Supplemental material to this article can be found at: http://dmd.aspetjournals.org/content/suppl/2017/08/21/dmd.117.076463.DC2 http://dmd.aspetjournals.org/content/suppl/2017/06/12/dmd.117.076463.DC1 1521-009X/45/9/1044–1048$25.00 https://doi.org/10.1124/dmd.117.076463 DRUG METABOLISM AND DISPOSITION Drug Metab Dispos 45:1044–1048, September 2017 Copyright ª 2017 by The American Society for Pharmacology and Experimental Therapeutics Age-dependent Protein Abundance of Cytosolic Alcohol and Aldehyde Dehydrogenases in Human Liver s Deepak Kumar Bhatt, Andrea Gaedigk, Robin E. Pearce, J. Steven Leeder, and Bhagwat Prasad Department of Pharmaceutics, University of Washington, Seattle, Washington (D.K.B., B.P.); Department of Clinical Pharmacology, Toxicology & Therapeutic Innovation, Children’s Mercy-Kansas City, Missouri and School of Medicine, University of Missouri-Kansas City, Kansas City, Missouri (A.G., R.E.P., J.S.L.) Received April 21, 2017; accepted June 5, 2017 ABSTRACT Hepatic cytosolic alcohol and aldehyde dehydrogenases (ADHs and the adult levels, respectively. For all proteins, the abundance steeply ALDHs) catalyze the biotransformation of xenobiotics (e.g., cyclo- increased during the first year of life, which mostly reached adult Downloaded from phosphamide and ethanol) and vitamin A. Because age-dependent levels during early childhood (age between 1 and 6 years). Only for hepatic abundance of these proteins is unknown, we quantified ADH1A protein abundance in adults (age > 18 year) was ∼40% lower protein expression of ADHs and ALDH1A1 in a large cohort of pediatric relative to the early childhood group. Abundances of ADHs and and adult human livers by liquid chromatography coupled with tandem ALDH1A1 were not associated with sex in samples with age > 1 year mass spectrometry proteomics. Purified proteins were used as cal- compared with males. Known single nucleotide polymorphisms had ibrators. Two to three surrogate peptides per protein were quantified no effect on the protein levels of these proteins. Quantification of dmd.aspetjournals.org in trypsin digests of liver cytosolic samples and calibrator proteins ADHs and ALDH1A1 protein levels could be useful in predicting under optimal conditions of reproducibility. Neonatal levels of ADH1A, disposition and response of substrates of these enzymes in younger ADH1B, ADH1C, and ALDH1A1 were 3-, 8-, 146-, and 3-fold lower than children. Introduction carboxyphosphamide, respectively. The ALDH-catalyzed detoxifica- at ASPET Journals on September 27, 2021 Age-dependent maturation of the expression or activity of drug metab- tion reaction competes with the activation reaction that converts olizing enzymes (DMEs) in humans is commonly observed (Hines, 2008). aldophosphamide to cytotoxic phosphoramide mustard, which is respon- Although substantial data exist on the ontogeny of major microsomal sible for the anticancer activity (de Jonge et al., 2005). Therefore, these DMEs, age-dependent regulation of cytosolic enzymes is not well studied. enzymes play a major role in determining the efficacy as well as off- Alcohol dehydrogenase 1 isoforms (ADH1A, ADH1B, and ADH1C) and target toxicity of CP. aldehyde dehydrogenase 1A1 (ALDH1A1) are high-abundant cytosol- Immature levels of DMEs and transporters are often associated with ic proteins in human liver (Edenberg, 2000; Sladek, 2003) and belong to reduced clearance and an increased risk of toxicity of xenobiotics in a family of NAD(P)+-dependent enzymes and primarily involved in the children (Hines, 2008). Although midterm fetal levels of hepatic ADH and biotransformation of primary alcohols to aldehydes and aldehydes to weak ALDH are significantly lower relative to adults (Smith et al., 1971), carboxylic acids, respectively (Sladek, 2003). For example, these enzymes ontogenic trajectories of these proteins across the entire age spectrum from play major roles in the metabolism of vitamin A (Kam et al., 2012; Arnold newborns to adults have not been established. Such data are important to et al., 2015), alcohol (Zakhari, 2006), and drugs such as cyclophosphamide predict the effect of age on in vivo clearance of ADHs and ALDH1A1 (CP) (de Jonge et al., 2005). Specifically, ALDH1A1 is responsible for substrates in children before clinical use. Therefore, we quantified ADH1A, converting retinaldehyde to retinoic acid, whereas ADH enzymes convert ADH1B, ADH1C, and ALDH1A1 in the hepatic cytosolic (HLC) fractions ethanol to acetaldehyde. CP is first converted by multiple cytochrome P450 isolated from a bank of tissues using our well-established proteomics enzymes to 4-hydroxy-CP, which exists in equilibrium with its open ring methodology (Prasad et al., 2016). We then studied the effect of age, sex, tautomer, aldophosphamide. Both 4-hydroxy CP and aldophosphamide are race, and genetic polymorphisms on protein expression of these enzymes. irreversibly deactivated by ADHs and ALDH1A1 to 4-keto CP and Although there are multiple ADHs (Estonius et al., 1996) and ALDHs (Stewart et al., 1996) expressed in various human tissues, this study is lim- ited to four major hepatic dehydrogenases (ADH1A, ADH1B, ADH1C, This work is primarily funded by grant from National Institutes of Health (NIH) and ALDH1A1). Integration of the protein abundance data into physio- Eunice Kennedy Shriver National Institute of Child Health and Human Develop- logically based pharmacokinetics modeling software would be useful to ment (Grant R01.HD081299-02). The National Institute of Child Health and Hu- predict accurate drug disposition in a pediatric population. man Development Brain and Tissue Bank for Developmental Disorders at the University of Maryland is funded by the NIH Contract HHSN275200900011C, Materials and Methods reference number, N01-HD-9-0011, and the Liver Tissue Cell Distribution System is funded by NIH Contract N01-DK-7-0004/HHSN267200700004C. Materials. Purified ADH1A, ADH1B, ADH1C, and ALDH1A1 protein https://doi.org/10.1124/dmd.117.076463. standards were procured from Abnova (Walnut, CA). Synthetic heavy labeled s This article has supplemental material available at dmd.aspetjournals.org. peptides (Supplemental Table 1S) were obtained from Thermo Fisher Scientific ABBREVIATIONS: ADH, alcohol dehydrogenases; ALDH, aldehyde dehydrogenase; CP, cyclophosphamide; DME, drug-metabolizing enzyme; HLC, human liver cytosol; LC-MS/MS, liquid chromatography coupled with tandem mass spectrometry. 1044 Hepatic Age-dependent ADH and ALDH Abundance 1045 (Rockford, IL). Ammonium bicarbonate (98% purity), bovine serum albumin multiple comparison test. Mann-Whitney test was used to analyze the effect of sex (BSA) and Pierce Trypsin protease (MS-grade) were purchased from Thermo on protein abundance. A nonlinear, allosteric sigmoidal model using eq. 1 was Fisher Scientific (Rockford, IL). Chloroform, ethyl ether, Optima MS-grade fitted to the continuous ontogeny protein abundance data until age 18 (GraphPad acetonitrile, methanol, and formic acid were purchased from Fischer Scientific Prism, San Diego, CA), as described previously (Boberg et al., 2017). (Fair Lawn, NJ). Adult 2 F Human Liver Cytosol Samples. Human cytosol samples of 129 pediatric liver F ¼ max birth  Ageh þ F ð1Þ h þ h birth donors were provided by Children’s Mercy-Kansas City (Kansas City, MO). Tissues Age50 Age were obtained from the National Institute of Child Health and Human Development Brain and Tissue Bank for Developmental Disorders at the University of Maryland Adultmax is the maximum average relative protein abundance; Age is the age in the Liver Tissue Cell Distribution System at the University of Minnesota and the years of the subject at the time of sample collection; Age50 is the age in years at University of Pittsburgh as well as Vitron (Tucson, AZ) and XenoTech LLC (Lenexa, which half-maximum adult protein abundance is obtained; F is the fractional KS). An additional 57 adult and 8 pediatric samples were available through the protein abundance in adult samples; Fbirth is the fractional protein abundance (of University of Washington School of Pharmacy liver bank. The samples were adult) at birth; and h is the exponential factor. classified based on the following age categories: neonatal (0–27 days; n =4),infancy The goodness of model fit was evaluated by visual inspection, 95% confidence 2 (28–364 days; n = 17), toddler/early childhood (1 year to , 6 years; n = 30), middle intervals (CIs) of the parameter estimates, and residual plots. Weights of 1/Y were childhood (6 years to , 12 years; n = 38), adolescence (12–18 years; n =48),and used. The Pearson regression test was used to test correlation between proteins adulthood (. 18 years; n = 57). The use of these samples has been classified as across entire population. A P value below 0.05 was considered statistically nonhuman subject research by the institutional review boards of the University of significant. The observed data were illustrated in graphs and tables using Microsoft Excel (Version 16, Redmond, WA) and GraphPad. Washington (Seattle, WA) and Children’s Mercy-Kansas City. Procurement and Downloaded from storage information were described previously (Pearce et al., 2016; Prasad et al., 2016; Shirasaka et al., 2016; Boberg et al., 2017). Detailed donor demographic Results and Discussion information is provided in Supplemental Table 2S. The HLC fraction was isolated from liver tissue samples by differential centrifugation as per established protocols The dynamic detection range was established for internal standard (Pearce