Liquid Chromatography–Mass Spectrometry-Based Metabolomics of Nonhuman Primates after 4 Gy Total Body Radiation Exposure: Global Effects and Targeted Panels

• Published in: Journal of proteome research 2019-05, Vol.18 (5), p.2260-2269
• Main Authors: Pannkuk, Evan L, Laiakis, Evagelia C, Gill, Kirandeep, Jain, Shreyans K, Mehta, Khyati Y, Nishita, Denise, Bujold, Kim, Bakke, James, Gahagen, Janet, Authier, Simon, Chang, Polly, Fornace, Albert J
• Format: Article
• Language: English
• ISSN: 1535-3893; 1535-3907
• DOI: 10.1021/acs.jproteome.9b00101

Rapid assessment of radiation signatures in noninvasive biofluids may aid in assigning proper medical treatments for acute radiation syndrome (ARS) and delegating limited resources after a nuclear disaster. Metabolomic platforms allow for rapid screening of biofluid signatures and show promise in differentiating radiation quality and time postexposure. Here, we use global metabolomics to differentiate temporal effects (1–60 d) found in nonhuman primate (NHP) urine and serum small molecule signatures after a 4 Gy total body irradiation. Random Forests analysis differentially classifies biofluid signatures according to days post 4 Gy exposure. Eight compounds involved in protein metabolism, fatty acid β oxidation, DNA base deamination, and general energy metabolism were identified in each urine and serum sample and validated through tandem MS. The greatest perturbations were seen at 1 d in urine and 1–21 d in serum. Furthermore, we developed a targeted liquid chromatography tandem mass spectrometry (LC–MS/MS) with multiple reaction monitoring (MRM) method to quantify a six compound panel (hypoxanthine, carnitine, acetylcarnitine, proline, taurine, and citrulline) identified in a previous training cohort at 7 d after a 4 Gy exposure. The highest sensitivity and specificity for classifying exposure at 7 d after a 4 Gy exposure included carnitine and acetylcarnitine in urine and taurine, carnitine, and hypoxanthine in serum. Receiver operator characteristic (ROC) curve analysis using combined compounds show excellent sensitivity and specificity in urine (area under the curve [AUC] = 0.99) and serum (AUC = 0.95). These results highlight the utility of MS platforms to differentiate time postexposure and acquire reliable quantitative biomarker panels for classifying exposed individuals.