Effect of dexamethasone on the metabonomics profile associated with phosphodiesterase inhibitor-induced vascular lesions in rats.

Metabonomics is the evaluation of the multiparametric metabolic response of biological systems to pathophysiological stimuli. High-resolution nuclear magnetic resonance (NMR) spectroscopy of biofluids coupled with pattern recognition-based chemometric analysis is an emerging approach to the study of metabonomics and may be used for the prediction of toxicity in vivo and for identification of surrogate markers of toxicity. Previously, we established that metabonomic analysis of urine samples has significant potential for identification of phosphodiesterase type 4 (PDE-4) inhibitor-induced vascular lesions in rats. It was not clear, however, whether the observed changes in metabonomics profile were related mechanistically to the pathogenesis of these vascular lesions or whether these changes were reflective primarily of the ensuing inflammatory response. In the present study, dexamethasone was used to suppress inflammation associated with vascular lesions induced in rats by the PDE-4 inhibitor CI-1018 and urine samples were evaluated for resultant changes in metabonomic profile. Female Wistar rats were given CI-1018 by gavage at 750 mg/kg with or without concurrent intraperitoneal administration of dexamethasone at 1 mg/kg for 4 days. Dexamethasone induced a characteristic lymphoid depletion and lymphocytolysis but no evidence of vascular lesions. Rats dosed with CI-1018 had mild vascular changes in liver and/or marked vascular lesions in mesentery characterized by medial necrosis, hemorrhage, and/or edema accompanied by perivascular mixed inflammatory cell infiltrates. Inflammatory infiltrates associated with these lesions were eliminated in rats given dexamethasone, yet minimal medial smooth muscle necrosis and degeneration still occurred, suggestive of etiologic changes rather than effects secondary to the inflammatory response. Principle component analysis of urine NMR spectra produced a clear pattern separation within 48 to 72 h between CI-1018-treated rats with vascular lesions and vehicle controls or rats given dexamethasone alone. There was no pattern separation, however, between rats given CI-1018 alone and rats given CI-1018 and dexamethasone concurrently, suggesting that CI-1018-induced urine spectral changes were associated with the vascular lesions, yet were independent of the inflammatory response. These findings provide new insight into the mechanism(s) of PDE inhibitor-induced vasculitis and support the potential use of metabonomics for developing reliable noninvasive methods for detecting vascular changes in rats.