Short Communication In Vitro and in Vivo Determination of Piperacillin Metabolism in Humans

Piperacillin metabolism and biliary excretion are different between humans and preclinical species. In the present study, piperacillin metabolites were characterized in bile and urine of healthy humans and compared with metabolites formed in vitro. Volunteers were administered 2 g of piperacillin IV; blood, urine, and duodenal aspirates (obtained via a custom-made oroenteric catheter) were collected. The metabolism of piperacillin in humans also was investigated in vitro using pooled human liver microsomes and sandwich-cultured human hepatocytes. Piperacillin and metabolites were estimated by high-performance liquid chromatography with tandem mass spectrometry detection. Piperacillin, desethylpiperacillin, and desethylpiperacillin glucuronide were detected in bile, urine, and human liver microsomal incubates. Similar to the in vivo results, desethylpiperacillin was formed and excreted into bile canaliculi of sandwich-cultured human hepatocytes. This is the first report of glucuronidation of desethylpiperacillin in vitro or in vivo. The clinical method employed in this study to determine biliary clearance of drugs also facilitates bile collection as soon as bile is excreted from the gallbladder, thereby minimizing the exposure of labile metabolites to the intestinal environment. This study exemplifies how a combination of in vitro and in vivo tools can aid in the identification of metabolites unique to the human species. The characterization of metabolic pathways that are unique to humans is very important in the drug development process. The formation of human-specific metabolites can be predicted using in vitro systems, including pooled human microsomes, suspended human hepatocytes or recombinant enzymes; nevertheless, these tools do not allow for mechanistic studies concerning the excretory route of these metabolites from the liver (canalicular versus basolateral excretion). Often, biliary excretion of drugs and metabolites can only be characterized using animal models, primarily rodent isolated perfused livers and bile duct cannulated animals. However, significant interspecies differences in the function and regulation of transport proteins and drug metabolizing enzymes have been reported, and these could compromise the ability to extrapolate from animal data to humans (Wang and LeCluyse, 2003; Zhang et al., 2005). Aspiration of bile from the duodenums of healthy volunteers can be used to determine biliary clearance of drugs (Ghibellini et al., 2006a). In addition, this method, combined with the sensitivity of mass spectrometry, may offer a new tool to identify metabolites formed in humans that might not be predicted from preclinical studies. Likewise, sandwich-cultured human hepatocytes (SCHH), a particularly powerful tool with respect to metabolic and transport capabilities, may be a useful in vitro tool to predict the formation and route of excretion of metabolites (LeCluyse et al., 2000; Hoffmaster et al., 2004). Piperacillin is a third-generation, broad-spectrum, penicillin derivative. The biliary excretion of piperacillin in animal models is extensive; 15% of the piperacillin dose was recovered in rat bile after single-pass isolated perfused liver experiments (Calhoun et al., 1987), and 37% of the dose was recovered in recirculating rabbit isolated perfused livers (Brogard et al., 1994). In contrast, the biliary excretion of piperacillin in healthy humans is negligible (Ghibellini et al., 2006b). Very little is known about the metabolism of piperacillin in humans; the liver-specific metabolism of this compound to desethylpiperacillin was reported previously using human liver homogenates (Minami et al., 1991), and this metabolite was identified in human urine (Komuro et al., 1997). No other metabolites have been described to date, and the excretion of piperacillin metabolites into human bile has not been characterized. During a clinical study designed to determine the biliary clearance of piperacillin in healthy volunteers, bile and urine samples were obtained from three subjects and analyzed by HPLC-MS-MS to estimate the contribution of the biliary route to the systemic clearance of the parent compound (Ghibellini et al., 2006b). Although desethylpiperacillin has been recovered previously in human urine and in human liver homogenates, the enzymes involved in the formation and the preferential route of excretion of desethylpiperacillin have not been demonstrated so far. Therefore, one of the goals of this study was to use mass spectrometry to evaluate whether sampling bile immediately upon secretion into the intestine would enable identification of potentially unstable drug metabolites. In addition, piperacillin metabThis work was supported by National Institutes of Health Grant R01-GM41935 and Grant RR00046 from the General Clinical Research Center program of the Division of Research Resources. G.G. was an American Foundation for Pharmaceutical Education Predoctoral Fellow. K.B. is a co-founder and Chair of the Scientific Advisory Board for Qualyst, Inc., which has exclusively licensed the sandwich-cultured hepatocyte technology for quantification of biliary excretion (B-CLEAR). None of the other authors has any conflict of interest. Article, publication date, and citation information can be found at http://dmd.aspetjournals.org. doi:10.1124/dmd.106.012278. ABBREVIATIONS: SCHH, sandwich-cultured human hepatocytes; MS-MS tandem mass spectrometry; HPLC, high-performance liquid chromatography; DEX, dexamethasone; HBSS, Hanks’ balanced salt solution; HLM, human liver microsome(s); UDPGA, uridine diphosphate glucuronic acid; DMEM, Dulbecco’s modified Eagle’s medium; BEI, biliary excretion index; IS, internal standard. 0090-9556/07/3503-345–349$20.00 DRUG METABOLISM AND DISPOSITION Vol. 35, No. 3 Copyright © 2007 by The American Society for Pharmacology and Experimental Therapeutics 12278/3185792 DMD 35:345–349, 2007 Printed in U.S.A. 345 at A PE T Jornals on A uust 7, 2017 dm d.aspurnals.org D ow nladed from olism also was evaluated using hepatic microsomes and sandwichcultured human hepatocytes to allow for the comparison of metabolites generated in vitro to those formed in vivo. Materials and Methods Chemicals and Reagents for In Vitro Studies. Insulin/transferrin/selenium (ITS ) was purchased from BD Biosciences (San Jose, CA). Dexamethasone (DEX), piperacillin, -NADPH, Hanks’ balanced salt solution (HBSS), and bovine serum albumin were purchased from Sigma Chemical Co. (St. Louis, MO). All other chemicals and reagents were of analytical grade and were available from commercial sources. Phase I Metabolism of Piperacillin. Mixed gender human liver microsomes (HLM; 2.5 mg/ml; XenoTech, LLC, Lenexa, KS) were incubated in triplicate in 100 mM potassium phosphate buffer, pH 7.4, 3.3 mM MgCl2, and a concentration of piperacillin within the range of the observed clinical Cmax (200 ), in a total incubation volume of 0.5 ml. After preincubation (5 min, 37°C), the reactions were initiated by the addition of 1 mM -NADPH; as a control, incubations without -NADPH were also performed. At 0, 5, 15, 30, 60, and 90 min, aliquots (50 l) were removed and the reactions stopped with an equal volume of ice-cold acetonitrile containing cimetidine as the internal standard (IS). Proteins were precipitated by centrifugation (3500g for 5 min), and the supernatant was analyzed immediately by HPLC-MS-MS. Liver Cytosolic Incubations. Individual incubations (total volume 0.5 ml) were composed of 0.25 mg/ml, mixed gender, pooled human liver cytosol (XenoTech LLC) in 100 mM phosphate buffer, pH 7.4, 3.3 mM MgCl2, 1 mM NADH, and piperacillin at three different concentrations (50, 100, and 200 ). After 5-min preincubation at 37°C, the reaction was initiated by the addition of NADH. Aliquots (50 l) were removed at 0, 2, 5, 10, 15, 30, and 60 min and quenched with 25 l of ice-cold acetonitrile. Proteins were precipitated, and 70 l of the supernatant was analyzed immediately using HPLC-MS-MS as described below. Glucuronidation of Piperacillin. Incubations were performed in duplicate using HLM (2.5 mg/ml; XenoTech LLC), 100 mM potassium phosphate buffer, pH 7.4, 3.3 mM MgCl2, 50 g of alamethicin/mg of protein, 5 mM saccharolactone, and 500 piperacillin in a total incubation volume of 0.5 ml. After preincubation (5 min, 37°C), the reactions were initiated by the addition of -NADPH (2 mM) and UDPGA (5 mM). Reactions were terminated at 0, 5, 15, 30, 60, and 90 min by quenching 50 l of the incubation mixture with an equal volume of ice-cold 80:20 (v/v) acetonitrile:1% acetic acid containing cimetidine (IS). After protein precipitation, the supernatant was analyzed by HPLC-MS-MS. Isolation and Culture of Human Hepatocytes. Human liver tissue was obtained by qualified medical staff from the University of North Carolina at Chapel Hill, School of Medicine, as waste from surgical resection. Donor consent and Institutional Review Board approval were obtained for all studies involving human liver tissue. Hepatocytes were isolated by modifications of the two-step collagenase digestion method (Hamilton et al., 2001). Hepatocytes were cultured for 6 days according to the methods described by Hoffmaster et al. (2004) with modifications, and the medium was replaced every 24 h. In brief, 1.5 10 hepatocytes/well were seeded on six-well Biocoat plates (BD Biosciences Discovery Labware, Bedford, MA) in 1.5 ml of Dulbecco’s modified Eagle’s medium (DMEM, without phenol red) supplemented with 5% fetal bovine serum, nonessential amino acids, L-glutamine, penicillin/streptomycin, and 1 M DEX. After cell attachment (2–6 h at 37°C in a humidified incubator with 95% air/5% CO2), medium was replaced with DMEM containing 0.1% (v/v) ITS , penicillin/streptomycin, nonessential amino acids, L-glutamine, and 0.1 M DEX; 6 to 12 h later, cells were overlaid with ice-cold medium containing 0.25 mg/ml Matrigel (BD Bioscience). Human Hepatocyte Studies. Four different SCHH preparations were used for these experiments. The biliary excretion index (BEI) was calculated according to the method described by Hoffmaster et al. (2004) using B-CLEAR technology (Qualyst, Inc., Research Triangle Park, NC). The BEI indicates the fraction of the total accumulation of drug that resides in the bile compartment. Accumulation of [H]taurocholate (1 M; 100 nCi) in cellular and cells plus bile compartments was determined over 10 min; piperacillin (300 M) accumulation was determined over 30 min. Only preparations with a [H]taurocholate BEI 50% were incubated with piperacillin. Because 30 min was not long enough to allow for the formation of phase II metabolites in SCHH of liver 1, additional experiments were performed with hepatocytes from livers 2, 3, and 4. Hepatocytes were incubated for 2 h in HBSS (livers 2, 3, and 4), and for 24 and 48 h (liver 4 only) in DMEM, both containing 500 M piperacillin, in an attempt to increase intracellular concentrations of piperacillin. To measure the accumulation of piperacillin in the presence or absence of intact canalicular spaces, hepatocytes were lysed with 70/30 (v/v) methanol/water. Samples were analyzed directly, or evaporated and reconstituted in 1/10 of the original volume containing cimetidine (IS) before analysis by HPLC-MS-MS. Nonspecific binding was accounted for by including a blank plate (Biocoat plus Matrigel overlay). Protein content in cell lysate was quantified with the BCA method (Smith et al., 1985) and was used to normalize accumulation. Because of the incompatibility of methanol with the protein assay, the average protein content for standard HBSS or Ca -free HBSS incubations with taurocholate was used to normalize piperacillin and metabolite content of the same liver preparation. Clinical Study. The design and conduct of the clinical study has been described previously (Ghibellini et al., 2006b). In brief, three male volunteers completed the study. The Clinical Research Advisory Committee and the Committee on the Protection of the Rights of Human Subjects (IRB) at the University of North Carolina at Chapel Hill School of Medicine approved all procedures. Subjects provided written informed consent before participation in the study. After an overnight fast, a custom-made oroenteric tube, described in detail previously (Ghibellini et al., 2004), was passed through the mouth and positioned in the duodenum. The distal end of the tube was fitted with a polyethylene balloon, which was inflated to occlude the intestine during bile collection. Subjects were administered 2 g of piperacillin (generous gift of American Pharmaceutical Partners, Los Angeles, CA) as a 15-min IV infusion, and blood, bile, and urine samples were collected at predetermined intervals over 6 h (bile) and 10 h (urine and plasma). Analysis of Piperacillin and Metabolites. Bile, urine, and samples from the in vitro experiments were analyzed for piperacillin and metabolites using HPLC-MS-MS. Bile and urine samples were diluted 1:7 with methanol containing cimetidine (0.5 g/ml) as the internal standard (IS). The Agilent 1100 HPLC system (Agilent Technologies, Palo Alto, CA) was coupled to a mass spectrometer with a TurboIonSpray source (API 4000; Applied Biosystems. Foster City, CA) as described previously (Ghibellini et al., 2006b). User controlled parameters were optimized via direct infusion for the detection of FIG. 1. Piperacillin, desethylpiperacillin, and desethylpiperacillin glucuronide in clinical samples. A, relative amounts of piperacillin, desethylpiperacillin, and desethylpiperacillin glucuronide in urine (left ordinate axis) and duodenal aspirates (right ordinate axis). Relative amounts are expressed as the mean S.D. of the analyte/IS peak area ratio multiplied by the collected volume. B, structure of piperacillin. 346 GHIBELLINI ET AL. at A PE T Jornals on A uust 7, 2017 dm d.aspurnals.org D ow nladed from