Characterization of oxycodone in vitro metabolism by human cytochromes P450 and UDP-glucuronosyltransferases.

The hepatic metabolism of oxycodone by cytochromes P450 (CYP) and the UDP-glucuronosyltransferases (UGT), the main metabolic enzymes of phase I and phase II, respectively, was assessed in vitro. The N-demethylation by CYP3A4/5 and the O-demethylation by CYP2D6 in human liver microsomes (HLM) followed Michaelis-Menten kinetics, with intrinsic clearances of 1.46µL/min/mg and 0.35µL/min/mg, respectively. Although noroxycodone and oxymorphone mainly contribute to the elimination of oxycodone, the simulated total in vivo clearance using in vitro phase I metabolism was underestimated. For the first time, metabolism of oxycodone by UGT was deeply investigated using HLM, recombinant enzymes and selective inhibitors. Oxycodone-glucuronide was mainly produced by UGT2B7 (Km=762±153µM, Vmax=344±20 peak area/min/mg) and to a lesser extent by UGT2B4 (Km=2454±497µM, Vmax=201±19 peak area/min/mg). Finally, the kinetics of the drug-drug interactions were assessed using two CYP and UGT cocktail approaches. Incubations of HLM with phase I and phase II drug probes showed that oxycodone mainly decreased the in vitro activities of CYP2D6, CYP3A4/5, UGT1A3, UGT1A6 and UGT2B subfamily with an important impact on UGT2B7.

Separation of substrates and closely related glucuronide metabolites using various chromatographic modes.

The aim of this study was to assess the retention and selectivity of a cocktail of 10 substrates of uridine diphosphate glucuronosyltransferase enzymes (UGTs) and their respective glucuronides using four chromatographic approaches. For this purpose, seven different stationary phases were employed in reversed phase liquid chromatography (RPLC), two in hydrophilic interaction liquid chromatography (HILIC), one in aqueous normal phase chromatography (ANPC) and four in subcritical fluid chromatography (SFC). Highly orthogonal separations were achieved with these chromatographic modes. Hydrophobic interactions mainly governed the retention of the substrates and their polar glucuronides in RPLC despite the use of different chemical stationary phase bonding, involving additional possible interactions. In ANPC, atypical separations and poor peak shapes were observed with the selected compounds. In HILIC and SFC conditions, the metabolites were more retained than the substrates because of the polarity increase related to the glucuronic acid moiety. For the latter, a very high proportion of organic solvent (up to 80%) was required to elute the glucuronides that often displayed poor peak shapes. Finally, the selectivity of nine chromatographic systems was compared for the separation of isomeric and diastereoisomeric compounds. The stationary phases used in RPLC mode were more selective towards the two positional isomers of morphine glucuronides since they possess distinct lipophilicity. HILIC and SFC columns were found to be promising for the separation of a critical diastereoisomers pair, namely epitestosterone-glucuronide and testosterone-glucuronide.

Inhibition screening method of microsomal UGTs using the cocktail approach.

A rapid method for the simultaneous determination of the in vitro activity of the 10 major human liver UDP-glucuronosyltransferase (UGT) enzymes was developed based on the cocktail approach. Specific substrates were first selected for each UGT: etoposide for UGT1A1, chenodeoxycholic acid for UGT1A3, trifluoperazine for UGT1A4, serotonin for UGT 1A6, isoferulic acid for UGT1A9, codeine for UGT2B4, azidothymidine for UGT2B7, levomedetomidine for UGT2B10, 4-hydroxy-3-methoxymethamphetamine for UGT2B15 and testosterone for UGT2B17. Optimal incubation conditions, including time-based experiments on cocktail metabolism in pooled HLMs that had been performed, were then investigated. A 45-min incubation period was found to be a favorable compromise for all the substrates in the cocktail. Ultra-high pressure liquid chromatography coupled to an electrospray ionization time-of-flight mass spectrometer was used to separate the 10 substrates and their UGT-specific glucuronides in less than 6 min. The ability of the cocktail to highlight the UGT inhibitory potential of xenobiotics was initially proven by using well-known UGT inhibitors (selective and nonselective) and then by relating some of the screening results obtained by using the cocktail approach with morphine glucuronidation (substrate highly glucuronidated by UGT 2B7). All the results were in agreement with both the literature and with the expected effect on morphine glucuronidation.

cIEF for rapid pKa determination of small molecules: a proof of concept.

A capillary isoelectric focusing (cIEF) method was developed for the determination of the ionization constants (pKa) of small molecules. Two approaches used to decrease the electroosmotic flow (EOF) were compared: (i) a hydroxypropylcellulose (HPC) coated capillary in aqueous medium and (ii) the addition of glycerol to act as a viscosifying agent. The cIEF method with the glycerol medium was selected, and the ionization constants of 22 basic and 21 acidic compounds, including 15 pharmaceutical drugs, were determined, resulting in pKa values from 3.5 to 7.4 and 6.4 to 9.3, respectively. cIEF offers an interesting alternative to other techniques for pKa determination with low sample consumption, high throughput and low cost.

High-throughput log P determination by MEEKC coupled with UV and MS detections.

A high-throughput screening method using MEEKC was developed for the determination of 1-octanol-water partition coefficients (log P(oct)). Two approaches were carried out to decrease determination times to about 20 min per compound: (i) a dynamically coated capillary was used to increase the EOF at low pH, allowing the measurement of log P(oct) of acidic compounds and (ii) a short-end injection was performed to reduce the capillary effective length. The analytical conditions were optimized to determine the lipophilicity of neutral, basic, and acidic compounds with log P(oct) ranging from 0 to 5. The developed method was first applied to a well-balanced set of 35 reference compounds, and second to a set of 21 acidic and 29 basic pharmaceutical compounds. Finally, determinations were achieved with MS detection, allowing a 20-fold throughput increase thanks to sample pooling. An atmospheric pressure photoionization source was selected to advantageously replace ESI as it was less affected by the non-volatile BGE additives used in MEEKC.