Dataset Information

Machine Learning Reveals Lipidome Dynamics in a Mouse Model of Ovarian Cancer

ABSTRACT: Ovarian cancer (OC) is one of the deadliest cancers affecting the female reproductive system. It presents little or no symptoms at the early stages, and typically unspecific symptoms at later stages. Of the OC subtypes, high-grade serous carcinoma (HGSC) is responsible for most OC deaths. However, very little is known about the metabolic course of this disease. In this longitudinal study, we investigated the temporal course of lipidome changes in a Dicer-Pten Double-Knockout (DKO) HGSC mouse model using machine and statistical learning approaches. Early progression of HGSC was marked by increased levels of phosphatidylcholines and phosphatidylethanolamines. In contrast, later stages were marked by more diverse lipids alterations, including fatty acids and their derivatives, triglycerides, ceramides, hexosylceramides, sphingomyelins, lysophosphatidylcholines, and phosphatidylinositols. These alterations provided evidence of perturbations in cell membrane stability, proliferation, and survival and candidates for early-stage and prognostic markers in humans.

ORGANISM(S): Mouse Mus Musculus

TISSUE(S): Blood

SUBMITTER: Samyukta Sah

PROVIDER: ST002276 | MetabolomicsWorkbench | Thu Sep 01 00:00:00 BST 2022

REPOSITORIES: MetabolomicsWorkbench

ACCESS DATA

Dataset's files

Source:

			Action	DRS
	mwtab	Other

Items per page:

1 - 1 of 1

Similar Datasets

Project description:Ovarian cancer (OC) ranks among the deadliest cancers in women. Lack of symptoms, rapid spread of metastases and common chemoresistance contribute to unfortunate fate of majority of OC patients, especially those having high-grade serous carcinoma of the ovary, fallopian tube and peritoneum (HGSC), the most common type of OC. Many HGSC patients have excess fluid in the peritoneum at the stage of diagnosis called ascites. Ascites is basically a tumor microenvironment (TME) in the liquid form containing various cells, proteins and also extracellular vesicles (EVs). EVs are small membrane-bound particles that convey proteins, lipids and nucleic acids between cells and their cargo reflects the cell of origin. EVs play important role in cancerogenesis and hold great promise as disease biomarkers as well as potential therapeutic targets. Small size and polydispersity of EVs bring various challenges to their isolation and characterization, including method-dependent enrichment of different EV subtypes as well as contaminants. We isolated EVs from ascites of 11 patients by two different methods (ultracentrifugation coupled to sucrose cushion and size-exclusion chromatography, qEV column) and analyzed them by mass spectrometry. We identified core ascitic EV proteins present in all patients that contain typical EV markers and are devoid of method-dependent contaminants. To cover interpatient heterogeneity, we expanded this “core proteome” with proteins found in majority of patients. Next, we compared them with proteins of EVs from related control fluids and uncovered proteins present only on EVs from HGSC patients. We believe this list of proteins contain both important players of HGSC progression as well as potential biomarkers. Using single cell RNA sequencing data, we mapped the origin of EVs to different types of cells present in malignant ascites. Our results suggest that EVs in ascites do not come predominantly from tumor cells, but rather from variety of non-malignant cell types including cancer-associated fibroblasts and tumor-associated macrophages, which presence in ascites was confirmed by flow cytometry. This emphasizes the recently appreciated role of TME in the progression of HGSC. To conclude, this is the first study attempting to link EV composition to the cell types producing it. As such it opens numerous avenues both for better understanding of EV role in tumor promotion/prevention and for the improved HGSC diagnostics.

Dataset Information

Machine Learning Reveals Lipidome Dynamics in a Mouse Model of Ovarian Cancer

Dataset's files

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets