Study On Glycoprotein Glycan Profiles Of Tissues And Saliva Of Patients With Lung Cancer And Construction Of Lung Cancer Diagnosis Models Based On Salivary Glycan Types

Background:Lung cancer is the most malignant disease with high incidence rate and mortality rate in China,which seriously endangers human health.As one of the important post-translational modifications of proteins,glycosylation is closely related to the occurrence and development of cancer.Studies have shown that there are changes in glycoprotein glycan structure in lung cancer tissues,serum and pleural effusion.However,it is not clear whether there are similar changes in glycoprotein glycans in different types of lung cancer tissues and whether there are similar changes in saliva of lung cancer patients.Therefore,in this study,lectin microarray technology and matrix assisted laser desorption / ionization time-of-flight mass spectrometry(MALDI-TOF / TOF-MS)were used to study the specific glycan structures related to lung cancer in tumor tissues and saliva of patients with squamous cell carcinoma(SCC),lung adenocarcinoma(ADC)and small cell lung cancer(SCLC).In order to provide theoretical basis for molecular mechanism of abnormal glycosylation of glycoproteins associated with lung cancer and to find biomarkers for diagnosis of lung cancer,Method:1.Using lectin microarrays to analyze the glycoprotein glycan profiles of tumor tissues and adjacent tissues from SCC patients,and screen abnormal glycans of glycoproteins associated with SCC,q RT-PCR and lectin histochemistry were used to verify the expression and distribution of the abnormal glycoprotein glycan.MALDI-TOF / TOF-MS technology was used to further analyze the N-glycan structure of SCC tumor tissues and adjacent tissues.2.Using lectin microarrays to detect the glycoprotein glycan profiles of tumor tissues and adjacent tissues from ADC patients,screen the abnormal glycoprotein glycans related to ADC,and verify the expression and distribution of the abnormal glycoprotein glycans by lectin blotting and lectin histochemistry.And further analyze the N-/ O-glycan structures of ADC tumor tissue and adjacent tissue by MALDI-TOF / TOF-MS.3.Using lectin microarrays to detect the salivary glycoprotein profiles of healthy volunteers(HV),SCC,ADC and SCLC patients,and screen abnormal salivary glycoprotein glycans associated with lung cancer.Combined with binary logistic stepwise regression analysis,the diagnostic models of SCC,ADC and SCLC were constructed,and the diagnostic accuracy of the models was evaluated by ROC curve.4.N-/ O-glycans were isolated from salivary glycoproteins from HV and patients with SCC,ADC and SCLC by PNGase F digestion and NaClO oxidation,respectively.Then,MALDITOF / TOF-MS was used to analyze their structure of N-/ O-glycans.5.Systematic comparative analysis of the differences of glycoprotein glycan profiles among tumor tissues,serum and saliva in patients with SCC,ADC and SCLC.Results:1.Through the analysis of glycoprotein glycan profiles of 7 pairs of SCC tumor tissues and their adjacent tissues,the results showed that the relative abundance of Fuc?-1,6Glc NAc(core fucose)in SCC tumor tissues was significantly higher than that in adjacent tissues,The relative abundance of Fuc?-1,6Glc NAc structure and the expression level of fut8 gene which regulates the synthesis of Fuc?-1,6Glc NAc in SCC tumor tissues were significantly higher than those in paracancerous tissues.Lectin histochemical results showed that Fuc?-1,6Glc NAc structure accumulated abnormally on the cell surface and cytoplasmic region of SCC tumor tissue.Nglycan structure identification results of SCC tissue showed that the expression level of Fuc?-1,6Glc NAc structure in SCC tumor tissue was significantly higher than that in adjacent tissues Five fucosylated N-glycans(m/z 2564.911,2816.025,3175.17,3273.218 and 3433.255)were found only in SCC tumor tissues.2.Through the analysis of glycoprotein glycan profiles of 7 pairs of ADC tumor tissues and their adjacent tissues,the results showed that the relative abundance of Fuc?1-6 Glc NAc,Fuc?1-3(Gal?1-4)Glc NAc structures recognized by AAL,Bisecting Glc NAc,bi-antennary Nglycans,tri-and tetra-antennary complex-type structures recognized by PHA-E + L were significantly higher in ADC tumor tissues than that in adjacent tissues.The results of lectin blotting were consistent with those of glycoprotein glycan profiles.Lectin histochemical results showed that the expression of complex type N-glycan was increased on the cell surface and cytoplasm of ADC tumor tissues and the identification results of N-/ O-glycan structures of ADC tissue showed that there were 12 N-glycans(e.g.,M / z1479.5471647.587 and 1869.711)and 8 O-glycans(e.g,M / Z 604.203,973.237 and 1005.320)were only found in ADC tumor tissues.3.The results of lectin microarrays showed that the changes of glycoprotein glycopatterns could distinguish SCC,ADC and SCLC patients.The model LC was used to determine whether there was lung cancer(AUC: 0.971,sensitivity: 0.953,specificity: 0.810)in the experimental set,and the model SCC(AUC: 0.930,sensitivity: 0.913,specificity: 0.842),Model ADC(AUC: 0.892,sensitivity: 0.877,specificity: 0.735)and model SCLC(AUC: 0.826,sensitivity: 0.882,specificity: 0.714)were used to determine whether or not to have SCC,ADC and SCLC achieved good diagnostic effect in the training set.In addition,the AUC values of model LC,model SCC,model ADC and model SCLC were all greater than 0.700 in the validation set.4.There were 23/38,27/35,37/20 and 35/27 N/O-glycans to be identified in salivary glycoproteins of patients with HV,SCC,ADC and SCLC,respectively.Among them,6 Nglycans(e.g.,m/z 1876.649,2012.719 and 2094.748)and 3 O-glycans(m/z 604.203774.272 and 958.307)were found only in SCC group,and five N-glycans(e.g,m/z 2061.73852303.814 and 2361.856,etc.)and four O-glycans(e.g.,m/z 1156.420,1213.440 and 1390.490)were only found in ADC group.Four N-glycans(e.g.,m/z1768.570,2319.809 and 2687.990)and 4 Oglycans(e.g.,m/z 1077.340,1208.396 and 1223.400)were only found in SCLC group.5.The results of comparative analysis of glycoprotein glycan profiles in tumor tissues,sera and saliva of patients with SCC / ADC / SCLC showed that there were 34 lectins to recognize glycoprotein glycans that had significant differences among tumor tissues,serum and saliva of patients with lung cancer.Among them,24 lectins(e.g.,STL,PNA,Con A)showed significant differences in tumor tissue,serum and saliva of SCC patients,27 lectins(e.g.,jacalin,HHL and DBA)showed significant differences in tumor tissue,serum and saliva of ADC patients,and21 lectins(e.g.,PHA-E,HHL and DBA)showed significant differences in tumor tissue,serum and saliva of SCLC patients.Conclusion:There were significant differences in glycoprotein glycans between lung cancer tissues and their adjacent tissues,and there were significant differences in glycoprotein glycans between the saliva of lung cancer patients and healthy volunteers.The glycoproteins isolated from lung cancer tissues and saliva of lung cancer patients had specifically expressed N/O-glycan structures.The lung cancer diagnosis model based on salivary glycoprotein glycopatterns has good ability to distinguish lung cancer patients.In addition,there were significant differences in glycoprotein glycan profiles among tissues,serum and saliva of patients with lung cancer.In this study,we systematically analyzed the glycoprotein glycan profiles in lung cancer tissues and saliva,and provided a theoretical basis for elucidating the molecular mechanism of abnormal glycosylation of glycoproteins associated with lung cancer and the discovery of glycan biomarkers for lung cancer diagnosis.