b\u003e The Schematic diagram summarizing the role of NEU4 in the desialylation of EGFR to remove the negative charges to promote the dimerization and hyperactivation of EGFR and the downstream dissemination-promoting signaling cascades.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe roles of NEU4 in malignancies are variable in different cancer types. In colon cancer and hepatocellular carcinoma (HCC), NEU4 has been reported to be a tumor suppressor. NEU4 hydrolyzes the sialyl-Lewis antigens on colon cancer cells and results in attenuated E-selectin-mediated cell adhesion, motility and growth (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e). And NEU4 expression is downregulated by hypoxia stimuli. They also observed the presence of some NEU4 proteins on the cell surface with unknown biologic functions (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e). Significant down-regulation of NEU4 was also found in highly metastatic hepatocellular carcinoma (HCC), correlating with high grades and poor outcomes (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e). NEU4 inhibits the motility and metastasis of HCC cells partially through removing the α2,3-linked sialic acid residues on CD44 to enhance its binding on the hyaluronic acid. On the contrary, the tumor-promoting function of NEU4 in neuroblastoma (NB) was also reported(\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e). NEU4 promotes the proliferation, survival and stemness of NB cells through the hyperactivation of the Wnt/β-catenin signaling pathway particularly by NEU4 isoform 1 (also known as NEU4 long or NEU4L) (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e). The functional diversity of NEU4 in malignancies may be caused by the presence of multiple, alternative splicing-generated isoforms of NEU4, which have different subcellular localizations. Previous studies of NEU4 are mostly focused on the mitochondrial- or lysosome-localized isoform 1 (NEU4L)(\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e), while the roles of endoplasmic reticulum (ER)-localized isoform 3 (also known as NEU4S)(\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e) and plasma membrane-localized isoform 2 of NEU4 are largely unexplored. Here, we identified both the isoform 2 and isoform 3 of NEU4 were key players in shaping the OC glylcome favoring the peritoneal dissemination. The ER-localized isoform 3 probably regulates the sialylation in the biosynthesis step of glycan chains and intrinsically influence the dissemination capacity of OC cells, while the cell-surface localized isoform 2 may instantly modify the glycan chains of the signaling receptors to adjust their reactivities in response to the environmental cues during the peritoneal dissemination of OC cells. It keeps an open question how different environmental stimuli regulate the alternative splicing of NEU4 in the disseminating OC cells to generate the isoform 2 and 3.\u003c/p\u003e \u003cp\u003eIn this study, we also identified membrane proteins whose sialylation was able to be modified by NEU4 in OC cells. To investigate the functional significance of NEU4-mediated desialylation, we chosed one tyrosine kinase receptor EGFR, whose amplification and overexpression were found in some epithelial OC patients and correlated with poorer clinical outcomes (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e). We showed that NEU4-mediated desialylation at N196 and N175 greatly contributed to the phosphorylation and activation of EGFR, possibly by eliminating the repulsive force of negatively charged terminal sialic acid on N196 and N175-glycan chains that opposed the dimerization of EGFR. However, when substituted with glycosylation-deficient EGFR, NEU4 isoform 2 was still able to further enhance the activation of ERK and AKT (\u003cb\u003eFig.\u0026nbsp;7F\u003c/b\u003e), indicating that NEU4 regulated more oncogenic signaling receptors. In addition to EGFR, we also identified 20 transmembrane signaling receptors including insulin-like growth factor 1 receptor (IGF1R), ephrin type A receptor 2 (EphA2), epithelial discoidin domain containing receptor 1 and 2 (DDR1 and DDR2) (\u003cb\u003eSupplementary Table\u0026nbsp;4\u003c/b\u003e). The involvement of IGF1/IGF1R signaling in the invasion, angiogenesis and cell survival of OC cells by the activation the downstream MAPK/ERK and PI3K/AKT pathways has been reported(\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e). As collagen-activated receptor tyrosine kinases, DDR1 and DDR2 regulate cell differentiation, proliferation, adhesion, migration, invasion, and matrix remodeling, whose dysregulation has also been found in ovarian cancer (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e). And the ligand-independent EphA2 signaling by S897 phosphorylation has been implicated in the adaptive chemotherapy resistance of OC during dissemination (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e). Therefore, the NEU4 might affect all these oncogenic signaling cascades by modifying the glycosylation of the key receptors.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003ePatient samples\u003c/h2\u003e \u003cp\u003eEight paired OC patient samples (including primary tumors and disseminated tumor nodules) were obtained from Tianjin Center Hospital of Gynecology Obstetrics from December 2017 to December 2019. Half of each sample was kept in 4% paraformaldehyde for Hematoxylin-eosin (H\u0026amp;E) and immunohistochemistry (IHC) staining, the other half was lysed in TRIeasy\u0026trade; LS Total RNA Extraction Reagent (Yeasen Biotechnology, Shanghai, China) for the extraction of protein and RNA. The human patient sample-based studies were performed in accordance with the ethics guidelines of the committee of Nankai University and Tianjin Center Hospital of Gynecology Obstetrics (2018KY032). All OC patients from Tianjin Center Hospital of Gynecology Obstetrics provided informed consents.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eCell culture\u003c/h2\u003e \u003cp\u003eCaov-3 cell lines was purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). HEK 293T and SK-OV-3 cell lines were purchased from American Type Culture Collection (ATCC). ID8 cells were purchased from Merck (Darmstadt, Germany). The identities of all cell lines were confirmed by STR analysis. And all cells were confirmed to be mycoplasma negative by using the MycAway\u0026trade; -Color One-Step Mycoplasma Detection Kit (Yeasen Biotechnology). The cells were maintained in an atmosphere of 5% CO\u003csub\u003e2\u003c/sub\u003e at 37\u0026deg;C in recommended medium. HEK 293T, Caov-3 and ID8 cells were cultured in Dulbecco\u0026rsquo;s Minimum Essential Medium (DMEM) (Biological Industries, Kibbutz Beit-Haemek, Isreal) supplemented with 10% of heat-inactivated fetal bovine serum (FBS) (Biological Industries), 2 mM of L-glutamine, 100 U/mL of penicillin and 100 \u0026micro;g/mL of streptomycin (Thermo-Fisher Scientific, Waltham, MA, USA). SK-OV-3 cells were cultured with McCoy's 5A Medium Modified.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003ePlasmid construction and stable cell line establishment\u003c/h2\u003e \u003cp\u003eThree transcript variants of human NEU4 isoforms were cloned by RT-PCR using cDNA generated from total RNAs of Caov-3 cells. The coding sequences of NEU4 isoforms were cloned into the lentiviral plasmid pLV-EF1α-MCS-IRES-Bsd (Biosettia Inc., San Diego, CA, USA), with the V5 tag-encoding sequence was added at the C-terminal just before the stop codon. The sequences of the primers are listed in the \u003cb\u003eSupplementary Table\u0026nbsp;5\u003c/b\u003e. The shRNA target sites were selected by using the Thermo-Fisher RNAi Designer (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://rnaidesigner.thermofisher.com/rnaiexpress/\u003c/span\u003e\u003cspan address=\"http://rnaidesigner.thermofisher.com/rnaiexpress/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and the shRNA templates were inserted into the lentiviral plasmid pLV-H1-EF1α-puro (Biosettia Inc.). The sequences of the shRNA templates are listed in the \u003cb\u003eSupplementary Table\u0026nbsp;5\u003c/b\u003e. All plasmids were further confirmed by DNA sequencing (Sangon, Beijing, China). The lentiviruses carrying these plasmids were produced in Lenti-293T cells (Biosettia Inc.) and were used to infect the OC cells at the multiplicity of infection (MOI) of 1.0 before the selection with 2.5-5.0 \u0026micro;g/mL of blasticidin (Sigma-Aldrich, Darmstadt, Germany) for the cells with the stable ectopic expression of target genes or 5 \u0026micro;g/mL of puromycin (Sigma-Aldrich) for the cells with stable knock-down of target genes. After 2-week-selection, the expression levels of target genes in the polyclonal cell populations were verified by western blot and quantitative RT-PCR analyses before the \u003cem\u003ein vitro\u003c/em\u003e studies or inoculation into the mice.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eProtein extraction and western blot\u003c/h2\u003e \u003cp\u003eCells were lysed by RIPA buffer [50 mM Tris-HCl pH 7.4, 150 mM NaCl, 1% (v/v) Triton X-100, 1% (w/v) sodium deoxycholate, 0.1% (w/v) SDS with protease inhibitor cocktail and phosphatase inhibitor cocktail (Sigma-Aldrich). The membrane proteins were extracted by using the Membrane Protein Extraction Kit (Bestbio, Nanjing, Jiangsu, China). The protein concentration was then measured by using the Pierce TM BCA Protein Assay Kit (Thermo-Fisher Scientific). Equal amount of proteins was resolved by SDS-PAGE (10%). The proteins in the gel were transferred onto the PVDF membrane and incubated with 5% defatted milk for 1 hour (h). The membrane was incubated with primary antibodies overnight at 4\u0026deg;C followed by the incubation with secondary antibodies (Proteintech, Wuhan, Hubei, China) for 1 h at room temperature. The primary antibodies used in this study are listed in the \u003cb\u003eSupplementary Table\u0026nbsp;2\u003c/b\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eQuantitative Real-time PCR\u003c/h2\u003e \u003cp\u003eThe total RNAs were extracted from cells by using the TRIeasy\u0026trade; LS Total RNA Extraction Reagent and reverse-transcribed into cDNA using M-MLV reverse transcriptase (Promega, Madison, WI, USA) according to the manufacturer\u0026rsquo;s protocol. Quantitative real time PCR analysis was performed by using SYBR Green SuperMix (Yeasen) on the Applied Biosystems-StepOnePlus Real-Time PCR system (Thermo-Fisher Scientific). The PCR program was as follows: initial denaturation at 95\u0026deg;C for 6 minutes (min), followed by 45 cycles of 95\u0026deg;C for 30 seconds (s) and 60\u0026deg;C for 45 s. And the relative expression of gene was calculated using the 2\u003csup\u003e\u0026minus;ΔΔCt\u003c/sup\u003e method with β-actin gene as the normalization control. The primer sequences are listed in the \u003cb\u003eSupplementary Table\u0026nbsp;5\u003c/b\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eHematoxylin and Eosin (H\u0026amp;E) and Immunohistochemistry (IHC) staining\u003c/h2\u003e \u003cp\u003eThe tissue samples were fixed with 4% (w/v) paraformaldehyde (Sigma-Aldrich) and dehydrated before being put into wax cylinder overnight for paraffin wax embedding. After consecutive sectioning, tissue sections of 5-\u0026micro;m thickness were stained with hematoxylin and eosin (OriGene, Rockville, MD, USA) for pathological analysis. For IHC analysis, sections were treated with 3% (v/v) hydrogen peroxide for quenching the endogenous peroxidase and heated for 1 h for the retrieval of antigens. After blocked with 5% (v/v) goat serum, the sections were incubated with primary antibodies, biotin-conjugated secondary antibodies (Vector Laboratories, Newark, CA, USA), and streptavidin-HRP (Vector Laboratories). Finally, DAB substrate (Zsgb-Bio, Beijing, China) was dropped onto sections for less than 2 min and hematoxylin was used to stain the cell nuclei. The primary antibodies used in this study are listed in the \u003cb\u003eSupplementary Table\u0026nbsp;6\u003c/b\u003e. The H score was calculated by multiplying positively stained area (P) with staining intensity (I), where the degrees for P were 0\u0026ndash;4 (0, \u0026lt;\u0026thinsp;5%; 1, 5\u0026ndash;25%; 2, 25\u0026ndash;50%;3, 50\u0026ndash;75%; 4, 75\u0026ndash;100%) and the degrees for I were 0\u0026ndash;3 (0, none; 1, weak; 2, moderate; 3, strong).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eImmunofluorescent (IF) staining\u003c/h2\u003e \u003cp\u003eCells were seeded on glass coverslips put in 24-well plate. The cells were washed once with cold phosphate buffered saline (PBS) and fixed by 4% paraformaldehyde for 10 min. Tissue sections were blocked with 5% goat serum before the sequential incubation with primary antibodies, secondary antibodies conjugated with Alexa Fluor-488 and Alexa Fluor-594 (Thermo-Fisher Scientific), and 4\u0026rsquo;,6-diamidino-2-phenylindole (DAPI) (Sigma-Aldrich) for the nuclear counterstain. Images were taken by using an Olympus FV1000 confocal microscope (Tokyo, Japan). The primary antibodies used in this study are listed in the \u003cb\u003eSupplementary Table\u0026nbsp;6\u003c/b\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eCell migration and invasion assay\u003c/h2\u003e \u003cp\u003eThe migration and invasion assays were carried with the Corning\u0026reg; Transwell\u0026reg; chambers with 8-\u0026micro;m pore size (Merck). For invasion assay, the transwell chambers were additionally coated with Matrigel\u0026trade; basement membrane matrix (BD Biosciences, San Jose, CA, USA). Ten thousand of ID8 cells or 5\u0026times;10\u003csup\u003e4\u003c/sup\u003e Caov-3 cells were mixed with FBS-free medium and added into top chambers. 600 \u0026micro;L medium complemented with 10% FBS was added to the lower chamber. To analyze the EGFR signaling, cells were seeded in the top chambers with 5 \u0026micro;M Osimertinib (MedChemExpress, Shanghai, China) added in FBS-free medium. After 6 h for ID8 or 18 h for Caov-3, cells migrated to the bottom surface of the transwell membranes were fixed in 4% paraformaldehyde and stained with crystal violet (Beyotime, Shanghai, China).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eCell proliferation assay\u003c/h2\u003e \u003cp\u003e1 \u0026times; 10\u003csup\u003e5\u003c/sup\u003e ID8 cells or 5 \u0026times; 10\u003csup\u003e5\u003c/sup\u003e Caov-3 cells were seeded in 6-well plates and maintained at 37\u0026deg;C in the incubator. Cells were detached by trypsinization with 0.25% trypsin\u0026ndash;0.02% ethylenediaminetetraacetic acid (EDTA) (Biological Industries) and counted every 24 h by using Countess 2 Automated Cell Counter (Thermo-Fisher Scientific) until the cells reached full confluency. The cell number at each time point was the average number from three wells.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eColony formation assay\u003c/h2\u003e \u003cp\u003e1\u0026times; 10\u003csup\u003e3\u003c/sup\u003e ID8 cells or 5\u0026times; 10\u003csup\u003e3\u003c/sup\u003e Caov-3 cells were seeded in 6-well plate and maintained at 37\u0026deg;C in incubator until colonies could be identified. The colonies on the plates were fixed in 4% paraformaldehyde for 1 h at 4\u0026deg;C, then stained with crystal violet (C0121, Beyotime). The number of colonies was counted to evaluate the proliferation of cells.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eLectin Blot and Lectin histochemistry\u003c/h2\u003e \u003cp\u003eTotal proteins were extracted from cells by using RIPA buffer and quantified by using the Pierce TM BCA Protein Assay Kit. The proteins were then resolved by SDS-PAGE and transferred on to PVDF membranes. After the PVDF membranes were blocked with 2.5% (w/g) oxidized bovine serum albumin (oBSA) for 4 h, the proteins with α2,6- or α2,3- sialylation modification were detected by the incubation with biotinylated Sambucus Nigra Lectin (SNA) or Maackia Amurensis Lectin (MAL II) (Vector Laboratories) for 1 h, respectively. Sialylated proteins were visualized by the incubation with HRP-conjugated streptavidin (Thermo-Fisher Scientific) for 1 h followed by ECL enhanced chemiluminescence reagents (EpiZyme Scientific, Shanghai, China). The Coomassie Brilliant Blue (CBB) staining of the membranes were used to estimate the total protein amounts. oBSA was prepared as follows: BSA (Solarbio, Beijing, China) was dissolved in 10 mM sodium metaperiodate (Aladdin, Shanghai, China) and maintained at 4 ˚C for 1 h before the dialysis against TBS-T buffer for 16 h at 4˚C to remove the sodium metaperiodate. All lectins were dissolved in 0.5% oBSA.\u003c/p\u003e \u003cp\u003eFor lectin histochemistry, 5-\u0026micro;m sections were blocked with 2.5% oBSA for 3 h and incubated with biotinylated SNA or MAL II overnight. After washing with TBST-T buffer for three times, sections were incubated with HRP-conjugated streptavidin for 1 h and visualized with 3,3\u0026rsquo;-diaminobenzidine (DAB) substrate (ZSGB-BIO, Beijing, China). The quantification was performed by the H score method calculated as described above in IHC segment.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eThe murine models of OC\u003c/h2\u003e \u003cp\u003eAll animal studies and procedures were approved by the Nankai University Animal Care and Use Committee. For the orthotopic murine model of OC, 5 \u0026times; 10\u003csup\u003e6\u003c/sup\u003e human Caov-3 cells (stably transfected with NEU4 isoforms or empty vector as the control) or murine ID8 cells (stably transfected with shNeu4 or shLacZ as a control) were suspended in 20 \u0026micro;L PBS and implanted into the surgically exposed right ovarian bursa of anaesthetized 6-week-old, female NOD-SCID mice or C57BL/6 mice (SPF Biotechnology, Beijing, China) by intrabursal injection. 75 days after ID8 injection or 120 days after Caov-3 injection, mice were sacrificed for OC development analysis. For the experimental dissemination murine model of OC, 1 \u0026times; 10\u003csup\u003e7\u003c/sup\u003e ID8 cells were suspended in 100 \u0026micro;L PBS and implanted into the peritoneal cavity of C57BL/6 mice by intraperitoneal injection. Then mice were sacrificed at day 60 after the injection. The ascites, primary tumors and peritoneal disseminated tumor nodules were collected.\u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003eThe CRISPR-Cas9 library screen in the murine model of OC\u003c/h2\u003e \u003cp\u003eThe CRISPR/Cas9 library containing 2,984 lentivirus-based sgRNA plasmids against 498 protein glycosylation\u0026ndash;related genes (4\u0026thinsp;~\u0026thinsp;6 sgRNAs per gene) and 1,000 non-targeting control sgRNA plasmids were selected from the genome-wide CRISPR/Cas9 library (GeCKO-v2.0) made by Feng Zhang\u0026rsquo;s laboratory (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e) (\u003cb\u003eSupplementary Table\u0026nbsp;3\u003c/b\u003e). The amplification and lentivirus preparation of CRIPSR-Cas9 library were performed as described by Feng Zhang. For the \u003cem\u003ein vivo\u003c/em\u003e screen, a total of 1\u0026times;10\u003csup\u003e7\u003c/sup\u003e of SK-OV-3 cells were infected in the presence of 8 \u0026micro;g/mL polybrene (Sigma-Aldrich) with the lentiviral particles at a multiplicity of infection of 0.3 to avoid multiple gene knock outs in one cell. The infected cells were selected by using 5 \u0026micro;g/mL puromycin (Sigma-Aldrich) after 72 h. After 14-day in vitro selection and expansion, a total of 2\u0026times;10\u003csup\u003e6\u003c/sup\u003e of SK-OV-3 cells were suspended in 20 \u0026micro;L PBS and implanted into the surgically exposed right ovarian bursa of anaesthetized 6-week-old, female NOD-SCID mice (SPF Biotechnology, Beijing, China) by intrabursal injection (n\u0026thinsp;=\u0026thinsp;3). Around 40 days post-implantation, the tumor tissues from the primary sites and peritoneal disseminated sites (including the tumor nodules on the abdominal wall, the out surfaces of the organs in the peritoneal cavity, and the malignant ascites) were dissected from all the mice. The tumor tissues were then minced and further dissociated with type I collagenase (Sigma-Aldrich) for \u003cem\u003ein vitro\u003c/em\u003e culture and expansion in the presence of 5 \u0026micro;g/mL puromycin for two weeks. The expanded cells from all the disseminated sites of different mice were pooled for the second round of intrabursal injection into three NOD-SCID mice (2\u0026times;10\u003csup\u003e6\u003c/sup\u003e SK-OV-3 cells per mouse). Around 30 d post-implantation, we sacrificed the mice and collected tumor cells for \u003cem\u003ein vitro\u003c/em\u003e expansion and the third round of \u003cem\u003ein vivo\u003c/em\u003e screen. After three rounds of screen, the pooled peritoneal disseminated tumor cells from the mice in third round screen and the pooled cells from the primary tumor sites in the first-round screen were collected separately for genomic DNA extraction, followed by PCR amplification of the sgRNA-coding region and deep DNA sequencing analysis as described by Feng Zhang (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003eLive imaging of the OC progression\u003c/h2\u003e \u003cp\u003e50 days after the intraperitoneal injection of ID8 cells into the peritoneal cavity, C57BL/6 mice were anaesthetized and 100 \u0026micro;L D-luciferin (Sigma-Aldrich) was injected intraperitoneally. Bioluminescence images was acquired by using the IVIS Lumina II Imager (Caliper Life Sciences, Hopkinton, MA, USA).\u003c/p\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003ch2\u003eMetabolic labeling of sialylated proteins by AC\u003csub\u003e4\u003c/sub\u003eManNAz\u003c/h2\u003e \u003cp\u003eCells were cultured in indicated medium with 50 \u0026micro;M of tetra acetyl-N-azidoacetylmannosamine (AC\u003csub\u003e4\u003c/sub\u003eManNAz) for 72 h at 37\u0026deg;C. For the detection of sialylated-proteins, cells were lysed with RIPA buffer and the protein concentrations were adjusted to 5 \u0026micro;g/\u0026micro;L. 20 \u0026micro;L of the cell lysates (~\u0026thinsp;100 \u0026micro;g of total proteins) were incubated with 2 \u0026micro;L of 5 mM EZ-Link\u0026reg; Phosphine-PEG3-Biotin (Thermo-Fisher Scientific) at 37\u0026deg;C for 2\u0026ndash;4 h. After the addition of SDS loading buffer and heat denaturation, the proteins were resolved by SDS-PAGE and transferred onto PVDF membranes. The membranes were blocked with 5% defatted milk for 1 h at room temperature and incubated with HRP-conjugated streptavidin (Thermo-Fisher Scientific). For the fluorescent labeling of sialylated-proteins in cells, cells were cultured in the medium containing 50 \u0026micro;M of DBCO-Cy5 (Sigma-Aldrich) for 1 h at 37\u0026deg;C after the 72-hour incubation with AC\u003csub\u003e4\u003c/sub\u003eManNAz.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section3\"\u003e \u003ch2\u003eIdentification of sialylated proteins\u003c/h2\u003e \u003cp\u003eAfter the metabolic labeling of sialylated proteins in cultured cells with AC\u003csub\u003e4\u003c/sub\u003eManNAz as described above, the membrane proteins were extracted by using the Mem-PER Plus Membrane Protein Extraction Kit (Thermo-Fisher Scientific) and the concentration was adjusted to 5 \u0026micro;g/\u0026micro;L. The membrane proteins were then incubated with Phosphine-PEG3-Biotin for 4 h to conjugate the biotin. Subsequently, the pre-washed streptavidin agarose beads (Yeasen Biotechnology) were added to the conjugates and incubated overnight at 4\u0026deg;C with rotation. After washed with the wash buffer (150 mM NaCl, 20 mM NaH\u003csub\u003e2\u003c/sub\u003ePO\u003csub\u003e4\u003c/sub\u003e, pH 7.4) three times, the biotin-conjugated membrane proteins were eluted with the elution buffer (0.1M glycine-HCl, pH 2.8) and resolved by SDS-PAGE. The protein bands were visualized by Coomassie blue staining before the identification by mass spectrometry analysis (LC-MS/MS) (BGI, Beijing, China).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section3\"\u003e \u003ch2\u003eImmunoprecipitation\u003c/h2\u003e \u003cp\u003eCell lysates were incubated with the EGFR antibody (\u003cb\u003eSupplementary Table\u0026nbsp;6\u003c/b\u003e) for 4 h before the addition of pre-washed protein G agarose beads (CWBIO, Beijing, China) overnight at 4\u0026deg;C with rotation. After three washes, the proteins bond to the beads was eluted by boiling in 1\u0026times; SDS loading buffer and resolved by SDS-PAGE for western blot to analyze the sialylation of EGFR.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003ePrism 8.0 software was used for statistical analysis. Quantitative data were presented as means\u0026thinsp;\u0026plusmn;\u0026thinsp;SEM, and the differences between the groups were analyzed using the Student\u0026rsquo;s t-test. Two-way ANOVA test was used to analyze the continuous variables. Pearson correlation analysis was performed to examine the relation between the level of p-EGFR and the NEU4 expression. Differences are considered statistically significant at *p\u0026thinsp;\u0026lt;\u0026thinsp;0.05; **p\u0026thinsp;\u0026lt;\u0026thinsp;0.01; ***p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; ns means not significant.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contributions:\u003c/h2\u003e \u003cp\u003eJ.S., J. L., Y. S., P. Q., L. W. and S. Y. conceived the project. J.S., R. Z., Y. L., B.T., and S.W. performed experiments. Y. L., Y. C, J. Y., T.H., Y. M, S.W., and X. S. helped with the design of experiments. J.S. performed the data analysis. Y. S. supervised the study. J.S., J. L., L. W., and Y. S. wrote the manuscript with comments from all authors.\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eCompeting Interest Statement:\u003c/strong\u003e \u003c/p\u003e\u003cp\u003eThe authors declare no potential conflict of interests.\u003c/p\u003e \u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eThis work was supported by the grants from the National Natural Science Foundation of China [32070752, 32200641] and Nankai University Undergraduate Innovation and Entrepreneurship Project [202110055083].\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e \u003cp\u003eThe data generated in this study are available upon request from the corresponding authors.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBast RC, Jr., Hennessy B, \u0026amp; Mills GB (2009) The biology of ovarian cancer: new opportunities for translation. Nat Rev Cancer 9(6):415\u0026ndash;428.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCortes-Guiral D, \u003cem\u003eet al.\u003c/em\u003e (2021) Primary and metastatic peritoneal surface malignancies. Nat Rev Dis Primers 7(1):91.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTan DS, Agarwal R, \u0026amp; Kaye SB (2006) Mechanisms of transcoelomic metastasis in ovarian cancer. Lancet Oncol 7(11):925\u0026ndash;934.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMereiter S, Balmana M, Campos D, Gomes J, \u0026amp; Reis CA (2019) Glycosylation in the Era of Cancer-Targeted Therapy: Where Are We Heading? Cancer Cell 36(1):6\u0026ndash;16.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChristiansen MN, \u003cem\u003eet al.\u003c/em\u003e (2014) Cell surface protein glycosylation in cancer. Proteomics 14(4\u0026ndash;5):525\u0026ndash;546.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchjoldager KT, Narimatsu Y, Joshi HJ, \u0026amp; Clausen H (2020) Global view of human protein glycosylation pathways and functions. Nat Rev Mol Cell Biol 21(12):729\u0026ndash;749.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEsposito M, \u003cem\u003eet al.\u003c/em\u003e (2019) Bone vascular niche E-selectin induces mesenchymal-epithelial transition and Wnt activation in cancer cells to promote bone metastasis. Nat Cell Biol 21(5):627\u0026ndash;639.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarthel SR, \u003cem\u003eet al.\u003c/em\u003e (2009) Alpha 1,3 fucosyltransferases are master regulators of prostate cancer cell trafficking. Proc Natl Acad Sci U S A 106(46):19491\u0026ndash;19496.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRodrIguez E, Schetters STT, \u0026amp; van Kooyk Y (2018) The tumour glyco-code as a novel immune checkpoint for immunotherapy. Nat Rev Immunol 18(3):204\u0026ndash;211.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarcos NT, \u003cem\u003eet al.\u003c/em\u003e (2004) Role of the human ST6GalNAc-I and ST6GalNAc-II in the synthesis of the cancer-associated sialyl-Tn antigen. Cancer Res 64(19):7050\u0026ndash;7057.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan de Wall S, Santegoets KCM, van Houtum EJH, Bull C, \u0026amp; Adema GJ (2020) Sialoglycans and Siglecs Can Shape the Tumor Immune Microenvironment. Trends Immunol 41(4):274\u0026ndash;285.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu Y, \u003cem\u003eet al.\u003c/em\u003e (2020) Integrated Proteomic and Glycoproteomic Characterization of Human High-Grade Serous Ovarian Carcinoma. Cell Rep 33(3):108276.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePan J, \u003cem\u003eet al.\u003c/em\u003e (2020) Glycoproteomics-based signatures for tumor subtyping and clinical outcome prediction of high-grade serous ovarian cancer. Nat Commun 11(1):6139.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiyagi T, Takahashi K, Hata K, Shiozaki K, \u0026amp; Yamaguchi K (2012) Sialidase significance for cancer progression. Glycoconj J 29(8\u0026ndash;9):567\u0026ndash;577.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePshezhetsky AV \u0026amp; Ashmarina LI (2013) Desialylation of surface receptors as a new dimension in cell signaling. Biochemistry (Mosc) 78(7):736\u0026ndash;745.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKakugawa Y, \u003cem\u003eet al.\u003c/em\u003e (2002) Up-regulation of plasma membrane-associated ganglioside sialidase (Neu3) in human colon cancer and its involvement in apoptosis suppression. Proc Natl Acad Sci U S A 99(16):10718\u0026ndash;10723.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiyagi T, Wada T, \u0026amp; Yamaguchi K (2008) Roles of plasma membrane-associated sialidase NEU3 in human cancers. Biochim Biophys Acta 1780(3):532\u0026ndash;537.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUemura T, \u003cem\u003eet al.\u003c/em\u003e (2009) Contribution of sialidase NEU1 to suppression of metastasis of human colon cancer cells through desialylation of integrin beta4. Oncogene 28(9):1218\u0026ndash;1229.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSanjana NE, Shalem O, \u0026amp; Zhang F (2014) Improved vectors and genome-wide libraries for CRISPR screening. Nat Methods 11(8):783\u0026ndash;784.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShalem O, \u003cem\u003eet al.\u003c/em\u003e (2014) Genome-scale CRISPR-Cas9 knockout screening in human cells. Science 343(6166):84\u0026ndash;87.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBerek JS, Renz M, Kehoe S, Kumar L, \u0026amp; Friedlander M (2021) Cancer of the ovary, fallopian tube, and peritoneum: 2021 update. Int J Gynaecol Obstet 155 Suppl 1:61\u0026ndash;85.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVarki A (2008) Sialic acids in human health and disease. Trends Mol Med 14(8):351\u0026ndash;360.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeyrantepe V, \u003cem\u003eet al.\u003c/em\u003e (2004) Neu4, a novel human lysosomal lumen sialidase, confers normal phenotype to sialidosis and galactosialidosis cells. J Biol Chem 279(35):37021\u0026ndash;37029.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYamaguchi K, \u003cem\u003eet al.\u003c/em\u003e (2005) Evidence for mitochondrial localization of a novel human sialidase (NEU4). Biochem J 390(Pt 1):85\u0026ndash;93.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBigi A, \u003cem\u003eet al.\u003c/em\u003e (2010) Human sialidase NEU4 long and short are extrinsic proteins bound to outer mitochondrial membrane and the endoplasmic reticulum, respectively. Glycobiology 20(2):148\u0026ndash;157.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDomcke S, Sinha R, Levine DA, Sander C, \u0026amp; Schultz N (2013) Evaluating cell lines as tumour models by comparison of genomic profiles. Nat Commun 4:2126.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVergara D, \u003cem\u003eet al.\u003c/em\u003e (2010) Epithelial-mesenchymal transition in ovarian cancer. Cancer Lett 291(1):59\u0026ndash;66.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSheng Q \u0026amp; Liu J (2011) The therapeutic potential of targeting the EGFR family in epithelial ovarian cancer. Br J Cancer 104(8):1241\u0026ndash;1245.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShiozaki K, Yamaguchi K, Takahashi K, Moriya S, \u0026amp; Miyagi T (2011) Regulation of sialyl Lewis antigen expression in colon cancer cells by sialidase NEU4. J Biol Chem 286(24):21052\u0026ndash;21061.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang X, \u003cem\u003eet al.\u003c/em\u003e (2021) NEU4 inhibits motility of HCC cells by cleaving sialic acids on CD44. Oncogene 40(35):5427\u0026ndash;5440.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSilvestri I, \u003cem\u003eet al.\u003c/em\u003e (2014) Sialidase NEU4 is involved in glioblastoma stem cell survival. Cell Death Dis 5:e1381.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTringali C, \u003cem\u003eet al.\u003c/em\u003e (2012) NEU4L sialidase overexpression promotes beta-catenin signaling in neuroblastoma cells, enhancing stem-like malignant cell growth. Int J Cancer 131(8):1768\u0026ndash;1778.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSiwak DR, \u003cem\u003eet al.\u003c/em\u003e (2010) Targeting the epidermal growth factor receptor in epithelial ovarian cancer: current knowledge and future challenges. \u003cem\u003eJ Oncol\u003c/em\u003e 2010:568938.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShao M, \u003cem\u003eet al.\u003c/em\u003e (2012) Targeting the Insulin Growth Factor and the Vascular Endothelial Growth Factor Pathways in Ovarian Cancer. Molecular Cancer Therapeutics 11(7):1576\u0026ndash;1586.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBali; James G. Kench; Lyndal S. Edwards; Patricia M. Vanden Bergh; Neville F. Hacker; Robert L. Sutherland; Philippa M. O\u0026rsquo;Brien VAH-SMG-GSMHJSRASMJDMHLHKA (2004)\u0026thinsp;\u0026lt;\u0026thinsp;Overexpression of the cell adhesion molecules DDR1, Claudin 3, and Ep-CAM in metaplastic ovarian epithelium and ovarian cancer.pdf\u0026gt;. \u003cem\u003eClin Cancer Res\u003c/em\u003e 10(13): 4427\u0026ndash;4436.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQuan J, Yahata T, Adachi S, Yoshihara K, \u0026amp; Tanaka K (2011) Identification of Receptor Tyrosine Kinase, Discoidin Domain Receptor 1 (DDR1), as a Potential Biomarker for Serous Ovarian Cancer. International Journal of Molecular Sciences 12(2):971\u0026ndash;982.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoyano-Galceran L, \u003cem\u003eet al.\u003c/em\u003e (2020) Adaptive RSK-EphA2-GPRC5A signaling switch triggers chemotherapy resistance in ovarian cancer. EMBO Mol Med 12(4):e11177.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"oncogene","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"onc","sideBox":"Learn more about [Oncogene](http://www.nature.com/onc/)","snPcode":"41388","submissionUrl":"https://mts-onc.nature.com/cgi-bin/main.plex","title":"Oncogene","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Ovarian carcinoma, NEU4, sialylation, transcoelomic metastasis, EGFR","lastPublishedDoi":"10.21203/rs.3.rs-3772327/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3772327/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eGlycosylation profoundly influences the interactions between cancer cells and microenvironmental stromal cells during the peritoneal disseminated metastasis of ovarian carcinoma (OC), which is the major cause of cancer-related death. Although the characteristic cancer glycoconjugates are widely used as biomarkers for cancer diagnosis, our knowledge about cancer glycome remains quite fragmented due to the technique limitations in analyzing glycan chains with tremendous structural and functional heterogeneity. Given the dysregulated cancer glycome is defined by the altered glycosylation machinery, here we performed a systematic loss-of-function screen on 498 genes involved in glycosylation for key regulators of OC dissemination. We identified neuraminidase 4 (NEU4), an enzyme capable of hydrolyzing terminal sialic acid from glycoconjugates, as a vital peritoneal dissemination-promoting modifier of OC glycome. In human patients with high-grade serous OC (HGSOC), increased NEU4 was detected in the disseminated OC cells when compared with that in the primary tumor cells, which significantly correlated with the worse survival. Among three alternative splice-generated isoforms of human NEU4, we revealed that only the plasma membrane-localized NEU4 isoform 2 (NEU4-iso2) and intracellular isoform 3 promoted the peritoneal dissemination of OC by enhancing the cell motility and epithelial-mesenchymal transition. We also identified NEU4-iso2-regulated cell surface glycoproteome and found that NEU4-iso2 desialylated the epithelial growth factor receptor (EGFR), in particular at N\u003csup\u003e196\u003c/sup\u003e residue, for the hyperactivation of EGFR and its downstream tumor-promoting signaling cascades. Our results provide new insights into how the OC glycome is dysregulated during OC progression and reveals a functionally important glycosite on EGFR for its abnormal activation in cancer.\u003c/p\u003e","manuscriptTitle":"NEU4-mediated desialylation ignites the oncogenic receptors for the dissemination of ovarian carcinoma","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-01-03 02:09:45","doi":"10.21203/rs.3.rs-3772327/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"revise","date":"2024-01-29T16:13:03+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"This content is not available.","date":"2024-01-23T16:08:52+00:00","index":2,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2023-12-30T10:59:59+00:00","index":2,"fulltext":"This content is not available."},{"type":"editorInvitedReview","content":"This content is not available.","date":"2023-12-29T07:35:35+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2023-12-28T18:45:40+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewersInvited","content":"","date":"2023-12-28T16:23:55+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2023-12-19T11:47:03+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2023-12-18T14:04:58+00:00","index":"","fulltext":""},{"type":"submitted","content":"Oncogene","date":"2023-12-18T14:04:57+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"oncogene","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"onc","sideBox":"Learn more about [Oncogene](http://www.nature.com/onc/)","snPcode":"41388","submissionUrl":"https://mts-onc.nature.com/cgi-bin/main.plex","title":"Oncogene","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"e91fd8cf-8ac8-49ce-837f-02e5914aebc3","owner":[],"postedDate":"January 3rd, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":27845428,"name":"Biological sciences/Cancer/Gynaecological cancer/Ovarian cancer"},{"id":27845429,"name":"Biological sciences/Cell biology/Glycobiology"}],"tags":[],"updatedAt":"2024-10-15T07:05:50+00:00","versionOfRecord":{"articleIdentity":"rs-3772327","link":"https://doi.org/10.1038/s41388-024-03187-x","journal":{"identity":"oncogene","isVorOnly":false,"title":"Oncogene"},"publishedOn":"2024-10-14 04:00:00","publishedOnDateReadable":"October 14th, 2024"},"versionCreatedAt":"2024-01-03 02:09:45","video":"","vorDoi":"10.1038/s41388-024-03187-x","vorDoiUrl":"https://doi.org/10.1038/s41388-024-03187-x","workflowStages":[]},"version":"v1","identity":"rs-3772327","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3772327","identity":"rs-3772327","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}