The lipid metabolome of kidney cancer.
16 Background: The commonest type of kidney cancer is CCRCC. Kidney cancer is known to accumulate lipids and a detailed knowledge of the lipid species present in these tumors could lead to a better understanding of the underlying aberrant metabolic pathways and suggest possible treatment strategies. We attempt to identify the lipidomic profile of CCRCC using a liquid chromatography MS-based approach (LC-MS). Methods: We utilized 6 fresh frozen representative samples of CCRCC and matching non-tumor areas of kidney from nephrectomy samples. Lipids and other non-polar cellular constituents were extracted from both CCRCC and control tissues by methyl-t-butyl ether /methanol. LC-MS based lipid profiling was performed on a Waters Q-ToF Premier MS coupled with Ultra Performance LC. The peak detection and alignment across all chromatograms were performed using the XCMS software (v 1.14.1, Scripps Center for Metabolomics). Statistical comparisons of the intensities of aligned peaks were performed using the XCMS-built-in Welch's t-test. The XCMS data was converted to log2 ratios (normal/tumor) in order to utilize the paired aspect of this data: each patient’s disease tissue was analyzed in conjunction with corresponding normal tissue. Significance was determined by controlling the family wise error rate (FWER) at 0.05 using a Bonferroni adjustment. All computations were performed using the R statistical software and the “limma” package. Results: The outcome of XCMS was converted to a table that contains fold change, p value and mass to charge ratio (m/z) for each peak, its corresponding retention time, and the integrated peak intensities from all samples. Controlling FWER at 0.05 using a Bonferroni scheme, we found eight statistically significant lipids. Preliminary attempts to identify the analytes included use of METLIN (Scripps Center for Metabolomics) and HMDB (Human metabolome database, Genome Alberta and Genome Canada) databases. The identified metabolites included phosphatidylcholines, cholesterol esters and triglycerides, as well as other lipid species. Conclusions: The lipid metabolomic profile varied significantly between CCRCC and control. Further studies are underway to confirm the identities and significance of the lipid species in detail.