A metabolic study on colon cancer using (1)h nuclear magnetic resonance spectroscopy

Abstract

Background. Colorectal carcinoma is the third cause of cancer deaths in the world. For diagnosis, invasive methods like colonoscopy and sigmoidoscopy are used, and noninvasive screening tests are not very accurate. We decided to study the potential of (1)HNMR spectroscopy with metabolomics and chemometrics as a preliminary noninvasive test. We obtained a distinguishing pattern of metabolites and metabolic pathways between colon cancer patient and normal. Methods. Sera were obtained from confirmed colon cancer patients and the same number of healthy controls. Samples were sent for (1)HNMR spectroscopy and analysis was carried out Chenomex and MATLAB software. Metabolites were identified using Human Metabolic Data Base (HDMB) and the main metabolic cycles were identified using Metaboanalyst software. Results. 15 metabolites were identified such as pyridoxine, orotidine, and taurocholic acid. Main metabolic cycles involved were the bile acid biosynthesis, vitamin B6 metabolism, methane metabolism, and glutathione metabolism. Discussion. The main detected metabolic cycles were also reported earlier in different cancers. Our observations corroborated earlier studies that suggest the importance of lowering serum LCA/DCA and increasing vitamin B6 intake to help prevent colon cancer. This work can be looked upon as a preliminary step in using (1)HNMR analysis as a screening test before invasive procedures.

Figure 1

Score plot of PLS after OSC shows very good separation of samples. Odd numbers indicate normal and even numbers patient samples.

Figure 2

Biplot of PLS after OSC showing differentiating metabolites. Odd numbers indicate normal and even number of patient samples.

Figure 3

Loads plot of metabolites below the axis show descending levels and above the axis show ascending levels of metabolites. Numbers indicate metabolites.

Figure 4

Summary plot for over representation analysis (ORA).