Short Communication Contrasting Influence of NADPH and a NADPH-Regenerating System on the Metabolism of Carbonyl-Containing Compounds in Hepatic Microsomes

Carbonyl containing xenobiotics may be susceptible to NADPHdependent cytochrome P450 (P450) and carbonyl-reduction reactions. In vitro hepatic microsome assays are routinely supplied NADPH either by direct addition of NADPH or via an NADPHregenerating system (NRS). In contrast to oxidative P450 transformations, which occur on the periphery of a microsome vesicle, intraluminal carbonyl reduction depends on transport of cofactors across the endoplasmic reticulum (ER) membrane into the lumen. Glucose 6-phosphate, a natural cofactor and component of the NRS matrix, is readily transported across the ER membrane and facilitates intraluminal NADPH production, whereas direct addition of NADPH has limited access to the lumen. In this study, we compared the effects of direct addition of NADPH and use of an NRS on the P450-mediated transformation of propiconazole and 11 -hydroxysteroid dehydrogenase type 1 (HSD1) carbonyl reduction of cortisone and the xenobiotic triadimefon in hepatic microsomes. Our results demonstrate that the use of NADPH rather than NRS can underestimate the kinetic rates of intraluminal carbonyl reduction, whereas P450-mediated transformations were unaffected. Therefore, in vitro depletion rates measured for a carbonylcontaining xenobiotic susceptible to both intraluminal carbonyl reduction and P450 processes may not be properly assessed with direct addition of NADPH. In addition, we used in silico predictions as follows: 1) to show that 11 -HSD1 carbonyl reduction was energetically more favorable than oxidative P450 transformation; and 2) to calculate chemical binding score and the distance between the carbonyl group and the hydride to be transferred by NADPH to identify other 11 -HSD1 substrates for which reaction kinetics may be underestimated by direct addition of NADPH. In vitro assays using hepatic microsomes are a powerful tool for both drug discovery applications and human health risk assessment (Obach, 1997). Microsomes are the metabolically active subcellular tissue fraction of the endoplasmic reticulum (ER) and exist primarily as vesicles. Extensive research with hepatic microsomes has focused on the cytochrome P450 (P450) superfamily of enzymes, which are located on the periphery of the microsome vesicle (Cribb et al., 2005). In vivo, the cytosolic pentose pathway primarily supplies P450 enzymes with NADPH for oxidative transformation. In vitro microsomal assays, which are devoid of the cytosolic fraction, are routinely supplied NADPH by either direct addition of NADPH or use of an NADPH-regenerating system (NRS), consisting of -NADP , glucose-6-phosphaste (G6P), and G6P dehydrogenase (G6PDH). In contrast to P450 enzymes, the susceptibility of a chemical to enzymatic processes inside the lumen (intraluminal) depends on the selective transport of cofactors across the ER membrane. Intraluminal carbonyl reduction is an important process for the normal function of various endogenous substrates, and many pharmaceutical agents and pesticides use carbonyl moieties in their mode of action. Although many cytosolic carbonyl-reducing enzymes have been identified, carbonyl reductase in hepatic microsomes is mainly attributed to 11 hydroxysteroid dehydrogenase type 1 (HSD1), a short-chain dehydrogenase-reductase. 11 -HSD1 is an intraluminal NADPH-dependent enzyme that has been well characterized for the interconversion of the glucocorticoid hormone cortisone to cortisol and is considered the primary microsomal reductase for carbonyl-containing therapeutics and pesticides (Oppermann, 2007). The majority of chemicals involved in intraluminal metabolism require protein-facilitated transport to cross the ER lipid bilayer. G6P, a natural cofactor and component of the NRS matrix, is readily transported across the ER membrane by a selective transporter protein (G6PT), whereas NADPH and NADP are relatively impermeable and have limited access to the lumen. G6P entering the lumen is used as a substrate by intraluminal hexose-6-phosphate dehydrogenase (H6PDH) for NADPH production from a separate NADP pool enclosed within the ER. The intraluminal production of NADPH is then directly coupled to 11 -HSD1 activity (Csala et al., 2006; Piccirella et al., 2006). We recently reported on the 11 -HSD1-mediated carbonyl reduction of the broad spectrum 1,2,4-triazole fungicide triadimefon to triadimenol (Kenneke et al., 2008). The current study focuses on the contrasting effects of NADPH and NRS on the kinetic transformation Article, publication date, and citation information can be found at http://dmd.aspetjournals.org. doi:10.1124/dmd.109.027615. □S The online version of this article (available at http://dmd.aspetjournals.org) contains supplemental material. ABBREVIATIONS: ER, endoplasmic reticulum; P450, cytochrome P450; NRS, NADPH-regenerating system; G6P, glucose-6-phosphate; G6PDH, G6P dehydrogenase; HSD1, 11 -hydroxysteroid dehydrogenase type 1; G6PT, glucose-6-phosphate transport protein; H6PDH, hexose-6phosphate dehydrogenase; PDB, Protein Data Bank; MOE, Molecular Operating Environment. 0090-9556/09/3709-1801–1805 DRUG METABOLISM AND DISPOSITION Vol. 37, No. 9 U.S. Government work not protected by U.S. copyright 27615/3506777 DMD 37:1801–1805, 2009 Printed in U.S.A. 1801 http://dmd.aspetjournals.org/content/suppl/2009/06/18/dmd.109.027615.DC1 Supplemental material to this article can be found at: at A PE T Jornals on M ay 4, 2017 dm d.aspurnals.org D ow nladed from of carbonyl-containing xenobiotics in hepatic microsomes. We examined the P450-mediated transformation of propiconazole and the intraluminal carbonyl reduction of cortisone and the xenobiotic triadimefon with direct addition of NADPH and NRS. In addition, in silico predictions based on chemical binding score and carbonyl to hydrideNADP distance were used to identify other 11 -HSD1 substrates that may yield discrepant reaction kinetics depending on the use of either NADPH or NRS. Materials and Methods Reagents. Triadimefon, triadimenol, and propiconazole were obtained from the U.S. Environmental Protection Agency (EPA) National Pesticide Standard Repository (Fort Meade, MD). NADP , NADPH, G6P, G6PDH, alamethicin, magnesium chloride (MgCl2), phosphate buffer (pH 7.4), cortisone, and cortisol were purchased from Sigma-Aldrich (St. Louis, MO). Microsomal Incubation Procedure. Frozen hepatic microsomes from male Sprague-Dawley rats at a concentration of 20 mg of microsomal protein/ml were purchased from In Vitro Technologies (Baltimore, MD) and stored at 80°C until use. All microsomal incubations (0.30 mg of protein) were conducted using the same lot of microsomes and assayed as described previously (Kenneke et al., 2008). In brief, metabolism assays were initiated by the addition of either 250 l of NRS or NADPH to the microsomal suspension. All alamethicin assays were conducted at 50 g/mg microsomal protein and were incubated for 10 min before initiation of the reaction (Piccirella et al., 2006). Abiotic microsomal controls were conducted with addition of the parent compound and omission of either NADPH or G6P from the NRS. Triadimefon Assays. All triadimefon and triadimenol standards were prepared in acetonitrile and stored in amber vials at 4°C. Substrate saturation kinetic profiles of triadimenol formation in the presence of either NADPH (1.8 mM) or NRS were conducted by varying the triadimefon stock solutions (0.3–170 M) at a 1% final acetonitrile concentration. A saturation kinetic plot of triadimenol formation with respect to varying NADPH concentration (0–4 mM) was conducted at 170 M triadimefon. Analysis of triadimenol formation while varying the cofactor composition was conducted in triplicate in the presence of triadimefon (170 M). Varying cofactor additions were also conducted with the 1,2,4-triazole, propiconazole (180 M), as a comparative control for P450-mediated transformation processes (Kenneke et al., 2008). Cortisone Assays. Cortisone and cortisol standards were dissolved in 50% ethanol/water and stored in amber vials at 4°C. The final concentration of ethanol in metabolism assays was less than 1%. A saturation kinetic profile of cortisol formation with respect to varying NADPH concentration (0–4 mM) was conducted using cortisone (137 M). Cortisol formation was analyzed in triplicate while varying the incubation matrix cofactor composition in the presence of cortisone (137 M). High-Performance Liquid Chromatography Analysis and Data Analysis. Analysis of triadimefon, triadimenol, cortisone, cortisol, and propiconazole metabolism samples was performed on an Agilent Series 1100 HPLC quaternary pump system (Agilent Technologies, Santa Clara, CA) as described previously (Kenneke et al., 2008). The Michaelis-Menten kinetic parameters Vmax and KM were determined based on nonlinear regression of product velocity formation at varying substrate concentrations (Sigma Plot; Systat Software, Inc., San Jose, CA). Molecular Docking. Triadimefon was docked into two catalytic systems: 1) human 11 -HSD1 [Protein Data Bank (PDB) accession codes: 2IRW] and 2) CYP3A4 (PDB accession code 2V0M) to quantitatively determine the differences in binding energetics between the two enzyme systems. Additional carbonyl-containing substrates (Table 1) were built in MOE and optimized using the MMFFx (Halgren and Nachbar, 1996) force-field as implemented in MOE. All docking experiments for substrates listed in Table 1 were performed in MOE using MOE-Dock on the 2BEL crystal structure with energies being reported in kJ/mol (see supplemental information).