Drug-induced impairment of mitochondrial fatty acid oxidation and steatosis: assessment of causal relationship with 45 pharmaceuticals.

Drug-induced liver injury (DILI) represents a major issue for pharmaceutical companies, being a potential cause of black-box warnings on marketed pharmaceuticals, or drug withdrawal from the market. Lipid accumulation in the liver also referred to as steatosis, may be secondary to impaired mitochondrial fatty acid oxidation (mtFAO). However, an overall causal relationship between drug-induced mtFAO inhibition and the occurrence of steatosis in patients has not yet been established with a high number of pharmaceuticals. Hence, 32 steatogenic and 13 non-steatogenic drugs were tested for their ability to inhibit mtFAO in isolated mouse liver mitochondria. To this end, mitochondrial respiration was measured with palmitoyl-L-carnitine, palmitoyl-CoA + L-carnitine, or octanoyl-L-carnitine. This mtFAO tri-parametric assay was able to predict the occurrence of steatosis in patients with a sensitivity and positive predictive value above 88%. To get further information regarding the mechanism of drug-induced mtFAO impairment, mitochondrial respiration was also measured with malate/glutamate or succinate. Drugs such as diclofenac, methotrexate and troglitazone could inhibit mtFAO secondary to an impairment of the mitochondrial respiratory chain, while dexamethasone, olanzapine and zidovudine appeared to impair mtFAO directly. Mitochondrial swelling, transmembrane potential and production of reactive oxygen species were also assessed for all compounds. Only the steatogenic drugs amiodarone, ketoconazole, lovastatin and toremifene altered all these 3 mitochondrial parameters. In conclusion, our tri-parametric mtFAO assay could be useful in predicting the occurrence of steatosis in patients. The combination of this assay with other mitochondrial parameters could also help to better understand the mechanism of drug-induced mtFAO inhibition.

Assessment of the inhibitory potential of anabolic steroids towards human AKR1D1 by computational methods and in vitro evaluation.

Exposure to xenobiotics can adversely affect biochemical reactions, including hepatic bile acid synthesis. Bile acids are essential for dissolving lipophilic compounds in the hydrophilic environment of the gastrointestinal tract. The critical micellar concentration of bile acids depends on the Δ4-reduction stereochemistry, with the 3-oxo-5ß-steroid-Δ4-dehydrogenase (AKR1D1) introducing the cis ring A/B conformation. Loss-of-function mutations in AKR1D1 cause hepatic cholestasis, which, if left untreated can progress into steatosis and liver cirrhosis. Furthermore, AKR1D1 is involved in clearing steroids with an A-ring Δ4-double bond. Here, we tested whether anabolic-androgenic steroids (AAS), often taken off-label at high doses, might inhibit AKR1D1, thereby potentially causing hepatotoxicity. A computational molecular model was established and used for virtual screening of the DrugBank database consisting of 2740 molecules, yielding mainly steroidal hits. Fourteen AAS were selected for in vitro evaluation, as such compounds can reach high hepatic concentrations in an abuse situation. Nandrolone, clostebol, methasterone, drostanolone, and methenolone inhibited to various extent the AKR1D1-mediated reduction of testosterone. Molecular modeling suggests that 9 out of 14 investigated AAS are competitive inhibitors. Moreover quantum mechanical calculations show that nadrolone and clostebol are substrates of AKR1D1 with different activation energy barriers for the hydrogen transfer from cofactor to the C5 position affecting their turnover. In this multidisciplinary approach, we established a molecular model of AKR1D1, identified several AAS as inhibitors, and described their binding mode. This approach may be applied to study other classes of inhibitors including non-steroidal compounds.