1.The Study Of HSP90β In The Expression And Mechanism Of Lung Cancer 2.Dissecting The Dynamic Immune Microenvironment Of Colorectal Liver Metastases Through Single-Cell Sequencing

Part Ⅰ The study of HSP90β in the expression and mechanism of lung cancerChapter 1 Integrative proteomic analysis of lung adenocarcinoma identifies HSP90β as a potential prognostic markerBackground:Lung adenocarcinoma is the most common histological subtype of lung cancer.Although research has made progress,there are still a large number of lung adenocarcinoma patients without available targeted therapy options.Since proteins are the function performers of cells,in-depth characterization of the proteome and signal transduction of lung adenocarcinoma will lay the foundation for a comprehensive understanding of the molecular mechanism of the disease and the development of new treatment methods.Aim:Through large samples of clinical data,integrating proteomics and phosphorylated proteomics,combined with clinical information,identify differential proteins that are significantly related to prognosis,and find prognostic markers for lung adenocarcinoma.Methods:In this part of the study,a total of 103 patients with lung adenocarcinoma were enrolled.Patients’ primary tumor tissues and paired adjacent normal lung tissues were included.Through mass spectrometry and phosphorylation mass spectrometry,the differential proteins and phospho-proteins between tumor tissue and normal lung tissue were identified.The protein was verified in independent samples by Western Blot and enzyme-linked immunosorbent assay.Results:We obtained the proteomic characterization of 103 patients with lung adenocarcinoma,and compared the proteomic characteristics of tumor and paired normal lung tissues in patients with lung adenocarcinoma and obtained some differential proteins and lung signature proteins that were significantly related to clinical prognosis.Some cancer-related phosphorylated proteins have also been obtained through phosphorylation proteomics,and some prognostic markers including HSP90β have been identified.Finally,experiments have verified the high levels and poor prognosis of HSP90β in lung adenocarcinoma tissue and the plasma.Conclusion:Different proteome and phosphorylation proteomic characteristics were found in the tumor tissues and the paired adjacent lung tissues of patients with lung adenocarcinoma.The differentially expressed molecule HSP90β is expected to become a prognostic marker of lung adenocarcinoma.Chapter 2 Development-related protein HSP90β affects the prognosis of non-small cell lung cancer through cell cycle and apoptosisBackground:The occurrence of tumor is considered to be similar to early embryonic development in many aspects.Using normal developmental samples as research models helps us understand the pathogenesis of tumor and identify potential biomarkers.Rhesus monkeys can be used as model animals to explore embryonic development and tumorigenesis.The use of time series data of rhesus monkey lung tissue can help screen out biomarkers of non-small cell lung cancer.There are few studies on the role and prognosis of HSP90β in non-small cell lung cancer,and there is no research on the cellular mechanism of HSP90β in lung cancer.Aim:To determine the expression of HSP90β in early lung development through time-series transcriptome data of rhesus monkey lung tissue,and to explore the mechanism of HSP90β through lung cancer cell lines.Methods:In this study,we used the transcriptome data of rhesus macaque lung tissue and human lung development chip data to screen out the regulatory molecules in the early stage of lung development,and then combined with plasma samples of non-small cell lung cancer patients to verify the results.Finally,the mechanism has been clarified through cell phenotyping experiments and functional studies in lung cancer cell lines.Results:We identified HSP90β involved in the cell cycle regulation of early lung development in our data on the lung development of rhesus monkeys.Then we confirmed the significant high expression of HSP90β in the data of human lung development and non-small cell lung cancer plasma specimens.In lung cancer cell lines,we found that knocking down HSP90β affects the cycle and apoptosis of lung cancer cells through phosphorylation of key proteins.Conclusion:HSP90β is highly expressed in tumor tissues and in the plasma of non-small cell lung cancer patients.It leads to lung cancer cell cycle arrest and increased cell apoptosis by phosphorylation of key proteins.Chapter 3 A lung cancer prognosis assessment model using HSP90β combined with CEA,CA125 and CYFRA21-1Background:It lacks symptoms and effective detection of biomarkers for the early stage of tumor.Most patients with lung cancer have been found on the advanced stage at the time of diagnosis.In order to improve the survival of lung cancer patients,it is urgent to find useful biomarkers for early detection,prognosis prediction and recurrence monitoring.Recursive partitioning analysis can be used as a hierarchical tool in clinical research to predict prognosis of patients.CEA,CA125 and CYFRA21-1 are all commonly used clinical markers for auxiliary diagnosis.Aim:To establish a lung cancer prognostic risk prediction model by combining the plasma protein data of HSP90β and the serum data of CEA,CA125 and CYFRA21-1 to predict the prognostic risk of lung cancer patients,and to evaluate whether this prediction model can be used as an independent prognostic factor for lung cancer patients.Methods:This study first integrated the plasma samples of 1162 lung cancer cases and evaluated the expression levels of HSP90β,and then detected the expression of HSP90βin lung cancer tissues and adjacent normal tissues by immunohistochemistry.Finally,we combined the plasma expression data of HSP90β and CEA,CA125 and CYFRA21-1 serum data and established a risk prediction decision tree model using the recursive decision tree algorithm to predict the prognostic risk of 671 lung cancer patients.Results:We found that HSP90β protein was significantly highly expressed in the plasma and tumor tissues of lung cancer patients.By constructing a recursive partitioning decision tree prediction model,the prognosis risk of lung cancer patients can be predicted,and this prediction model can also be used as an independent prognostic factor for lung cancer patients.Conclusion:The recursive partitioning decision tree prediction model constructed by using HSP90β combined with CEA,CA125 and CYFRA21-1 can predict the prognosis risk of lung cancer patients and can be used as an independent prognostic factor for lung cancer patients.Part Ⅱ Dissecting the dynamic immune microenvironment of colorectal liver metastases through single-cell sequencingBackground:Colorectal cancer is one of the most common cancer in the world,and its incidence is increasing year by year.Colorectal liver metastases is the most common cause of death in patients with colorectal cancer.Elucidating the mechanism of colorectal liver metastases is an important direction of current research and is of great significance for guiding clinical treatment.Single-cell sequencing technology is a very useful tool that can be used to analyze the characteristics of multiple cell types in and around tumors.The immune repertoire can evaluate the diversity of the immune system in many ways,and can further evaluate the anti-tumor immune response of human body.At present,the research of single-cell sequencing and immune repertoire in tumors is becoming more and more mature.Aim:Using the sequencing of single-cell transcriptome and immune repertoire in colorectal liver metastases,to study the landscape of immune dynamic changes in patients with colorectal liver metastases during treatment,and to find key events that affect the prognosis.Methods:This part of the study enrolled the peripheral blood of 5 patients with colorectal liver metastases before surgery,seven days after surgery,and at the first evaluation of efficacy after chemotherapy.At the same time,the primary tumors and adjacent tissues as well as metastatic tumors and adjacent tissues resected during surgery were collected.The peripheral blood of the patient is separated to obtain mononuclear cells and polymorphonuclear cells,and the patient’s tissues are prepared with a single cell suspension,and then the CD45-positive cells are sorted.The cells separated from peripheral blood and the cells after tissue sorting were sequenced for single-cell transcriptome and TCR/BCR immune repertoire,and then the data was integrated and analyzed.Results:We obtained the immune landscape of 5 patients with colorectal liver metastases during treatment,and found that the same patient’s peripheral blood at different time points,as well as the primary and metastatic tissues directly shared T cell clones,while rarely shared between different patients.The diversity of the T-cell immune repertoire in the peripheral blood of patients increased significantly after surgery,and the diversity decreased to the level before surgery after chemotherapy.CD4 and CD8 effector T cell subsets in the peripheral blood increased significantly after surgery and chemotherapy,and the expression of effector molecules increased,showing an active cellular immune response.The proportion of CD4 effector T cells in tumor tissues is higher than that in adjacent tissues,while the proportion of CD8 memory T cells in adjacent tissues is higher than that in tumor tissues.The primary tumor tissue of colorectal cancer contains a large number of effector plasma cells,while the adjacent tissues and metastatic tissues are mainly B cells and memory B cells.In addition,a subset of neutrophils with high expression of FCGRT appeared in the peripheral blood of patients after chemotherapy.Conclusion:The diversity of T cell immune repertoire in patients with colorectal liver metastases increased significantly after surgery,and the proportion of effector T cells and effector molecules increased,showing an active adaptive immune response.The primary and metastatic tissues of colorectal liver metastases,as well as tumors and adjacent tissues have different immune characteristics,and the distribution of the same cell subsets is also different.