Nuclear Magnetic Resonance Metabolomic Profiling of Mouse Kidney, Urine and Serum Following Renal Ischemia/Reperfusion Injury

PLoS One. 2016 Sep 22;11(9):e0163021. doi: 10.1371/journal.pone.0163021. eCollection 2016.

Abstract

Background: Ischemia/reperfusion (I/R) is the most common cause of acute kidney injury (AKI). Its pathophysiology remains unclear. Metabolomics is dedicated to identify metabolites involved in (patho)physiological changes of integrated living systems. Here, we performed 1H-Nuclear Magnetic Resonance metabolomics using urine, serum and kidney samples from a mouse model of renal I/R.

Methods: Renal 30-min ischemia was induced in 12-week-old C57BL/6J male mice by bilaterally clamping vascular pedicles, and was followed by 6, 24 or 48-hour reperfusion (n = 12/group). Sham-operated mice were used as controls. Statistical discriminant analyses, i.e. principal component analysis and orthogonal projections to latent structures (OPLS-DA), were performed on urine, serum and kidney lysates at each time-point. Multivariate receiver operating characteristic (ROC) curves were drawn, and sensitivity and specificity were calculated from ROC confusion matrix (with averaged class probabilities across 100 cross-validations).

Results: Urine OPLS-DA analysis showed a net separation between I/R and sham groups, with significant variations in levels of taurine, di- and tri-methylamine, creatine and lactate. Such changes were observed as early as 6 hours post reperfusion. Major metabolome modifications occurred at 24h post reperfusion. At this time-point, correlation coefficients between urine spectra and conventional AKI biomarkers, i.e. serum creatinine and urea levels, reached 0.94 and 0.95, respectively. The area under ROC curve at 6h, 24h and 48h post surgery were 0.73, 0.98 and 0.97, respectively. Similar discriminations were found in kidney samples, with changes in levels of lactate, fatty acids, choline and taurine. By contrast, serum OPLS-DA analysis could not discriminate sham-operated from I/R-exposed animals.

Conclusions: Our study demonstrates that renal I/R in mouse causes early and sustained metabolomic changes in urine and kidney composition. The most implicated pathways at 6h and 24h post reperfusion include gluconeogenesis, taurine and hypotaurine metabolism, whereas protein biosynthesis, glycolysis, and galactose and arginine metabolism are key at 48h post reperfusion.

Grants and funding

PT is Senior Research Associate of the Fonds de la Recherche Scientifique (FNRS) and FJ is a MD Postdoctoral Fellow of the Fonds de la Recherche Scientifique (FNRS), and received support from FNRS (Research Credit #3309), University of Liège (Fonds Spéciaux à la Recherche; Fonds Leon Frédéricq) and from the Belgian Royal Academy of Medicine (Fondation O. Dupont).