Mendelian randomization combined with multi-omics explores the relationship between heart failure and cancer.

Background: Whether there is an association between HF (HF) and cancer has not been conclusively established, and it is not clear whether patients with cancer can share similar hospitalization strategies and outcomes with patients with HF. Methods: Genome-wide association summary statistics were performed using a two-sample Mendelian randomization (MR) method for HF patients and cancer patients from the GWAS directory, with co-localization and Summary Data-Based Mendelian Randomization (SMR) analyses to identify HF-associated genes, and transcriptomic analyses to analyze the roles of these genes in the clinical diagnosis and targeted therapies of multiple cancer types. Results: Two-sample MR analysis showed that increased risk of HF was associated with decreased risk of cervical, brain, breast, colorectal, lung, and skin cancers, and co-localization combined with SMR analysis identified ABO and SURF1 as HF-associated genes, and transcriptomic analyses showed that ABO is a risk factor for HF and a protective factor against cancer, whereas SURF1 is a protective factor against HF and a protective factor against cancer. Conclusion: There was no causal relationship between heart failure and cancers (Cervical, brain, breast, colorectal, lung and skin cancers) risk factors, however there was a trend toward a negative causal relationship between heart failure and cancers (Cervical, brain, breast, colorectal, lung and skin cancers) occurrence.

IQGAP3 promotes the progression of glioma as an immune and prognostic marker.

Background: IQGAP3 plays a crucial role in regulating cell proliferation, division, and cytoskeletal organization. Abnormal expression of IQGAP3 has been linked to various tumors, but its function in glioma is not well understood. Methods: Various methods, including genetic differential analysis, single-cell analysis, ROC curve analysis, Cox regression, Kaplan-Meier analysis, and enrichment analysis, were employed to analyze the expression patterns, diagnostic potential, prognostic implications, and biological processes involving IQGAP3 in normal and tumor tissues. The impact of IQGAP3 on immune infiltration and the immune microenvironment in gliomas was evaluated using immunofluorescence. Additionally, the cBioPortal database was used to analyze copy number variations and mutation sites of IQGAP3. Experimental validation was also performed to assess the effects of IQGAP3 on glioma cells and explore underlying mechanisms. Results: High IQGAP3 expression in gliomas is associated with an unfavorable prognosis, particularly in wild-type IDH and 1p/19q non-codeleted gliomas. Enrichment analysis revealed that IQGAP3 is involved in regulating the cell cycle, PI3K/AKT signaling, p53 signaling, and PLK1-related pathways. Furthermore, IQGAP3 expression may be closely related to the immunosuppressive microenvironment of glioblastoma. BRD-K88742110 and LY-303511 are potential drugs for targeting IQGAP3 in anti-glioma therapy. In vitro experiments showed that downregulation of IQGAP3 inhibits the proliferation and migration of glioma cells, with the PLK1/PI3K/AKT pathway potentially playing a crucial role in IQGAP3-mediated glioma progression. Conclusion: IQGAP3 shows promise as a valuable biomarker for diagnosis, prognosis, and immunotherapeutic strategies in gliomas.

Highly efficient identification of nucleocytoplasmic O-glycosylation by the TurboID-based proximity labeling method in living cells.

Glycosylation is a ubiquitous posttranslational modification and plays an important role in many processes, such as protein stability, folding, processing, and trafficking. Among glycosylation types, O-glycosylation is difficult to analyze due to the complex glycan composition, low abundance and lack of glycosidases to remove the O-glycans. Many methods have been applied to analyze the O-glycosylation of membrane glycoproteins and secreted glycoproteins since the synthesis of O-glycosylation occurred in the Golgi apparatus. In recent years, some O-glycosylation has been reported in the nucleus. In this work, we present a proximity labeling strategy based on TurboID by combining core 1 ß1-3 galactosyltransferase (C1GalT1), which has been reported in the nucleus, to characterize nucleocytoplasmic O-glycosylation in living HeLa cells. The O-glycosylated protein C1GalT1 was biotinylated by the proximity labeling method in living HeLa cells overexpressing C1GalT1 fused by TurboID and enriched by streptavidin-coated beads. Following digestion with trypsin and mass spectrometry analysis, 68 high-confidence and 298 putative O-glycosylated sites were identified on 366 peptides mapped to 267 proteins. These results indicated that the proximity labeling method is a highly efficient technique to identify O-glycosylation. Furthermore, the finding of abundant O-glycosylation from nucleocytoplasmic proteins indicates a new pathway of O-glycosylation synthesis in cells.

Role of transcription factor FOXM1 in diabetes and its complications (Review).

Diabetes mellitus is a chronic metabolic disease commonly associated with complications such as cardiovascular disease, nephropathy and neuropathy, the incidence of which is increasing yearly. Transcription factor forkhead box M1 (FOXM1) serves an important role in development of diabetes and its complications. The present study aimed to review the association between FOXM1 with pathogenesis of diabetes and its complications. FOXM1 may be involved in development and progression of diabetes and its complications by regulating cell biological processes such as cell cycle, DNA damage repair, cell differentiation and epithelial­mesenchymal transition. FOXM1 is involved in regulation of insulin secretion and insulin resistance, and FOXM1 affects insulin secretion by regulating expression of insulin­related genes and signaling pathways; FOXM1 is involved in the inflammatory response in diabetes, and FOXM1 can regulate key genes associated with inflammatory response and immune cells, which in turn affects occurrence and development of the inflammatory response; finally, FOXM1 is involved in the regulation of diabetic complications such as cardiovascular disease, nephropathy and neuropathy. In summary, the transcription factor FOXM1 serves an important role in development of diabetes and its complications. Future studies should explore the mechanism of FOXM1 in diabetes and find new targets of FOXM1 as a potential treatment for diabetes and its complications.

AANL6 is a new efficient tool to probe non-reducing N-acetylglucosamine of N-linked glycans.

Terminal N-acetylglucosamine (GlcNAc) N-linked glycosylation is a truncated N-glycosylated modification that has been reported to be involved in various diseases, such as autoimmune diseases, cancers, and neurodegenerative diseases. New and simple tools will be always valuable for further characterization of the functions of this kind of glycosylation. Our previous paper proved that an optimized lectin created from Agrocybe aegerita GlcNAc selective lectin (AANL) named AANL6, can effectively identify O-GlcNAcylation, which is terminal GlcNAc O-linked glycosylation. We speculated that AANL6 could also be used to identify terminal GlcNAc N-linked glycosylation. Using therapeutic monoclonal antibodies as a model of terminal GlcNAc N-glycosylated proteins, we proved that AANL6 could selectively identify terminal GlcNAc N-linked glycosylation. The ratio of terminal GlcNAc N-linked glycosylation was increased by enrichment with AANL6 in human serum. Using cell membrane proteins as a complex sample, we found that AANL6 bound to the sperm surface, which expresses abundant terminal GlcNAc N-glycans, but did not bind to some tumor cell surfaces such A549 and MCF-7 cells, which is rich in high mannose glycoforms. In conclusion, AANL6 was identified as a powerful tool to probe terminal GlcNAc N-linked glycosylation and would be valuable for uncovering the function of this glycosylation.

Three-Dimensional Digitalized Virtual Planning of Free Anterior Tibial Artery Perforator Flap for Repairing Soft Tissue Defects in Extremities.

BACKGROUND: Traditional research methods have limited the application of anterior tibial artery perforator flap due to incomplete knowledge of the perforator. This study aimed to investigate the feasibility of three-dimensional digitalized virtual planning of free anterior tibial artery perforator flap for repairing soft tissue defects in extremities. METHODS: A total of 11 patients with soft tissue defects in extremities were included. The patient underwent computed tomography angiography (CTA) of bilateral lower limbs, and then the three-dimensional models of bones, arteries, and skin were constructed. Septocutaneous perforators with appropriate length and diameter were selected to design anterior tibial artery perforator flaps in software, and the virtual flaps were superimposed onto the patient's donor site in a translucent state. During the operation, the flaps were dissected and anastomosed to the proximal blood vessel of the defects as designed. RESULTS: Three-dimensional modeling showed clear anatomical relationships between bones, arteries, and skin. The origin, course, location, diameter, and length of the perforator obtained during the operation were consistent with those observed preoperatively. Eleven anterior tibial artery perforator flaps were successfully dissected and transplanted. Postoperative venous crisis occurred in one flap, partial epidermis necrosis occurred in another flap, while the remaining flaps completely survived. One flap was treated with debulking operation. The remaining flaps maintained aesthetic appearance, which did not affect the function of the affected limbs. CONCLUSIONS: Three-dimensional digitalized technology can provide comprehensive information on anterior tibial artery perforators, thus assisting in planning and dissecting patient-specific flaps for repairing soft tissue defects in extremities.

OGT Binding Peptide-Tagged Strategy Increases Protein O-GlcNAcylation Level in E. coli.

O-GlcNAcylation is a single glycosylation of GlcNAc mediated by OGT, which regulates the function of substrate proteins and is closely related to many diseases. However, a large number of O-GlcNAc-modified target proteins are costly, inefficient, and complicated to prepare. In this study, an OGT binding peptide (OBP)-tagged strategy for improving the proportion of O-GlcNAc modification was established successfully in E. coli. OBP (P1, P2, or P3) was fused with target protein Tau as tagged Tau. Tau or tagged Tau was co-constructed with OGT into a vector expressed in E. coli. Compared with Tau, the O-GlcNAc level of P1Tau and TauP1 increased 4~6-fold. Moreover, the P1Tau and TauP1 increased the O-GlcNAc-modified homogeneity. The high O-GlcNAcylation on P1Tau resulted in a significantly slower aggregation rate than Tau in vitro. This strategy was also used successfully to increase the O-GlcNAc level of c-Myc and H2B. These results indicated that the OBP-tagged strategy was a successful approach to improve the O-GlcNAcylation of a target protein for further functional research.

Study of FOXO1-interacting proteins using TurboID-based proximity labeling technology.

BACKGROUND: Protein‒protein interactions (PPIs) are the foundation of the life activities of cells. TurboID is a biotin ligase with higher catalytic efficiency than BioID or APEX that reduces the required labeling time from 18 h to 10 min. Since many proteins participate in binding and catalytic events that are very short-lived, it is theoretically possible to find relatively novel binding proteins using the TurboID technique. Cell proliferation, apoptosis, autophagy, oxidative stress and metabolic disorders underlie many diseases, and forkhead box transcription factor 1 (FOXO1) plays a key role in these physiological and pathological processes. RESULTS: The FOXO1-TurboID fusion gene was transfected into U251 astrocytes, and a cell line stably expressing FOXO1 was constructed. While constructing the FOXO1 overexpression plasmid, we also added the gene sequence of TurboID to perform biotin labeling experiments in the successfully fabricated cell line to look for FOXO1 reciprocal proteins. Label-free mass spectrometry analysis was performed, and 325 interacting proteins were found. A total of 176 proteins were identified in the FOXO1 overexpression group, and 227 proteins were identified in the Lipopolysaccharide -treated group (Lipopolysaccharide, LPS). Wild-type U251 cells were used to exclude interference from nonspecific binding. The FOXO1-interacting proteins hnRNPK and RBM14 were selected for immunoprecipitation and immunofluorescence verification. CONCLUSION: The TurboID technique was used to select the FOXO1-interacting proteins, and after removing the proteins identified in the blank group, a large number of interacting proteins were found in both positive groups. This study lays a foundation for further study of the function of FOXO1 and the regulatory network in which it is involved.

Efficient TurboID-based proximity labelling method for identifying terminal sialic acid glycosylation in living cells.

TurboID, a proximity labelling method based on mutant biotin ligase, is an efficient new technique for recognizing protein-protein interactions and has been successfully applied to living cells. Sialic acid is typically the terminal monosaccharide attached to many glycoproteins and plays many important roles in many biological processes. However, the lack of enrichment methods for terminal sialic acid glycosylation in vivo hinders the identification and analysis of this glycosylation. Here, we introduce TurboID to identify terminal sialic acid glycosylation in living cells. SpCBM, the carbohydrate-binding domain of sialidase from Streptococcus pneumoniae, is fused with TurboID and overexpressed in HeLa cells. After streptavidin-based purification and detection by mass spectrometry, 31 terminal sialic acid N-glycosylated sites and 1359 putative terminal sialic acid glycosylated proteins are identified, many of which are located in the cytoplasm and nucleus.

Identification of novel combined biomarkers in the diagnosis of multiple myeloma.

PURPOSE: Multiple myeloma (MM) is a haematological malignant disease with a clonal proliferation of plasma cells, and timely surveillance is helpful to improve the survival rate of patients with MM. However, there is a lack of simple and effective biomarkers for the diagnosis, prognosis, and residual disease evaluation of MM. MATERIAL & METHODS: In the detection cohort, we used the samples from six newly diagnosed MM patients and six control subjects. Plasma proteins were labelled with dimethyl reagents and enriched by lectin AANL6, then the deglycosylated peptides were identified by LC-MS/MS. Differentially expressed proteins were used for further exploration. In the validation cohort, we used 90 newly diagnosed patients with MM and 70 cases of unrelated diseases as controls. The diagnosis performance was analysed by ROC analysis using SPSS. RESULTS: In this study, we show, using lectin blots with AANL6, that glycosylation levels were higher in MM patients than in controls. After AANL6 enrichment, we detected 58 differentially expressed proteins using quantitative proteomics. We further validated one candidate Fibulin-1 (FBLN1). Using an Elisa assay, we showed that FBLN1 expression was increased in plasma of 90 cases of MM, and which was significantly correlated with DKK1 expression. ROC analysis showed that these two markers had a 95.7% specificity for determining the diagnosis of MM. CONCLUSION: These data suggest that the MM cases display increased glycosylation after AANL6 enrichment and that the combined expression of FBLN1 and DKK1 can be used as an effective diagnostic biomarker.

Synergistic effects of AAGL and anti-PD-1 on hepatocellular carcinoma through lymphocyte recruitment to the liver.

OBJECTIVE: Therapy for hepatocellular carcinoma (HCC) is a major challenge, and targeted therapies provide only a modest benefit in terms of overall survival. Treatment with antibodies to programmed cell death protein 1 (PD-1)/PD-L1 can restore the functions of tumor-infiltrating T cells in HCC and has shown clinical efficacy in 20% of patients with advanced HCC. Novel approaches are urgently needed to treat HCC and to augment the efficacy of immunotherapy. METHODS: Tumor-bearing mice were treated with Agrocybe aegerita galectin (AAGL) alone or in combination with anti-PD-1, and the tumor sizes and lifespans of mice were determined. Transcriptome analysis, cytokine analysis, flow cytometry analysis of the number and proportion of immune cell subsets in the liver and spleen, and molecular and cellular analyses of tumors were used to define the underlying mechanisms. RESULTS: AAGL significantly inhibited the growth of liver tumors in a dose-dependent manner. Furthermore, AAGL increased the expression of multiple cytokines and chemokines in tumor-bearing mouse livers; this effect was associated with the activation and migration of T cells and macrophages, in agreement with the in vitro results. Importantly, the aggregation of T cells and macrophages induced by AAGL in tumor-bearing mouse livers clearly enhanced the response to PD-1 blockade immunotherapy. CONCLUSIONS: The results showed that AAGL induced the activation and migration of lymphocytes to the liver, and that the combination of AAGL and anti-PD-1 may be a promising strategy for HCC treatment.

CBS homogenization mutation strategy narrows the glycan binding profile of a GlcNAc-specific lectin AANL.

Glycosylation plays important roles in many cellular processes, such as signal transduction, cell cycle progression and transcriptional regulation. However, the identification and analysis of glycosylation are severely hampered by the low specificity or avidity of antiglycan antibodies and lectins. We have reported that a lectin AANL, which has high specificity for terminal GlcNAc glycans and contains six carbohydrate binding sites (CBSs), was used to enrich O-GlcNAcylated peptides. To further improve AANL binding specificity, we designed a CBS-homogenization strategy and restructured six mutant lectins, known as AANL1-AANL6. Affinity chromatography with GlcNAc and isothermal titration calorimetry analysis indicated that the two mutants (AANL3 and AANL6) all maintained GlcNAc binding activity. AANL6 and AANL3 showed higher specificity for terminal GlcNAc glycans than AANL, as shown by the hemagglutination assay, cell binding assays and glycan microarray analysis, and AANL6 exhibited the highest specificity. The binding activity of AANL6 for O-GlcNAcylated peptides was shown by surface plasmon resonance assays. By AANL6 affinity chromatography enrichment and mass spectrometry analysis, 79 high-confidence and 21 putative O-GlcNAcylated sites were identified on 85 peptides mapped onto 54 proteins. Most of these sites were new sites compared with reported data. These results indicate that the enrichment capacity of AANL6 is higher than that of wild-type AANL. In conclusion, the CBS-homogenization mutation strategy was successful, and AANL6 was identified as a powerful tool for O-GlcNAcylation enrichment. Our research suggests that the CBS-homogenization strategy is valuable for improving the specificity of lectins with multiple CBSs.

Dramatically changed immune-related molecules as early diagnostic biomarkers of non-small cell lung cancer.

Non-small cell lung cancer (NSCLC) is the main type of lung cancer, with a low 5-year survival rate because of the absence of effective clinical biomarkers for early diagnosis. Based on the immunosurveillance theory, we proposed that changes in the immune system are more pronounced than tumour-associated antigens during the early stage of cancer. Therefore, a new strategy was designed to screen early diagnostic biomarkers from peripheral leukocytes in early-stage NSCLCs with transcriptome sequencing. A total of 358 immune-related differentially expressed genes were identified between early-NSCLC patients and healthy individuals. Orosomucoid-1 (ORM1, a acute phase protein), the total ORM and chitotriosidase-1 (involved in degradation of chitobiose) were selected for further verification in 210 serum samples by western blotting, ELISA and nephelometry immunoassay (based on immuno-scatter turbidmetry). Receiver operating characteristic curve analysis show that ORM1 and total ORM have excellent diagnostic efficacies, with area under the curve of 0.862 and 0.920, respectively, which significantly distinguished very early-NSCLC (IA) from healthy samples. Flow cytometry results showed that CD15+ neutrophils made up 73% of ORM1+ peripheral leukocytes. In mouse lung cancer model, serum ORM1, but not liver ORM1, changed significantly in the early stage of NSCLC. ORM1 expression in peripheral leukocytes was regulated by TGF-ß and mediated by the TGF-ß/Smad signalling pathway. Our results indicated that combined ORM and TGF-ß could be a promising clinical biomarker in the diagnosis of early NSCLC.

Critical role of IL-1ß in the pathogenesis of Agrocybe aegerita galectin-induced liver injury through recruiting T cell to liver.

Acute liver failure (ALF) can be the consequence of various etiologies, which immune response plays a pivotal role in the pathogenesis. For the diversity of etiologies, more animal models are still needed in this field. Here, we developed a new acute liver injury mouse model induced by a fungal lectin AAGL (Agrocybe aegerita galectin). Intravenous injection of AAGL could induce the infiltration and activation of T, NKT and NK cells in liver and T cell played an important role in the pathogenesis. However, compared with the widely used concanavalin A model, AAGL model showed different immune mechanism. Transcriptome analysis of live tissue suggested that inflammation mediated by chemokine and cytokine signaling pathway was different between AAGL and Con A model. Fluorescent quantitative PCR verification assay showed that IL-1ß was expressed much higher in AAGL-treated mice and anti-IL-1ß could ameliorate AAGL-induced liver injury by inhibiting NF-κB and p38 signaling pathway. The expression of CXCL9 which was responsible for T cell infiltration in liver was also inhibited in AAGL model. We found a critical role of IL-1ß in the pathogenesis of AAGL model through recruiting T cells to liver, which highlighted that IL-1ß antibody might be a candidate therapy for ALF.

Identification a novel clinical biomarker in early diagnosis of human non-small cell lung cancer.

Non-small cell lung cancer (NSCLC) is a malignant tumor with high morbidity and mortality. The clinical biomarkers currently used for the early diagnosis of lung cancer have poor sensitivity and specificity. Therefore, it is urgent to identify sensitive biomarkers for the early detection of NSCLC to improve the patient survival of patients. In our previously study, we identified glycoprotein alpha-1-antichymotrypsin (AACT) as an early biomarker of NSCLC. In this study, serum glycopeptides were enriched using the high-GlcNAc-specific binding lectin, AANL/AAL2, for further quantitative proteomics analysis using LC-MS/MS. A total of 55 differentially expressed proteins were identified by using demethylation labelling proteomics. Serum paraoxonase/arylesterase 1 (PON1) was selected for validation by western blotting and lectin-ELISA in samples from 120 enrolled patients. Our data showed that AANL-enriched PON1 has better diagnostic performance than total PON1 in early NSCLC, since it differed between early Stage I tumor samples and tumor-free samples (healthy and benign). Combining AANL-enriched PON1 with carcinoembryonic antigen (CEA) significantly improved the diagnostic specificity of CEA. Moreover, combined AANL-enriched PON1 and AANL-enriched AACT was significantly different between early NSCLC samples and tumor-free samples with an AUC of 0.940, 94.4% sensitivity, and 90.2% specificity. Our findings suggest that combined AANL-enriched PON1 and AANL-enriched AACT is a potential clinical biomarker for the early diagnosis of NSCLC.

Identification and quantification of proteins at adsorption layer of emulsion stabilized by pea protein isolates.

This study adopted the method of quantitative proteomics to analyze the adsorbed proteins in oil-in-water emulsions stabilized by pea protein isolate (PPI). Adsorbed proteins were precipitated by an optimized precipitation method and precipitates were labeled and subjected to a reversed-phase high performance liquid chromatography coupled to tandem mass spectrometry (RPLC-ESI-MS/MS) for protein identification and quantification. In total, 77 proteins were identified, of which 49 proteins with significant differences were observed. There were 25 upregulated proteins (fold change > 1) and 24 downregulated proteins (fold change < 1). The interfacial adsorption abilities of these proteins were compared according to the classification of protein families. The results showed that all isoforms of vicilins exhibited high adsorption abilities at the oil-water interface. Compared with vicilin, convicilin showed opposite adsorption capacity. Different legumin families showed significantly different affinities on the oil-water interface. In contrast to albumin-1, albumin-2 was preferentially adsorbed to the interface. The amino acid sequence alignment and hydropathy profile analysis of these proteins showed that the proteins well-balanced between hydrophobic and hydrophilic amino acid groups displayed high interfacial activity. In contrast, a long hydrophilic or hydrophobic fragment could adversely influence protein interfacial activity. This study provides an insight into the interfacial behaviors of proteins by supplying detailed quantitative information of interfacial layer.

AANL (Agrocybe aegerita lectin 2) is a new facile tool to probe for O-GlcNAcylation.

O-linked N-acetylglucosamine (O-GlcNAcylation) is an important post-translational modification on serine or threonine of proteins, mainly observed in nucleus or cytoplasm. O-GlcNAcylation regulates many cell processes, including transcription, cell cycle, neural development and nascent polypeptide chains stabilization. However, the facile identification of O-GlcNAc is a major bottleneck in O-GlcNAcylation research. Herein, we report that a lectin, Agrocybe aegerita GlcNAc-specific lectin (AANL), also reported as AAL2, can be used as a powerful probe for O-GlcNAc identification. Glycan array analyses and surface plasmon resonance (SPR) assays show that AANL binds to GlcNAc with a dissociation constant (KD) of 94.6 µM, which is consistent with the result tested through isothiocyanate (ITC) assay reported before (Jiang S, Chen Y, Wang M, Yin Y, Pan Y, Gu B, Yu G, Li Y, Wong BH, Liang Y, et al. 2012. A novel lectin from Agrocybe aegerita shows high binding selectivity for terminal N-acetylglucosamine. Biochem J. 443:369-378.). Confocal imaging shows that AANL co-localizes extensively with NUP62, a heavily O-GlcNAcylated and abundant nuclear pore glycoprotein. Furthermore, O-GlcNAc-modified peptides could be effectively enriched in the late flow-through peak from simple samples by using affinity columns Sepharose 4B-AANL or POROS-AANL. Therefore, using AANL affinity column, we identified 28 high-confidence O-linked HexNAc-modified peptides mapped on 17 proteins involving diverse cellular progresses, including transcription, hydrolysis progress, urea cycle, alcohol metabolism and cell cycle. And most importantly, major proteins and sites were not annotated in the dbOGAP database. These results suggest that the AANL lectin is a new useful tool for enrichment and identification of O-GlcNAcylated proteins and peptides.

Identification of GlcNAcylated alpha-1-antichymotrypsin as an early biomarker in human non-small-cell lung cancer by quantitative proteomic analysis with two lectins.

BACKGROUND: Non-small-cell lung cancer (NSCLC) is the main type of lung cancer with high mortality rates in worldwide. There is a need to identify better biomarkers to detect NSCLC at an early stage as this will improve therapeutic effect and patient survival rates. METHODS: Two lectins (AAL/AAGL and AAL2/AANL), which specifically bind to tumour-related glycan antigens, were first used to enrich serum glycoproteins from the serum of early NSCLC patients, benign lung diseases subjects and healthy individuals. The samples were investigated by using iTRAQ labelling and LC-MS/MS. RESULTS: A total of 53 differentially expressed proteins were identified by quantitative proteomics and four glycoproteins (AACT, AGP1, CFB and HPX) were selected for further verification by western blotting. Receiver operating characteristic analysis showed AACT was the best candidate for early NSCLC diagnosis of the four proteins, with 94.1% sensitivity in distinguishing early tumour Stage (IA+IB) from tumour-free samples (healthy and benign samples, HB). The GlcNAcylated AACT was further detected by lectin-based ELISA and has better advantage in clinical application than total AACT. The GlcNAcylated AACT can effectively differentiate Stage I from HB samples with an AUC of 0.908 and 90.9% sensitivity at a specificity of 86.2%. A combination of GlcNAcylated AACT and carcinoembryonic antigen (CEA) was able to effectively differing Stage I from HB samples (AUC=0.914), which significantly improve the specificity of CEA. The combination application also has the better clinical diagnostic efficacy in distinguishing cancer (NSCLC) from HB samples than CEA or GlcNAcylated AACT used alone, and yielded an AUC of 0.817 with 93.1% specificity. CONCLUSIONS: Our findings suggest that the GlcNAcylated AACT will be a promising clinical biomarker in diagnosis of early NSCLC.