Identification of metabolomic biomarkers for drug-induced acute kidney injury in rats.

Nephrotoxicity is a common side effect observed during both nonclinical and clinical drug development investigations. The present study aimed to identify metabolomic biomarkers that could provide early and sensitive indication of nephrotoxicity in rats. Metabolomic analyses were performed using capillary electrophoresis-time-of-flight mass spectrometry on rat plasma collected at 9 and 24 h after a single dose of 2-bromoethylamine or n-phenylanthranilic acid and at 24 h after 7 days of repeated doses of gentamicin, cyclosporine A or cisplatin. Among a total of 169 metabolites identified, 3-methylhistidine (3-MH), 3-indoxyl sulfate (3-IS) and guanidoacetate (GAA) were selected as candidate biomarkers. The biological significance and reproducibility of the observed changes were monitored over time in acute nephrotoxicity model rats treated with a single dose of cisplatin, with the glomerular filtration rate monitored by determination of creatinine clearance. Increased plasma levels of 3-MH and 3-IS were related to a decline in glomerular filtration due to a renal failure. In contrast, the decrease in plasma GAA, which is synthesized from arginine and glycine in the kidneys, was considered to reflect decreased production due to renal malfunction. Although definitive validation studies are required to confirm their usefulness and reliability, 3-MH, 3-IS and GAA may prove to be valuable plasma biomarkers for monitoring nephrotoxicity in rats.

Monitoring of the human serum albumin carbonylation level through determination of guanidino group content.

Carbonylation of the protein amino, guanidine, and thiol groups with α-oxoaldehydes (which are produced in higher quantities in diabetes, uremia, oxidative stress, aging, and inflammation) is one of the important causes of vascular complications. For monitoring of the human serum albumin (HSA) carbonylation level, a spectrophotometric method based on the formation of colored adduct between guanidine group and thymol-sodium hypobromite reagent in the alkaline medium was investigated. Beer's law is obeyed in the concentration range of Arg and protein guanidine groups from 1 to 40 mM. Precision of the method (relative standard deviation) was in the range of 0.9 to 2%. Accuracy was examined by the standard addition method (recovery ~100%). The method was applied for monitoring of the carbonylation level of HSA with methylglyoxal in vitro and of HSA isolated (using affinity chromatography) from sera of 21 patients with type 2 diabetes and 12 healthy persons. The content of guanidine groups in HSA isolated from diabetics (19.64 ± 1.07 mM/mM albumin) was significantly lower (P < 0.001) in comparison with a control group (21.87 ± 1.02 mM/mM albumin). The method is simple and fast, has good accuracy and precision, and is suitable for clinical practice as well for in vitro protein carbonylation experiments.

The effect of high protein diet on urea and guanidino compound levels in renal insufficient mice.

Nephrectomy in mice provokes a decrease in creatinine clearance (CTN(Cl)) and an increase in urea and specific guanidino compound (GC) concentrations in blood and other tissues. Our purpose was to investigate the influence of high protein diet (HPD) on CTN(Cl), urea and GC levels in NX mice. Mice were nephrectomized or sham-operated and subdivided in groups to study five diet conditions. At the end of each experiment, 10 days and 30 days postsurgery, urine and blood were collected for determination of urea and GCs, including creatinine. HPD resulted in significantly higher CTN(Cl) values in sham-operated mice than those observed in mice under normal protein diet, 10 days as well as 30 days postnephrectomy. HPD induced significant increases in plasma urea, guanidinosuccinic acid, argininic acid and a-keto-delta-guanidinovaleric acid concentration 10 days postsurgery but not 30 days postsurgery. HPD coincided with significantly higher excretion of urea, guanidinosuccinic acid, alpha-keto-delta-guanidinovaleric acid, creatine, argininic acid and gamma-guanidinobutyric acid in sham-operated and nephrectomized mice 10 days postsurgery. Our results show that HPD induces supplementary (to nephrectomy) increases of urea and GCs in the early postsurgery period but not in the later phase.

Biochemical and histopathological changes in nephrectomized mice.

Renal failure is characterized by the retention of nitrogenous metabolites such as urea, creatinine (CTN) and other guanidino compounds (GCs), uric acid, and hippuric acid, which could be related to the clinical syndrome associated with renal insufficiency. A model of renal failure has been developed in male C57BL x Swiss-Webster mice using nephrectomy (NX) and/or arterial ligation. A sham group (group A) and two nephrectomized groups, group B (one kidney removed) and group C (one kidney removed and ligation of the contralateral anterior artery branch), were studied. Ten days postsurgery, morphological and functional indices of renal failure were investigated. Nephrectomized mice manifested features of renal failure like polyuria and wasting. CTN clearance (CTN[Cl]) decreased by +/-26% in group B and +/-33% in group C as compared with the control values. Marked increases in the plasma concentration of guanidinosuccinic acid ([GSA] fourfold) and guanidine ([G] twofold) were observed in the experimental animals. CTN and alpha-keto-delta-guanidinovaleric acid (alpha-keto-delta-GVA) reached levels of, respectively, 1.5-fold and twofold those of controls. Urinary GSA excretion increased and guanidinoacetic acid (GAA) excretion decreased about twofold in group C. GSA increases (2.6-fold) were also observed in the brain in group C, in addition to a significant increase of G (2.5-fold) and gamma-guanidinobutyric acid ([GBA] 1.5-fold). Finally, the extent of NX was found to be 45.2% in group B and 71.4% in group C. Light microscopy revealed an expansion and increase in cellularity of the mesangium of the glomeruli, particularly in group C. A significant correlation (r = .574, P < .0001) was found between CTN(Cl) and the degree of NX as calculated from the remaining functional area. These data suggest that the model can be used as a tool for further pathophysiological and/or behavioral investigations of renal failure.