Investigation Of Biomarkers In Gastric Cancer Using ¹⁸O-labeling Quantitative Proteomics Coupled With Laser Capture Microdissection

Background: Gastric cancer (GC) is the fourth most common cancer and the second leading cause of cancer-related death behind lung cancer in the world. There are marked geographic variations in gastric cancer incidence, with the highest rates in China, Korea, Japan and South America. As an etiologically multi-factorial disease, carcinogenesis of gastric cancer may result from combined effects of helicobacter pylori, genetic and environmental factors. Although numerous efforts have been made to reveal the molecular mechanism of GC carcinogenesis, it remains poorly understood. And due to lack of suitable and specific biomarkers for early detection and monitoring of disease progress, most patients are diagnosed at an advanced stage. If being diagnosed in early stage, after operation, patients' five-year survival rates can reach 90% to 95%. But in advanced stage, the five-year survival rates are only 20-30%, mostly because of local recurrence and distant metastasis. In this regard, the identification of specific biomarkers is therefore desperately needed for gastric cancer.Proteomics has been introduced to the field of cancer research as an effective approach at identifying differentially expressed proteins associated with cancer development and progression. The high throughput proteomics approaches may uncover new biomarkers and therapeutic targets for cancer as well as reveal possible molecular mechanism underlying the disease. Because of the traditional two-dimensional gel electrophoresis (2DE) has a number of limitations, which include difficulties in analyzing membrane proteins and difficulties in analyzing proteins that have extreme isoelectric points and sizes, the reproducibility and sensitivity are lower. So it is necessary for developing higher throughout and more accurate proteomic technologies. Quantitative proteomics which are based on stable isotope labeling and mass spectrum are developing quickly. Compared with other isotopic labeling strategies, ¹⁸O labeling method has several advantages including simple labeling procedures, cheap reagents.Using clinical tissue samples from patients may be the most direct and persuasive way to find biomarkers and therapeutic targets for cancers by a proteomic approach. A major obstacle, however, to the analysis of tumor specimens is tissue heterogeneity. For identifying biomarkers more accurately by proteomic analysis, it is important to obtain homogeneous cell populations from a heterogeneous tissue. Laser capture microdissection (LCM) is a technique developed recently, by which specific cells may be separated from tissue slides, even from different stages and sites in a same sample. It is a very important technique in molecular pathology and tumor genomics study for its success in solving the question of cellular heterogeneity.To search for gastric adenocarcinoma biomarkers, we try to use LCM to purify the target cells from gastric adenocarcinoma (GA) and non-malignant gastric epithelial tissues (NMGET). ¹⁸O labeling quantitative proteomics would be preformed to identify the differentially expressed proteins between gastric adenocarcinoma cells (GAC) and normal gastric epithelial cells (NGEC). Furthermore, clincopathological significances of the differential proteins would be studied.Chapter 1 Identification of differentially expressed proteins by ¹⁸O stable isotope labeling quantitative proteomics and LCMObjective: To separate and identify differentially expressed proteins of GA. Methods: Firstly, we performed LCM to purify both GAC and NGEC from paired surgical specimens of human GA. Then, the proteins extracted from these cells were prefractionated by one-dimensional SDS-polyacrylamide gel electrophoresis (1D-SDS-PAGE). The fractionated proteins were trypsined and then labeled with ¹⁸O-water and ¹⁶O-water respectively. The two group labeled peptides were then mixed and analyzed by Nano-RPLC-MS/MS. Results: A total of 306 proteins were identified and quantified. Among these proteins, 78 proteins were differentially expressed between GAC and NGEC (42 proteins are up-regulated in GAC and 36 proteins are down-regulated). Conclusions:The procedure of ¹⁸O stable isotope labeling quantitative proteomics coupled with LCM could identify and quantify the differential proteins of GA accurately, which provides a new choice for investigating biomarkers of solid tumor.Chapter 2 Validation of differential proteins and study the clincopathological significancesObjective: To validate differential proteins and study the clincopathological significances of these differential proteins. Methods: Western blotting was used to detect the expression levels of five differential proteins (Moesin, Periostin, AnnexinA2, AnnexinA4 and RKIP) in GAC and NGEC. Immunohistochemistry (IHC) were performed to detected the expression of RKIP, S100A9, AnnexinA2 and AnnexinA4 in 70 cases of NMGET, 118 cases of GA, and matched positive lymph node (LN) specimens of primary GC. Results: (1) The results of western blotting showed that the express levels of five proteins. Moesin, Periostin, AnnexinA2, AnnexinA4 were higher in GAC than NGEC. RKIP was downexpressed in GAC versus NGEC. (2) The results of IHC showed, in the groups of NMGET, GA, and matched LN tissue, the positive expressed rates of RKIP were 97.14% (68/70),45.76% (54/118),0 (0/35), respectively; and positive expression of RKIP protein was closely correlated with deeper invasion, poor histological differentiation, TNM stage and lymphoid node metastasis in gastric cancer (P< 0.01). (3) in the groups of NMGET, GA, and matched LN tissue, the positive expressed rates of S100A9 were 32.86% (23/70),61.02% (72/118),82.86% (29/35), respectively; and expression of S100A9 has a positive correlation with TNM stage and lymphoid node metastasis in gastric cancer, but a negative correlation with histological differentiation (P< 0.05); and the expression of S100A9 was higher in intestinal type caner than diffuse type. (4) In the groups of NMGET and GA, the positive expressed rates of AnnexinA2 were 42.86% (30/70), 59.32% (70/118), respectively; and positive expression of AnnexinA2 were closely correlated with deeper invasion, poor histological differentiation, TNM stage and lymphoid node metastasis in gastric cancer (P< 0.05). (5) In the groups of NMGET and GA, the positive expressed rates of AnnenxinA4 were 38.57% (27/70), 55.93% (66/118), respectively; and positive expression of AnnexinA4 were closely correlated with deeper invasion, poor histological differentiation, TNM stage and lymphoid node metastasis in gastric cancer (P< 0.05). Conclusions: These results of western blotting and IHC were well consistent with the ¹⁸O quantitative proteomics data, which suggested that ¹⁸O quantitative proteomics are accurate and reliable. RKIP, S100A9, AnenxinA2 and AnnexinA4 proteins might affect the biological behavior of GC and be potential biomarkers for differentiation, invasion and metastasis of GC.