NMR based metabolic profiling of liver cancer

Yuliana Suryani, Purdue University

Abstract

1H and 31P magic angle spinning (MAS) NMR spectroscopy in conjunction with multivariate statistical analysis and 1H and 31P heterostatistical total correlation NMR spectroscopy (1H-31P HET-STOCSY) were used to investigate changes in the hepatic metabolome correlated with the development of human hepatocellular carcinoma (HCC) from tumor, tumor adjacent, and liver tissue from patients without HCC (control). HCC tumor tissue showed lower levels of lipids, glucose, phosphodiesters (PDE), and elevated levels of lactate and total choline when compared to matched, tumor adjacent samples, which indicates that the tumor cells are associated with enhanced glycolysis, and altered lipid and phospholipid metabolism. Moreover, 1H- 31P HET-STOCSY analysis, which identifies signals from the same metabolites or metabolites from the same metabolic pathway, showed differences between tumor adjacent and control hepatic tissue, indicating a progressive change in the hepatic metabolome as the liver advances from normal to HCC. Metabolic profiling of human serum from patients with hepatitis C (no cancer), hepatitis C with HCC, and control samples using 1H NMR and multivariate statistical analysis also provides a powerful method to discover biomarkers that differentiate the two disease groups from healthy controls. A number of metabolites were identifed and quantified, such as pyruvate, lactate, choline, alanine, tyrosine, and other amino acids. The alterations of these metabolite levels were linked to several perturbed metabolic pathways, including glycolysis and amino acid metabolism. These results are valuable for understanding the metabolic basis of tumorigenesis, and could aid in identifying patients with an increased risk of developing HCC.

Degree

M.S.

Advisors

Raftery, Purdue University.

Subject Area

Analytical chemistry|Oncology

COinS