gms | German Medical Science

Background: Kidney cancer is the twelfth most common cancer worldwide with about 340.000 new cases and almost 150.000 cases of death per year [1], [2]. Among the different types of kidney cancer, clear cell renal cell carcinoma (ccRCC) represents the most prevalent form (> 70%) [3]. The symptoms of ccRCC are usually mild and as a result the disease is often diagnosed in an advanced state. In order to facilitate diagnosis and prognosis of ccRCC already at early stages, a continuous search for metabolic biomarkers is required. Here, we present an analytical approach for the comprehensive metabolomic profiling of small molecules and lipids in kidney tissue. The method was established and validated based on porcine tissue and, as proof of concept, applied to a small cohort of human ccRCC tissue samples.

Methods: Porcine kidney was obtained as fresh food product and was used for method development and validation. Five fresh frozen ccRCC samples and corresponding normal tissue were used for cancer-specific metabolomic profiling and were derived from patients who underwent partial or radical nephrectomy. Metabolites and lipids were recovered from tissue samples (10-45 mg) by a two-step extraction protocol. Tissue homogenization and extraction of polar metabolites were performed in methanol/water (aqeous extract) by a bead-beating approach. Lipids were recovered by consecutive extraction of the pellet with methanol/methyl tert-butyl ether (organic extract). Metabolites in aqueous extracts were separated by hydrophilic liquid interaction chromatography whereas compounds in organic extracts were separated by reversed phase chromatography. Pooled tissue extracts were used as quality control (QC) samples and have been analyzed throughout the analytical batch. Mass spectrometric analysis was performed in negative and positive ionization mode by quadrupole time of flight mass spectrometry (QTOF-MS). For data processing, analysis and evaluation, Mass Hunter Qualitative Analysis, Profinder and Mass Profiler Professional were used.

Preliminary Results: Reproducibility of tissue extraction and metabolite analysis was assessed by the analysis of multiple individually prepared porcine kidney samples. More than 1000 metabolic features including amino acids, nucleotides, small organic acids, phospholipids, sphingolipids, glycerolipids and fatty acids could be reproducible (CV<30%) detected in each extract (organic and aqueous) and in each ionization mode (positive and negative ion). Principle component analysis (PCA) revealed QC sample clustering substantiating analytical reproducibility. As proof of concept, the validated protocol was applied for metabolomic profiling of kidney tissue derived from ccRCC patients. Based on unsupervised multivariate statistics, a differentiation between cancerous and normal tissue for the small metabolites profile as well as for the lipid profile could be observed. On the basis of accurate mass, retention time and fragment spectra information several key metabolic features, such as triglycerides, which contribute to the differentiation of malignant and normal kidney tissue, could be identified.

Conclusion: Metabolomic profiling of kidney tissue extracts enables differentiation between ccRCC and normal kidney tissue samples based on the lipid and small molecule metabolomic profiles. Further studies on larger and independent sample groups are necessary to confirm and validate our preliminary findings. In summary, the presented approach provides a first basis for comprehensive metabolomics studies in human kidney and thus offers great potential for the discovery of novel, diagnostic and prognostic biomarkers that will facilitate personalized ccRCC therapy.